01.06.2015 Views

Full PDF - The Bone & Joint Journal

Full PDF - The Bone & Joint Journal

Full PDF - The Bone & Joint Journal

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Canadian Orthopaedic Association (COA)<br />

Canadian Orthopaedic Research Society (CORS)<br />

Canadian Orthopaedic Residents’ Association (CORA)<br />

Whistler, British Columbia, Canada: 3–6 July 2009<br />

President: Dr. Peter O’Brien<br />

Correspondence should be addressed to CEO Doug C. Thomson. Email: doug@canorth.org<br />

CORS PAPER SESSION 1:<br />

CARTILAGE AND MUSCLE<br />

1. INFLAMMATION CAUSES MUSCLE<br />

INJURY IN COMPARTMENT SYNDROME:<br />

AN EXPERIMENTAL STUDY<br />

Abdel-Rahman Lawendy 1 , David W. Sanders 2 ,<br />

Aurelia Bihari 3 , Amit Badhwar 3<br />

1<br />

University of Western Ontario; 2 Victoria<br />

Hospital, London Health Sciences Centre;<br />

3<br />

Lawson Health Research Institute<br />

Purpose: Compartment syndrome is a limb-threatening<br />

complication of skeletal trauma. Both ischemia and<br />

inflammation may be responsible for tissue necrosis<br />

in compartment syndrome (CS). In this study, normal<br />

rodents were compared with neutropenic animals to<br />

determine the importance of inflammation as a mechanism<br />

of cellular damage using techniques of intravital<br />

videomicroscopy (IVVM) and histochemical staining.<br />

Method: Forty Wistar rats were randomised. Twenty<br />

animals served as a control (group C). Twenty rats were<br />

rendered neutropenic using cyclophosphamide (250mg/<br />

kg) (group N). Animals were anaesthetised with 5 %<br />

isoflurane. Elevated intracompartmental pressure was<br />

induced by saline infusion into the anterior hindlimb<br />

compartment and maintained at 30–40 mmHg for 0,<br />

15, 45 or 90 minute time intervals. Following fasciotomy,<br />

the EDL muscle was analyzed using IVVM to<br />

quantify tissue injury, capillary perfusion, and inflammatory<br />

response.<br />

Results: <strong>The</strong> proportion of injured cells decreased in<br />

group N compared to group C at all time intervals<br />

of EICP (p0.05).<br />

Conclusion: This study demonstrates the importance<br />

of inflammation as a cause of injury in compartment<br />

syndrome. <strong>The</strong>re was a 50% decrease in injury in neutropenic<br />

animals compared to controls after 90 minutes<br />

of elevated intracompartmental pressure. Microvascular<br />

perfusion analysis demonstrated a time-dependent<br />

decrease in capillary perfusion in both neutropenic<br />

and control animals. Blocking of the inflammatory<br />

response via neutropenia was protective against tissue<br />

injury. <strong>The</strong>se results provide evidence toward a potential<br />

therapeutic benefit for anti-inflammatory treatment of<br />

elevated intra-compartmental pressure.<br />

2. CHONDROITINASE ABC AND ACUTE<br />

ELECTRICAL STIMULATION ARE<br />

BENEFICIAL FOR MUSCLE REINNERVATION<br />

AFTER A SCIATIC NERVE TRANSECTION IN<br />

RAT<br />

Frédérick-Charles Cloutier, Dominique Rouleau,<br />

Eric Beaumont, Michael Atlan * , Pierre H.<br />

Beaumont<br />

Hôpital Sacré-Coeur de Montréal; *France<br />

Purpose: Nerve re-generation and functional recovery<br />

are often incomplete after a peripheral nerve lesion. <strong>The</strong><br />

aim of this study was to determine if the injection of<br />

chondrotinase ABC at the lesion site, one hour of electrical<br />

stimulation, and the combination of these treatments<br />

at the time of repair are effective in promoting<br />

nerve regeneration and muscle re-innervation.<br />

Method: A complete right sciatic nerve section was done<br />

on 32 female Sprague-Dawley rats. End-to-end microsuture<br />

repair was performed and fibrin glue was added.<br />

Five groups were studied: 1- Sutures and Fibrine glue<br />

(S+F), 2- S+F and chondrotinase ABC, 3- S+F and electrical<br />

stimulation, 4- S+F and chondrotinase and electrical<br />

stimulation, 5 uninjured nerve. Video kynematic,<br />

EMG, muscle strengh and axonal count were used to<br />

asses nerve recovery at 150 days post-repair.<br />

Results: Side video kinematics was performed and a<br />

larger excursion of the hip-ankle-toe angle during walking<br />

was showed in groups 2, 3, and 4. (p


244 COA/CORS/CORA<br />

cells within the MSC population. It is also present on<br />

chondrocytes and is thought to play a critical role in cartilage<br />

matrix generation and homeostasis. We hypothesized<br />

that a CD44+ purified subpopulation of MSCs will<br />

possess enhanced chondrogenic potential and be more<br />

suitable for articular cartilage regeneration.<br />

Method: <strong>Bone</strong> marrow aspirates were collected from<br />

orthopaedic patients undergoing iliac crest bone grafting.<br />

Human MSCs were isolated and cultured using<br />

standard techniques. Flow cytometry was utilized to<br />

identify the cell surface antigens characteristic of the<br />

MSC population. FACS was utilized to isolate the CD44<br />

positive cells based on antigenic recognition, generating<br />

a CD44 positive population and a CD44 negative population.<br />

To confirm the multilineage potential of the isolates,<br />

defined media and culture conditions were utilized<br />

to differentiate both groups into osteocytes, adipocytes<br />

and chondrocytes. Real time polymerase chain reaction<br />

was utilized to quantify and compare the essential markers,<br />

collagen II, collagen I and aggrecan, in the stem cell<br />

derived chondrocytes. <strong>The</strong> CD44 enriched and CD44<br />

depleted populations were compared.<br />

Results: <strong>The</strong> cells isolated possessed a cell morphology<br />

and surface antigen profile consistent with a MSC<br />

population. In addition, both experimental groups demonstrated<br />

multipotent ability. Real time PCR analysis<br />

of the chondrogenic cells demonstrated that the CD44<br />

positive population expressed collagen II and aggrecan<br />

at a significantly higher level than the CD44 negative<br />

population.<br />

Conclusion: To date no group has successfully identified<br />

a relationship between a MSC subpopulation and the<br />

multipotent progenitors responsible for generating cartilage.<br />

This work demonstrated that there are MSC subpopulations<br />

with different potential for chondrogenic<br />

expression and represents an important step towards<br />

identifying MSC subpopulations with enhanced cartilage<br />

formation potential.<br />

5. MECHANISM OF DECREASED<br />

EXPRESSION OF TYPE X COLLAGEN<br />

IN HUMAN MESENCHYMAL STEM<br />

CELLS CULTURED ON NITROGEN-RICH<br />

PLASMA POLYMERS: IMPLICATION OF<br />

CYCLOOXYGENASE-1<br />

Fackson Mwale 1 , Hong Tian Wang 2 , Pierre-Luc<br />

Girard-Lauriault 3 , Michael R. Wertheimer 3 , John<br />

Antoniou 1 , Alain Petit 2<br />

1<br />

McGill University; 2 Lady Davis Institute; 3 École<br />

Polytechnique<br />

Purpose: Recent evidence indicates that a major drawback<br />

of current cartilage and intervertebral disc (IVD)<br />

tissue engineering is that human mesenchymal stem cells<br />

(MSCs) from osteoarthritic patients rapidly express<br />

type X collagen (COL10A1), a marker of late-stage<br />

chondrocyte hypertrophy associated with endochondral<br />

ossification. We recently discovered that a novel<br />

atmospheric-pressure plasma-polymerized thin film substrate,<br />

named “nitrogen-rich plasma-polymerized ethylene”<br />

(PPE:N), is able to inhibit COL10A1 expression<br />

in committed MSCs. However, the cellular mechanisms<br />

implicated in the inhibition of COL10A1 expression by<br />

PPE:N surfaces are unknown.<br />

Method: Human mesenchymal stem cells (MSCs) were<br />

obtained from aspirates from the intramedullary canal<br />

of donors (60-80 years of age) undergoing total hip<br />

replacement for osteoarthritis. <strong>Bone</strong> marrow aspirates<br />

were processed and MSCs were cultured on commercial<br />

polystyrene (PS control) and on PPE:N surfaces in the<br />

presence of different kinases and cyclooxygenase inhibitors<br />

for 3 days. Total RNA was extracted with TRIzol<br />

reagent (Invitrogen, Burlington, ON) and the expression<br />

of COL10A1, cyclooxygenase-1 (COX-1), and 5-<br />

lipoxygenase (5-LOX) genes was measured by real-time<br />

quantitative RT-PCR.<br />

Results: Results showed that a non-specific inhibitor of<br />

cyclooxygenases reduced the expression of COL10A1.<br />

In contrast, inhibitors of protein kinases stimulated<br />

the expression of COL10A1. Furthermore, potent and<br />

selective inhibitors of COX-1 and 5-LOX also reduced<br />

the expression of COL10A1. However, COX-2 and 12-<br />

LOX inhibitors had no significant effect on the expression<br />

of COL10A1. COX-1 gene expression was also<br />

decreased when MSCs were incubated on “S5” PPE:N<br />

surfaces. Interestingly, MSCs did not express 5-LOX.<br />

Conclusion: PPE:N surfaces suppress COL10A1<br />

expression through the inhibition of COX-1 which is<br />

directly implicated in the synthesis of prostaglandins.<br />

<strong>The</strong> decreased expression of COX-1 and COL10A1 in<br />

human MSCs cultured on PPE:N is therefore in agreement<br />

with the induction of the osteogenic capacity of<br />

rat bone marrow and bone formation by systemic or<br />

local injection of PGE2 in rats. However, PGE2 and<br />

other prostaglandins inhibited COL10A1 expression in<br />

chick growth plate chondrocytes. This suggests that the<br />

effect of prostaglandins on COL10A1 expression may<br />

be cell-specific or may be dependent on pre-existing<br />

patho-physiological conditions.<br />

6. HYPOXIC REGULATION OF<br />

CHONDROCYTE DIFFERENTIATION AND<br />

ITS APPLICATION TO CARTILAGE REPAIR<br />

Richard C. Smith, Brenton Short, Paul W.<br />

Clarkson, Bassam A. Masri, Michael Underhill<br />

UBC<br />

Purpose: Chondral injuries of the knee are commonly<br />

seen at arthroscopy, yet there is no consensus on the most<br />

appropriate treatment method. However, untreated cartilage<br />

injury predisposes to osteoarthritis contributing to<br />

pain and disability. For cell-based cartilage repair strategies,<br />

an ex-vivo expansion phase is required to obtain<br />

sufficient numbers of cells needed for therapy. Although<br />

recent reports demonstrated the central role of oxygen<br />

for the function and differentiation of chondrocytes,<br />

little is known of the effect of physiological low oxygen<br />

concentrations during the expansion of the cells and<br />

whether this alters their chondrogenic capacity.<br />

Method: Initial studies of chondrocyte expansion were<br />

performed in mature mice, with cells expanded at<br />

either atmospheric oxygen tension (21%) or 5% 02 in<br />

monolayer cultures. Chondrogenic differentiation was<br />

subsequently assessed via micromass culture. Having<br />

determined that oxygen tension influences murine chondrocyte<br />

expansion and differentiation, similar studies<br />

were conducted using adult human chondrocytes taken<br />

from knee arthroplasty off-cuts, with mRNA expression<br />

of select genes involved in the chondrogenic program<br />

analyzed by q-PCR.<br />

Results: Cellular morphology was improved in hypoxic<br />

culture, with a markedly more fibroblastic appearance<br />

seen after greater than 2 passages in 21% O2. Micromass<br />

cultures maintained in hypoxic conditions demonstrated<br />

stronger staining with Alcian blue, indicating stronger<br />

expression of cartilaginous glycosaminoglycans. Collagen<br />

type II mRNA expression was two-fold higher in cells<br />

expanded at 5% as compared to expansion at 21% O2.<br />

Micromass cultures grown at 21% O2 showed up to a<br />

twofold increase in the tissue content of glycosaminoglycans<br />

when formed with cells expanded at 5% instead of<br />

21% O2. However, no differences in the mRNA expression<br />

or staining for collagen type II protein were observed<br />

in these micromass cultures. Hypoxia (5% O2) applied<br />

during micromass cultures gave rise to tissues with low<br />

contents of glycosaminoglycans.<br />

Conclusion: In-vivo, chondrocytes are adapted to a<br />

hypoxic environment. Taking this into account, applying<br />

5% O2 in the expansion phase in the course of cellbased<br />

cartilage repair strategies, may result in a repair<br />

tissue with higher quality by increasing the content of<br />

glycosaminoglycans.<br />

7. THE MOLECULAR MECHANISMS OF<br />

COMPARTMENT SYNDROME<br />

Greg W. McGarr, David W. Sanders, Amit<br />

Badhwar<br />

Lawson Health Research Institute<br />

Purpose: Compartment syndrome is a severe complication<br />

of skeletal trauma. Intravital microscopy (IVVM)<br />

has demonstrated an inflammatory response to compartment<br />

syndrome (CS). <strong>The</strong> molecular mechanisms<br />

underlying this inflammatory response are unknown.<br />

<strong>The</strong> purpose of this study was threefold. First, a broad<br />

inflammatory cytokine profile was examined to determine<br />

the molecules responsible for white cell recruitment.<br />

As well, skeletal muscle expression of white cell<br />

adhesion molecules including P-Selectin, E-Selectin,<br />

Mac-1 and ICAM-1 were examined to assess the extent<br />

of white cell activation in target tissues. Finally, skeletal<br />

muscle apoptosis was measured to determine the magnitude<br />

of cell death.<br />

Method: Normal and neutropenic rats were randomised<br />

to either compartment syndrome or control groups. CS<br />

Animals were treated with 45 minutes of elevated intracompartmental<br />

pressure (EICP) of the hindlimb. Fasciotomy<br />

was then performed, followed by 60 minutes<br />

of reperfusion. Control animals experienced no EICP.<br />

Blood was collected from carotid arterial lines used for<br />

pressure monitoring. Skeletal muscle tissue samples were<br />

collected from the EDL following reperfusion. Blood<br />

samples were obtained from carotid arterial lines and<br />

skeletal muscle was collected following reperfusion. A<br />

Multiplex assay was used to examine serum levels of 24<br />

proinflammatory cytokines/chemokines. Skeletal muscle<br />

mRNA levels of P-Selectin, E-Selectin, Mac-1 and<br />

ICAM-1 were evaluated using real-time PCR. Finally,<br />

skeletal muscle apoptosis was measured by DNA laddering<br />

and a caspase-3 assay.<br />

Results: Neutropenic CS animals demonstrated a continuous<br />

increase in TNF-alpha levels, peaking at 700+/-<br />

350pg/ml by 60 minutes of reperfusion. TNF-alpha<br />

values for other groups did not increase. A 104-fold<br />

increase in ICAM-1 mRNA levels was observed in neutropenic<br />

CS rats while other groups showed no significant<br />

increase. <strong>The</strong>re was no significant increase in any<br />

group for P-Selectin, E-Selectin, or Mac-1.<br />

Conclusion: This study is the first to attempt to describe<br />

the molecular inflammatory response in CS. Neutropenic<br />

CS animals demonstrated an upregulation in TNFalpha<br />

and ICAM-1 mRNA levels. This likely represents<br />

an attempt to generate an inflammatory response in<br />

the neutropenic animals. Additional data at incremental<br />

timepoints is necessary to further characterize the<br />

molecular mechanisms. However, both TNF-alpha and<br />

ICAM-1 appear to be important in the mechanism of<br />

inflammatory activation in compartment syndrome.<br />

8. INSIGHTS INTO MESENCHYMAL STEM<br />

CELL DIFFERENTIATION TO ADULT<br />

CHONDROCYTES -HUMAN FACILITATIVE<br />

GLUCOSE TRANSPORTERS (SLC2A<br />

PROTEINS)? A NEW GENERATION OF<br />

PHYSIOLOGICAL MARKERS<br />

Andrei R. Manolescu, David Cinats, Charles<br />

Secretan, Deborah O’Neill, Chris Cheeseman,<br />

Keith Bagnall, Nadr M. Jomha<br />

University of Alberta<br />

Purpose: Differentiation of BM-MSCs into adult chondrocytes<br />

represents a complex physiological mechanism<br />

and full characterization of each individual stage<br />

through which the BM-MSC differentiate into adult<br />

chondrocytes is not yet understood.<strong>The</strong> physiological<br />

micro-environment of the chondrocytes is intensely<br />

hypoxic which triggers over-expression SLC2A proteins<br />

(GLUTs) in their membranes as a compensatory mechanism<br />

for energy production within the glycolytic cycle.<br />

Method: We cultured and differentiated BM-MSC, and<br />

adult chondrocytes in hypoxic (5% O2 tension) and<br />

normoxic (20% O2) conditions. Within this cell populations<br />

we screened for the presence of the 12 GLUT<br />

genes as well as quantification of the variation of the<br />

12 GLUTs gene translation by simple pcr and rt-pcr.<br />

<strong>The</strong> expression profile of the GLUT proteins was investigated<br />

using western blot analysis and immunohistochemistry.<br />

Functional characterization of the GLUTs<br />

expressed in the different cell populations was carried<br />

out by the means of radio-isotope labeled hexose fluxes<br />

done accordingly to the substrate specificity and kinetic<br />

properties particular to each SLC2A isoforms.<br />

Results: Our data showed that the functional genotype<br />

and phenotype of the adult chondrocyte and hypoxic<br />

BM-MSC comprised an extensive expression of fructose-transporting<br />

GLUTs as opposed to the glucoseonly<br />

transporting isoforms expression in normoxic<br />

J BONE JOINT SURG [BR] 2011; 93-B:SUPP III


COA/CORS/CORA 245<br />

BM-MSC. <strong>The</strong> flux data showed clear similarities in<br />

functional GLUT profiles between BM-MSC cultured in<br />

hypoxic conditions, adult chondrocytes. Investigation<br />

of the uptake of a panel of five individual sugars (glucose,<br />

fructose, 2-deoxy-gluose, 3-orthomethyl-glucose<br />

and galactose) in these cellular populations under both<br />

hypoxic and normoxic conditions and in the presence<br />

and absence of Cytochalasin B (a GLUT1-specific inhibitor)<br />

showed that SLC2A class II transporters (GLUTs 5,<br />

7, 9 and 11) play a more important role in the uptake of<br />

sugars by the normal hypoxic chondrocytes when compared<br />

to the ubiquitously-expressed GLUT1.<br />

Conclusion: Use of this approach allows the correct<br />

culturing conditions to be identified that would select<br />

for those chondrocyte precursors from the total BM-<br />

MSC population that would have the best potential for<br />

producing viable articular cartilage. In addition, specific<br />

substrates for GLUTs isoforms could be used for physiologic,<br />

non-invasive and real time imaging of cartilage,<br />

BM-MSC and cartilage autograft by means of Positron<br />

Emission Tomography.<br />

9. THE EFFECT OF REMAINING AT<br />

CONFLUENCE ON THE CHONDROCYTIC<br />

PHENOTYPE<br />

Krishna Maragh, Jenn Bater, Charles Secretan,<br />

Keith M. Bagnall, Nadr M. Jomha<br />

University of Alberta<br />

Purpose: Current techniques for articular cartilage<br />

repair remain sub-optimal. <strong>The</strong> best technique involves<br />

the introduction of cultured chondrocytes into the<br />

injury site. Experimental results of current chondrocyte<br />

culture and expansion techniques (passaging) have<br />

shown phenotypic alteration resulting in fibroblastlike<br />

cells. <strong>The</strong>refore, treatment methods that propose<br />

the transplantation of cultured chondrocytes might be<br />

transplanting fibroblast-like cells instead of chondrocytes.<br />

This experiment explored the difference in genetic<br />

expression of chondrocytes left at confluence compared<br />

to chondrocytes that were passaged as performed in current<br />

culture techniques. It was hypothesized that chondrocytes<br />

left at confluence would maintain their collagen<br />

I and collagen II gene expression over time.<br />

Method: Fresh normal human articular cartilage was<br />

collected from deceased donor patients. <strong>The</strong> matrix was<br />

digested and the chondrocytes were plated in monolayer<br />

to create two groups. <strong>The</strong> first group was cultured and<br />

passaged 2? at confluence seven times. <strong>The</strong> second group<br />

was cultured at confluence and left for seven weeks,<br />

with medium changes every 3-4 days without passaging.<br />

At weekly intervals RNA was extracted from cells in<br />

both groups and analyzed with real time PCR, probing<br />

specifically for the genes responsible for the production<br />

of collagen I, collagen II, aggrecan, and GAPDH. This<br />

was done in duplicate.<br />

Results: Collagen II gene expression was maintained<br />

over seven weeks in cells left at confluence but was<br />

decreased in passaged cells. Collagen I gene expression<br />

decreased over seven weeks in cells left at confluence,<br />

but remained the same in passaged cells. Aggrecan gene<br />

expression remained the same in both groups.<br />

Conclusion: Current culture and expansion techniques<br />

that employ passaging (as used in clinical scenarios)<br />

result in significant alterations in gene expression that<br />

are inconsistent with the current definition of a “chondrocyte”.<br />

Culturing chondrocytes at confluence can<br />

produce gene expression more similar to native chondrocytes<br />

but even these cells have expression of collagen<br />

type I that should not be present in chondrocytes. <strong>The</strong><br />

results of this study suggest that further investigation is<br />

required to develop chondrocyte culture and expansion<br />

techniques that minimize the de-differentiation of chondrocytes<br />

by maintaining collagen II gene expression and<br />

eliminating/preventing collagen I gene expression.<br />

CORS PAPER SESSION 2: BONE<br />

10. AN IN-VIVO EVALUATION OF THE<br />

EFFECT OF A HYDROXYAPATITE COATING<br />

WITH AND WITHOUT THE USE OF BMP-7<br />

ON EXTRACORTICAL BONE BRIDGING<br />

USING A CANINE SEGMENTAL DEFECT<br />

MODEL<br />

Neil Saran, Robert É. Turcotte, Renwen Zhang*<br />

McGill University Health; *Stryker Orthopaedics<br />

Purpose: Extracortical bone bridging and ingrowth<br />

have been shown to reduce stresses on the stem and<br />

cement mantle of tumor endoprostheses. <strong>The</strong> purpose<br />

of this study was to assess the effect of bone morphogenetic<br />

protein 7 (BMP-7) delivered by Peri-Apatiteâ<br />

(PA, Stryker Orthopaedics) hydroxyapatite coating on<br />

porous segmental replacement prostheses.<br />

Method: Eighteen mature mongrel canines were<br />

implanted with unilateral segmental replacement prostheses<br />

made of a cobalt-chromium (Co-Cr) alloy and coated<br />

with two layers of sintered Co-Cr alloy beads (diameter<br />

600 to 800mm). <strong>The</strong> control group consisted of a plain<br />

porous coated segmental prosthesis without any PA coating.<br />

Group 2 consisted of a PA-coated segmental prosthesis<br />

coated with buffer solution. Group three consisted of<br />

a PA-coated segmental prosthesis loaded with rhBMP-7<br />

(Stryker Biotech) in a buffer solution carrier. Group 1<br />

had the implant only. Group 2 had the buffer solution<br />

evenly applied to the porous coat and group 3 had 2.9<br />

mg of BMP-7 in liquid buffer solution evenly applied.<br />

<strong>The</strong> canines were allowed to fully bear weight without<br />

restrictions. <strong>The</strong> femurs were retrieved at twelve weeks<br />

for radiographic and histologic analysis.<br />

Results: Gross and radiographic data of the retrieved<br />

specimens showed that all six PA-coated implants augmented<br />

with BMP-7 had complete bone bridging; only<br />

one of the PA-coated implants and only two of the plain<br />

porous implants were completely bridged. <strong>The</strong>re was<br />

a greater percentage of bone apposition for the BMP-<br />

7 augmented PA-coated group compared to both the<br />

plain (p=0.0026) and the PA-coated (p=0.0001). <strong>The</strong>re<br />

was no difference in bone formation or bone apposition<br />

between the plain and PA-coated groups. Histology<br />

revealed greater depth of bone ingrowth in the BMP-7<br />

augmented PA-coated group as compared to the plain<br />

(p


246 COA/CORS/CORA<br />

13. EFFECT OF THE HVEGF TRANSFER ON<br />

ENDOGENOUS VEGF MRNA EXPRESSION<br />

IN A RAT OSTEOBLAST OR FIBROBLAST<br />

CULTURE MODEL<br />

Claire Li, Ru Li, Michael D. McKee, Emil H.<br />

Schemitsch<br />

University of Toronto<br />

Purpose: Vascular Endothelial Growth Factor (VEGF)<br />

plays an important role in promoting angiogenesis and<br />

osteogenesis during fracture repair. Our previous studies<br />

have shown that cell-based VEGF gene therapy accelerates<br />

bone healing of a rabbit tibia segmental bone defect<br />

in-vivo, and increases osteoblast proliferation and<br />

mineralization in-vitro. <strong>The</strong> aim of this project was to<br />

examine the effect of exogenous human VEGF (hVEGF)<br />

on the endogenous rat VEGF messenger RNA (mRNA)<br />

expression in a cell-based gene transfer model.<br />

Method: <strong>The</strong> osteoblasts were obtained from the rat<br />

periosteum. <strong>The</strong> fibroblasts were obtained from the rat<br />

dermal tissue. <strong>The</strong> cells were then cultured to reach 60%<br />

confluence and transfected with hVEGF using Superfect.<br />

Four groups were: 1) osteoblast-hVEGF, 2) fibroblasthVEGF,<br />

3) Osteoblasts alone, and 4) Fibroblasts only.<br />

<strong>The</strong> cultured cells were harvested at 1, 3 and 7 days<br />

after the transfection. <strong>The</strong> total mRNA was extracted<br />

(TRIZOL); both hVEGF and rat VEGF mRNA were<br />

measured by reverse transcriptase- polymerase chain<br />

reaction (RT-PCR) and quantified by VisionWorksLS.<br />

Results: <strong>The</strong> hVEGF mRNA was detected by RT-PCR<br />

from transfected osteoblasts after three days of gene<br />

transfection. <strong>The</strong> hVEGF mRNA expression in transfected<br />

fibroblasts increased exponentially at days 1, 3 and<br />

7 after the transfection. We compared the endogenous<br />

rat VEGF mRNA expression level of the osteoblasts or<br />

fibroblasts that were transfected with hVEGF with the<br />

cells without the transfection. <strong>The</strong> hVEGF transfected<br />

osteoblasts had a greater rat VEGF mRNA expression<br />

than the non-transfected osteoblasts. Furthermore,<br />

when hVEGF was transfected to the rat fibroblasts, the<br />

endogenous mRNA expression level measured was also<br />

greater than that from the non-transfected fibroblasts.<br />

Rat VEGF mRNA expression increased in the first three<br />

days of the hVEGF transfection, but the expression level<br />

was reduced at Day 7.<br />

Conclusion: <strong>The</strong>se results suggest that cell-based hVEGF<br />

gene therapy enhances endogenous rat VEGF mRNA<br />

expression in both osteoblasts and fibroblasts.<br />

14. VASCULAR ENDOTHELIAL GROWTH<br />

FACTOR REGULATES OSTEOBLAST CELL<br />

DEATH IN OSTEOPOROTIC VERTEBRAL<br />

FRACTURE<br />

John Street 1,2 , Brian Lenehan 1,2 , Charles G.<br />

Fisher 2 , Marcel Dvorak 2<br />

1<br />

University College Cork, Ireland; 2 UBC<br />

Purpose: Apoptosis of osteoblasts and osteoclasts regulates<br />

bone homeostasis. Vertebral osteoporotic insufficiency<br />

fractures are characterised by pathological rates<br />

of osteoblast apoptosis. Skeletal injury in humans results<br />

in ‘angiogenic’ responses primarily mediated by vascular<br />

endothelial growth factor(VEGF), a protein essential for<br />

bone repair in animal models. Osteoblasts release VEGF<br />

in response to a number of stimuli and express receptors<br />

for VEGF in a differentiation dependent manner. This<br />

study investigates the putative role of VEGF in regulating<br />

the lifespan of primary human vertebral osteoblasts<br />

(PHVO) in-vitro.<br />

Method: PHVO were cultured from biopsies taken at<br />

time of therapeutic vertebroplasty and were examined<br />

for VEGF receptors. Cultures were supplemented with<br />

VEGF(0–50ng/mL), a neutralising antibody to VEGF,<br />

mAB VEGF(0.3ug/mL) and Placental Growth Factor<br />

(PlGF), an Flt-1 receptor-specific VEGF ligand(0–100<br />

ng/mL) to examine their effects on mineralised nodule<br />

assay, alkaline phosphatase assay and apoptosis. <strong>The</strong><br />

role of the VEGF specific antiapoptotic gene target BCl2<br />

in apoptosis was determined.<br />

Results: PHVO expressed functional VEGF receptors.<br />

VEGF 10 and 25 ng/mL increased nodule formation<br />

2.3- and 3.16-fold and alkaline phosphatase release 2.6<br />

and 4.1-fold respectively while 0.3ug/mL of mAB VEGF<br />

resulted in approx 40% reductions in both. PlGF 50ng/<br />

mL had greater effects on alkaline phosphatase release<br />

(103% increase) than on nodule formation (57%<br />

increase). 10ng/mL of VEGF inhibited spontaneous and<br />

pathological apoptosis by 83.6% and 71% respectively,<br />

while PlGF had no significant effect. Pretreatment with<br />

mAB VEGF, in the absence of exogenous VEGF resulted<br />

in a significant increase in apoptosis (14 versus 3%).<br />

BCl2 transfection gave a 0.9% apoptotic rate. VEGF 10<br />

ng/mL increased BCl2 expression four fold while mAB<br />

VEGF decreased it by over 50%.<br />

Conclusion: VEGF is a potent regulator of osteoblast<br />

life-span in-vitro. This autocrine feedback regulates survival<br />

of these cells, mediated via the KDR receptor and<br />

expression of BCl2 antiapoptotic gene. This mechanism<br />

may represent a novel therapeutic model for the treatment<br />

of osteoporosis.<br />

15. CHARACTERIZATION OF RAT AND<br />

MOUSE FORELIMB COMPRESSION MODELS<br />

FOR STUDIES OF WOVEN BONE REPAIR IN<br />

RESPONSE TO FATIGUE DAMAGE<br />

Thomas Karakolis, Gregory R. Wohl<br />

McMaster University<br />

Purpose: <strong>Bone</strong> fatigue damage can lead to stress fractures<br />

and may play a role in fragility fractures. <strong>The</strong> rat<br />

forelimb compression model has been used to examine<br />

biological responses and gene expression associated with<br />

woven bone repair after fatigue damage. Development a<br />

similar mouse model would enable the use of genetically<br />

modified mice to study molecular mechanisms associated<br />

with bone repair.<br />

Method: Following approval from our Central Animal<br />

Facility, forelimbs of male retired breeder C57BL/6 mice<br />

and Sprague Dawley rats (n=31 each) were loaded in<br />

axial compression across the carpus and olecranon. First,<br />

both forelimbs (postmortem, n=6 each) were monotonically<br />

loaded to determine failure load. Next, both forelimbs<br />

of animals (postmortem, n=5 each) were loaded<br />

cyclically to sub-fracture load (67% of monotonic load<br />

for mice, 55% for rats) until fatigue failure. Following<br />

analysis of fatigue displacement histories, right forelimbs<br />

(post-mortem, n=10 each) were loaded cyclically to a set<br />

displacement short of the expected failure displacement<br />

(mice–30%; rats–55%). Non-loaded left forelimbs served<br />

as controls. Three-point bending tests were performed on<br />

the ulnae; mechanical properties were compared between<br />

fatigued and non-loaded limbs. Finally, right forelimbs<br />

(n=10 each) were cyclically loaded in anaesthetised (2.5%<br />

isofluorane) animals to 30% (mice) and 55% (rats) of<br />

failure displacement. Animals recovered for seven days;<br />

microCT imaging and three-point bend tests were performed<br />

on the ulnae.<br />

Results: Ultimate forelimb failure loads were 5.63 ±<br />

0.47 N (mouse) and 57.1 ± 5.8 N (rat). Measured from<br />

the 10th cycle, fatigue failure occurred at displacements<br />

of 1.68 ± 0.21 mm (mouse) and 2.96 ± 0.22 mm (rat). In<br />

three-point bending, fatigue damaged ulnae failed at significantly<br />

lower loads versus control (mouse -51.6%; rat<br />

-32.1%). After seven days healing, bone cross-sectional<br />

area was significantly greater (microCT) and mechanical<br />

properties partially recovered (-13.8% versus control).<br />

Conclusion: Rat and mouse forelimb fatigue loading<br />

models have been developed to induce repeatable<br />

bone damage. Observed differences in fatigue behaviour<br />

necessitated different loading parameters between<br />

models. Following seven days of healing, recovery of<br />

mechanical strength accompanied woven bone formation<br />

(demonstrated by microCT). Further work will<br />

compare the biological, woven bone, response between<br />

the mouse and rat forelimb models.<br />

16. THE EFFECT OF INTRAMEDULLARY<br />

REAMING ON A DIAPHYSEAL BONE<br />

DEFECT OF THE TIBIA<br />

Paul R. T. Kuzyk, John E.D. Davies, Emil H.<br />

Schemitsch<br />

University of Toronto<br />

Purpose: <strong>The</strong> purpose of this study was to relate the<br />

extent of reaming to bone formation occurring around<br />

a critical sized defect in the tibia.<br />

Method: Eleven canines were allocated into 2 groups:<br />

empty (N=5) or iliac crest autograft (N=6). All tibiae<br />

were reamed to 7.0 mm and fixed with a 6.5 mm statically<br />

locked intramedullary nail after creation of an 8.0<br />

mm diaphyseal defect. <strong>The</strong> extent of reaming of the<br />

canal was dependent on the cross-sectional area of the<br />

tibia as all tibiae were reamed to 7.0 mm. Fluorescent<br />

markers were given at different times: calcein green (6<br />

weeks), xylenol orange (9 weeks), and tetracycline (11<br />

and 14 weeks). Animals were sacrificed at 15 weeks and<br />

perfused with a barium compound. Radiography, Micro<br />

CT, brightfield microscopy and fluorescent microscopy<br />

were used for analysis.<br />

Results: <strong>Bone</strong> and vasculature volume within the defect<br />

were reported as a percentage of the total volume of<br />

the defect. Linear regression analysis of percent bone<br />

volume (dependent variable) and canal area (independent<br />

variable) provided a Pearson correlation coefficient<br />

of 0.925 (p=0.025) for the empty group and 0.244<br />

(p=0.641) for the autograft group. Linear regression<br />

analysis of percent vasculature volume (dependent variable)<br />

and canal area (independent variable) provided a<br />

Pearson correlation coefficient of 0.784 (p=0.117) for<br />

the empty group and -0.146 (p=0.783) for the autograft<br />

group. <strong>Bone</strong> formation at osteotomy sites was defined<br />

as the distance from the original osteotomy site to the<br />

tip of newly formed bone. Linear regression analysis of<br />

bone formation at the osteotomy sites (dependent variable)<br />

and canal area (independent variable) provided a<br />

Pearson correlation coefficient of 0.132 (p=0.832) for<br />

the empty group and -0.937 (p=0.006) for the autograft<br />

group. <strong>Bone</strong> formation rates were reported as the distance<br />

between the fluorescent labels. <strong>Bone</strong> formation<br />

rate was less within the endosteum, cortex and periosteum<br />

with extensive reaming in empty samples.<br />

Conclusion: Our results suggest that the acute management<br />

of tibia fractures with bone defects should involve<br />

limited reaming. This does not apply when the defect is<br />

autografted. Limited reaming may be defined by the crosssectional<br />

area of the tibia in ratio to that of the reamer.<br />

17. USE OF CO-REGISTERED HIGH-<br />

RESOLUTION COMPUTED TOMOGRAPHY<br />

SCANS BEFORE AND AFTER SCREW<br />

INSERTION AS A NOVEL TECHNIQUE FOR<br />

BONE MINERAL DENSITY DETERMINATION<br />

ALONG SCREW TRAJECTORY<br />

Marlis Sabo, Steven I. Pollmann, Kevin R. Gurr,<br />

Christopher Bailey, David W. Holdsworth<br />

University of Western Ontario<br />

Purpose: <strong>Bone</strong> mineral density (BMD) is an important<br />

factor in the performance of orthopaedic instrumentation<br />

both in and ex-vivo, and until now, there has not existed<br />

a reliable technique for determining BMD at the precise<br />

location of such hardware. This paper describes such a<br />

technique using cadaveric human sacra as a model.<br />

Method: Nine fresh-frozen sacra had solid and hollow<br />

titanium screw placed into the S1 pedicles from a posterior<br />

approach. High-resolution micro-computed tomography<br />

(CT) was performed on each specimen before and<br />

after screw placement. All images were reconstructed<br />

with an isotropic spatial resolution of 0.308 mm, reoriented,<br />

and the pre-screw and post-screw scans were registered<br />

and transformed using a six-degree rigid-body<br />

transformation matrix. Once registered, two points,<br />

corresponding to the center of the screw at the cortex<br />

and at the screw tip, were determined in each scan.<br />

<strong>The</strong>se points were used to generate cylindrical regions of<br />

interest (ROI) with the same trajectory and dimensions<br />

as the screw. BMD measurements were obtained within<br />

each of the ROI in the pre-screw scan. To examine the<br />

effect of artefact on BMD measurements around the<br />

J BONE JOINT SURG [BR] 2011; 93-B:SUPP III


COA/CORS/CORA 247<br />

titanium screws, annular ROI of 1 mm thickness were<br />

created expanding from the surface of the screws, and<br />

BMD was measured within each in both the pre- and<br />

post-screw scans.<br />

Results: <strong>The</strong> registration process was accurate, with<br />

an error of 0.2 mm. Four specimens were scanned five<br />

times with repositioning, and error in BMD measurements<br />

was ± 2%. BMD values in the cylindrical ROI<br />

corresponding to screw trajectories were not statistically<br />

different from side to side of each specimen (p = 0.23).<br />

Artefact-related differences in BMD values followed an<br />

exponential decay curve as distance from the screws<br />

increased, approaching a low value of approximately 20<br />

mg HA/cc, but not disappearing completely.<br />

Conclusion: CT in the presence of metal creates artefact,<br />

making measured BMD values near implants unreliable.<br />

This technique is accurate for determination of<br />

BMD, non-destructive, and eliminates the problem of<br />

this metal artefact through the use of co-registration of<br />

a pre- and post-screw scan. This technique has applications<br />

both in-vitro and in-vivo.<br />

18. TREATMENT OF FRACTURE NON-<br />

UNION USING RECOMBINANT BMP-7:<br />

SINGLE CENTRE EXPERIENCE<br />

Inder Gill, Vinod Kolimarala, Richard<br />

Montgomery<br />

James Cook, University Hospital, Middlesbrough,<br />

UK<br />

Purpose: To analyse the results of the use of Recombinant<br />

<strong>Bone</strong> Morphogenic Protein (BMP-7) for treatment<br />

of fracture nonunions at our institution.<br />

Method: From 2001 to 2006, 23 patients with fracture<br />

non-union were treated with BMP-7 for bone healing.<br />

<strong>The</strong>re were 14 male and nine females. <strong>The</strong> mean age of<br />

patients was 45 years (Range 21-76 yrs). <strong>The</strong>re were 11<br />

femoral, nine tibial and three humerus fractures. <strong>The</strong>re<br />

were four open injuries. <strong>The</strong> average number of operations<br />

before BMP-7 insertion was 2.5 (Range 0-6). <strong>The</strong><br />

mean time between the injury and BMP insertion was<br />

52 months (Range 5-312). Nine (40%) patients had<br />

previous autologous bone graft inserted without union.<br />

4 patients had BMP-7 insertion on its own. In another<br />

4 patients it was mixed with allograft. In the rest of 15<br />

patients BMP-7 was mixed with autologous bone graft. 2<br />

patients needed BMP-7 insertion on 2 separate occasions.<br />

In all except 1 patient the original fixation of the fracture<br />

had to be revised using various appropriate methods.<br />

Results: All the fracture went on to unite within an average<br />

of seven months (Range 4-16). <strong>The</strong>re were no complications<br />

from the use of BMP-7.<br />

Conclusion: Use of recombinant BMP-7, bone graft and<br />

stable fixation lead to fracture union in all our patients.<br />

We believe that the use of BMP-7 improved the chances<br />

of fracture healing in persistent non-unions and it is safe<br />

and easy to use.<br />

CORS PAPER SESSION 3:<br />

MECHANICS AND MATERIALS<br />

19. EXAMINATION OF RADIOGRAPHIC<br />

FEATURES AND LURCH; A MEASURE OF<br />

ASYMMETRIC GAIT, AMONG PATIENTS<br />

AWAITING TOTAL HIP ARTHROPLASTY<br />

Kyle A. R. Kemp, Michael J. Dunbar, Lori A.<br />

Livingston, Allan Hennigar<br />

Dalhousie University<br />

Purpose: Despite their inclusion within clinical practice,<br />

standardized radiographs may not accurately project an<br />

individual’s level of function and mobility. <strong>The</strong> purpose<br />

of this study is to examine the potential relationship<br />

between established radiographic features and lurch; a<br />

functional measure of asymmetric gait, in a group of<br />

patients who will receive total hip arthroplasty (THA).<br />

Method: Thirty-two patients (16 females, 16 males) identified<br />

as hip replacement candidates were recruited, with<br />

a mean age of 57.0 years. Lurch was obtained using the<br />

Walkabout Portable Gait Monitor (WPGM); a wireless,<br />

tri-axial accelerometry device. <strong>The</strong> independent variables<br />

were comprised of the Kellgren-Lawrence Scale, and a<br />

collection of standard radiographic features, as adopted<br />

by the American Academy of Orthopaedic Surgeons<br />

(AAOS), the National Institutes of Health (NIH), and the<br />

World Health Organization (WHO). Radiographs were<br />

blinded, and the surgeon completing the rating scale was<br />

unaware of patient’s lurch values. Age-adjusted regression<br />

analyses were used to examine the potential association<br />

between each radiographic feature and lurch.<br />

Results: Increased amounts of lurch (i.e. functional<br />

impairment) were independently associated with higher<br />

Kellgren-Lawrence Scale scores (p=.047), increased <strong>Joint</strong><br />

Space Narrowing in the mid-portion of the joint (zone<br />

2; p=.004), the presence of acetabular wear (p=.045),<br />

an increased severity of subchondral femoral head cysts<br />

(p=.004), and higher surgeon-rated Visual Analog Scale<br />

scores for overall severity of joint degeneration (p=.008).<br />

Lurch was not significantly associated with the remaining<br />

10 features which were examined. Further analyses<br />

revealed that lurch was not significantly associated with<br />

certain demographic factors, including sex, Body Mass<br />

Index, and co-morbid health conditions.<br />

Conclusion: Although the Kellgren-Lawrence scale was<br />

associated with an objective measure of gait, our results<br />

indicate that other radiographic features may provide<br />

a more accurate prediction of gait performance among<br />

this patient population. As lurch appears to be a robust<br />

objective measure of physical impairment, which is<br />

unaffected by BMI and co-morbidities, we believe that<br />

portable triaxial accelerometers can likely be used to<br />

conveniently collect objective gait data. This functional<br />

data may be used to supplement clinical efforts to screen<br />

and prioritize appropriate hip arthroplasty patients.<br />

20. MOTION-BASED JOINT COORDINATE<br />

SYSTEMS FOR THE ELBOW: A NEW<br />

METHOD FOR REDUCING VARIABILITY OF<br />

FLEXION KINEMATICS<br />

Louis M. Ferreira, Graham J.W. King*, James A.<br />

Johnson<br />

University of Western Ontario; *St. Joseph’s<br />

Health Care<br />

Results: Repeatability of creating motion-based JCS<br />

was less than 1 mm and 1º in all directions. <strong>The</strong> interspecimen<br />

standard-deviations of position and orientation<br />

measurements were smaller for the motion-based<br />

than for the anatomy-based JCS in every direction and<br />

for every specimen (p


248 COA/CORS/CORA<br />

23. IMAGE-BASED NAVIGATION IMPROVES<br />

THE POSITIONING OF THE HUMERAL<br />

COMPONENT IN TOTAL ELBOW<br />

ARTHROPLASTY<br />

Colin P. McDonald, James A. Johnson, Terry M.<br />

Peters*, Graham J.W. King<br />

St. Joseph’s Health Care London; *Robarts<br />

Research Institute<br />

Purpose: This study evaluated the accuracy of humeral<br />

component alignment in total elbow arthroplasty. An<br />

image-based navigated approach was compared against<br />

a conventional non-navigated technique. We hypothesized<br />

that an image-based navigation system would<br />

improve humeral component positioning, with navigational<br />

errors less than or approaching 2.0mm and 2.0°.<br />

Method: Eleven cadaveric distal humeri were imaged<br />

using a CT scanner, from which 3D surface models<br />

were reconstructed. Non-navigated humeral component<br />

implantation was based on a visual estimation of the<br />

flexion-extension (FE) axis on the medial and lateral<br />

aspects of the distal humerus, followed by standard<br />

instrumentation and positioning of a commercial prosthesis<br />

by an experienced surgeon. Positioning was based<br />

on the estimated FE axis and surgeon judgment. <strong>The</strong> stem<br />

length was reduced by 75% to evaluate the navigation<br />

system independent of implant design constraints. For<br />

navigated alignment, the implant was aligned with the<br />

FE axis of the CT surface model, which was registered<br />

to landmarks of the physical humerus using the iterative<br />

closest point algorithm. Navigated implant positioning<br />

was based on aligning a 3D computer model calibrated<br />

to the implant with a 3D model registered to the distal<br />

humerus. Each alignment technique was repeated for<br />

a bone loss scenario where distal landmarks were not<br />

available for FE axis identification.<br />

Results: Implant alignment error was significantly lower<br />

using navigation (P


COA/CORS/CORA 249<br />

and by rotating the cups, such that a 900 abduction<br />

angle and a 00 anteversion angle were achieved. <strong>The</strong><br />

grid used was divided to quadrants, and subdivided into<br />

radial thirds of the average rim radius. <strong>The</strong> correspondence<br />

of left and right density maps was investigated by<br />

comparing the average bone density in corresponding<br />

zones and across the population.<br />

Results: High bone densities were found around the roof<br />

of the acetabulum aligning with the femoral mechanical<br />

axis during standing. <strong>The</strong> highest average bone density<br />

were found to be the superior and posterior walls of<br />

the acetabulum, corresponding to regions 8, 9, and 12<br />

compared to other regions of the acetabuli (P


250 COA/CORS/CORA<br />

32. COMPARISON BETWEEN CT-<br />

TOMASD AND CT-OAM FOR ASSESSING<br />

OSTEOARTHRITIC AND NORMAL TIBIAE<br />

James D. Johnston*, Bassam A. Masri, David R.<br />

Wilson<br />

*University of Saskatchewan; UBC<br />

Purpose: Subchondral cortical and trabecular bone mineral<br />

density (BMD) may increase and/or decrease during<br />

different stages of osteoarthritis (OA) disease progression.<br />

2D in-vivo imaging studies examining direct associations<br />

between increased proximal tibial BMD and<br />

knee OA offer conflicting results, which may be due to<br />

the inherent limitations of 2D BMD imaging tools. Our<br />

objective was to compare existing and novel 3D imaging<br />

techniques for distinguishing subchondral bone properties<br />

in OA and normal cadaveric tibiae.<br />

Method: Eight intact cadaver knees from five donors<br />

(4M:1F; age: 77+/-10) were repositioned and scanned<br />

three times using QCT (0.5mm isotropic resolution,<br />

0.15mSv dosage). BMD was assessed using 1) computed<br />

tomography absorptiometry (CT-OAM) which uses<br />

maximum intensity projections to assesses peak density<br />

values within subchondral bone, and 2) our novel computed<br />

tomography topographic mapping of subchondral<br />

density (CT-TOMASD) technique, which uses surface<br />

projections to assess both cortical and trabecular bone<br />

density at specific depths from the subchondral surface.<br />

Average BMD at normalized depths of 0-2.5mm, 2.5–<br />

5.0mm, and 5.0-10mm from the surface were assessed<br />

using CT-TomasD. Regional analyses were performed<br />

consisting of: (1) medial/lateral (M/L) BMD ratio, and<br />

(2) BMD of a 10mm diameter core identified as having<br />

the maximum regional BMD. Each bone was assessed<br />

for OA using a modified-KL scoring system: Normal<br />

(mKL=0); Early-OA (1-2); and Late-OA (3-4).<br />

Results: OA was identified in four compartments of<br />

three tibiae (1 late OA+valgus, 1 late OA+varus, 1 early<br />

OA+neutral). Larger density differences between OA<br />

and normal knees were noted using CT-TOMASD compared<br />

with CT-OAM. CT-TomasD demonstrated that<br />

the two knees with late OA demonstrated M/L BMD<br />

ratios differing by more than 3.4 SD compared with<br />

normals, with peak cores higher than normals across<br />

all depths. <strong>The</strong> knee with early OA and neutral alignment<br />

demonstrated M/L ratios less than normals while<br />

core differences were highest proximally, with density<br />

becoming lower than normals with increasing depth.<br />

Conclusion: CT-TomasD demonstrated larger differences<br />

between OA and normal subjects when compared<br />

with CT-OAM differences. This may be due to CT-<br />

OAM primarily assessing peak density within the thin<br />

subchondral cortical endplate; a region demonstrating<br />

fairly uniform peak densities within a limited range.<br />

33. IN VITRO ACTIVITY OF FUSIDIC<br />

ACID AND VANCOMYCIN IN PMMA<br />

BONE CEMENT FOR THE TREATMENT<br />

OF METHICILLIN-RESISTANT<br />

STAPHLYOCOCCUS AUREUS<br />

Fay Leung, Clive P. Duncan, Helen Burt, John<br />

Jackson<br />

UBC<br />

Purpose: This study investigates the synergistic use of<br />

fusidic acid with vancomycin, and linezolid in polymethylmethacrylate<br />

(PMMA) cement for the treatment<br />

of orthopedic MRSA and MRSE infections. Alone, Vancomycin<br />

is typically eluted in limited quantities from<br />

cement. <strong>The</strong> purpose of this study was to 1) combine FA<br />

and Vancomycin, and Linezolid alone in PMMA cement<br />

and characterize antibiotic elution, and 2) to improve<br />

drug release using polyethylene glycol (PEG) and NaCl<br />

in PMMA cement.<br />

Method: Standardized 1g pellets of Palacos cement were<br />

manufactured containing Vancomycin and FA or Linezolid<br />

at increasing concentrations in three batches: without<br />

additive, with increasing concentrations of PEG,<br />

and with increasing concentrations of NaCl. <strong>The</strong> pellets<br />

were incubated in phosphate buffered saline and sampled<br />

at regular intervals. Drug analysis was performed<br />

with high pressure liquid chromatograpy.<br />

Results: Total drug release at 2.5% loading of Vancomycin<br />

alone was 0.84% and of FA was 2.35%. Linezolid<br />

showed comparable release profiles. Vancomycin and<br />

FA combined yeilded Vancomycin release of 6.2% and<br />

FA of 8.4%. <strong>The</strong> addition of 30% PEG increased release<br />

of Vancomycin and Fusidic Acid by six-fold. <strong>The</strong> addition<br />

of 18% NaCl increased total Vancomycin release<br />

by 11-fold but had no effect on FA release.<br />

Conclusion: Linezolid, Vancomycin and FA can be combined<br />

in PMMA and have favorable release profiles.<br />

<strong>The</strong> addition of PEG and NaCl dramatically increases<br />

the release of antibiotics, with the exception of FA and<br />

NaCl. <strong>The</strong>se strategies may be useful in the management<br />

of MRSA/MRSE infections.<br />

34. PREVALENCE OF CAM TYPE FAI<br />

MORPHOLOGY IN 200 ASYMPTOMATIC<br />

VOLUNTEERS<br />

Paul-Edgar Beaulé, Kalesha Hack, Gina DiPrimio,<br />

Kawan Rakhra<br />

University of Ottawa<br />

Purpose: A growing body of literature confirms that<br />

idiopathic OA is frequently caused by subtle, and often<br />

radiographically occult, abnormalities at the femoral<br />

head-neck junction or acetabulum that result in abnormal<br />

contact between the femur and acetabulum. This<br />

condition, known as femoroacetabular impingement, is<br />

a widely accepted cause of early OA of the hip. MRI is<br />

the imaging modality that is most sensitive in detecting<br />

cam morphology. <strong>The</strong>re is currently little published data<br />

regarding the prevalence of abnormalities of the femoral<br />

head-neck junction in patients without hip pain or previous<br />

hip pathology. <strong>The</strong> primary aim of this project is<br />

to examine the incidence of cam morphology in a population<br />

without hip pain or pre-existing hip disease using<br />

non-contrast MRI.<br />

Method: Two hundred asymptomatic volunteers underwent<br />

magnetic resonance imaging targeted to both hips.<br />

Subjects were examined at the time of MRI to document<br />

internal rotation of the hips at 90 degrees flexion<br />

and to assess for a positive impingement sign. <strong>The</strong> mean<br />

age was 29.4 years (range 21.4-50.6); 77.5% were Caucasian<br />

and 55.5% female. <strong>The</strong> Nötzli alpha angle was<br />

measured on oblique axial images through the middle of<br />

the femoral neck for each hip. A value greater than 50<br />

degrees was considered consistent with cam morphology.<br />

Measurements were performed independently by<br />

two musculoskeletal radiologists.<br />

Results: Twenty-six percent of volunteers had at least<br />

one hip with cam morphology: 20% had an elevated<br />

alpha angle on either the right or the left side, and 6%<br />

had bilateral deformity. <strong>The</strong> average alpha angle was<br />

42.6 degrees on the right (SD=7.9) and 42.4 degrees<br />

on the left (SD=7.7). Internal rotation was negatively<br />

correlated with alpha angle (p55.5 on the Dunn view and Pincer impingement was<br />

defined by the presence of either acetabular retroversion<br />

or coxa profunda. Statistical analysis was done using<br />

the two tailed paired t-test, chi-square test and intraclass<br />

correlation coefficient. Odds Ratios were calculated<br />

using conditional logistic regression.<br />

Results: Eighty-eight patients (77.8%) had bilateral<br />

deformity and 27% had symptoms in both hips. Mean<br />

α angles were higher for bilateral impingement deformity<br />

than for the impingement side only when unilateral<br />

deformity was present (72.10 versus 64.50, p


COA/CORS/CORA 251<br />

37. HIP ARTHROSCOPY MAY CAUSE<br />

CHONDROLYSIS<br />

Javad Parvizi, Orhan Bican, Kevin Bozic * , Chris<br />

Peters †<br />

Rothman Institute; * University of California;<br />

†<br />

University of Utah<br />

Purpose: Hip arthroscopy has been used at an increasing<br />

frequency over the last few years. Majority of patients<br />

undergoing hip arthroscopy are young and active individuals<br />

who seek definitive therapy for a painful hip condition<br />

and wish to avoid undergoing a hip replacement.<br />

Although relatively successful, complications following<br />

hip arthroscopy occur. This multi-institutional study<br />

presents a worrisome and previously unrecognized complication<br />

of hip arthroscopy, namely chondrolysis that<br />

lead to accelerated development of end-stage arthritis.<br />

Method: Using the computerized database in each<br />

institution, patients undergoing total hip arthroplasty<br />

between 1999-2008 who had received hip arthroscopy<br />

prior to arthroplasty were identified. 39 patients were<br />

identified to have undergone at least one hip arthroscopy<br />

on the affected hip prior to arthroplasty. <strong>The</strong>re<br />

were21 female and 18 male patients. <strong>The</strong> mean age of<br />

patients at the time of hip arthroscopy was 45.9 years.<br />

Data regarding demographics, comorbidities, preoperative<br />

diagnosis, number of previous procedures, and the<br />

details of the surgical procedure were compiled. Radiographs<br />

were evaluated.<br />

Results: <strong>The</strong> median time from arthroscopy to arthroplasty<br />

was 14.8 months (range 2.2 months to 7 years).<br />

Fourteen patients (35%) underwent THA within 12<br />

months of a previous hip arthroscopy and in nine of<br />

these patients the indication for hip arthroscopy was<br />

labral tear debridement. <strong>The</strong>se patients despite having<br />

none to minimal arthritis at the time of arthroscopy<br />

developed accelerated arthritis within a year that necessitated<br />

hip replacement.<br />

Conclusion: Hip arthroscopy can cause accelerated<br />

degenerative arthritis of the hip in some patients. We<br />

believe the subset of our patients who developed arthritis<br />

so early after hip arthroscopy may have suffered<br />

chondrolysis and/or chondral injury during the arthroscopy<br />

that resulted in progressive and aggressive arthritis<br />

of the hip within 12 months. Hip arthroscopy should<br />

be reserved for a select group of patients. All measures<br />

to minimize the possibility of chondrolysis and/or chondral<br />

injury should be exercised.<br />

COA/CORS COMBINED PAPER SESSION 5:<br />

SPINE/TRAUMA<br />

38. BIOMECHANICAL CHARACTERIZATION<br />

OF CERVICAL SPINE DISLOCATION IN AN<br />

INNOVATIVE SPINAL CORD INJURY (SCI)<br />

MODEL<br />

Qingan Zhu, Jie Liu, Tim Bhatnagar, Wolfram<br />

Tetzlaff, Thomas Oxland<br />

UBC<br />

Purpose: Recent studies have shown differences in short<br />

term spinal cord pathology between spinal column<br />

injury mechanisms, such as contusion and fracturedislocation.<br />

Such differences may exist at longer time<br />

points, and thus survival studies are needed in the dislocation<br />

models. A more in-depth characterization of the<br />

dislocation model is needed for development of a mildmoderate<br />

cervical spine dislocation model in a rat that is<br />

suitable for survival studies. Specifically, our objective in<br />

this study was to determine the dislocation displacement<br />

that produces initial spinal column failure in a Sprague-<br />

Dawley rat model and to validate a consistent injury at<br />

the desired dislocation in-vitro and in-vivo.<br />

Method: For the dislocation model, the dorsal ligaments<br />

and facets at C4-C5 were removed to mimic the<br />

type of posterior element fracture and ligament injury<br />

commonly seen in a bilateral fracture-dislocation. C3<br />

and C4 were clamped together and held stationary<br />

while the clamp holding C5 and C6 was connected to<br />

an electromagnetic actuator and displaced dorsally to<br />

produce the injury while force and displacement were<br />

recorded. Twenty-eight isolated cervical spine specimens<br />

of Sprague-Dawley rats were used to determine dislocation<br />

displacement at initial spinal column failure. <strong>The</strong><br />

C4-C5 segment sustained dislocation (>3mm) injury<br />

at 0.05mm/s (n=11), 100mm/s (n=4) and 1000mm/s<br />

(n=13). Initial spinal column failure was defined at with<br />

maximum force during the dislocation. A dislocation<br />

displacement of 1.4mm was applied to 7 isolated specimens<br />

and 4 anesthetized rats at 430mm/s. <strong>The</strong> spinal<br />

column failure was inspected up to 3 days after injury,<br />

as well as hemorrhage of spinal cord in-situ.<br />

Results: <strong>The</strong> dislocation displacement at in-vitro spinal<br />

column failure was 0.95mm±0.32 and not significantly<br />

different among specimens at the three dislocation<br />

speeds. Under a dislocation displacement of 1.4mm,<br />

rupture of the C4-C5 disc occurred in all in-vitro<br />

(0.67mm±0.38) and in-vivo (0.65mm±0.17) cases. SCI<br />

hemorrhage at epicenter was observed in 3 of 4 cases.<br />

Conclusion: <strong>The</strong> initial spinal column failure in an innovative<br />

SCI model occurs at displacement between 0.65mm<br />

and 0.95mm. Dislocation displacement of 1.4mm results<br />

in spinal column failure consistently and SCI hemorrhage,<br />

and may be suitable for survival studies.<br />

39. MAVERICK TOTAL DISC REPLACEMENT<br />

IN THE LOWER LUMBAR SPINE ADJACENT<br />

TO A LONG SPINAL FUSION: AN IN VITRO<br />

BIOMECHANICAL STUDY OF KINEMATICS<br />

Qingan Zhu, Claire Jones, Tim Schwab,<br />

Chad Larson, Eyal Itshayek, Lawrence Lenke,<br />

Washington University; Peter Cripton<br />

UBC<br />

Purpose: A long spinal fusion across the thoracolumbar<br />

region is sometimes applied in scoliosis. Adjacent level<br />

degeneration below these constructs has been documented.<br />

Treatment with an artificial disc replacement<br />

below the fusion has been proposed to prevent degeneration<br />

there. <strong>The</strong>re is currently little data detailing the<br />

expected biomechanics of this situation. <strong>The</strong> objective<br />

of this study was to evaluate range of motion (ROM)<br />

and helical axis of motion (HAM) changes due to oneand<br />

two-level Maverick total disc replacement adjacent<br />

to a long spinal fusion.<br />

Method: A multidirectional flexibility testing protocol<br />

with compressive follower preload was used to test seven<br />

human cadaveric spine specimens (T8-S1). A continuous<br />

pure moment ±5.0 Nm was applied in flexion-extension<br />

(FE), lateral bending (LB) and axial rotation (AR), with<br />

a compressive follower preload of 400 N. <strong>The</strong> motion<br />

of each vertebra was monitored with an optoelectronic<br />

camera system. <strong>The</strong> test was completed for the intact<br />

condition and after each surgical technique: (1) T8-L4<br />

fusion and facet capsulotomy at L4-L5 and L5-S1; (2)<br />

L4-L5 Maverick; (3) L5-S1 Maverick. Maverick total<br />

disc replacement and fusion with the CD Horizon<br />

system was performed. Repeated measures ANOVA<br />

was used to analyze changes in ROM and HAM of the<br />

L4-L5 and L5-S1 segments.<br />

Results: Following L4-L5 Maverick replacement, L4-<br />

L5 ROMs tended to decrease slightly (on average from<br />

6.2°±2.8° to 5.1°±3.8° in FE, 1.1°±1.1° to 0.9°±0.5° in<br />

LB and 1.3°±0.9° to 1.0°±0.6° in AR). With two-level<br />

Maverick implantation, L5-S1 ROMs tended to increase<br />

slightly in FE (from 6.6°±2.6° to 7.1°±3.9°), and to<br />

decrease slightly in LB (from 1.5°±0.9° to 1.0°±0.3°) and<br />

AR (from 1.5°±1.5° to 1.1°±0.6°), compared to the fused<br />

condition. As a trend, HAM location shifted posteriorly<br />

in FE and AR, and inferiorly in LB following Maverick<br />

replacement. However, neither ROM nor HAM at these<br />

two segments showed any significant change due to the<br />

implantation of one-or two-level Maverick total disc<br />

replacement in any of the three directions.<br />

Conclusion: <strong>The</strong> present results suggested that lower<br />

lumbar segments with Maverick disc replacement exhibited<br />

intact-like kinematics in both extent and quality of<br />

motion.<br />

40. CERVICAL SPINAL MOTION THAT<br />

WOULD OTHERWISE BE SAFE, CAN<br />

CAUSE SPINAL CORD COMPRESSION IN A<br />

STENOTIC SPINE<br />

James C. Boak, Philippe Gedet, Marcel Dvorak*,<br />

Stephen Ferguson*, Peter Cripton<br />

UBC; *University of Bern<br />

Purpose: <strong>The</strong> average age of people suffering spinal cord<br />

injuries in many countries is shifting toward an older<br />

population, with a disproportionate number occurring<br />

in the spondylotic cervical spine. <strong>The</strong>se injuries<br />

are typically due to low energy impacts, such as a fall<br />

from standing height. Since a stenotic spinal canal (a<br />

common feature of a spondylotic cervical spine) can<br />

cause myelopathy when the spine is flexed or extended,<br />

traumatic flexion or extension likely causes the injury<br />

during the low energy impact. However, this injury<br />

mechanism has not been observed experimentally.<br />

Method: To better understand this injury mechanism an<br />

in-vitro study, using six whole cervical porcine spines,<br />

was conducted. <strong>The</strong> following techniques were combined<br />

to directly observe spinal cord compression in a<br />

stenotic spine during physiologic and super-physiologic<br />

motion:<br />

• A radio-opaque surrogate cord, with material properties<br />

matched to in-vivo specimens, replaced the real<br />

spinal cord.<br />

• Sagittal plane X-rays imaged the surrogate cord in the<br />

spine during testing.<br />

• Varying levels of canal stenosis were simulated by a<br />

M8 machine cap screw that entered the canal from<br />

the anterior by drilling through the C5 vertebral<br />

body.<br />

• Pure moment loading and a compressive follower<br />

load were used to replicate physiologic and superphysiologic<br />

motion.<br />

Results: Initial results show that a stenotic occlusion that<br />

removes all extra space in the canal in the neutral posture,<br />

without compressing the cord, can lead to spinal<br />

cord compression within physiologic ranges of flexion<br />

and extension. <strong>The</strong> spinal cord can also be compressed<br />

during slightly super-physiologic flexion and extension<br />

with only 25% canal occlusion. Physiologic loads and<br />

motions in the same spines did not cause cord compression<br />

when canal occlusion was 0%.<br />

Conclusion: <strong>The</strong>se results support the hypothesis that cervical<br />

spinal canal stenosis increases the risk of spinal cord<br />

injury because spinal cord compression was observed<br />

during motions and loads that would be safe for a nonstenotic<br />

spine. <strong>The</strong>se results are limited primarily due to<br />

the use of a porcine spine. However, this new stenosis<br />

model and experimental technique will be applied to invitro<br />

human spine specimens in future work.<br />

41. THE POSITIVE EFFECTS OF POLY-N-<br />

ACETYL GLUCOSAMINE ON HUMAN<br />

INTERVERTEBRAL DISC CELL METABOLISM<br />

IN VITRO<br />

Deepthi Gorapalli 1,2 , Albert J.M. Yee 1,2 , Aiguo<br />

Zhang 1 , Marina Demcheva 3 , Cari Whyne 1,2 ,<br />

J. Vournakis 3 , A. Seth 1<br />

1<br />

Sunnybrook Health Sciences Centre, 2 University<br />

of Toronto, 3 Marine Polymers Technologies Inc<br />

Purpose: <strong>The</strong>re is interest in biologic strategies that can<br />

potentially treat degenerative disc disease (DDD). A new<br />

deacetylated derivative of a marine diatomic glycosaminoglycan<br />

(DEAC) was developed and incorporated<br />

into two sulphated hydrogel formulations; Gel 1 and<br />

2. <strong>The</strong>se materials were proposed to have a reparative<br />

effect on damaged tissue. Biochemical studies were conducted<br />

using primary human disc cell (HDC) cultures.<br />

Method: HDCs were isolated from surgical specimens<br />

by sequential enzymatic digestion (pronase and collagenase).<br />

Time-course in-vitro studies were conducted<br />

on cell cultures treated with DEAC, Gel 1 or Gel 2 (28<br />

day period). Proteoglycan content (alcian blue), cellular<br />

viability/proliferation (MTT assay), and type collagen<br />

II, aggrecan expression (RT-PCR, immunohistochemistry)<br />

was assessed.<br />

Results: When compared to controls, the DEAC, Gel 1<br />

and 2 treated HDC groups showed significant increases<br />

J BONE JOINT SURG [BR] 2011; 93-B:SUPP III


252 COA/CORS/CORA<br />

in proteoglycan content as early as day 12. <strong>The</strong> greatest<br />

effect was observed with Gel 1 (78.4±1.9 fold greater<br />

optical density compared to control, p < 0.05). <strong>The</strong><br />

amount of proteoglycan quantified on DEAC treated<br />

HDCs on day 28 was 27.7±0.09 times higher than<br />

control (p


COA/CORS/CORA 253<br />

necrosis versus apoptosis to cellular injury. Existing<br />

studies have only considered indomethacin administration<br />

prior to induction of compartment syndrome.<br />

<strong>The</strong> purpose of this study was to determine the effect<br />

of timing of indomethacin administration on muscle<br />

damage in compartment syndrome, and to assess apoptosis<br />

as a cause of tissue demise.<br />

Method: Twenty-four Wistar rats were randomized to<br />

elevated intracompartmental pressure (EICP) for either<br />

45 or 90 minutes (30mm Hg). In the 45 min group,<br />

indomethacin was withheld (group 1), given prior to<br />

induction of EICP (group 2) or given 15 min prior to<br />

fasciotomy (group 3). In the 90 min group, indomethacin<br />

was withheld (group 4) or provided 30 or 60 minutes<br />

prior to fasciotomy (groups 5 and 6). Intravital<br />

microscopy and histochemical staining assessed capillary<br />

perfusion, cell damage and inflammatory activation<br />

within EDL muscle. Apoptosis was assessed using<br />

ELISA staining for caspase-3. Groups were compared<br />

with one-way ANOVA (p


254 COA/CORS/CORA<br />

51. CEMENTED AND CEMENTLESS TOTAL<br />

HIP ARTHROPLASTY: RESULTS OF A<br />

RANDOMISED CONTROLLED TRIAL AT 17<br />

TO 21 YEARS FOLLOW-UP<br />

Keegan Au*, Kristoff Corten † , Robert B. Bourne † ,<br />

Cecil H. Rorabeck † , Andreas Laupacis ‡ , Li Ka<br />

Shing ‡<br />

*London Health Sciences Centre; † University<br />

Hospital, University of Western Ontario;<br />

‡<br />

Knowledge Institute of St. Michael’s Hospital<br />

Purpose: A randomised controlled trial comparing fixation<br />

of a Mallory-Head prosthesis for total hip arthroplasty<br />

(THA) with and without cement was performed<br />

with average 19 years (range 17-21 years) of continuous<br />

follow-up.<br />

Method: Two hundred and fifty patients were randomised<br />

to undergo THA using either a Mallory-Head<br />

THA prosthesis designed to be inserted with cement or<br />

one designed for cementless insertion. Both patients and<br />

those involved in outcome assessment were blinded to<br />

the type of implantation. Patients were followed yearly<br />

after the first post-operative year for outcomes including<br />

mortality, revision arthroplasty, and health-related quality<br />

of life assessment scores.<br />

Results: Primary THA was performed with cement<br />

in 124 patients and without cement in 126 patients.<br />

Mean age at the time of surgery was 64 years, and 48%<br />

were female. During the period of review, there were<br />

78 (31%) deaths in the cohort, and 75 (30%) patients<br />

underwent revision surgery. Kaplan-Meier survivorship<br />

analysis revealed significantly increased revision rates in<br />

cemented compared with cementless THA using failure<br />

of either component (p=0.01) or femoral component<br />

(p


COA/CORS/CORA 255<br />

approach=140.0, p=0.053). <strong>The</strong> cup abduction angle<br />

(CA) was slightly different between the two groups (MI<br />

RAH 42.5°, lateral approach=39.2°, p=0.03). More<br />

patients had cup abduction angles in the 45°-55° range<br />

(p=0.009) in the MI HR group but none had a cup angle<br />

over 55° of abduction in either group. On the femur<br />

side, component positioning was comparable.<br />

Conclusion: Based on our early results, the anterior-<br />

Hueter approach is a reasonable alternative to more<br />

extensile surgical approaches. Like any MI approach to<br />

hip surgery, great care has to be taken not to put the<br />

cups too vertical. Further long-term studies as well as<br />

comparisons to other approaches such as the posterior<br />

approach will determine if the anterior approach can be<br />

recommended for hip resurfacing.<br />

56. COMPUTER NAVIGATED HIP<br />

RESURFACING: AN EVALUATION OF<br />

ACCURACY AND LEARNING CURVE<br />

Michael Olsen, Edward T. Davis*, James P.<br />

Waddell, Emil H. Schemitsch<br />

St. Michael’s Hospital, *Royal Orthopaedic<br />

Hospital, Birmingham, UK<br />

Purpose: Hip resurfacing is a technically demanding<br />

alternative to total hip arthroplasty. Placement of the<br />

initial femoral guidewire utilizing traditional mechanical<br />

jigs may lead to preparatory errors and a high degree<br />

of variability in final implant stem-shaft angle (SSA).<br />

Intra-operative computer navigation has the potential<br />

to decrease preparatory errors and provide a reliable<br />

method of femoral component placement. <strong>The</strong> current<br />

study evaluated the accuracy and learning curve of 140<br />

consecutive navigated hip resurfacing arthroplasties.<br />

Method: Between October 2005 and May 2007, 140 consecutive<br />

Birmingham Hip Resurfacings were performed<br />

on 132 patients (107 male, 25 female). <strong>The</strong> mean age of<br />

the cohort was 51.2 years (range 25-82). Indications for<br />

surgery included osteoarthritis (n=136) and avascular<br />

necrosis (n=4). Pre-operative templating was performed<br />

using digital AP unilateral hip radiographs. Neck-shaft<br />

angles (NSA) were digitally measured and relative<br />

implant stem-shaft angles planned. <strong>The</strong> central guidewire<br />

was drilled and verified intra-operatively using an imageless<br />

navigation system. Implant stem-shaft angles were<br />

assessed using 3 month post-operative radiographs.<br />

Results: Pre-operative templating determined a mean NSA<br />

of 132.2 degrees (SD 5.3 degrees, range 115-160). <strong>The</strong><br />

planned SSA was a relative valgus alignment of 9.5 degrees<br />

(SD 2.6 degrees). <strong>The</strong> post-operative SSA differed from the<br />

planned SSA by 2.5 degrees (SD 1.9 degrees, range 0-8).<br />

<strong>The</strong> final SSA measured within ±5 degrees of the planned<br />

SSA in 89% of cases. Of the remaining 11% of cases, all<br />

measurements erred in valgus. No cases of neck notching<br />

or varus implant alignment occurred in the series. <strong>The</strong><br />

mean navigation time for the entire series was 18 minutes<br />

(SD 6.6 minutes, range 10-50). A learning curve was<br />

observed with respect to navigation time, with a significant<br />

decrease in navigation time between the first 20 cases and<br />

the remainder of the series. <strong>The</strong>re was no evidence of a<br />

learning curve for implant placement accuracy.<br />

Conclusion: Imageless computer navigation shows<br />

promise in optimizing preparation of the femoral head<br />

and reducing the introduction of mechanical preparatory<br />

factors that predispose to femoral neck fracture.<br />

Navigation may afford the surgeon an accurate and<br />

reliable method of femoral component placement with<br />

negligible learning curve.<br />

COA PAPER SESSION 7:<br />

TRAUMA LOWER EXTREMITY 1<br />

57. EVEN UNDISPLACED FRACTURE NECK<br />

OF FEMUR CAN RESULT INTO POOR<br />

OUTCOME DEPENDING ON ASSOCIATED<br />

CO-MORBIDITES: A STUDY OF AO<br />

CANNULATED SCREWS DONE IN 315<br />

PATIENTS<br />

Vishal Upadhyay, Ajay Sahu*, Charalambos P.<br />

Charalambous, N. Harshawardena, Heath P.<br />

Taylor, Mark Farrar<br />

Poole General Hospital, UK; *Stepping Hill<br />

Hospital, UK<br />

Purpose: <strong>The</strong> aim of this study was to analyze the outcome<br />

of AO cannulated screws for undisplaced fracture<br />

neck of femur and find out the correlation in outcome<br />

with respect to co-morbidities in a general trauma unit<br />

in UK.<br />

Method: A retrospective study was conducted using<br />

data from electronic patient record,clinical coding<br />

information,clinic letters and GP’s. 315 patients who<br />

underwent AO screws for fracture neck of femur during<br />

2000 to 2004 were included. We looked into age, place<br />

of living, classification, mechanisn of injury, comorbidities,<br />

mobility before fracture, allergy, addictions,<br />

whether patient was anticoagulated, delay for theatre<br />

with reasons, length of stay in hospital, complications<br />

and treatment for complications. We assessed reasons<br />

for other admissions later on, need and type of another<br />

operation, consequently developed comorbidities,<br />

patient getting fracture of other side and its treatment,<br />

time and cause of death if happened?<br />

Results: <strong>The</strong>re were 81 males and 234 females in the<br />

study. Mean age of patients was 72 years (range 50-96<br />

years). Non-union occurred in 19 patients (6%) and<br />

avascular necrosis occurred in 49 patients (15.5%).<br />

Reoperation with an arthroplasty was required in 69<br />

patients (21.9 %). <strong>The</strong> incidence of avascular necrosis<br />

with internal fixation at 1 year was 31 (9.8%). Fifty-one<br />

(16%) patients died in 2 year period. <strong>The</strong> age, walking<br />

ability of the patient, and associated co-morbidities<br />

were of statistical significance in predicting fracture<br />

healing complications. We correlated our complications<br />

with comorbidities and found them more in patients<br />

with end-stage renal failure, steroid intake, osteoporosis<br />

and diabetes mellitus etc.<br />

Conclusion: <strong>The</strong> rate of fracture healing complications<br />

and reoperations in patients with undisplaced fractures<br />

was high in our series with two year follow up. It was<br />

even higher in patients with age greater than 80 years<br />

and some specific comorbidities. We should also consider<br />

co-morbidities and age before deciding for internal<br />

fixation rather than only the fracture configuration<br />

(Treat patient not the X-rays). Outcome is multifactorial<br />

and depends on many predictive factors. Each patient<br />

should be evaluated carefully and we should treat the<br />

physiological age and not the chronological age.<br />

58. A STUDY ANALYSING THE OUTCOME<br />

OF AO CANNULATED SCREWS FOR THE<br />

FRACTURE NECK OF FEMUR IN PATIENTS<br />

WITH DIABETES MELLITUS<br />

Vishal Upadhyay, Ajay Sahu, Ravi Mahajan*,<br />

Heath Taylor, Mark Farrar<br />

Poole Hospital NHS, Trust, UK; *Royal<br />

Bournemouth Hospital NHS Trust, UK<br />

Purpose: <strong>The</strong> aim of the study was to analyze the outcome<br />

of AO cannulated screws for fractures neck of<br />

femur in patients with Diabetes mellitus.<br />

Method: Sixty-two patients aged 50 years or more (17<br />

males & 45 females) who underwent AO screws for<br />

fracture neck of femur over seven years (1999–2005)<br />

and followed-up for a minimum of two years formed<br />

the study population. A retrospective review of data<br />

from electronic patient record (EPR), clinical coding,<br />

clinic & GP letters was made. Age, residential placement,<br />

Garden’s classification of fracture, mode of injury,<br />

associated other co morbidities, pre-admission mobilisation<br />

status, allergies, addictions and anticoagulation<br />

status details were collected.<br />

Results: <strong>The</strong> mean age of patients was 67 years (range<br />

52–96 yrs). Eleven patients died in two years time.<br />

Forty-one patients were less than 75 years of age and 21<br />

patients were more than 75 years of age. All the patients<br />

more than 75 years of age had undisplaced intracapsular<br />

fractures. Thirteen patients were type I and 49 patients<br />

were type II diabetic. Non-union and avascular necrosis<br />

occurred in nine (17%) & 13 (26%) patients respectively.<br />

Revision surgery in the form of total hip replacement or<br />

hemiarthroplasty were performed in 21 (41%) cases. <strong>The</strong><br />

incidence of avascular necrosis following osteosynthesis<br />

at one year was 14%. Age, control of diabetes, postoperative<br />

complications, pre-fracture mobilization status<br />

etc. Complications like wound infection were more principally<br />

in patients who had poorly-controlled diabetes.<br />

Conclusion: Patients with diabetes mellitus have metabolic<br />

bone disease due to vasculitis. This increases the<br />

risk of complications associated with fracture fixation<br />

such as non-union, cut-through and avascular necrosis<br />

(AVN). <strong>The</strong> complications and revision surgery rate<br />

was high in patients with displaced fractures and with<br />

poorly controlled diabetes. Comorbidities like diabetes<br />

and patient’s age were also strong predictors of healing<br />

in addition to fracture configuration. Looking at very<br />

high complication and re-operation rate, our recommendation<br />

in patients with diabetes is primary hemiarthroplasty<br />

irrespective of femoral head displacement, if<br />

there age is more than 75 years.<br />

59. HIP FRACTURE COMPLICATIONS AND<br />

30-DAY MORTALITY OUTCOMES WITHIN<br />

ACUTE CARE IN CANADA FROM 2001–02<br />

TO 2003–04<br />

Gisele M. Carriere, Pierre Guy*<br />

Statistics Canada; *UBC<br />

Purpose: Decubitus ulcers and post-operative infections<br />

significantly impact patients` outcome and resource utilization.<br />

<strong>The</strong> purpose of this study is to report incidence<br />

of post-surgical infection, decubitus ulcer and associations<br />

to 30-day in-hospital mortality among elderly<br />

Canadians admitted for hip fracture.<br />

Method: Statistics Canada`s national Health Person-<br />

Oriented Information database of linked acute care<br />

hospital discharges was queried for fiscal 2001–02,<br />

2002–03, 2003–04 creating a cohort of 67,434 hip fracture<br />

patients aged 60+. Demographics, comorbidities<br />

(enhanced Charlson Index), fracture type and treatment<br />

were used in logistic regression models to report odds<br />

ratios for outcomes.<br />

Results: Women were 76% of the cohort, median age<br />

was 82 yrs. Decubitus ulcer was detected in 2.3% of<br />

hip fracture patients. Increased risk was indentified<br />

for trochanteric fractures (OR 1.14, p< .05), dementia<br />

(OR 1.25, p< .05) and increasing age (OR: 1.02,<br />

p


256 COA/CORS/CORA<br />

60. PERCUTANEOUS REDUCTION AND<br />

FIXATION OF ACETABULUM FRACTURES IN<br />

ELDERLY PATIENTS<br />

Joshua Gary, Kelly Lefaivre*, Frank Gerold,<br />

Michael Hay † , Charles M. Reinert, Adam J. Starr<br />

UTSW; *UBC; † UTS<br />

Purpose: Acetabular fractures in elderly patients are<br />

difficult problems with various treatment options. Our<br />

institution treats many of these patients with percutaneous<br />

acetabular fixation. We reviewed medical records<br />

and contacted patients to determine the rate of conversion<br />

to total hip arthroplasty.<br />

Method: Our institutional trauma database was searched<br />

for all patients age 60 and older who had been treated<br />

with percutaneous screw fixation for an acetabular fracture.<br />

Seventy-nine consecutive patients (80 fractures)<br />

were identified. Medical records were examined to obtain<br />

peri-operative and follow-up information regarding the<br />

hospital course and conversion to total hip arthroplasty.<br />

A survivorship anaylsis was created with conversion to<br />

total hip arthroplasty as the censored event, and standard<br />

Kaplan-Meier curves were constructed. Five categorical<br />

variables were used to test for differences in survival of<br />

the native hip: age, sex, simple versus complex fracture<br />

pattern, closed versus limited open reduction, and occurrence<br />

of a medical complication.<br />

Results: Seventy-five fractures had adequate clinical<br />

follow-up with a mean of 3.9 years (range 0.5 – 11.9<br />

years). Average blood loss was 69 cc and there were no<br />

postoperative infections. 19/75 (25%) were converted<br />

to total hip arthroplasty at a mean time of 1.4 years<br />

after the index procedure. Survivorship analysis demonstrated<br />

a cumulative survival of 65% at 11.9 years<br />

of follow-up. <strong>The</strong>re were no conversions to arthroplasty<br />

beyond 4.7 post-operatively. <strong>The</strong>re were no statistically<br />

significant associations between conversion to<br />

arthroplasty and age, sex, closed versus limited open<br />

reduction, simple versus complex fracture pattern, and<br />

occurrence of a medical complication.<br />

Conclusion: Percutaneous fixation is a viable treatment<br />

option for patients age 60 or greater with acetabular fractures.<br />

Rates of conversion to total hip arthroplasty are<br />

comparable to other treatment methods and if conversion<br />

is required, soft tissues are preserved for future surgery.<br />

61. MODIFIED STOPPA APPROACH FOR<br />

ACETABULAR FRACTURES OF THE<br />

ELDERLY PATIENT<br />

G. Yves Laflamme, Benoit Benoit, Stéphane Leduc,<br />

Jonah Hébert-Davies*<br />

Hôpital du Sacré-Coeur; *Université de Montreal<br />

Purpose: <strong>The</strong> age of patients presenting with acetabular<br />

fracture has increased over the last ten years. Older<br />

patients tend to have patterns involving the anterior<br />

column with comminution of the quadrilateral plate.<br />

Our goal was to investigate the appropriateness of<br />

open reduction and internal fixation using the modified<br />

Stoppa approach for geriatric acetabular fractures.<br />

Method: A retrospective review of patients over the age<br />

of 60 having presented to an academic level I trauma<br />

center over the course of four years. Twenty patients<br />

were identified and treated using the modified Stoppa<br />

approach with plating of the quadrilateral surface.<br />

Patients were evaluated clinically using both SF-36 and<br />

Harris Hip Score. Records and radiographs (using criteria<br />

described by Matta) were reviewed retrospectively.<br />

Results: All patients were followed for a minimum<br />

of two years with no lost at follow-up. Mean age for<br />

patients at time of intervention was 68 years. Average<br />

blood lost was 800cc and surgical time was 130 minutes<br />

(range, 55-210). <strong>The</strong>re was one traumatic injury<br />

to the obturator nerve and two patients were noted to<br />

have temporary weakness of the hip adductors postoperatively.<br />

Average Harris Hip Score and the SF-36<br />

were improved significantly (p


COA/CORS/CORA 257<br />

65. ENDSTAGE ARTHRITIS FOLLOWING<br />

TIBIA PLATEAU FRACTURES: 10 YEAR<br />

FOLLOW UP<br />

Ramin Mehin, Peter O’Brien, Penny Brasher,<br />

Henry M. Broekhuyse, Piotr Blachut, Robert N.<br />

Meek, Pierre Guy<br />

UBC<br />

Purpose: Problem: Tibia plateau fractures may lead to<br />

end-stage post-traumatic arthritis that requires reconstructive<br />

surgery. <strong>The</strong> incidence of this problem is<br />

unknown but has been estimated at 20–40% by studies<br />

that were limited by small sample sizes, potential followup<br />

bias, and the limitations of using radiographic arthritis<br />

as a chosen outcome (not correlated to function). <strong>The</strong><br />

use of administrative data bases to follow the care of a<br />

large number patients for robust end points such as surgery,<br />

offers an opportunity to address these limitations.<br />

Purpose: to determine the minimum ten year incidence<br />

of post-traumatic arthritis necessitating reconstructive<br />

surgery following tibia plateau fractures.<br />

Method: We queried our prospectively collected Orthopedic<br />

Trauma Data base to identify operatively treated<br />

patients with tibia plateau fractures. <strong>The</strong>se cases were<br />

cross-referenced with the data from our Province’s<br />

administrative health database and tracked over time<br />

for the performance of reconstructive knee surgery.<br />

Each individual’s exposure/follow-up period was limited<br />

by end of health plan coverage on record or date of<br />

death from vital statistics data. <strong>The</strong> minimum follow-up<br />

was ten years.<br />

Results: Between 1987 and 1994, 378 patients with a<br />

tibia plateau fracture were treated at our institution.<br />

<strong>The</strong> average age was 46 years (SD=18, range 14-87),<br />

while 56% of patients were males. Seventeen out-of-<br />

Province residents were excluded, along with forty-six<br />

others whose “Medical Services Plan” numbers could<br />

not be identified. Of which seven were WCB patients<br />

and one who was affiliated with the military. <strong>The</strong> study<br />

cohort therefore consisted of 311 patients with 314 tibia<br />

plateau fractures. Four individuals (1.3%) we treated<br />

tibia plateau fractures have required reconstructive knee<br />

surgery for end-stage post-traumatic knee arthritis at 10<br />

years. Of these 3 of 4 were type VI fractures and 1 of<br />

4 was open.<br />

Conclusion: Patients who require surgical treatment of<br />

tibia plateau fractures may be counseled on their longterm<br />

risk of requiring reconstructive knee surgery for<br />

endstage knee arthritis based on a clinical study. Based<br />

on our findings, the proportion of those who have<br />

required a total knee surgery, ten years following their<br />

injury, is lower than previously published.<br />

66. OSTEOTOMY FOR FEMORAL OR TIBIAL<br />

SHAFT MALUNION IN PATIENTS WITH<br />

END-STAGE OSTEOARTHRITIS OF THE<br />

KNEE<br />

Arvindera Ghag, Pierre Guy, Peter J. O’Brien,<br />

Henry M. Broekhuyse, Robert N. Meek, Piotr A.<br />

Blachut<br />

UBC<br />

Purpose: Femoral and tibial shaft malunion may predispose<br />

to knee osteoarthritis but may also pose a problem<br />

for knee reconstruction; malposition of total knee<br />

prostheses being a known cause of early failure. Limb<br />

realignment may prove to be beneficial prior to proceeding<br />

with arthroplasty. <strong>The</strong> purpose of this study was to<br />

evaluate the outcome and effect of shaft osteotomy prior<br />

to total knee arthroplasty (TKA).<br />

Method: A search of the trauma database between 1987<br />

and 2006 was conducted. Twenty-two osteotomies were<br />

performed on 21 patients with femoral or tibial shaft<br />

malunion who had been considered for TKA. Mean<br />

age at osteotomy was 54 years and mean follow-up 86<br />

months. Time intervals between surgical procedures and<br />

Knee Society scores were calculated. Patients were surveyed<br />

regarding pain relief and functional improvement.<br />

Results: Femoral osteotomy improved mean Knee Society<br />

knee scores from 47 to 76 and function scores from<br />

34 to 61. Tibial osteotomy improved knee scores from<br />

53 to 82 and function scores from 28 to 50. Four osteotomies<br />

were complicated by nonunion and required<br />

further intervention. Osteotomy subjectively improved<br />

pain and function for a mean of 56 months. Femoral<br />

and tibial shaft osteotomy delayed TKA in 45% (10<br />

cases) for a mean period of just over 6.5 years (89 and 73<br />

months for femoral and tibial osteotomy respectively).<br />

Pre and post Knee society scores were: Femur: knee 56<br />

to 88, function 41 to 72; Tibia: knee 65 to 85, function<br />

25 to 57. One TKA was revised after 11 months due to<br />

valgus malalignment and was complicated by a wound<br />

infection. <strong>The</strong>re were no other infections or wound<br />

complications. <strong>The</strong> procedure additionally relieved pain<br />

and improved function in the remaining 12 joints, not<br />

yet requiring arthroplasty.<br />

Conclusion: Femoral and tibial shaft osteotomy may<br />

delay and possibly avoid TKA, relieve pain and improve<br />

function in patients who present with malunion and<br />

end-stage knee arthritis. <strong>The</strong> complication rate and clinical<br />

results of TKA following shaft osteotomy appear<br />

to be similar to primary TKA.This treatment strategy<br />

should be considered in younger patients with post traumatic<br />

osteoarthritis where significant femoral or tibial<br />

deformity is present.<br />

COA PAPER SESSION 8:<br />

RECONSTRUCTIVE UPPER EXTREMITY 1<br />

67. ASSOCIATION BETWEEN THE UPWARD<br />

MIGRATION INDEX AND FUNCTIONAL<br />

AND QUALITY OF LIFE OUTCOMES IN<br />

ARTHROSCOPIC ROTATOR CUFF REPAIR<br />

Peter Lapner, Emilio Lopez, Felipe Pereira, Salah<br />

Elfatori, David Simon<br />

University of Ottawa<br />

Purpose: <strong>The</strong> upward migration index (UMI) is a useful<br />

radiographic parameter for assessment of disorders of the<br />

rotator cuff. Utility of the UMI as a prognostic indicator<br />

for outcome following cuff repair has not been previously<br />

studied. <strong>The</strong> objective of this study was to determine if<br />

an association exists between the pre-operative UMI and<br />

the improvement in clinical and quality of life outcome<br />

scores following arthroscopic rotator cuff repair.<br />

Method: Patients with a full thickness tear of the rotator<br />

cuff who underwent an arthroscopic repair of the<br />

cuff were selected for review. Eighty-four patients were<br />

included in the series. Mean patient age was 55 (range 25-<br />

78). <strong>The</strong> UMI was measured by MRI, and patients were<br />

divided into three groups: < 1.25 (GROUP A), 1.25-1.35<br />

(GROUP B) and > 1.35 (GROUP C). Outcome variables<br />

were the non-weighted Constant-Murley score, ASES and<br />

the WORC assessed at baseline, 6 month and 12 months<br />

post-operatively. <strong>The</strong> paired t-test was used to carry out<br />

comparisons in follow up and one-way ANOVA was<br />

used to carry out comparisons between groups.<br />

Results: <strong>The</strong>re were 9 patients in group A, 33 in group<br />

B and 42 in group C. <strong>The</strong> improvement in scores from<br />

baseline to 1 year were as follows: ASES; 21.1 (A),<br />

32.6 (B), and 38.4 (C); Constant 21.4 (A) 19.8 (B), and<br />

24.2 (C) and WORC 31.9 (A), 42.7 (B), and 44 (C).<br />

Statistically significant improvements were observed<br />

in all groups in all outcome measures from baseline to<br />

6 months and from 6 months to 1 year. Although the<br />

differences were not statistically significant (p>0.05),<br />

a trend toward greater improvement in outcomes was<br />

observed with higher upward migration indices.<br />

Conclusion: A lower UMI was associated with less<br />

improvement in functional and quality of life outcomes<br />

following arthroscopic rotator cuff repair, although<br />

these differences were not statistically significant.<br />

Patients with a low UMI demonstrated a significant<br />

improvement in functional and quality of life scores<br />

following surgery. In isolation, a low UMI should not<br />

represent a significant contraindication to treatment by<br />

arthroscopic rotator cuff repair.<br />

68. DOES PHYSICAL EXAMINATION OF THE<br />

SHOULDER PREDICT PATIENT-REPORTED<br />

FUNCTIONAL OUTCOME IN PATIENTS<br />

WITH PREVIOUS PROXIMAL HUMERUS<br />

FRACTURES?<br />

Gerard Slobogean, Akin Famuyide, Vanessa<br />

Noonan, Peter J. O’Brien<br />

UBC<br />

Purpose: To quantify how well the physical examination<br />

of the shoulder predicts patient-reported functional<br />

outcome in a cohort of patients with previous proximal<br />

humerus fractures.<br />

Method: Potential subjects were identified from a recent<br />

study cohort of proximal humerus fracture patients<br />

treated within the past six years. <strong>The</strong> cohort consisted<br />

of all fracture types and treatment modalities. Participants<br />

underwent a focused physical examination of<br />

their injured shoulder containing the components of the<br />

Constant-Murley shoulder score: range of motion for<br />

forward flexion, abduction, internal rotation, external<br />

rotation, and abduction strength measured by an IsoBex<br />

muscle strength analyzer. Participants also completed<br />

the following patient-reported functional outcome questionnaires:<br />

Disabilities of Arm, Shoulder, Hand (DASH),<br />

American Shoulder and Elbow Surgeons Standardized<br />

Shoulder Assessment Form (ASES), Simple Shoulder<br />

Test (SST), and Oxford Shoulder Score (OSS). Forwardand<br />

backward-stepwise linear regression was used to<br />

assess the relationship between the functional outcomes<br />

and the physical exam measurements.<br />

Results: Thirty-one subjects with a mean age of 70 ±<br />

8 years participated. Sixteen patients were previously<br />

treated with ORIF and 15 were treated with sling<br />

immobilization. <strong>The</strong> mean physical examination measures<br />

were: flexion 117° ± 31°, abduction 117° ± 37°,<br />

internal rotation 7° ± 2°, external rotation 7° ± 4°, and<br />

strength 6 ± 5 Newtons. <strong>The</strong> mean functional outcome<br />

scores were: DASH 21 ± 19, ASES 82 ± 17, SST 8 ± 3,<br />

and Oxford 20 ± 8. Using linear regression, adjusted<br />

R-squared statistics suggest components of the physical<br />

exam can explain 38% of the Oxford, 50% of the<br />

DASH, 58% of the SST, and 70% of the ASES variance.<br />

Abduction strength was a significant predictor for all<br />

functional outcomes. Combinations of flexion, abduction,<br />

or internal rotation were also significant predictors<br />

depending on the outcome instrument being modeled.<br />

Conclusion: Physical exam of the shoulder accounts for<br />

differing amounts of patient-reported functional outcome<br />

variance. Abduction strength is the most consistent<br />

predictor of functional outcome within this cohort<br />

of proximal humerus fracture patients.<br />

69. IMPACT OF COMPENSATION CLAIMS<br />

ON SURGICAL OUTCOME IN PATIENTS<br />

WITH ROTATOR CUFF RELATED<br />

PATHOLOGIES<br />

Richard M. Holtby, Helen Razmjou, Iona<br />

MacRitchie<br />

Holland Orthopaedic and Arthritic Centre<br />

Purpose: <strong>The</strong>re is controversial information on recovery<br />

of patients with compensable injuries. <strong>The</strong> purpose of this<br />

matched case-control study was to examine the impact of<br />

an active compensation claim following a work-related<br />

shoulder injury on reporting disability as measured by subjective<br />

and objective outcomes at 1 year post-operatively.<br />

Method: Data of 506 consecutive patients who had undergone<br />

a decompression or rotator cuff repair were reviewed.<br />

One hundred and fourteen patients were on compensation<br />

related to their shoulder problems. Patients were matched<br />

with a historical control group (patients without a compensation<br />

claim) based on age (4 age groups:


258 COA/CORS/CORA<br />

each group, 42 patients (58%) had undergone a fullthickness<br />

repair and 65 (61%) had surgeries related to<br />

impingement syndrome. Paired and independent t-tests<br />

showed that both groups improved significantly regardless<br />

of their claim status. However, the compensation<br />

group had a significantly lower level of improvement<br />

than the non-compensation group. An analysis of covariance<br />

which adjusted for pre-operative differences in<br />

disability scores showed that an active claim was indeed<br />

a strong predictor of follow up scores.<br />

Conclusion: This is the first study that has used a<br />

matched case-control design to control for potential<br />

confounding factors in injured worker population. Our<br />

results indicate that although patients with work-related<br />

injuries demonstrate a lower level of recovery, they still<br />

benefit from surgery.<br />

70. CONTACT PRESSURES AND<br />

GLENOHUMERAL TRANSLATIONS<br />

FOLLOWING SUBACROMIAL<br />

DECOMPRESSION: HOW MUCH IS<br />

ENOUGH?<br />

Patrick Denard, Timothy Bahney, Robert M. Orfaly<br />

OHSU<br />

Purpose: Determine the ideal form of subacromial<br />

decompression.<br />

Method: Six cadaveric shoulders with intact rotator<br />

cuffs (RTC) underwent “smooth & move (SM),” limited<br />

acromioplasty with coracoacromial ligament (CAL) preservation,<br />

and CAL resection. Glenohumeral translation<br />

was measured in four directions utilizing electromagnetic<br />

spatial sensors. Peak RTC pressure was measured during<br />

arm abduction utilizing pressure film sensors.<br />

Results: Anterosuperior translation was unchanged<br />

after SM or acromioplasty, but increased from 2mm at<br />

baseline to 4mm following CAL resection with the arm<br />

at 300 abduction (p=0.03). <strong>The</strong>re were no significant<br />

changes in other directions of translation following any<br />

procedure. In neutral humeral rotation RTC pressure<br />

was unchanged after SM (p=1.00). Pressure decreased<br />

64% after a limited acromioplasty (p=0.04), and 72%<br />

after CAL resection (p=0.03). <strong>The</strong>re was a trend towards<br />

increased abduction at which peak pressure occurred<br />

following CAL resection (760 compared to 620;p=0.11)<br />

In external rotation, RTC pressure decreased 26%<br />

following SM, 52% after limited acromioplasty, and<br />

64% after CAL resection, but values were not statistically<br />

changed (p=0.52, p=0.08, and p=0.06). Similarly,<br />

abduction angle at which peak pressure was reached<br />

increased but was statistically insignificant after SM<br />

(720; p=0.75), limited acromioplasty (750; p=0.11),<br />

and CAL resection (790; p=0.08). In internal rotation,<br />

RTC pressure decreased 32% following the SM, 59%<br />

following the limited acromioplasty, and 58% following<br />

CAL resection, but none reached statistical significance<br />

(p=0.52, p=0.26, p=0.17). Abduction angle of peak<br />

pressure was unchanged after SM (670; p=0.63) and<br />

limited acromioplasty (670; p=0.63), but increased following<br />

CAL resection (620 vs. 790; p=0.04).<br />

Conclusion: A CAL resection leads to increased anterosuperior<br />

instability. “Smooth and move” or acromioplasty<br />

can safely be performed without increasing<br />

translation. Rotator cuff pressure did not significantly<br />

decrease after SM. Rotator cuff pressure was significantly<br />

decreased to a similar degree following a limited<br />

acromioplasty or a CAL resection. A limited acromioplasty<br />

with preservation of the CAL may offer the greatest<br />

decrease in cuff pressures without the undesirable<br />

effect of increased translation. However, statistical<br />

significance was affected by high anatomic variability.<br />

<strong>The</strong>refore, the choice between “smooth & move” and<br />

acromioplasty to decrease contact pressure is likely best<br />

to be individualized based on acromial morphology.<br />

71. THE EFFECT OF POSTERIOR CAPSULAR<br />

TIGHTNESS ON PRESSURE IN THE<br />

SUBACROMIAL SPACE<br />

Peter Lapner, Philippe Poitras, Othman Ramadan,<br />

Stephen Kingwell, Donald Russell*<br />

University of Ottawa; *Carleton University<br />

Purpose: Subacromial impingement syndrome is a painful<br />

condition which occurs during overhead activities as<br />

the rotator cuff is compressed between the greater tuberosity<br />

and the acromion. Unrecognized secondary causes<br />

of impingement syndrome may lead to treatment failure.<br />

Posterior capsular tightness, believed to alter shoulder<br />

joint kinematics, is often cited as a secondary cause but<br />

scientific evidence is lacking. <strong>The</strong> objective of this study<br />

was to evaluate the effect of posterior capsular tightness<br />

on pressure in the subacromial space.<br />

Method: Ten fresh-frozen cadaver shoulder specimens<br />

were mounted on a custom testing apparatus. With the<br />

scapula fixed, the deltoid and cuff muscles were loaded<br />

statically with a constant ratio to elevate the humerus<br />

in the scapular plane under physiologic loading conditions.<br />

For each treatment (intact capsule, 1cm and 2cm<br />

plication), pressure in the subacromial space and glenohumeral<br />

kinematics were recorded during elevation. <strong>The</strong><br />

treatment order was randomly assigned to each specimen.<br />

Peak pressure and translation of the humeral head center<br />

were compared using a repeated measures ANOVA.<br />

Results: Peak subacromial pressures (mean±sd) were similar<br />

between treatment groups: 345±152 kPa, 410±213<br />

kPa and 330±164 kPa for the intact, 1cm and 2cm plication<br />

respectively (p>0.05). No significant differences were<br />

found for superior or antero-posterior translations of the<br />

humeral head at the peak pressure position (p>0.05).<br />

Conclusion: Posterior capsular tightness, as a sole variable,<br />

did not contribute significantly to increased pressure<br />

in the subacromial space or to increased anterior<br />

or superior humeral head translation during abduction.<br />

Clinically, posterior capsular tightness may occur in<br />

association with impingement syndrome but may not<br />

play a significant role in causation.<br />

72. CONTACT AREA, CONTACT PRESSURE<br />

AND LOAD-FAILURE OF 3 ROTATOR CUFF<br />

REPAIR METHOD: A BIOMECHANICAL<br />

ANALYSIS<br />

Frédéric Balg, Josianne Lepine, Nicolas Huppe,<br />

Eve Langelier, Denis Rancourt<br />

Université de Sherbrooke<br />

Purpose: Comparer la technique de réparation de la<br />

coiffe des rotateurs par haubanage tendineux en simple<br />

rangée aux techniques transosseuse et double-rangée par<br />

rapport à la surface et la pression de contact à l’interface<br />

tendon-os, et la force de rupture.<br />

Method: Pour tester la pression et la surface de contact,<br />

les techniques de réparation ont été faite sur 2 spécimens<br />

cadavériques (tête humérale et sus-épineux) chaque. Un<br />

film Prescale pressure-sensing a été interposé entre les<br />

tendons et l’os pendant 2 minutes avec une tension de<br />

120N sur les tendons. Les films ont été numérisés pour<br />

l’analyse avec le logiciel ImageJ. La force de rupture a été<br />

testé sur un modèle Sawbones d’humérus proximal. Des<br />

tendons synthétiques en fibre de nylon et polyesther dans<br />

du silicone ont été créés pour les propriétés d’un tendon<br />

proportionnellement à la rigidité du Sawbones. La force<br />

a été appliquée à 135° jusqu’à rupture sur 2 montages<br />

par technique de réparation. La suture transosseuse<br />

utilisait 2 fils Orthocord dans 2 tunnels transosseux. La<br />

suture double rangé a été faite avec 2 ancres Spiralok<br />

médialement et 2 ancres Versalok latéralement avec des<br />

fils Orthocord. Le haubanage tendineux a été fait avec 2<br />

ancres Panalok RC latéralement dans la zone corticale.<br />

Results: La surface de contact du haubanage de 17mm2<br />

était significativement plus basse que de la suture transosseuse<br />

à 48mm2 (p=0.002) et double-rangée à 86mm2<br />

(p=0.001). La différence entre transosseux et double<br />

rangée était significative (p=0.029). La pression de contact<br />

du haubanage de 0.353MPa était significativement<br />

plus basse que de la suture transosseuse à 0.441MPa<br />

(p=0.002) et double-rangée à 0.567MPa (p=0.003). La<br />

différence entre transosseux et double rangée était significative<br />

(p-0.029). La force de rupture du haubanage<br />

de 106N était significativement plus basse que de la<br />

suture transosseuse à 249N (p=0.03) et double-rangée<br />

à 316N (p=0.04). La différence entre transosseux et<br />

double rangée n’était pas significative.<br />

Conclusion: Le haubanage tendineux ne reproduit pas<br />

l’empreinte anatomique du sus-épineux sur la grande<br />

tubérosité ni une pression de contact adéquate en plus<br />

d’avoir une force de rupture plus faible. Malgré son coût<br />

plus élevé, la suture par double rangée est supérieure à la<br />

technique transosseuse ou simple rangée.<br />

73. PATIENT REPORTED ACTIVITIES<br />

AFTER SHOULDER REPLACEMENT:<br />

HEMIARTHROPLASTY VERSUS TOTAL<br />

SHOULDER ARTHROPLASTY<br />

Peter Zarkadas*, Thomas Throckmorton, Diane<br />

Dahm, John Sperling, Robert Cofield<br />

*Lions Gate Hospital; Mayo Clinic<br />

Purpose: <strong>The</strong> indication to perform a total shoulder<br />

arthroplasty (TSA) versus a hemiarthroplasty is guided<br />

by a patient’s intended level of activity after surgery. It<br />

is unclear what activities patients actually perform following<br />

shoulder replacement, therefore, the purpose of<br />

this study was to compare the self-reported activities of<br />

patients following either a TSA or hemiarthroplasty.<br />

Method: Two groups of 75 patients each, following<br />

TSA or hemiarthroplasty, were matched for a variety<br />

of demographic variables. A mailed activity questionnaire<br />

asked patients to report their level of pain,<br />

motion, strength, and a choice of 70 different activities.<br />

Reported activities were classified as high (i.e. tennis) or<br />

low (i.e. fishing) demand, and categorized as household<br />

(i.e. cooking), yard work (i.e. gardening), sporting (i.e.<br />

golf), or musical (i.e. piano).<br />

Results: Ninety-six (64%) patients completed the survey,<br />

50 in the TSA group (27F:19M, avg. 53.2 yrs), and 46<br />

in the HA group (29F:21M, avg. 53.5 yrs). Pain was<br />

not different between groups (3.6/10 TSA: 3.9/10 HA),<br />

yet a significant difference was reported in forward flexion<br />

(145° TSA: 120° HA, P


COA/CORS/CORA 259<br />

6 weeks, and 3, 6, 12, 18, and 24 months. <strong>The</strong> primary<br />

endpoint was the WOOS score at 2 years.<br />

Results: One hundred and sixty-one patients were<br />

consented and randomized for the study. <strong>The</strong>re were<br />

80 patients in the cemented and 81 patients in the<br />

uncemented group. At baseline, the groups were alike<br />

with regards to demographics and baseline evaluations.<br />

<strong>The</strong> WOOS scores at post-operative intervals of<br />

12, 18 and 24 months showed a significant difference<br />

(p=0.009, 0.001, 0.028 respectively) in favour of the<br />

cemented group. <strong>The</strong> cemented group also had better<br />

strength (3 m p=0.038, 12 m p= 0.036, 18 m p=0.051,<br />

24 m p=0.053) and forward flexion (6m p=0.031, 12<br />

m p=0.04). As expected, the operative time was significantly<br />

less for the uncemented group (C = 2.26h +/-.63;<br />

U = 1.69h +/- 1.9, p= 0.03).<br />

Conclusion: <strong>The</strong>se findings provide the first evidence<br />

that cemented fixation of the humeral head provides<br />

better quality of life, strength and ROM than uncemented<br />

fixation. This was a Tier 1 Project of the JOINTs<br />

Canada group.<br />

75. BONY INCREASED-OFFSET REVERSE<br />

SHOULDER ARTHROPLASTY (BIO RSA):<br />

A BIOLOGLIC SOLUTION TO SCAPULAR<br />

NOTCHING, PROSTHETIC INSTABILITY<br />

AND LIMITED SHOULDER ROTATION<br />

Ryan Bicknell*, Pascal Boileau, Yannick<br />

Roussanne, Nicolas Brassart, Chris Chuinard<br />

*Queen’s University; University of Nice<br />

Purpose: We hypothesized that lateralization of the RSA,<br />

with a glenoid bone graft taken from the osteotomised<br />

humeral head, would prevent those problems without<br />

increasing torque on the glenoid component by keeping<br />

the center of rotation within the glenoid. <strong>The</strong> objectives<br />

of this study were to describe the results of the first 12<br />

patients that underwent a bony increased-offset RSA<br />

(BIO RSA).<br />

Method: Thirty-six shoulders in 34 consecutive patients<br />

with cuff tear arthritis (mean age 72 years, range 52-<br />

86 years) received a BIO RSA, consisted of a RSA<br />

incorporating an autogenous humeral head bone graft<br />

placed beneath the glenoid baseplate. A baseplate with a<br />

lengthened central peg (+25 mm) was inserted in the glenoid<br />

vault, securing the bone graft beneath the baseplate<br />

and screws. All patients underwent clinical and radiographic<br />

(computed tomography) review at a minimum<br />

1-year follow-up.<br />

Results: All patients were satisfied or very satisfied and<br />

all had no or slight pain. Mean active elevation increased<br />

from 72° to 142° (p


260 COA/CORS/CORA<br />

concerns identified by PSI-P. This analysis will guide us<br />

in the development of a new and more comprehensive<br />

instrument for evaluating ankle outcomes following<br />

fusion or replacement.<br />

79. PREFERENCE-BASED QUALITY OF LIFE<br />

IN END-STAGE ANKLE ARTHRITIS, TOTAL<br />

ANKLE ARTHROPLASTY, AND ANKLE<br />

ARTHODESIS<br />

Gerard Slobogean, Alastair S.E. Younger, Carlo<br />

A. Marra, Kevin J. Wing, Murray J. Penner, Mark<br />

Glazebrook*<br />

UBC; *Dalhousie University; On Behalf of the<br />

Canadian Orthopaedic Foot and Ankle Society<br />

Purpose: To describe the pre- and one-year post-operative<br />

preference-based, health related quality of life (health<br />

state values) among a cohort of subjects with end-stage<br />

ankle arthritis treated with total ankle arthroplasty or<br />

ankle arthrodesis. This short-term study is not intended<br />

to compare the efficacy of arthoplasty and arthrodesis.<br />

Method: <strong>The</strong> Short-Form 36 (SF-36) was prospectively<br />

completed by subjects enrolled in the Canadian Orthopaedic<br />

Foot and Ankle Society Multicentered Ankle<br />

Arthritis Outcome Study between 2003 and 2005. Preference-based<br />

quality of life was assessed pre-operatively<br />

and at one-year post-procedure using health state values<br />

(HSVs) derived from the SF-36 transformation described<br />

by Brazier (SF-6D). <strong>The</strong> SF-6D scores are anchored at<br />

1.0 (full health) and at 0 (death). Basic patient demographic<br />

and treatment information was also collected.<br />

<strong>The</strong> decision to perform arthroplasty or arthrodesis was<br />

made by the attending surgeon.<br />

Results: Two hundred four of the 214 eligible subjects<br />

had complete preoperative SF-36 data to allow transformation<br />

to SF-6D values. One-year follow-up was<br />

available for 114 of the participants. <strong>The</strong> mean age at<br />

surgery of the included subjects was 58.9 +/- 13.3 years.<br />

Of the patients with one-year follow-up, 56% were male<br />

and 59% had received total ankle arthroplasty. <strong>The</strong>se<br />

demographics did not differ from the original preoperative<br />

cohort. <strong>The</strong> mean SF-6D score among all subjects<br />

with end-stage ankle arthrosis was 0.66 (95% CI 0.65<br />

– 0.68). At one-year, the mean HSVs of the total ankle<br />

arthroplasty and ankle arthrodesis groups were 0.73<br />

(95% CI 0.71 – 0.76) and 0.73 (95% CI 0.70 – 0.75),<br />

respectively. <strong>The</strong> reported pre-operative scores describe<br />

health states below normative data for the US population<br />

(0.76 +/- 0.01 for females, ages 55-64).<br />

Conclusion: <strong>The</strong>se are the first available HSVs for a<br />

cohort of patients with end-stage ankle arthritis treated<br />

with total ankle arthroplasty or ankle arthrodesis. <strong>The</strong>se<br />

data demonstrate an improvement in preference-based<br />

quality of life following ankle arthroplasty or arthrodesis.<br />

At one-year follow-up, patient reported HSVs<br />

approach age-matched US norms.<br />

80. RSA RESULTS FOR AN UN-CEMENTED<br />

MOBILE-BEARING TOTAL ANKLE<br />

ARTHROPLASTY SYSTEM<br />

Andrea Veljkovic, Jason Fong, Allan Henigar,<br />

David R. Wilson, Michael J. Dunbar, Mark<br />

Glazebrook<br />

Dalhousie University<br />

Purpose: Radiostereometric Analysis (RSA) is used to<br />

measure migration and inducible displacement (ID) of<br />

orthopedic implant devices to allow early prediction<br />

of implant failure (eg. Aseptic loosening). Total Ankle<br />

Arthroplasty (TAA) is used for the treatment of endstage<br />

ankle arthritis. First generation TAA implant have<br />

meet with widespread failures while some second generation<br />

TAA implants are showing improved results. In<br />

this study RSA is used to evaluate the biomechanical<br />

properties of a new third generation TAA implant in an<br />

attempt to set a standard for the biomechanical evaluation<br />

on TAA implants in-vivo.<br />

Method: Patients undergoing TAA were enrolled consecutively<br />

(n=12; 7 males; mean age=59 years; mean<br />

BMI=29) and had 8 markers (0.08mm) inserted in both<br />

the tibia and talus during surgery. Standing, weightbearing<br />

RSA exams were performed at 3 and 6 months<br />

and compared to concurrent supine exams to determine<br />

component ID.<br />

Results: For tibial components: at six months the components<br />

had translated posteriorly (0.3mm±0.5) and<br />

proximally (0.5mm±0.2), tilted into varus (0.5°±1.3),<br />

and tilted posteriorly (0.4°±0.8). <strong>The</strong> magnitudes of<br />

ID for the tibial components were moderate (mean <<br />

0.2mm and 0.5°, standard deviation < 0.3mm and 2.2°<br />

in each direction). For talar components: at six months<br />

the components had translated distally (0.28mm±0.35),<br />

rotated internally (0.21°±1.32) and tilted posteriorly<br />

(0.15°±0.90). <strong>The</strong>re was varus/valgus tilt measured in<br />

the talar components but there was no consistent direction<br />

of migration (0.03°±1.4). At six months the magnitudes<br />

of ID for the talar components were small (mean<br />

< 0.1mm and 0.25°, standard deviation < 0.2mm and<br />

0.6° in each direction).<br />

Conclusion: An RSA methodology has been established<br />

to predict stability.<br />

81. ACHILLES INSERTIONAL<br />

TENDINOPATHY TREATED BY A<br />

POSTERIOR MIDLINE APPROACH: A SAFE<br />

PROCEDURE<br />

Stéphane Leduc*, Michael P. Clare † , Scott<br />

Swanson, Arthur K. Walling<br />

*Université de Montréal; † Florida Orthopaedic<br />

Institute; Florida Orthopaedic Institute<br />

Purpose: Insertional calcific Achilles tendinosis is a painful,<br />

frequently disabling, condition. <strong>The</strong> longitudinal<br />

and radial alignment of the angiosomes of the posterior<br />

region of the leg makes a straight posterior midline<br />

approach logical. <strong>The</strong> safety of the posterior midline<br />

approach and the outcome of a central tendon splitting<br />

approach associated with a Strayer procedure to treat<br />

this condition was evaluated.<br />

Method: A retrospective review of a consecutive cohort<br />

of a single surgeon was performed. All patients had<br />

failed conservative treatment and all patients were primary<br />

cases. Forty-seven patients (48 heels) were treated<br />

over a 11-year period for chronic insertional Achilles<br />

tendinosis. All patients underwent a midline posterior<br />

splitting approach, debridment of the bursae, resection<br />

of the haglund deformity, partial Achilles detachment,<br />

debridement, reinsertion with bone anchor associated<br />

with a proximal gatrocnemius recession (strayer procedure)<br />

through a second midline incision. <strong>The</strong> average<br />

age was 59 years old (39-75), co-morbidities included<br />

four smokers and one diabetic patient. <strong>The</strong> average followup<br />

was 54 months (15-144). All patients answered<br />

pre-op and latest follow up AOFAS questionnaire, satisfaction<br />

rate and complications were reviewed.<br />

Results: Satisfaction rate was 100%. AOFAS score<br />

improved significantly from 59 (36-80) preop to 97 (90-<br />

100) at the latest follow-up. Complications included<br />

one superficial infection and one sural nerve paresthesia.<br />

<strong>The</strong>re were no major complications.<br />

Conclusion: Achilles insertional tendinopathy treated by<br />

a posterior midline approach is a safe and reliable procedure.<br />

<strong>The</strong> procedure was associated with high patient<br />

satisfaction rate and excellent outcome.<br />

82. OPENING-WEDGE VERSUS<br />

PROXIMAL CHEVRON OSTEOTOMY<br />

FOR HALLUX VALGUS WITH INCREASED<br />

INTERMETATARSAL ANGLE<br />

Peter Copithorne, Timothy R. Daniels*, Mark<br />

Glazebrook<br />

Dalhousie University; *University of Toronto;<br />

Purpose: For patients with moderate to severe hallux<br />

valgus with increased intermetatarsal angle, correction<br />

with a proximal first metatarsal osteotomy is indicated.<br />

<strong>The</strong> purpose of this study is to compare the opening-wedge<br />

osteotomy of the proximal first metatarsal the proximal<br />

chevron osteotomy in the treatment of moderate to severe<br />

hallux valgus with increased intermetatarsal angle.<br />

Method: This prospective, randomized, multi-centered<br />

study is being conducted at three centers in Canada.<br />

Approximately 75 adult patients with hallux valgus<br />

are being randomized to either the proximal metatarsal<br />

opening-wedge osteotomy with plate fixation or the<br />

proximal chevron osteotomy. Patient functional scores<br />

using the SF-36, American Orthopaedic Foot and Ankle<br />

Society (AOFAS) forefoot metatarsophalangeal interphalangeal<br />

score and Visual Analogue Scale (VAS) for<br />

pain, activity & patient satisfaction, are assessed prior<br />

to surgery and 3, 6, 12 and 24 months. Surgeon preference<br />

is being evaluated based on a questionnaire and<br />

actual surgical times. Radiologic measurements (intermetatarsal<br />

angle correction, hallux valgus angle correction,<br />

sagital talus-first metatarsal (Meary’s) angle,<br />

metatarsal length and union) will also be assessed.<br />

Results: Preliminary results demonstrate that patients<br />

who undergo the opening-wedge osteotomy have less<br />

pain at 3 months (ave.VAS pain reduction 2.9, SE±1.0)<br />

than those with the chevron (ave. VAS pain reduction<br />

2.4, SE±1.2). VAS for activity demonstrates greater<br />

improvements with the chevron osteotomy at 3 months<br />

(0.8, SE±0.8) versus the opening-wedge (0.1, SE±1.0).<br />

AOFAS scores improve on average 18.3 (SE±8.6) with<br />

the opening wedge compared to 20.8 (SE±7.4) with<br />

the chevron at 3 months. Average hallux valgus angle<br />

correction for opening-wedge and chevron osteotomies<br />

are 11.0 degrees (SE±2.5) and 19.0 degrees (SE±3.1)<br />

respectfully. Average intermetatarsal angle correction<br />

for opening-wedge and chevron osteotomies are 6.5<br />

(SE±1.3) and 4.3 (SE±1.7) respectfully. Both procedures<br />

are effective at maintaining metatarsal length. <strong>The</strong><br />

opening-wedge osteotomy takes on average 60.9 minutes<br />

(SE±3.9) to complete compared to 69.1 minutes<br />

(SE±5.1) for the chevron ostetotomy. Surgeon response<br />

to the new opening-wedge osteotomy is favorable.<br />

Conclusion: Opening-wedge and proximal chevron<br />

osteotomies have comparable pain, function and radiographic<br />

outcomes. Opening wedge osteotomy is technically<br />

less demanding and requires less surgical time.<br />

83. RELATION TIBIO-PERONIERE DISTALE<br />

SUR VUE FLUOROSCOPIQUE LATERALE<br />

Benoit Benoit, Stéphanie Grenier, G. Yves<br />

Laflamme, Dominique Rouleau, Stéphane Leduc<br />

Université de Montreal<br />

Purpose: Lors de la réduction chirurgicale des fractures<br />

de la cheville avec instabilité syndesmotique, le<br />

chirurgien se fie généralement sur les vues de mortaise<br />

et antéro-postérieure. Toutefois, une subluxation ou<br />

luxation antérieure du péroné par rapport au tibia distal<br />

peu survenir et passer inaperçu (trois exemples cliniques<br />

prouvés par CT Scan post-opératoire), spécialement lors<br />

de la pose de vis syndesmotique(s). La présente étude a<br />

pour but d’établir la relation radiologique précise sur<br />

une vue latérale fluoroscopique entre les tibia et péroné<br />

distaux qui permettra au chirurgien de confirmer en<br />

peropératoire que l’articulation tibio-péronière distale<br />

est bel et bien réduite.<br />

Method: Les chevilles normales de trente volontaires<br />

sans antécédent de traumatisme ou de maladie de<br />

la cheville ont été imagées sous une vue latérale fluoroscopique<br />

parfaite, avec un Mini C-Arm. Les images<br />

ont été analysées et comparées entre elle afin d’établir<br />

une relation radiologique fiable et reproductible entre le<br />

tibia et le péroné distaux.<br />

Results: Dans les trente cas, il y avait intersection du<br />

milieu de la cicatrice physaire et du cortex antérieur du<br />

péroné. Cette relation a été trouvée statistiquement significative.<br />

Conclusion: La réduction chirurgicale parfaite de<br />

l’articulation tibio-péronière distale peut être confirmée<br />

avec une vue latérale fluoroscopique de la cheville. Le<br />

cortex antérieur du péroné doit toucher le milieu de la<br />

cicatrice physaire.<br />

J BONE JOINT SURG [BR] 2011; 93-B:SUPP III


COA/CORS/CORA 261<br />

84. OPERATIVE VERSUS NON-OPERATIVE<br />

TREATMENT OF ACHILLES TENDON<br />

RUPTURES: A RANDOMISED CONTROLLED<br />

TRIAL<br />

Kevin Willits, Nicholas G.H. Mohtadi*, Crystal<br />

Kean, Dianne Bryant, Annunziato Amendola †<br />

<strong>The</strong> University of Western Ontario; *University of<br />

Calgary; † University of Iowa<br />

Purpose: <strong>The</strong> purpose of this randomised controlled<br />

trial was to compare outcomes of operative and nonoperative<br />

management of Achilles tendon ruptures.<br />

Method: Patients with acute complete Achilles tendon<br />

ruptures were randomised to receive open suture repair<br />

followed by graduated rehabilitation or graduated rehabilitation<br />

alone. <strong>The</strong> primary outcome measure was rerupture<br />

rate. Assessments at three and six months, and<br />

one and two years included a modified Leppelhati score<br />

(no strength data), range of motion, calf circumference,<br />

and isokinetic strength at one and two years. We report<br />

the two year findings.<br />

Results: Two centres randomized 145 patients (118 males<br />

and 27 females), mean age 40.9±8.8 years (22.5 – 67.2)<br />

to operative (n=73) and non-operative (n=72) treatment.<br />

Fourteen were lost to follow-up. Re-rupture occurred in<br />

three patients in both groups. <strong>The</strong> mean modified Leppelhati<br />

score (out of 85) was 78.2±7.7 in the operative<br />

group and 79.7±7.0 in the non-operative group, which<br />

was not significant (-1.5 95%CI -6.4 to 3.5, p=0.55).<br />

Mean side-to-side difference in plantar flexion and calfcircumference<br />

in the operative group was -2.0±3.2° and -<br />

1.4±1.2cm, and in the non-operative group -0.9±3.0°and<br />

-1.6±1.8cm respectively. Mean isokinetic plantar flexion<br />

strength was 62.4±24.2 for the operative and 56.7±19.3<br />

for the non-operative group, which was not significant<br />

(5.7, 95%CI -3.1 to 14.5, p=0.20). <strong>The</strong>re were a greater<br />

number of serious adverse events in the operative group,<br />

including pulmonary embolus in one patient, deep vein<br />

thrombosis in one and deep infections requiring irrigation<br />

and debridement in three.<br />

Conclusion: This study suggests that non-operative<br />

management of Achilles tendon ruptures utilizing an<br />

accelerated rehabilitation programme may produce<br />

comparable results with fewer adverse events.<br />

85. TO ASSESS MODERATE-TERM<br />

OUTCOMES OF SILASTIC JOINT<br />

REPLACEMENTS OF THE FIRST<br />

METATARSOPHALANGEAL JOINT<br />

Vishal Upadhyay*, Ravi H. Mahajan, Ajay Sahu,<br />

Usman Butt, Amir Khan, Rakesh B. Dalal<br />

*Poole General Hospital, UK; Stepping Hill<br />

Hospital<br />

Purpose: To assess moderate-term outcomes of silastic<br />

joint replacements of the first metatarsophalangeal<br />

joint.<br />

Method: <strong>The</strong> thirty-two patients (37 feet) that had silastic<br />

implants inserted were reviewed at an average of 2 years<br />

and 4 months (ranging 7 months to 5 years and 4 months).<br />

<strong>The</strong> mean patient age was 63 years. <strong>The</strong>se patients<br />

answered a subjective questionnaire, had their feet examined<br />

clinically and radiographically and a pre-operative<br />

and post-operative AOFAS score was calculated for each.<br />

Results: <strong>The</strong> follow-up assessment revealed that every<br />

patient described that their pain had decreased after surgery<br />

and 17 feet (46%) were completely pain free. <strong>The</strong>re<br />

was a significant improvement in patients’ subjective pain<br />

scores after surgery (t value =


262 COA/CORS/CORA<br />

89. TOTAL KNEE ARTHROPLASTY: DOES A<br />

DIFFERENCE BETWEEN GENDERS EXIST?<br />

Camilo Resterpo, Javad Parvizi, Peter F. Sharkey,<br />

Aidin Eslam Pour, Craig T. Haytmanek, Nathan<br />

Roberts, Richard H. Rothman<br />

Rothman Institute<br />

Purpose: Recently an orthopedic manufacturer has introduced<br />

a gender specific knee design implying that there is<br />

a substantial anatomical difference between the genders.<br />

If such concept is true then TKA prosthesis implanted<br />

in the female population over the last decades, by definition,<br />

must have suboptimal outcome when compared<br />

to the male patients. <strong>The</strong> purpose of this study was to<br />

examine the functional outcome, the incidence of complications,<br />

and the need for revision between the two<br />

genders receiving the same knee design.<br />

Method: <strong>The</strong> study selected a matched group of 150 men<br />

and 250 women undergoing TKA at our instituion. <strong>The</strong><br />

patients were matched for age, BMI, pre-op diagnosis,<br />

comorbidities, race, mode of fixation, and the type of<br />

implant. Other demographic, surgical, and medical factors<br />

between the two genders were similar. Both pre-op<br />

and post-op functional scores were compared between<br />

the two groups. Pre-op and post-op radiographic images<br />

were assessed for implant fit.<br />

Results: <strong>The</strong>re was a significant improvement in functional<br />

outcome as measured by Knee Society score,<br />

WOMAC, and SF-36 for all patients. <strong>The</strong> improvement<br />

in functional outcome was not different between<br />

the two groups. <strong>The</strong> incidence of complications, reoperations,<br />

and need for revision between the two genders<br />

was also not significantly different.<br />

Conclusion: Total knee arthroplasty continues to be<br />

an effective surgical procedure. Both genders appear<br />

to enjoy relief of pain and improvement of function<br />

equally. Based on this retrospective study the use of nongender<br />

specific knee prosthesis did not seem to result in<br />

suboptimal outcome in female patients. <strong>The</strong>re appears<br />

to be little merit in introduction of gender specific knee<br />

designs when previous non-gender specific prosthesis<br />

appeared to function well in both genders.<br />

90. THE RELATIONSHIP BETWEEN<br />

BODY HABITUS AND LEPTIN IN A KNEE<br />

OSTEOARTHRITIS POPULATION<br />

Rajiv Gandhi, Mark Takahashi, Khalid Syed, J<br />

Roderick Davey, Nizar N. Mahomed<br />

University of Toronto<br />

Purpose: Synovial fluid (SF) leptin has been shown to<br />

have an association with cartilage degeneration. Our<br />

objective was to examine the relationship between different<br />

measures of body habitus and SF leptin levels in<br />

an end stage knee osteoarthritis (OA) population.<br />

Method: Sixty consecutive patients with knee OA were<br />

surveyed prior to surgery for demographic data. Body<br />

habitus was assessed with the body mass index (BMI),<br />

waist circumference (WC) and waist-hip ratio (WHR).<br />

SF and serum samples were analyzed for leptin and<br />

adiponectin using specific ELISA. Non-parametric correlations<br />

and linear regression modeling was used to<br />

identify the relationship between the measures of body<br />

habitus and SF leptin levels.<br />

Results: Females had greater levels of leptin than males<br />

in both the serum and SF. Significant correlations were<br />

found between SF leptin levels and BMI and WC (R2<br />

0.44 and 0.38 respectively, p< 0.05). Regression modeling<br />

showed that female gender and WC were independent<br />

predictors of a greater SF leptin level independent<br />

of age, BMI, and presence of diabetes.(p


COA/CORS/CORA 263<br />

compartmental knee arthroplasty (UKA) and total knee<br />

arthroplasty (TKA) are all viable options. Gait analysis<br />

is one tool available to clinically assess knee kinematics,<br />

and may prove to be a good way of predicting functional<br />

outcomes of these different surgical procedures.<br />

<strong>The</strong> purpose of this study was to compare the knee<br />

kinematics, function, and quality of life of patients that<br />

underwent either a medial opening wedge HTO, UKA,<br />

or TKA for primary medial compartment OA.<br />

Method: A matched prospective cohort study of patients<br />

between the ages of 45 and 65 who had undergone an<br />

HTO, UKA, or TKA for primary medial compartment<br />

knee OA was undertaken over a 3-year period. Primary<br />

outcome measures were gait variables, namely knee<br />

adduction moments, as measured through gait analysis.<br />

Secondary measures included quality of life (WOMAC),<br />

functional performance tests (six minute walk and timedup-and-go),<br />

self-reported functional ability (LEFS), and<br />

general health (SF-36). Gait and functional performance<br />

tests were evaluated preoperatively and at 6, 12, and<br />

24 months postoperatively. Self-reported quality of life,<br />

function and general health were assessed preoperatively<br />

and at 3, 6, 12, and 24 months post-operatively.<br />

Results: Twenty HTOs, 19 medial UKAs, and 17 TKAs<br />

were matched for Kellgren-Lawrence grade of medial<br />

OA, age at surgery, and body mass index. Significant<br />

differences were observed between the three groups in<br />

step length and peak adduction moments at 24 months.<br />

Significant differences were observed in preoperative<br />

WOMAC pain and function scores, KOOS pain scores,<br />

and LEFS, but no significantly different outcome measures<br />

were observed postoperatively. Lateral Blackburne-Peel<br />

and modified Insall-Salvati ratios were the<br />

only significant radiographic differences observed<br />

between groups at 24 months.<br />

Conclusion: To our knowledge, no gait analysis study<br />

exists comparing the medial opening wedge HTO to<br />

UKA or TKA. <strong>The</strong> results of this study suggest that most<br />

gait variables except step length and knee adduction<br />

moments are similar between groups. Moreover, except<br />

for patellar height, there were no major functional or<br />

radiographic differences between these groups.<br />

95. OUTPATIENT UNICOMPARTMENTAL<br />

KNEE ARTHROPLASTY WITH FEMORAL<br />

NERVE BLOCK<br />

Geoffrey Dervin, Holly Evans, Susan Madden,<br />

Peter R. Thurston<br />

University of Ottawa<br />

Purpose: Unicompartmental replacement for medial<br />

compartment arthrosis of the knee has become popular<br />

with eligible patients because of the shortened recovery<br />

time, decreased tissue damage and easier future revision.<br />

Contemporary multimodal anesthesia has added<br />

the potential to safely perform this as outpatient surgery<br />

reducing inpatient bed burden. We describe our initial<br />

pilot experience with this approach.<br />

Method: <strong>The</strong> first 25 patients who fulfilled the criteria<br />

developed underwent same day surgery for unicompartmental<br />

arthroplasty for medial (19) or lateral (3) compartment<br />

replacement with either the Oxford knee (20)<br />

or the Uniglide (2). All patients were treated with an<br />

indwelling femoral nerve catheter supplied by Ropivacaine<br />

through a constant release pump (Stryker) which<br />

was discontinued at 48 hours. Home care support was<br />

made available in first 72 hours by way of RN and<br />

physiotherapy visits and mandatory use of walker or<br />

crutches for the first 48 hours.<br />

Results: Patients in this cohort were universally very satisfied<br />

with the model of post-op care as described and<br />

particularly pleased to avoid a hospital stay. Eighty percent<br />

of those who were offered this model chose it. <strong>The</strong><br />

use of narcotic oral medication was consistently about<br />

50% less than that observed to similar inpatients treated<br />

without catheter, and eight patients had complete opioid<br />

sparing experience. <strong>The</strong>re were no complications related<br />

to the catheter, in particular serious falls or longer term<br />

neurologic sequelae. <strong>The</strong> clinical results were very good<br />

and equal to those who were in patients.<br />

Conclusion: Outpatient unicompartmental replacement<br />

can be performed safely recognizing the decreased surgical<br />

trauma and pain stimuli associated with UKR and a<br />

relatively younger and healthier cohort screened for this<br />

alternative. <strong>The</strong>se patients are amongst the most satisfied<br />

with their perioperative course and all would do the<br />

same again if given the chance. Other models of analgesia<br />

could be considered, though the catheter does seem<br />

to have a large opioid sparing effect that likely contributed<br />

to patient well being and satisfaction.<br />

96. MINIMAL INCISION SURGERY AS A RISK<br />

FACTOR FOR EARLY FAILURE OF TOTAL<br />

KNEE ARTHROPLASTY<br />

Robert L. Barrack*, R. Stephen J. Burnett † , C.<br />

Lowry Barnes‡, Derk Miller*, John C. Clohisy*,<br />

William J. Maloney §<br />

*Washington University Department of<br />

Orthopaedic Surgery; † Division of Orthopaedic<br />

Surgery, Victoria BC Canada; ‡ University of<br />

Arkansas Department of Orthopaedic Surgery;<br />

§<br />

Stanford University Department of Orthopaedic<br />

Surgery<br />

Purpose: A study was undertaken to determine the current<br />

prevalence of revisions of total knee arthroplasty<br />

(TKA) following minimal incision surgery (MIS) and to<br />

compare revisions of MIS TKA procedures to revisions of<br />

TKA performed following a standard surgical approach.<br />

Method: A consecutive series of revision TKA performed<br />

at three centers by five surgeons over a three<br />

year time period was reviewed. Revisions performed<br />

for infection and re-revisions were excluded. Review of<br />

clinical and radiographic data determined incision type,<br />

gender, age, time to revision, and primary diagnosis at<br />

time of revision.<br />

Results: Two hundred and thirty-seven first time revision<br />

TKAs were performed of which 44 (18.6%) had<br />

been a MIS primary TKA and 193 (81.4%) had been a<br />

standard primary TKA. Patients with MIS were younger<br />

(62.1 years versus 66.2 years, p=.02). <strong>The</strong>re was a trend<br />

towards a higher percentage of females in the MIS<br />

group (75% versus 63%), although this difference was<br />

not significant (p=0.12). Most striking was the difference<br />

in time to revision which was significantly shorter<br />

for the MIS group (14.8 months versus 80 months,<br />

p


264 COA/CORS/CORA<br />

egories were reviewed for each patient. Sacral fractures<br />

were graded based on severity. <strong>The</strong> age, ISS, and six categories<br />

of AIS were recorded for each patient. A statistical<br />

analysis was performed to test the associations between<br />

fracture characteristics and injury severity.<br />

Results: All patients but three had one or more rami<br />

fractures, and all but two had a sacral fracture. Of the<br />

98 anterior sacral injuries, there were nine (9.2%) buckles,<br />

39 (39.8%) simple fractures, and 50 (51.0%) comminuted<br />

fractures. Of these 98 anterior sacral injuries,<br />

47 (48.0%) were complete, passing through the sacrum<br />

and exiting the posterior cortex. Increasing severity of<br />

anterior sacrum fracture was associated with the presence<br />

of a complete sacral fracture (p value 3mm gap). Post-operative<br />

measurement of step and gap by CT scan were 1.2 mm<br />

(SD) and 2.3 mm (SD) respectively. Using CT scans, eight<br />

patients were found to have either an inadequate reduction,<br />

intra-articular hardware or retained fragments.<br />

Computed tomography demonstrated 2 times more step<br />

and gap compared to plain radiographs.<br />

Conclusion: Post-operative CT was found to be more<br />

sensitive than plain radiographs to assess the quality of<br />

acetabular fracture reduction. Plain radiographs detected<br />

only 1 out of 8 cases where further operative intervention<br />

may have been beneficial. Given the consequences<br />

of missing an unacceptable reduction, intra-articular<br />

hardware, or retained intra-articular fragments, it is<br />

recommended that all fractures should be assessed postoperatively<br />

with CT unless the patient is not a candidate<br />

for further surgery for reasons independent of reduction<br />

quality. <strong>The</strong> benefits of post-operative CT imaging in<br />

acetabular fractures likely outweigh the cost and radiation<br />

exposure associated with its use.<br />

102. THE EFFECT OF FEMORAL NECK<br />

CUT, CABLE TENSION AND MUSCLE<br />

FORCES ON THE STABILITY OF GREATER<br />

TROCHANTER REATTACHMENT<br />

Kajsa Duke, G. Yves Laflamme*, Yvan Petit<br />

École de Technologie Superieure; *Hôpital du<br />

Sacré-Coeur de Montreal;<br />

Purpose: Greater trochanter reattachment is frequently<br />

accomplished using cable grip type systems. <strong>The</strong>re is a<br />

relatively high failure rate for these systems, the mechanisms<br />

of which are unclear. One possible source of<br />

instability could be femoral neck cut location. Another<br />

concern is the effect of variability in cable tension. <strong>The</strong><br />

objective is to create a femur implant model which<br />

allows for variation in cable tension, common muscle<br />

forces and the placement of the femoral neck cut in<br />

order to analyse trochanter fragment fixation.<br />

Method: A finite element model (FEM) of a femur with<br />

simulated greater trochanter osteotomy (30º) was combined<br />

with the femoral component of a hip prosthesis<br />

and a greater trochanter reattachment system with 4<br />

cables (Cable-Ready®, Zimmer). A total of 18 simulations<br />

were modeled in a full factorial design using three<br />

independent variables; cable tightening (178N, 356<br />

N and 534 N), muscle forces (rest, walking and stair<br />

climbing) and femoral neck cut (10 mm and 15 mm<br />

above the lesser trochanter). Displacement of the fragment,<br />

in terms of both gap and shear components, as<br />

well, stress in the bone were investigated.<br />

Results: <strong>The</strong> location of the femoral neck cut reduced<br />

contact surface area by 20% and had the largest influence<br />

on displacement (0.24 mm). Pivoting of the fragment<br />

was observed with a maximum gap (0.38 mm) and<br />

maximum total displacement (0.41 mm) at the bottom of<br />

the fragment. This was observed during stair climbing,<br />

while the cables were tightened to 177.9 N and with the<br />

femoral neck cut at 10 mm. Increased tightening of the<br />

cables provided no significant reduction in fragment displacement.<br />

However, higher cable tension significantly<br />

increased the stress in the bone (8 MPa and 26 MPa for<br />

cable tension of 178 N and 534 N respectively).<br />

Conclusion: Placement of the femoral neck cut closer<br />

to the lesser trochanter significantly increased fragment<br />

displacement. Preservation of the contact surface area is<br />

recommended. Excessive cable tightening did not reduce<br />

fragment movement and only exacerbated bone stress.<br />

Caution must be used to not over tighten the cables.<br />

This model can be used to test and compare the performance<br />

of new implant designs.<br />

103. PROGNOSTIC BASELINE FACTORS FOR<br />

PREDICTING RE-OPERATIONS IN PATIENTS<br />

WITH TIBIAL SHAFT FRACTURES<br />

Sprint Investigators*, Emil H. Schemitsch<br />

*McMaster University; St. Michael’s Hospital/<br />

University of Toronto<br />

Purpose: Accurate prediction of re-operation following<br />

tibial nailing may facilitate optimal patient care. We<br />

recently completed the SPRINT trial, a large, multicentre<br />

trial of reamed versus non-reamed intramedullary<br />

nails in 1226 patients with tibial shaft fractures.<br />

Using the SPRINT data, we conducted an investigation<br />

of baseline and surgical patient characteristics to<br />

determine if they are associated with increased risk of<br />

re-operation within one year.<br />

Method: Using multivariable logistic regression analysis,<br />

we investigated 15 characteristics for association with<br />

increased risk of re-operations. Because the primary<br />

SPRINT analysis found that reamed nailing reduced<br />

events in patients with closed but not open fractures, we<br />

considered both open and closed as well as treatment<br />

status in our model.<br />

Results: We found an increased risk of re-operation in<br />

patients with a high energy mechanism of injury (odds<br />

ratio, OR=1.57, 95% CI 1.05 to 2.35), stainless steel<br />

versus titanium nail (OR=1.52, 95% CI 1.10 to 2.13),<br />

fracture gap (OR=2.40, 95% CI 1.47 to 3.94) and<br />

post-operative weightbearing (OR=1.63, 95% 1.003 to<br />

2.64). Open fractures increased the risk of re-operation<br />

in patients who received a reamed nail (OR=3.26, 95%<br />

CI 2.01 to 5.28) but not in patients who received a nonreamed<br />

nail (OR=1.50, 95% CI 0.92 to 2.47). Patients<br />

with open fractures who had either wound management<br />

without any additional procedures, or delayed primary<br />

closure, had a decreased risk of re-operation when<br />

compared to patients who required subsequent reconstruction<br />

(respectively, OR=0.18, 95% CI 0.09 to 0.35;<br />

OR=0.29 95% CI 0.14 to 0.62).<br />

Conclusion: To ensure optimal patient care surgeons<br />

should consider the characteristics identified in our<br />

analysis to reduce risk of re-operation.<br />

104. REVISION OF PROVISIONAL<br />

STABILIZATION IN PILON FRACTURES<br />

REFERRED FROM OUTSIDE INSTITUTIONS<br />

David Barei, Michael Gardner, Sean Nork,<br />

Stephen Benirschke<br />

Harborview Medical Center<br />

Purpose: Pilon fractures demonstrate complex osseous<br />

and soft tissue injury. Protocols involving immediate<br />

tibial reduction and external fixation, with or without<br />

J BONE JOINT SURG [BR] 2011; 93-B:SUPP III


COA/CORS/CORA 265<br />

fibular fixation, then delayed definitive fixation result<br />

in decreased complications. Our purpose was to evaluate<br />

the treatment course of pilon fractures provisionally<br />

stabilised at outside institutions and subsequently transferred,<br />

focusing on the incidence and reasons for revision<br />

procedures, and subsequent complication rates.<br />

Method: An institutional trauma database was retrospectively<br />

reviewed, demonstrating 668 pilon fractures<br />

treated at our institution between 2000-2007. Of these,<br />

39 patients with 42 fractures had a temporising surgical<br />

procedure prior to referral. Demographics, injury characteristics,<br />

reason for revision, and subsequent complications<br />

were determined. Clinical follow-up averaged<br />

60 weeks (range, 1 to 281).<br />

Results: Mean age was 41 years (range, 18-78). Twentytwo<br />

fractures (52%) were open; 38 (90%) demonstrated<br />

a fractured fibula. Referral occurred an average of 5.8<br />

days (range, 1-20) after initial stabilization. Pre-transfer<br />

fixation was revised in 40 fractures (95%). Reasons<br />

for revision included tibial malreduction (33 fractures,<br />

83%), fibular malreduction (4 fractures, 10%), pins in<br />

the proposed incision (5 fractures, 13%), or loose pins<br />

(3 fractures, 8%). Of the 34 fractures with distal pins, 24<br />

(71%) required revision for pin malposition, loosening,<br />

drainage, talar placement, or extraosseous placement.<br />

Late complications occurred in 14 fractures (33%),<br />

including deep infection in 10 (24%), and non-union in<br />

3 (7%). Twenty-three patients (55%) required additional<br />

procedures following definitive fixation, including 9 soft<br />

tissue coverage procedures and 3 amputations.<br />

Conclusion: <strong>The</strong> majority of patients with pilon fractures<br />

treated with provisional stabilisation followed by referral<br />

to our institution required revision prior to definitive<br />

fixation. This resulted in many avoidable additional procedures,<br />

and a higher complication rate than recent contemporary<br />

controls. <strong>The</strong> authors recommend that, when<br />

possible, the initial and definitive management of these<br />

injuries be performed at the accepting institution.<br />

105. ORIF OF HIGH-ENERGY PILON<br />

FRACTURES: VIOLATING THE 7-CM SKIN<br />

BRIDGE RULE<br />

Geoffrey Wilkin*, Steven Papp, Wade Gofton,<br />

Allan Liew<br />

*Queen’s University; University of Ottawa<br />

Purpose: <strong>The</strong> purpose of this study was to review our<br />

results in patients with pilon fractures treated with<br />

ORIF in which surgical planning involved multiple skin<br />

incisions, ensuring that the distance incisions overlapped<br />

was less than the distance between them. We hypothesized<br />

that soft-tissue complications would be minimal<br />

despite incisions placed 85% of cases. In contrast, there was a statistically<br />

significant association between a Weber C fracture<br />

and older age and the presence of a SIMPLE PM fracture.<br />

<strong>The</strong>se factors being potentially “protective” from<br />

joint comminution.<br />

Conclusion: We have defined and quantified the PM<br />

articular lesions which require anatomic reduction<br />

and fixation, beyond what has been published. We<br />

have defined clinical and radiographic criteria which,<br />

because higly associated with COMPLEX lesions, could<br />

1) prompt surgeons to order further imaging (CT) to<br />

better delineate the lesion, and 2) draw his/her attention<br />

to potentially malaligned fragments at the time surgery.<br />

COA PAPER SESSION 12: ONCOLOGY<br />

107. CORE NEEDLE BIOPSY IS HIGHLY<br />

ACCURATE IN DIAGNOSING BONE AND<br />

SOFT-TISSUE TUMOURS<br />

Piya Kiatisevi*, Torsten Nielsen, Malcolm Hayes,<br />

Peter L. Munk, Amy E. LaFrance, Paul Clarkson,<br />

Bassam A. Masri<br />

*Lerdsin General Hospital, Bangkok, Thailand;<br />

UBC<br />

Purpose: Core needle biopsy is increasingly accepted for<br />

the diagnosis of bone and soft-tissue tumours. Advantages<br />

over open biopsy include reduced morbidity, time<br />

and cost; however diagnostic accuracy remains a concern.<br />

Our objective was to assess and compare the diagnostic<br />

accuracy of core needle, open, and fine needle biopsies.<br />

Method: We reviewed 286 cases collected in a prospective<br />

database between 2004 and 2007. Of these, 229<br />

had core needle, 32 open, and 25 fine needle biopsies.<br />

230 had soft-tissue lesions, 56 had bone lesions. <strong>The</strong><br />

results of these biopsies were compared to the final<br />

resection diagnosis for accuracy and, where inaccurate,<br />

any effects on management.<br />

Results: Ninety-two percent of the core needle, 100% of<br />

the open and 72% of the fine needle biopsies had adequate<br />

tissue to make a diagnosis. Of the adequate specimens,<br />

the accuracy of core/open/fine needle biopsy was<br />

96%, 97% and 94% for determining malignant versus<br />

benign; of the correctly identified malignant lesions<br />

97%, 100% and 80% were accurate for histological<br />

grade; and 79%, 84%, 59% for histological subtype.<br />

Conclusion: Core needle biopsy yields diagnostic results<br />

comparable to open biopsy for determining malignancy<br />

and grade in bone and soft-tissue tumours. Fine needle<br />

biopsy has a high inadequate sampling rate and should<br />

not be used for diagnosing bone and soft-tissue tumours.<br />

Given the reduced cost and morbidity associated with<br />

core needle biopsies we believe they should be used<br />

routinely for diagnosis where possible, and open biopsy<br />

reserved for situations where an inadequate specimen is<br />

obtained or core biopsy is not feasible.<br />

108. GAMMA PROBE GUIDED SURGERY<br />

FOR BENIGN BONE TUMOURS: SURGICAL<br />

RESULTS WITH AVERAGE FOLLOW-UP OF 5<br />

YEARS<br />

Nanjundappa S. Harshavardhana, Brian J.C.<br />

Freeman*, Alan C. Perkins † , Ujjwal K. Debnath<br />

Queen’s Medical Centre; *Royal Adelaide<br />

Hospital; † University of Nottingham;<br />

Purpose: Intra-op localisation of small nidus in Osteiod<br />

osteoma and Osteoblastomas is often difficult resulting in<br />

failed excision with persistent pain. We report two year<br />

follow-up results of the efficacy and reliability of using an<br />

intra-operative gamma probe in conjunction with fluoroscopy<br />

to aid resection in primary and revision surgeries.<br />

Method: Eight patients (6M; 2F) with a diagnosis of<br />

osteoid osteoma (7) and osteoblastoma (1) were seen<br />

at our centre. <strong>The</strong> mean age at presentation was 20.9<br />

years (9–31y). <strong>The</strong> tumour was localised to cervical (2),<br />

thoracic(4) and lumbar (2) posterior elements. All had<br />

back or neck pain of varying duration with a mean of 20<br />

months (6-48mo). Three patients had failed treatments<br />

including CT-guided radiofrequency ablation in one and<br />

surgical excision under fluoroscopy in two. No case had<br />

previously utilised an intra-op gamma probe for localisation.<br />

All patients had work-up with plain X-rays, CT,<br />

MRI and 99 m Technetium bone scan to identify and<br />

localise the lesion. A pre-requisite for use of intra-op<br />

gamma probe was a positive pre-op bone scan. On the<br />

day of surgery, 600 MBq Tech HMDP (hydroxy-methylene-di-phosphate)<br />

was administered IV 3 hours prior<br />

to surgery. Fluoroscopy was used to confirm anatomical<br />

level, permanent mark made on skin and area exposed<br />

surgically. A 5 mm cadmium telluride (Cd Te) probe<br />

(which converts gamma radiation into electrical signal)<br />

and rate meter were used to scan the area containing<br />

lesion and counts per second(cps) recorded. <strong>The</strong> tumour<br />

nidus was then excised and cps from tumour bed and<br />

excised specimen recorded.<br />

Results: <strong>The</strong> mean follow-up was 5.85 years (2–12.3y).<br />

<strong>The</strong> mean cps for osteoid osteoma pre-excision was<br />

203.8 (60-515), which fell to 72.5 (10-220) post-excision.<br />

<strong>The</strong> cps reduced from 373 to 40.5 post-operatively<br />

for Osteoblastoma. Complete excision was recorded<br />

every time and all patients reported characteristic disappearance<br />

of pre-operative pain. All had discontinued<br />

analgesic medication and returned to normal activity by<br />

three months. All patients were followed-up regularly<br />

when they filled NDI, ODI and SF-36.<br />

Conclusion: Gamma probe guided surgical excision<br />

facilitates accurate localisation of lesion, is less invasive<br />

and most importantly confirmation of complete excision<br />

of the tumour nidus consistently every time.<br />

J BONE JOINT SURG [BR] 2011; 93-B:SUPP III


266 COA/CORS/CORA<br />

109. LOCAL RECURRENCE OF SOFT TISSUE<br />

SARCOMA FOLLOWING PRIMARY TUMOUR<br />

MANAGEMENT BY MUSCULOSKELETAL<br />

ONCOLOGY TEAMS<br />

Sarantis Abatzoglou, Abdurahman Adoubali*,<br />

Cindy Wong, Marc Isler, Robert É. Turcotte<br />

McGill University Health; *Hopital Maisonneuve-<br />

Rosemont<br />

Purpose: Management of local recurrence (LR) remains<br />

unclear. Optimal management of primary tumour by<br />

specialised teams minimises this risk. However, previous<br />

treatments may impact on the available options when<br />

LR is encountered. We thus studied the outcome of this<br />

population with recurrent STS.<br />

Method: Retrospective review was carried based on our<br />

prospective sarcoma databases. DFSP and ALT were<br />

excluded. Among 618 primarily managed STS we found<br />

35 cases of local recurrences (5.7%). Median f-up after<br />

LR was 14 mos (0-98).<br />

Results: Twenty were female. Mean age was 54 (Range<br />

15 – 92). 22 involved lower limb, 11 upper limb and 2<br />

the trunk. Mean delay from original surgery was 23 mos<br />

(3-75) and the mean size of LR was 4.7 cm (0.4-28.0<br />

cm). Primary tumours were superficial in 4 and deep in<br />

31 while recurrences were found superficial in 8 and deep<br />

in 26. Most frequent histology was MFH 8, Leiomyosarcoma<br />

6, Liposarcoma, synovial sarcoma and MPNST had<br />

4 each. 84% were high grade. Only 23 showed no metastatic<br />

disease at time LR was diagnosed. All 5 pts without<br />

initial RT got RT for their LR. 7 pts with therapeutic level<br />

of RT to the primary tumour got full course of RT as well<br />

for their LR. 11 did not undergo surgery. 6/18 who had<br />

initial RT underwent amputation as opposed to 0/6 who<br />

did not. Trend to amputate was for younger age, deep<br />

and large tumour and previous RT. Ultimatly, 21(60%)<br />

locally recurrent tumours showed metastatic disease; 6<br />

prior diagnosis of LR, 6 concomitantly and 9 after with<br />

an average delay of 17 months (1-24). 6 pts developed<br />

additional local recurrences.<br />

Conclusion: Although infrequent local recurrence correlates<br />

with impaired outcome. Albeit challenging, limb salvage<br />

and additional radiotherapy remain possible despite<br />

optimal multi modality management of the initial tumour.<br />

110. SURGICAL MANAGEMENT OF<br />

RECURRENT GIANT CELL TUMOUR<br />

OF LONG BONES. A LONG-TERM<br />

RETROSPECTIVE STUDY<br />

Frank M. Klenke, Doris E. Wenger, Carrie Y.<br />

Inwards, Franklin H. Sim<br />

Mayo Clinic<br />

Purpose: Giant cell tumor (GCT) of bone is a rare,<br />

usually benign, primary skeletal lesion. <strong>The</strong> disease’s<br />

clinical course may be complicated by local recurrence<br />

subsequent to surgical treatment or the development of<br />

benign pulmonary metastases. Intra-lesional curettage is<br />

the standard treatment of primary GCT of bone. However,<br />

the value of intralesional procedures in recurrent<br />

GCT has not been well established.<br />

Method: Forty-six patients with recurrent GCT of long<br />

bones treated between 1983 and 2005 were followed<br />

retrospectively. Minimum follow-up was three years;<br />

mean follow-up was 11.1 (±4.8) years.<br />

Results: Wide resections were performed in 18 patients.<br />

Intralesional, joint preserving procedures were performed<br />

in 28 patients. Subsequent recurrence occurred<br />

in nine patients (20%). Wide resection was performed<br />

if joint salvage was not achievable due to expansion of<br />

the tumor. Reconstructions following wide resection<br />

included arthroplasty (n=4), osteoarticular allograft<br />

(n=3), APC (n=1) and fibular autograft reconstruction<br />

of the wrist (n=3). Amputations were performed in two<br />

patients. Patients undergoing wide resections for local<br />

recurrence had a significantly smaller risk of subsequent<br />

recurrence as compared to patients treated with intralesional<br />

surgery (6% versus 32%, hazard ratio: 0.28,<br />

p


COA/CORS/CORA 267<br />

Musculoskeletal Tumor Society (MSTS) 1987 scores<br />

demonstrated one excellent, 4 good, and 5 fair results.<br />

MSTS 1993 scores averaged 71.4 ± 17.2 percent and<br />

Toronto Extremity Salvage Scores (TESS) averaged 61.7<br />

± 21.8. <strong>The</strong>re were no significant differences between<br />

the functional scores for patients with femoral or sciatic<br />

nerve resections (P=1.0).<br />

Conclusion: Femoral nerve resection appears more<br />

morbid than anticipated. <strong>The</strong> falls to which patients were<br />

prone, even years after surgery, subject them to ongoing<br />

long-term risks for fractures and other injuries. Nervespecific<br />

functional outcomes should be considered when<br />

counseling patients prior possible resection of the femoral<br />

nerve for involvement by a soft tissue sarcoma.<br />

114. IMMEDIATE TISSUE TRANSFER MAY<br />

BE SUPERIOR TO PRIMARY WOUND<br />

CLOSURE IN INTERNAL HEMIPELVECTOMY<br />

PROCEDURES<br />

Arvindera Ghag, Kyle Winter, Erin Brown, Amy E.<br />

LaFrance, Paul Clarkson, Bassam A. Masri<br />

UBC<br />

Purpose: Resection of pelvic sarcoma with limb preservation<br />

(internal hemipelvectomy) is a major undertaking.<br />

Resection requires large areas of soft-tissue to be<br />

removed. Because of wound complications, we manage<br />

these defects with immediate tissue transfer (ITT) at<br />

the time of resection when a large defect is anticipated.<br />

This study compares the outcomes of ITT with primary<br />

wound closure (PWC).<br />

Method: Twenty patients undergoing 22 separate procedures<br />

(1995-2007) were identified in our prospectively<br />

maintained database. Demographics, tumour type,<br />

operative data and complications, and functional scores<br />

(MSTS-1993, TESS) were collected.<br />

Results: Twelve defects were managed with ITT, nine<br />

with pedicled myocutaneous vertical rectus abdominis<br />

(VRAM) flaps (one received double VRAM flaps due to<br />

the large defect), two with tensor fascia lata (TFL) rotation<br />

flaps (one augmented by local V-Y advancement,<br />

the other with gluteus maximus rotation flap) and one<br />

received latissimus dorsi free tissue transfer. Four wound<br />

complications necessitated operative intervention in this<br />

group: two debrided VRAM flaps went on to heal and<br />

the two TFL flaps required revision: one to VRAM<br />

flap and the other to a latissimus dorsi free flap which<br />

ultimately suffered chronic infection and hindquarter<br />

amputation was performed. Ten defects were managed<br />

with PWC, and 5 wound complications occurred, all<br />

five suffered infection, one developed hematoma and<br />

one dehisced. One wound resolved with debridement,<br />

two healed after revision to pedicled gracilis and gluteus<br />

maximus myocutaneous flaps. Two patients were converted<br />

to hindquarter amputation due to chronic infection.<br />

Functional scores were collected on 8 of 12 living<br />

patients, at time of writing. <strong>The</strong> mean TESS scores were<br />

83 and 73 in the ITT and PWC groups. Five patients in<br />

the ITT and 3 in the PWC group were deceased.<br />

Conclusion: Soft-tissue closure following pelvic sarcoma<br />

resection remains a difficult challenge, and our experience<br />

reflects that. <strong>The</strong>re were fewer wound complications<br />

(33% v 50%) and slightly better function with ITT<br />

than PWC, but this was not statistically significant due<br />

to the small size of our study. Although small, this study<br />

suggests ITT should be considered whenever a large soft<br />

tissue defect is anticipated.<br />

115. SINGLE DOSE ANTIBIOTIC<br />

PROPHYLAXIS FOR LOWER LIMB<br />

ARTHROPLASTY<br />

Inder Gill, Ajay Malviya, Scott Muller, Mike Reed<br />

Northumbria Healthcare NHS Trust, UK<br />

Purpose: To assess the infection rate following Lower<br />

Limb Arthroplasty using single dose gentamicin antibiotic<br />

prophylaxis compared to a traditional three doses<br />

of cephalosporin.<br />

Method: All patients undergoing Total Hip and Knee<br />

replacements over six months (October 2007 to March<br />

2008) at three participating hospitals were prospectively<br />

followed to assess perioperative infection rates<br />

using Surgical Site Surveillance(SSI) criteria. All patients<br />

received single dose antibiotic prophylaxis using intravenous<br />

Gentamicin 4.5mg/kg. This was compared with<br />

previous data collected over a 6 month period (Jan to<br />

Mar 2007 and Oct to Dec 2005) from the same hospitals<br />

using 3 doses of Cefuroxime 750mg. Return to<br />

theatre data was collected independently after introduction<br />

of gentamicin to compare with previous data. <strong>The</strong><br />

change in creatinine level postoperatively was also measured<br />

in a selected group of patients.<br />

Results: Four hundred and eight patients underwent<br />

Total Hip Replacements (THR) and 458 patients Total<br />

Knee Replacements (TKR) during the study period. This<br />

was compared with 414 and 421 patients who underwent<br />

THRs and TKRs respectively during a previous<br />

six month period. SSI was detected in 9 THRs(2.2%)<br />

and 2 TKRs(0.44%) in the study group as compared<br />

to 13 THRs(3.1%) and 12 TKRs(2.9%) in the control<br />

group. <strong>The</strong> infection rates in THRs were not significantly<br />

different between the 2 groups(p value–0.52)<br />

but were significantly reduced in the study group for<br />

TKRs(p value–0.005). <strong>The</strong> rate of Clostridium difficile<br />

infection was reduced within the hospital with the use<br />

of gentamicin, although other measures to reduce its<br />

incidence were also introduced. <strong>The</strong> return to theatre<br />

was 1.64%(23/1402) after introduction of Gentamicin<br />

as compared with 1.05%(21/2005) [p value–0.092]<br />

before this. This was a cause for concern although not<br />

significant. <strong>The</strong> day1 postoperative creatinine level<br />

increased by more than 30 units in 6% of patients on<br />

Gentamicin.<br />

Conclusion: This study shows that the use of single dose<br />

prophylaxis using Gentamicin is effective for Lower<br />

Limb Arthroplasty. However, be wary of increased rate<br />

of return to theatre and the rise in creatinine level following<br />

use of gentamicin. Further period of evaluation<br />

and study is needed before it is recommended for routine<br />

use in present or modified form.<br />

COA PAPER SESSION 13:<br />

SPORTS UPPER EXTREMITY<br />

116. ARTHROSCOPIC TREATMENT<br />

OF MULTIDIRECTIONAL SHOULDER<br />

INSTABILITY IN ATHLETES: A<br />

RETROSPECTIVE ANALYSIS OF 2 TO 5 YEAR<br />

CLINICAL OUTCOMES<br />

Randy Mascarenhas*, Champ L. Baker † , Alex J.<br />

Kline ‡ , Anikar Chhabra § , Mathew Pombo ‡ , James<br />

P. Bradley ‡<br />

University of Manitoba; † Hughston Sports<br />

Medicine; ‡ University of Pittsburgh; § Canyon<br />

Orthopaedics<br />

Purpose: <strong>The</strong>re are few reports in the literature detailing<br />

the arthroscopic treatment of multidirectional instability<br />

of the shoulder. <strong>The</strong> purpose of this study was to evaluate<br />

the results of arthroscopic methods in the treatment<br />

of athletes with symptomatic multidirectional instability<br />

of the shoulder.<br />

Method: Forty patients (43 shoulders) with multidirectional<br />

instability of the shoulder were treated via<br />

arthrscopic means and were evaluated at a mean of<br />

33.5 months post-operatively. <strong>The</strong> mean patient age<br />

was 19.1 years (range 14 to 39). <strong>The</strong>re were 24 male<br />

patients and 16 female patients. Patients were evaluated<br />

with the ASES and WOSI scoring systems. Stability,<br />

strength, and range of motion were evaluated with<br />

patient-reported scales.<br />

Results: <strong>The</strong> mean ASES score postoperatively was 91.4<br />

out of 100. <strong>The</strong> mean WOSI post-operative percentage<br />

score was 91.1 out of 100. Ninety-one percent of<br />

patients had full or satisfactory range of motion, 98%<br />

had normal or slightly decreased strength, and 86% of<br />

patients were able to return to their sport with little or<br />

no limitation.<br />

Conclusion: Arthroscopic methods can provide an effective<br />

treatment for symptomatic multidirectional instability<br />

in an athletic population.<br />

117. GENERALIZED LIGAMENTOUS LAXITY<br />

AS A PREDISPOSING FACTOR FOR PRIMARY<br />

ANTERIOR SHOULDER DISLOCATION<br />

Jaskarndip Chahal, Tom McCarthy, Jeff Leiter*,<br />

Daniel B. Whelan †<br />

University of Toronto; *Pan Am Clinic, University<br />

of Manitoba; †St. Michael’s Hospital<br />

Purpose: To determine whether generalized ligamentous<br />

laxity is a predisposing factor for primary traumatic<br />

anterior shoulder dislocation in young, active patients.<br />

Method: Prospective case series with age and sex<br />

matched controls. <strong>The</strong> Hospital Del Mar Criteria was<br />

utilized to measure generalized ligamentous laxity. Fiftyseven<br />

(n=57) consecutive individuals (age


268 COA/CORS/CORA<br />

lar to previous reports, and resulted in improved HRQL<br />

and shoulder ROM. <strong>The</strong> WOSI score was better able<br />

to detect problems in HRQL related to instability than<br />

either the ASES or Constant score.<br />

119. DOES THE WESTERN ONTARIO<br />

SHOULDER INSTABILITY (WOSI) INDEX<br />

DISTINGUISH BETWEEN OPERATIVE<br />

AND NON-OPERATIVE PATIENTS WITH<br />

SHOULDER INSTABILITY: A CASE-CONTROL<br />

STUDY<br />

Nicholas G.H. Mohtadi, Jocelyn N. Fredine,<br />

Heather N. Hannaford, Denise S. Chan, Treny M.<br />

Sasyniuk*<br />

University of Calgary; *LifeMark Health<br />

Purpose: Shoulder instability is a common problem<br />

affecting patients in their most active years resulting in<br />

an impact on their quality of life. <strong>The</strong> WOSI is a validated,<br />

disease-specific (shoulder instability) evaluative<br />

quality of life measure. It has not been tested for its ability<br />

to discriminate between those who require surgical<br />

care and those who do not. <strong>The</strong> purpose of this study<br />

is to determine if the WOSI can discriminate between<br />

surgical and non-surgical patients and between patients<br />

with different types of shoulder instability.<br />

Method: Sixty patients with a confirmed diagnosis of<br />

shoulder instability were included as cases. Twenty<br />

had documented multidirectional instability requiring<br />

surgery: Group 1 Surgical MDI – 20 patients had<br />

documented recurrent traumatic anterior dislocations<br />

requiring surgery: Group 2 Surgical Anterior – 20<br />

patients were first time anterior dislocators who were<br />

followed for a minimum one year who had no further<br />

recurrences and did not require surgery: Group 3 Non-<br />

Surgical First Time Anterior – <strong>The</strong> cases were compared<br />

to 60 age and gender matched control patients with no<br />

history of shoulder problems: Group 4 Control – WOSI<br />

scores were analyzed using a one-way ANOVA.<br />

Results: <strong>The</strong> WOSI scores were as follows: Group 1<br />

Surgical MDI- mean 30.5 (95% CI 23.1-37.8); Group 2<br />

Surgical Anterior- mean 39.8 (95% CI 33.1-46.5); Group<br />

3 Non-Surgical First time Anterior- mean 76.2 (95%<br />

CI 66.4-86.0) and Group 4 Control- mean 96.6 (95%<br />

CI 95.8-97.4). Based on the 95% Confidence Intervals,<br />

there were statistically significant differences between the<br />

two surgical groups (Group 1 Surgical MDI and Group 2<br />

Surgical Anterior) compared to the non-surgical patients<br />

(Group 3 Non Surgical First Time Anterior) and the controls<br />

(P=0.000). <strong>The</strong>re is a trend to discriminate between<br />

the two surgical groups (P=0.079).<br />

Conclusion: <strong>The</strong> WOSI Index clearly discriminates<br />

between surgical and non-surgical patients with shoulder<br />

instability, and the control population with normal<br />

shoulders. <strong>The</strong>re is a trend to discriminate between MDI<br />

and recurrent anterior traumatic dislocators.<br />

120. THE UNSTABLE PAINFUL SHOULDER<br />

(UPS): AS A CAUSE OF PAIN FROM<br />

UNRECOGNIZED INSTABILITY IN THE<br />

YOUNG ATHLETE<br />

Ryan Bicknell*, Chris Chuinard, Scott Penington,<br />

Frédéric Balg, Pascal Boileau<br />

*Queen’s University; University of Nice<br />

Purpose: Shoulder pain in the young athlete is often a<br />

diagnostic challenge. It is our experience that this pain<br />

can be related to a so-called “unstable painful shoulder”<br />

(UPS), defined as instability presenting in a purely painful<br />

form, without any history of instability but with anatomical<br />

(soft tissue or bony) ‘roll-over’ lesions. <strong>The</strong> objectives<br />

are to describe the epidemiology and diagnostic criteria<br />

and to report the results of surgical treatment.<br />

Method: A prospective review was performed of 20<br />

patients (mean age 22 ± 8 years). Inclusion criteria: a<br />

painful shoulder and “roll-over lesions” on imaging or<br />

at surgery. Exclusion criteria: a dislocation/subluxation;<br />

associated pathology; previous shoulder surgery.<br />

Results: Most patients were male (60%), athletes (85%)<br />

and involved the dominant arm (80%). All patients<br />

denied a feeling of instability and only complained<br />

of deep, anterior pain. Most had a history of trauma<br />

(80%). All patients had rehabilitation without success<br />

and 30% had subacromial injections. All had to stop<br />

sports. Most (85%) had anterior or inferior hyperlaxity.<br />

All had pain with an anterior apprehension test and<br />

relieved by relocation test. ‘Roll-over’ lesions included:<br />

labrum detachment (90%), capsular distension (75%),<br />

HAGL lesion (10%), glenoid fracture (20%) or Hill-<br />

Sachs (40%). Time from symptoms to surgery was 25<br />

± 23 months. All patients had arthroscopic treatment.<br />

Mean follow-up was 38 ± 14 months. Eighteen patients<br />

(90%) were very satisfied/satisfied. None had pain at<br />

rest, but one (5%) had pain with apprehension test.<br />

<strong>The</strong>re was no change in elevation, external or internal<br />

rotation (p>0.05). <strong>The</strong>re were no cases of instability.<br />

Rowe and Duplay scores improved (p


COA/CORS/CORA 269<br />

to 57.1 days (p < 0.01). ROM increased in both groups<br />

non-significantly from pre-op to 3 months follow up in<br />

the study group.<br />

Conclusion: <strong>The</strong> results of this study show that the<br />

fibrin-PRP clot enhances the healing of the labrum to<br />

the glenoid. We have significantly fewer failures, less<br />

pain, quicker time to discharge, and faster functional<br />

recovery. By suturing a fibrin-PRP clot between the<br />

labrum and glenoid, recovery and healing of the tear<br />

occurs quicker and more reliably.<br />

124. INJURY OF THE SUPRASCAPULAR<br />

NERVE DURING ARTHROSCOPIC REPAIR OF<br />

SUPERIOR LABRAL TEARS: AN ANATOMIC<br />

STUDY<br />

Holman Chan, Martin Bouliane, Lauren Beaupré<br />

University of Alberta Hospital<br />

Purpose: Due to its proximity to the glenohumeral joint,<br />

the suprascapular nerve may be at risk of iatrogenic nerve<br />

injury during arthroscopic labral repair. Our primary objective<br />

is to evaluate the risk of suprascapular nerve injury<br />

during standard drilling techniques utilized in arthroscopic<br />

superior labral repairs. Secondarily, we evaluated the correlation<br />

between this risk and scapular size.<br />

Method: Forty-two cadaveric shoulders were dissected<br />

to isolate their scapulae. A surgical drill and guide<br />

was used to create suture anchor holes in 3 locations<br />

in the superior rim of the glenoids as typically done in<br />

arthroscopic superior labral repairs. <strong>The</strong> orientation<br />

of these drill holes correspond to common shoulder<br />

arthroscopic portals. <strong>The</strong> suprascapular nerve was then<br />

dissected from the suprascapular notch to the spinoglenoid<br />

notch. <strong>The</strong> presence of drill perforations through<br />

the medial cortex of the glenoid vault was recorded<br />

along with the corresponding hole depth and distance<br />

to the suprascapular nerve.<br />

Results: Medial glenoid vault perforations occurred in<br />

8/21(38%) cadavers with a total of 18/126(14%) perforations.<br />

<strong>The</strong> suprascapular nerve was in line of the drill<br />

path in 5/18(28%) perforations. Female specimens and<br />

smaller scapulae had a statistically higher risk of having<br />

a perforation (p


270 COA/CORS/CORA<br />

total laxity of 11.2 ± 1.5. <strong>The</strong>re was no significant effect<br />

of sequential olecranon excision on elbow kinematics<br />

or stability with the elbow in the vertical or horizontal<br />

positions. <strong>The</strong> elbows became grossly unstable after<br />

resection of greater than 75% of the olecranon.<br />

Conclusion: A progressive increase in the varus-valgus<br />

laxity of the elbow was seen with sequential excision<br />

of the olecranon. Laxity of the elbow was increased<br />

with excision of 75% of the olecranon, likely due to<br />

the loss of the bony congruity and attachment site of<br />

the posterior band of the medial collateral ligament.<br />

Gross instability resulted when 87.5% or greater was<br />

removed, likely due to damage to the anterior band of<br />

the medial collateral ligament as it inserts on the sublime<br />

tubercle of the ulna. Rehabilitation of the elbow<br />

with the arm in the dependant position should be considered<br />

following excision of the olecranon; varus and<br />

valgus orientations should be avoided. <strong>The</strong> contribution<br />

of the olecranon to elbow stability may be even more<br />

important in patients with associated ligament injuries<br />

or fractures of the elbow.<br />

129. PROXIMAL ULNAR ANATOMY:<br />

IMPORTANCE OF THE DORSAL<br />

ANGULATION (PUDA)<br />

Dominique Rouleau*, George Athwal, Kenneth<br />

J. Faber<br />

*Hôpital Sacré-Coeur de Montréal; University of<br />

Western Ontario<br />

Purpose: Recognition of the proximal ulna dorsal angulation<br />

(PUDA) is important for anatomic reduction<br />

of proximal ulnar fractures or osteotomies, especially<br />

when using newer straight precontoured proximal ulnar<br />

plates. <strong>The</strong> purpose of this study was to characterize the<br />

PUDA in 50 patients with bilateral elbow radiographs.<br />

Method: Bilateral elbow radiographs (100 radiographs)<br />

were magnified four times using commercial software.<br />

<strong>The</strong> PUDA was measured from the intersection of lines<br />

tangent to the subcutaneous border of the olecranon and<br />

the proximal ulnar shaft. <strong>The</strong> olecranon tip-to-apex distance<br />

of the PUDA was also measured. Three orthopaedic<br />

surgeons independently examined the radiographs<br />

and intra/inter-observer reliability was calculated using<br />

Intra-Class-Correlation (ICC).<br />

Results: A PUDA was present in 96% of radiographs.<br />

<strong>The</strong> average PUDA was 5.7° (range, 0°to14°). <strong>The</strong><br />

Pearson Correlation coefficient for a side-to-side comparison<br />

was 0.86(p


COA/CORS/CORA 271<br />

programs and other cost containment measures in<br />

orthopaedic surgery.<br />

134. THE 80 HOUR WORK WEEK: IMPACT<br />

ON RESIDENT SURGICAL EXPOSURE AND<br />

NATIONAL IN-TRAINING EXAM SCORES<br />

John M. Froelich, Joseph C. Milbrandt, D.<br />

Gordon Allan*<br />

Southern Illinois University School of Medicine;<br />

*Orthopaedic Center of Illinois<br />

Purpose: <strong>The</strong> current study examines the impact of the<br />

80-hour work week on the number of surgical cases performed<br />

by PGY2 – PGY5 Orthopedic residents. We also<br />

evaluated Orthopaedic In-training Exam (OITE) scores<br />

during the same time period.<br />

Method: Data were collected from the ACGME<br />

national database for 3 academic years prior to and 5<br />

years after July 1, 2003. CPT surgical procedure codes<br />

logged by all residents three years prior to and five years<br />

following implementation of the 80-hour work week<br />

were compared. <strong>The</strong> average raw OITE scores for each<br />

class obtained during the same time period were also<br />

evaluated. Data were reported as the mean ± standard<br />

deviation (SD) and group means were compared using<br />

independent t-tests.<br />

Results: No statistical difference was noted in the<br />

number of surgical procedure codes logged prior to or<br />

after institution of the 80-hour week during any single<br />

year of training. However, an increase in the number<br />

of CPT codes logged in the PGY-3 year after 2003 did<br />

approach significance (457.7 vs. 551.9, p=0.057). <strong>The</strong>re<br />

was a statistically significant increase in total number of<br />

cases performed (464.4 vs. 515.5 p=0.048). No statistically<br />

significant difference was noted in the raw OITE<br />

scores before or after work hour restrictions for our<br />

residents or nationally.<br />

Conclusion: We found no statistical difference for each<br />

residency class in the average number of cases performed<br />

or OITE scores. We also found no statistical difference<br />

in the national OITE scores. Our data suggest that the<br />

impact of the 80 hour work has not had a detrimental<br />

effect in these two resident training measurements.<br />

135. THE CANADIAN ORTHOPAEDIC<br />

MEDICOLEGAL CLIMATE: GLOBAL<br />

WARMING OR ISOLATED COOLING?<br />

Carol R. Hutchison, Claude Martin*<br />

University of Calgary; *Canadian Medical<br />

Protective Association<br />

Purpose: Litigation continues to be a concern in orthopaedic<br />

surgery despite suggestions on how to contain<br />

liability. <strong>The</strong> purpose of this study was to characterize<br />

orthopaedic litigation in Canada from 1997-2006.<br />

Method: This study reviewed all closed claims reported<br />

to the Canadian Medical Protective Association (CMPA)<br />

for 1997-2006 in which orthopaedic surgeons were<br />

named. <strong>The</strong>re were 11,983 closed legal actions involving<br />

CMPA members (> 73,000 physicians), and 1,353<br />

involved orthopaedic surgeons. A careful review of<br />

closed legal actions is a recognized tool for risk identification,<br />

assessment and management. <strong>The</strong> CMPA identifies<br />

any critical incidents within the closed legal files. A<br />

critical incident is defined as any omission or commission<br />

in the evaluation or management which led to the<br />

problem(s) that triggered the legal action. Each closed<br />

legal action can have more than one critical incident.<br />

Results: Performance, diagnostic and communication<br />

issues were the most frequently identified problems.<br />

<strong>The</strong>se three areas account for 55% of the critical incidents<br />

identified. Performance related issues accounted<br />

for 395 critical incidents (29%). Diagnostic issues,<br />

including deficient histories and general evaluations,<br />

were identified in 281 cases (21%). Communicationrelated<br />

critical incidents included those concerning<br />

informed consent. <strong>The</strong> lack of informed consent was<br />

a common allegation, proven in 71 cases. In 439 cases<br />

(32%) there was no identifiable critical incident for the<br />

orthopaedic surgeon involved. Seventy-eight per cent of<br />

patients experienced minor or no disability and 22%<br />

experienced major disability or death. Events related<br />

to tibia trauma and knee arthroscopy formed the two<br />

major categories of claims. Patient care areas of high<br />

risk include the operating room and outpatient clinic.<br />

Overall, 31% of legal actions against orthopaedic surgeons<br />

had outcomes in favour of the plaintiffs, compared<br />

with 33% of all CMPA members’ claims.<br />

Conclusion: Although the likelihood for an orthopaedic<br />

surgeon to be sued in Canada has decreased over the last<br />

10 years, the percentage of legal cases resolved in favour<br />

of plaintiffs has remained stable. Performance-related<br />

deficiencies, delays in diagnosis, and failures in communication<br />

represent areas of high medico-legal risk. Suggestions<br />

for risk management are provided to further<br />

decrease adverse events and the medico-legal risks for<br />

Canadian orthopaedic surgeons.<br />

COA PAPER SESSION 15: SPINE 1<br />

136. PREOPERATIVE PREDICTORS FOR<br />

POSTOPERATIVE CLINICAL OUTCOME IN<br />

LUMBAR DISCECTOMY<br />

Yangmin Zeng, Travis Marion, Pamela Leece,<br />

Eugene Wai*<br />

University of Ottawa, Faculty of Medicine;<br />

*University of Ottawa, Division of Orthopaedic<br />

Surgery<br />

Purpose: Persistent radiculopathy secondary to lumbar<br />

disc herniation is a common problem that greatly compromises<br />

quality of life. In North America, lumbar<br />

discectomies are among the most common elective surgical<br />

procedures performed. <strong>The</strong>re is still much debate<br />

about when conservative or surgical treatments should<br />

be offered to patients. Although the related literature is<br />

comprehensive, there are limited systematic reviews on<br />

the prognostic factors predicting the outcome of lumbar<br />

discectomy. <strong>The</strong> purpose of this review is to define the<br />

preoperative factors predicting clinical outcome after<br />

lumbar discectomy.<br />

Method: We conducted a computerized literature search<br />

using Ovid Medline and the Cochrane Central Register<br />

of Controlled Trials. We included randomized controlled<br />

trials or prospective studies dealing with lumbar<br />

disc surgery. <strong>The</strong> preoperative predictors had to be<br />

clearly identified and correlated with outcome measures<br />

in terms of pain, disability, work capacity, analgesia<br />

consumption, or a combination of these measures. We<br />

assessed the articles as high or low quality studies using<br />

the Newcastle-Ottawa Quality Assessment Scale, and<br />

summarized the results of High Quality Studies.<br />

Results: A total of 39 articles were included. <strong>The</strong> two<br />

most prominent negative predictors were Workers’<br />

Compensation status and depression according to 6<br />

studies. Poor predictors reported in 4 articles were<br />

female gender, increasing age, and prolonged duration<br />

of leg or back pain. Lower education level, smoking,<br />

and higher levels of psychological complaints were<br />

negative predictors in 3 articles. A positive Lasègue sign<br />

was a positive predictor in 7 articles. Absence of back<br />

pain, positive patient expectations, and higher income<br />

were good prognostic factors in 3 studies. Patients with<br />

contained herniations did worse than those who had<br />

uncontained disc extrusions and sequestrations according<br />

to 4 studies. <strong>The</strong> level of herniation was not a predictive<br />

factor in 7 studies.<br />

Conclusion: Workers’ Compensation, depression, greater<br />

back versus leg pain, increasing age, female gender, contained<br />

herniations, and prolonged symptoms predict<br />

unfavourable postoperative outcomes after lumbar discectomy.<br />

Positive Lasègue sign, higher income, uncontained<br />

herniations, and positive patient expectations<br />

predict favourable postoperative outcomes. <strong>The</strong> level of<br />

herniation is not an established prognostic factor. <strong>The</strong><br />

results of this review provide a preliminary framework<br />

for patient selection for lumbar disc surgery.<br />

137. PRE-OPERATIVE CT IMAGING OF THE<br />

CROSS-SECTIONAL AREA OF PERISPINAL<br />

MUSCULATURE AS A PREDICTOR OF<br />

POSTERIOR LUMBAR FUSION SURGERY<br />

OUTCOMES<br />

Travis E. Marion, Yangmin Zeng, Eugene Wai*<br />

University of Ottawa, Faculty of Medicine;<br />

*University of Ottawa, Division of Orthopaedic<br />

Surgery<br />

Purpose: Perispinal core muscle strength has been theorized<br />

to be an important component in the pathogenesis<br />

of back pain. Recent research has demonstrated a<br />

strong association between preoperative perispinal musculature,<br />

adjusted for fatty infiltration and prospective<br />

outcomes and improvements in back pain in patients<br />

undergoing lumbar laminectomy without fusion. <strong>The</strong><br />

purpose of this study is to determine if a similar relationship<br />

exists in patients undergoing elective posterior<br />

lumbar fusion and decompression (PLFD) surgery.<br />

Method: A retrospective observational study of prospectively<br />

collected outcomes data was conducted in<br />

which pre-operative function and patient variables of<br />

those undergoing PLFD were derived from a functional<br />

status questionnaire and medical records. ImageJ Digital<br />

Imaging Software was utilized to measure the total<br />

(CSA) and percentage of fatty infiltration of the psoas,<br />

multifidus, and erector spinae muscles in pre-operative<br />

L4 axial CT images. Pre-operative and post-operative<br />

lateral images were evaluated for degree of post-operative<br />

adjacent level degeneration. Follow-up consisted of<br />

a functional status questionnaire. Outcomes measured<br />

were improvements in back pain, leg pain, and Oswestry<br />

disability scores.<br />

Results: Twenty-three patients were analyzed with a<br />

mean follow-up of 2 years (range 1 – 5 years). Outcomes<br />

improved following surgery. <strong>The</strong>re were strong to moderate<br />

correlations between percentage of fat in the preoperative<br />

posterior spinal muscles and improvements<br />

in leg pain (r = 0.63, p = < 0.001) and improvements<br />

in back pain (r = 0.41, p = 0.05). <strong>The</strong>re was a moderate<br />

trend towards greater adjacent level degeneration (r<br />

= 0.37, p = 0.1) in patients with higher percentage of<br />

fat in the pre-operative posterior spinal muscles. <strong>The</strong>re<br />

was a strong relationship between greater adjacent level<br />

degeneration and pre-operative disability as measured<br />

by the Oswestry (r = 0.62, p = 0.03).<br />

Conclusion: <strong>The</strong> results demonstrate that a potential<br />

relationship exists between pre-operative fatty infiltration<br />

of posterior perispinal muscles and post-operative<br />

outcomes, and adjacent level degeneration following<br />

lumbar fusion surgery. This suggests that perispinal<br />

muscle atrophy and conditioning may play a role in<br />

these outcomes. Results may be used for prognostication,<br />

surgical candidate selection, and interventional<br />

strategies.<br />

138. CAUDAL EPIDURAL STEROID<br />

INJECTIONS FOR LUMBO-SACRAL<br />

RADICULAR PAIN… DOES IT REALLY MAKE<br />

A DIFFERENCE?<br />

Girish N. Swamy, Lynn DeLoughery,<br />

Rajendranath Bommireddy, Zdenek Klezl, Denis<br />

Calthorpe<br />

Derbyshire Royal Infirmary<br />

Purpose: <strong>The</strong> management of radicular pain due to<br />

lumbar or sacral nerve root compromise remains controversial.<br />

Caudal epidural steroid injections are widely<br />

employed although there is little hard evidence to confirm<br />

their efficacy. This empirical treatment still remains<br />

a matter of personal choice and experience. To investigate<br />

the clinical effectiveness of caudal epidural steroid<br />

injections (CESIs) in the treatment of sciatica and<br />

to identify potential predictors (clinical subgroups) of<br />

response to CESIs.<br />

Method: Prospective study. All patients with corresponding<br />

radicular pain received a course of three caudal epidural<br />

steroid injections, two weeks apart. All patients<br />

reviewed at three months interval in a dedicated epidural<br />

follow up clinic and one-year postal and telephonic<br />

follow-up. Exhaustive epidural database maintained.<br />

VAS scores documented both axial and limb pain for<br />

J BONE JOINT SURG [BR] 2011; 93-B:SUPP III


272 COA/CORS/CORA<br />

actual and comparative analysis. ODI and HADS were<br />

recorded prior to treatment, at three months follow-up<br />

and one year. Main outcome measures: <strong>The</strong> primary<br />

outcome measure was the Oswestry Disability Questionnaire<br />

(ODQ). <strong>The</strong> Visual analogue score (VAS) and<br />

the Hospital Anxiety and Depression Scores (HADS)<br />

were also employed in all cases.<br />

Results: In the largest single series to date, we report<br />

on 928 consecutive patients, with three months followup<br />

and 354 patients with 12 months follow-up. Fiftyeight<br />

percent were females, 24% smoked and 4.1% had<br />

ongoing litigation due to their pain. <strong>The</strong> mean age was<br />

56 years with BMI ranging from 17 to 50 (mean=28).<br />

Ten (0.6%) patients required subsequent surgical intervention<br />

due to disc herniation.<strong>The</strong> mean VAS, ODI and<br />

HADS improved significantly at three months and oneyear<br />

results were encouraging.<br />

Conclusion: Significant improvement in both axial and<br />

limb pain in the short and intermediate terms achieved<br />

facilitating onward referral for physical therapy. Subgroups<br />

predicting poor outcome are identified. Positive<br />

primary care feedback encourages further recruitment.<br />

139. THE TREATMENT OF LUMBAR<br />

INTERMITENT NEUROGENIC<br />

CLAUDICATION USING THE XSTOP IPD<br />

DEVICE: A PROSPECTIVE CINICAL AND<br />

FUNCTIONAL OUTCOME STUDY<br />

Bernard E. Rerri, Temilolu O. Opadele*<br />

Brantford General Hospital; *McMaster<br />

University, Hamilton<br />

Purpose: Lumbar spinal stenosis is the most common<br />

indication for spine surgery in the elderly. XStop IPD is<br />

an attractive alternative to traditional laminectomy or<br />

laminectomy with fusion as it avoids a longer procedure<br />

and anaesthesia with significantly less blood loss. <strong>The</strong><br />

purpose of this study is to prospectively evaluate clinical<br />

outcomes, complications and functional evaluation of<br />

symptom severity, physical function and patient satisfaction<br />

following XStop IPD procedure<br />

Method: Preoperative and postoperative clinical data as<br />

well as SF 36, visual analog scale and Roland Morris<br />

questionnaire data collected on 16 consecutive patients<br />

over 60 years undergoing XStop IPD at L3-4 and L4-<br />

5 levels or both levels. All patients had symptomatic<br />

lumbar spine stenosis with intermittent neurogenic claudication.<br />

Evaluations were made pre-operatively and<br />

post-operatively at 3, 6, 12 and 24 months. All patients<br />

had clinical radiographic data as well as data on visual<br />

analog scale SF 36 and the Roland Morris back questionnaire.<br />

Results: Patients ages ranged from 58 to 86 years with<br />

an average age of 74.25 years. In 75 percent of patients<br />

there were two or more significant co-morbidities with<br />

18.75 percent requiring 2 level surgery. Four of the 16<br />

patients had lumbar degenerative scoliosis with cobb<br />

angle less than 25 degrees. 50.25% the patients had<br />

grade I spondylolisthesis. No patient had previous spine<br />

surgery. In 31.25 percent of patients there was a history<br />

of diabetes. BMI ranged from 20 to 40. Seventy<br />

five percent of patients were discharged home within 24<br />

hours. Ninety percent of patients reported relief of their<br />

leg pain at their first follow up visit within two weeks<br />

of the surgery.<strong>The</strong>re were no significant complications.<br />

One-year follow up in six patients demonstrated<br />

improvements in VAS, Roland Morris criteria and SF 36<br />

while the remaining patients have up to nine months of<br />

follow-up clinical data.<br />

Conclusion: We present our early results of this prospective<br />

study. <strong>The</strong>re were significant improvements in<br />

functional outcomes. We therefore recommend the use<br />

of XStop IPD for elderly patients with multiple co-morbidities<br />

suffering from symptomatic lumbar spine stenosis<br />

with neurogenic claudication.<br />

140. UNDERSTANDING PATIENT AND<br />

PHYSICIAN PREFERENCES FOR SURGERY<br />

ON THE DEGENERATIVE LUMBAR SPINE<br />

S. Samuel Bederman*, Nizar N. Mahomed, Hans<br />

J. Kreder, Warren J. McIsaac † , Peter C. Coyte ‡ ,<br />

James G. Wright<br />

*Department of Orthopaedic Surgery, University<br />

of California, San Francisco; Division of<br />

Orthopaedic Surgery, University of Toronto;<br />

†<br />

Department of Family and Community Medicine,<br />

University of Toronto; ‡ Department of Health<br />

Policy Management and Evaluation, University of<br />

Toronto<br />

Purpose: Surgery for degenerative lumbar spinal conditions<br />

offers tremendous benefit for patients with moderate/severe<br />

symptoms failing non-operative treatment.<br />

<strong>The</strong>re is little appreciation among referring family physicians<br />

(FPs) on factors that identify the ideal surgical<br />

candidate. Differences in preferences between patients<br />

and physicians leads to wide variation in referrals and<br />

impedes the shared decision-making process. Our purpose<br />

was to identify the dominant clinical factors influencing<br />

patient, FP, and surgeon preferences for lumbar<br />

spinal surgery.<br />

Method: We used conjoint analysis, a rigorous method<br />

for eliciting preferences, in a mailed survey to all orthopaedic<br />

and neurosurgeons, a random sample of FPs,<br />

and patients in Ontario to determine the importance<br />

that respondents place on decisions for lumbar spinal<br />

surgery. We identified six clinical factors (walking tolerance,<br />

duration of pain, pain severity, neurological symptoms,<br />

typical onset, and dominant location of pain) and<br />

presented 16 hypothetical vignettes to participants who<br />

rated, on a six-point-scale, their preference for surgery.<br />

Data were analyzed using random-effects ordered probit<br />

regression models and relative importance of each clinical<br />

factor was reported.<br />

Results: We obtained responses from 131 surgeons, 202<br />

FPs, and 164 patients. We demonstrated that despite<br />

wide variations in overall responses, all six clinical factors<br />

were highly associated with surgical preference<br />

(p


COA/CORS/CORA 273<br />

clinical results in carefully selected patients. <strong>The</strong> radiographic<br />

assessment confirmed preservation of movement<br />

at the replaced disc during flexion and extension<br />

of the lumbar spine.<br />

143. PROSPECTIVE RANDOMISED<br />

CONTROLLED STUDY COMPARING A DBM-<br />

CASO4 COMPOSITE GRAFT AND BONE<br />

MARROW ASPIRATE WITH AUTOLOGOUS<br />

ILIAC CREST BONE GRAFT IN ONE-<br />

LEVEL AND TWO-LEVEL LUMBAR AND<br />

LUMBOSACRAL SPINAL FUSIONS<br />

David I. Alexander, William M. Oxner, Alex M.<br />

Soroceanu, Adrienne Kelly, Donna Shakespeare<br />

Dalhousie University<br />

Purpose: <strong>The</strong> current gold standard for spinal arthrodesis,<br />

autologous bone graft harvested from the iliac crest,<br />

has several disadvantages including donor site morbidity,<br />

blood loss, delayed wound healing, and increased<br />

operative time. Our study explores a Demineralized<br />

<strong>Bone</strong> Matrix-Calcium Sulfate(DBM-CaSO4) composite<br />

graft with autologous bone marrow aspirate (BMA),<br />

and compares it to autologous iliac crest bone graft in<br />

lumbar and lumbosacral spinal fusions.<br />

Method: A total of 80 patients were recruited for the<br />

study and randomised, via a computer-generated randomisation<br />

schedule, to autologous iliac crest bone graft<br />

(control) or DBM-CaSO4 composite graft with BMA<br />

(study) groups. Patients were evaluated at three-months,<br />

six-months, 12-months and 24-months post-operatively<br />

with questionnaires to evaluate clinical outcome<br />

(Oswestry disability questionnaire (ODI), visual analogue<br />

pain scales (VAS), and validated SF-36) and with<br />

posteroanterior and lateral x-rays of the spine to evaluate<br />

radiological outcome.<br />

Results: At 24-months post-operatively, there were no<br />

statistical differences seen between the two groups based<br />

on the clinical outcomes measured. Average ODI values<br />

were 27.19 for the control group versus 22.68 for the<br />

study group (p > 0.05). <strong>The</strong> average back VAS pain for<br />

the control group was 3.50 versus 3.51 for the study<br />

group (p > 0.05). <strong>The</strong> SF-36 score was 89.22 for the control<br />

group versus 91.56 for the study group (p > 0.05).<br />

<strong>The</strong> average operative time was 115.7 minutes for the<br />

control group versus 104.2 minutes for the study group<br />

(p: 0.014). Average calculated blood loss was 571.9 cc for<br />

the control group versus 438.2 cc for the study group (p:<br />

0.025). <strong>The</strong> Lenke score was 1.92 for the control group<br />

versus 2.66 for the study group (p: 0.004).<br />

Conclusion: At two year follow-up, radiographic fusion<br />

was slightly higher in the ICBG. However, clinical outcomes<br />

were equivalent in both groups. Moreover, the<br />

DBM-CaSO4 and BMA composite graft offered the<br />

advantages of decreased blood loss and shorter operative<br />

time. <strong>The</strong>refore, the DBM-CaSO4 and BMA composite<br />

graft represents a viable alternative to autologous<br />

iliac crest bone graft in carefully selected patients undergoing<br />

spinal arthrodesis.<br />

144. VERTEBRAL FRACTURE RISK AND<br />

TOKUHASHI SCORE VALIDATION IN<br />

PATIENTS WITH METASTATIC BREAST<br />

CANCER<br />

John Townley, Cari Whyne*, Michael R.<br />

Hardisty*, Liying Zhang, Mark Clemons † , Albert<br />

J.M. Yee ‡<br />

University of Toronto; *Sunnybrook Health<br />

Sciences Centre, Toronto; † University Health<br />

Network, Toronto; ‡ University of Toronto Spine<br />

Program, Toronto<br />

Purpose: To identify local and systemic risk factors<br />

for the development of pathologic fractures and determine<br />

the value of the Tokuhashi Score in patients with<br />

known asymptomatic lytic spinal metastases secondary<br />

to breast cancer.<br />

Method: A prospective cohort study was carried out<br />

on 51 patients with lytic spinal metastases secondary<br />

to breast cancer identified as having either purely lytic<br />

or mixed disease. <strong>The</strong> Tokuhashi Score, developed to<br />

estimate life expectancy for patients with symptomatic<br />

spinal metastases being considered for surgery, was calculated<br />

for each of the 51 patients. <strong>The</strong> score consists of<br />

six parameters each of which is rated from 0-2. Initial<br />

and follow up CT images and pain and function data<br />

were obtained every four months for one year. A final<br />

review of patient charts was performed two years later<br />

to determine if each patient was still alive.<br />

Results: Tumour burden was predominantly blastic and<br />

mixed rather than lytic. <strong>The</strong>re was no progression of<br />

lytic tumour burden over the 12-month period, however<br />

there was progression of blastic tumour load. Eleven<br />

compression fractures occurred in seven patients; no<br />

burst fractures occurred during the study. No correlation<br />

between tumour burden (lytic, blastic or both) and<br />

risk of fracture was found. A weak correlation between<br />

bone mineral density and length of time elapsed from<br />

diagnosis of metastatic disease and fracture risk was<br />

found. Pain and functional data results were not related<br />

to tumour load. Tokuhashi score did correlate with survival,<br />

however actual survival in our population was<br />

far longer than that found in previous studies. Negative<br />

progesterone status was found to be negatively associated<br />

with life expectancy.<br />

Conclusion: Metastatic vertebral disease in breast cancer<br />

patients has a predominantly blastic and mixed appearance<br />

with current pharmacologic therapies. Pathologic<br />

fracture risk appears to be more related to bone mineral<br />

density than tumour burden in this population.<br />

Tokuhashi score does correlate with life expectancy in<br />

patients with relatively asymptomatic spinal metastases.<br />

Having a progesterone receptor negative tumour has a<br />

significantly negative impact on life expectancy.<br />

145. VALIDATION OF THE ENNEKING<br />

ONCOLOGIC CLASSIFICATION IN THE<br />

MANAGEMENT OF PRIMARY TUMOURS OF<br />

THE SPINE: A COHORT STUDY<br />

Davor D. Saravanja, Charles G. Fisher, Marcel<br />

Dvorak, Michael Boyd*, Paul Clarkson †<br />

CNOSP; *Hongbin Zhang, Research Institute;<br />

†<br />

Orthopaedic Surgery<br />

Purpose: Oncologic management of primary bone<br />

tumors of the spine is inconsistent, controversial and<br />

open to individual interpretation. Tumor margin violation<br />

intraoperatively increases local recurrence and mortality.<br />

<strong>The</strong> purpose of this study is to determine whether<br />

applying Enneking’s principles to the surgical management<br />

of primary bone tumors of the spine significantly<br />

decreases local recurrence and/or mortality.<br />

Method: A prospective and retrospective multicenter<br />

Cohort Study: Inclusion of patients undergoing en<br />

bloc or intralesional resection of primary tumors of the<br />

spine at four separate quaternary care centers, between<br />

January 1994 and January 2008. Patients were staged,<br />

using the Enneking system, prior to surgery and baseline<br />

demographic and surgical variables were recorded. Outcomes<br />

measured were disease local recurrence, or death.<br />

<strong>The</strong> results were statistically analyzed for significance.<br />

Results: One hundred-fifty patients with primary tumors<br />

of the spine were recruited. Average age was 47.0 (range<br />

8 to 83). Sixty-two patients were identified to have local<br />

recurrence. A statistically significant decrease in local<br />

recurrence (p=0.0001) was observed in favor of en bloc<br />

resection. In patients with local recurrence there was a<br />

significant increased risk of mortality, (p


274 COA/CORS/CORA<br />

Median urine Co(mg/day): 0.44(poly), 4.55(28mm),<br />

5.42(36mm)). (Median serum Cr(mg/L): 0.17(poly),<br />

1.29(28mm), 0.91(36mm). Median erythrocyte Cr(mg/<br />

L): 1.10(poly), 1.10(28mm), 1.20(36mm). Median urine<br />

Cr(mg/day): 0.27(poly), 1.92(28mm), 2.02(36mm)).<br />

Conclusion: Both cobalt and chromium ion measurements<br />

were significantly elevated in the blood and urine<br />

of the patients randomized to receive the metal-on-metal<br />

bearings at all time intervals. <strong>The</strong>re were no differences<br />

seen between the 28mm and 36mm metal-on-metal<br />

bearings, keeping all other variables identical. <strong>The</strong><br />

larger diameter bearing therefore provides the potential<br />

clinical advantages of improved range of motion and<br />

stability, while providing a similar metal ion profile.<br />

While reduced wear is seen with larger diameter metalon-metal<br />

bearings in-vitro, we could not demonstrate a<br />

reduction in blood or urine metal ion levels in-vivo.<br />

148. RANDOMISED CONTROLLED<br />

TRIAL COMPARING METAL ON METAL<br />

VERSUS METAL ON POLYETHYLENE<br />

ARTICULATION IN THA<br />

Rajeshkumar Kakwani, Chris Wainwright,<br />

Gautam Tawari, Shankar Kashyap, A. Roysam,<br />

A. Nanu<br />

Gateshead Heathcare<br />

Purpose: A single blind prospective randomised controlled<br />

trial comparing the Metal-on-polyethylene articulation<br />

with the metal-on-metal articulation in THA.<br />

Method: <strong>The</strong> clinical and radiological findings of the<br />

consecutive patients who were enrolled in the RCT at<br />

the participating centres were recorded prospectively.<br />

<strong>The</strong> clinical evaluation was performed with the Harris<br />

scoring system as well as the Oxford Hip Scoring Sheet.<br />

<strong>The</strong> computer randomised option was revealed to the<br />

operative surgeon only after the patient was anaesthetised,<br />

during the recruitment period (June 1998 to July<br />

2004). Of the total of 378 patients, 2 died prior to the<br />

final review and 63 were lost to follow-up. <strong>The</strong> final<br />

study group contained 315 patients, with 159 patients<br />

in the metal-on-polyethylene group and 156 patients in<br />

the metal-on-metal group.<br />

Results: <strong>The</strong> indication for the hip arthroplasty for<br />

majority (309 patients) was primary osteoarthritis. <strong>The</strong><br />

average age at the time of the surgery was 68.2 years<br />

and the average duration of follow-up was 85 months<br />

(42-115). <strong>The</strong>re was an improvement of the Oxford hip<br />

scores from an average of 37 per-operatively to 16 postoperatively.<br />

<strong>The</strong> Harris hip scores also improved from<br />

an average of 47.0 pre-operatively to 87.3 post-operatively.<br />

<strong>The</strong> patient groups were statistically similar with<br />

respect to age, sex and duration of follow-up, and the<br />

final outcome scores revealed no statistical difference<br />

between the two groups.<br />

Conclusion: <strong>The</strong> clinical results obtained with the use of<br />

the articulation are comparable to those obtained by the<br />

metal-on-polyethylene articulation encouraging the use<br />

of this alternative bearing surface.<br />

149. RECONSTRUCTION OF ACETABULAR<br />

HIP BIOMECHANICS WITH SURFACE<br />

REPLACEMENT ARTHROPLASTY<br />

Kristoff Corten, Ward Bartels*, Guy Molenaers,<br />

Jos Vander Sloten*, Paul Broos, Johan Bellemans,<br />

Jean-Pierre Simon<br />

University Hospital Leuven; *Catholic University<br />

Leuven<br />

Purpose: Precise biomechanical reconstruction of the<br />

hip joint by a hip arthroplasty is essential for the success<br />

of this procedure. With the increasing use of surface<br />

replacement arthroplasty (SRA), there is a need for<br />

better understanding of the key factors that influence<br />

the anatomical and the biomechanical parameters of the<br />

resurfaced hip joint. <strong>The</strong> goal of this study was to examine<br />

the influence of SRA on the vertical and horizontal<br />

offset of the hip.<br />

Method: Twenty-one hips from 12 embalmed cadavers<br />

were resurfaced with a Birmingham Hip resurfacing.<br />

<strong>The</strong> thickness of the acetabular bone was measured pre-<br />

and post-reaming in 6 acetabular zones. Radiographs<br />

were taken before and after the procedure with a scaling<br />

marker. For statistical analysis, the paired Student’s T-<br />

test with a confidence interval of 95% and a significant<br />

p-value of p


COA/CORS/CORA 275<br />

revised for recurrent infection. Currently no patients are<br />

suspected of having a recurrence of infection.<br />

Conclusion: Retention of a well-fixed femoral cement<br />

mantle during two-stage revision for infection and subsequent<br />

in-cement reconstruction is safe with a cure rate<br />

of 93%. Advantages include a shorter operating time,<br />

reduced loss of bone stock, improved component fixation<br />

and a technically easier second stage procedure.<br />

153. ACETABULAR REVISIONS USING<br />

ANTI PROTRUSION (ILIOISCHIAL)<br />

CAGE AND TRABECULAR METAL CUP<br />

CONSTRUCTS FOR SEVERE ACETABULAR<br />

BONE LOSS ASSOCIATED WITH PELVIC<br />

DISCONTINUITY? PRELIMINARY RESULTS<br />

WITH 1 TO 6 YEARS FOLLOW-UP<br />

David Backstein, Yona Kosashvili, Oleg Safir,<br />

Dror Lakstein, Matthew MacDonald, Allan E.<br />

Gross<br />

Arthroplasty Division, University of Toronto<br />

Purpose: Pelvic discontinuity associated with bone loss<br />

is a complex challenge in acetabular revision surgery.<br />

Reconstruction with anti protrusion cages, Trabecular<br />

Metal (Zimmer, Warsaw, Indiana) cups and morselized<br />

bone (Cup-Cage) constructs is a relatively new technique<br />

used by the authors for the past 6 years. <strong>The</strong> purpose<br />

of the study was to examine the clinical outcome<br />

of these patients.<br />

Method: Thirty-two consecutive acetabular revision<br />

reconstructions in 30 patients with pelvic discontinuity<br />

and bone loss treated by cup cage technique between<br />

January 2003 and September 2007 were reviewed.<br />

Average clinical and radiological follow up was 38.5 ±<br />

19 months (range 12 – 68, median 34.5). Failure was<br />

defined as component migration > 5mm.<br />

Results: In 29 (90.6%) patients there was no clinical or<br />

radiographic evidence indicative of loosening at latest<br />

follow up. Harris Hip Scores improved significantly<br />

(p0.05). We found<br />

a deep infection rate of 3.1% in the ALBC group and<br />

2.2% in the PBC group (p=0.27). Adjusted analysis<br />

showed that ALBC was not predictive of a lower infection<br />

rate at 1 year (p=0.84).<br />

Conclusion: ALBC did not reduce the incidence of deep<br />

infection following primary TKA at 1 year follow up.<br />

Further studies are needed to define any high risk groups<br />

for which ALBC might be beneficial.<br />

158. METABOLIC SYNDROME AND THE<br />

INCIDENCE OF DEEP VEIN THROMBOSIS<br />

FOLLOWING TOTAL KNEE ARTHROPLASTY<br />

Rajiv Gandhi, Fahad Razak, Peggy Tso, J<br />

Roderick Davey, Nizar N. Mahomed<br />

University of Toronto<br />

Purpose: Metabolic syndrome (MS) is defined as central<br />

adiposity, elevated fasting glucose, hypertension, and dyslipidemia<br />

defined as high triglyceride and low high-density<br />

lipoprotein (HDL) cholesterol. MS is associated with<br />

a systemic proinflammatory and prothrombotic state. We<br />

asked if patients with MS undergoing total knee arthroplasty<br />

(TKA) have an increased risk for symptomatic<br />

deep vein thrombosis at 3 months follow-up (DVT).<br />

Method: 1,460 consecutive patients were reviewed<br />

from our joint registry undergoing primary, unilateral<br />

TKA between the years of 1998-2006. Demographic<br />

variables of age, gender, comorbidity, and education<br />

were retrieved. Metabolic syndrome (MS) was defined<br />

as body mass index above 30 kg/m2, diabetes, hypertension,<br />

and hypercholesterolemia. Logistic regression<br />

was used to examine the relationship of MS on the<br />

incidence of DVT.<br />

J BONE JOINT SURG [BR] 2011; 93-B:SUPP III


276 COA/CORS/CORA<br />

Results: <strong>The</strong> overall incidence of symptomatic DVT was<br />

4.4% (65/1460). Patients with MS had an increased incidence<br />

of DVT as compared to those without MS(15.5%<br />

vs 3.4%). Adjusted analysis showed that MS increased<br />

the risk of symptomatic DVT by 3.2(95% CI [1.0,15.4],<br />

p=0.04) times compared to those without MS.<br />

Conclusion: Hospital protocols developed for prophylactic<br />

anti-coagulation following TKR should give special<br />

consideration to patients with MS.<br />

159. THE RECORD4 STUDY: ORAL,<br />

ONCE-DAILY RIVAROXABAN COMPARED<br />

TO SUBCUTANEOUS, TWICE-DAILY<br />

ENOXAPARIN FOR PREVENTION OF<br />

VENOUS THROMBOEMBOLISM AFTER<br />

TOTAL KNEE REPLACEMENT<br />

William D. Fisher 1 , Michael Gent 2 , Bruce L.<br />

Davidson 3 , Michael R. Lassen 4 , Louis M.<br />

Kwong 5 , Fred D. Cushner 6 , Paul A. Lotke 7 , Frank<br />

Misselwitz 8 , Tiemo J. Bandel 8 , Alexander G.G.<br />

Turpie 2<br />

1<br />

McGill University Health Centre; 2 McMaster<br />

University; 3 Weill Cornell Medical College;<br />

4<br />

Hoersholm Hospital; 5 Harbor-UCLA<br />

Medical Center; 6 Insall Scott Kelly Center for<br />

Orthopaedics and Sports Medicine; 7 University<br />

of Pennsylvania Medical Center; 8 Bayer Health<br />

Care AG<br />

Purpose: Venous thromboembolism (VTE) after major<br />

orthopaedic surgery remains an important clinical<br />

problem. Convenient, oral antithrombotic agents that<br />

are both effective and safe could improve adherence<br />

to guidelines for VTE prophylaxis. Recently, the focus<br />

has been on the development of oral agents that target<br />

a single step in the coagulation cascade and Factor Xa<br />

is a pivotal step. Rivaroxaban is an oral, direct Factor<br />

Xa inhibitor. Four international phase III trials (the<br />

RECORD programme) were undertaken to investigate<br />

the safety and efficacy of once-daily rivaroxaban for<br />

thromboprophylaxis after major orthopaedic surgery.<br />

<strong>The</strong> results of RECORD3 showed that rivaroxaban<br />

was more effective than enoxaparin 40 mg once daily<br />

after total knee replacement (TKR), with a 48% risk<br />

reduction in VTE and all cause mortality. RECORD4<br />

was designed to determine the efficacy and safety of 10<br />

mg rivaroxaban od compared to 30 mg bid enoxaparin<br />

after total knee replacement (TKR).<br />

Method: This study randomized 3148 patients to either<br />

rivaroxaban (10 mg od started 6–8 hours after surgery)<br />

or enoxaparin (30 mg bid s.c. started 12–24 hours after<br />

surgery) for 10–14 days. <strong>The</strong> primary efficacy outcome<br />

was the composite of asymptomatic deep vein thrombosis<br />

(DVT) detected by mandatory, bilateral venography<br />

and symptomatic DVT, non-fatal pulmonary embolism<br />

(PE), and all-cause mortality up to day 13±4. Secondary<br />

outcomes included major VTE (composite of proximal<br />

DVT, non-fatal PE, and VTE-related death) and symptomatic<br />

VTE. Safety outcomes included on-treatment<br />

major and non-major bleeding.<br />

Results: Rivaroxaban provided a 31% relative risk<br />

reduction in the incidence of the primary efficacy outcome<br />

when compared to enoxaparin (6.9% vs 10.1%,<br />

respectively; p=0.012). <strong>The</strong> corresponding rates for<br />

major VTE were 1.2% and 2.0%, respectively (p=0.124)<br />

and for symptomatic VTE were 0.7% and 1.2%, respectively<br />

(p=0.187). <strong>The</strong>re were no significant differences<br />

in bleeding incidence observed between rivaroxaban<br />

and enoxaparin (major bleeding: 0.7% vs 0.3%, respectively,<br />

p=0.110; clinically relevant non-major bleeding:<br />

2.6% vs 2.0%, respectively, p=0.279).<br />

Conclusion: Rivaroxaban 10 mg od is the first oral<br />

thromboprophylactic agent to significantly reduce the<br />

incidence of VTE after TKR compared to enoxaparin<br />

30 mg bid, with a similar, low rate of bleeding.<br />

160. ALIGNMENT TO ACCP PROPHYLAXIS<br />

GUIDELINES AND VTE OUTCOMES IN THR<br />

AND TKR PATIENTS<br />

Rita Selby 1 , Bijan Borah 2 , Heather McDonald 3 ,<br />

Joe Henk 2 , Mark Crowther 4 , Phil Wells 5<br />

1<br />

Division of Hematology, University of Toronto;<br />

2<br />

i3Innovus, Eden Prarie, USA; 3 Bayer Inc.,<br />

Canada; 4 Department of Hematology, St. Joseph’s<br />

Healthcare, Hamilton, Canada; 5 Division of<br />

Hematology and Clinical Epidemiology, Ottawa<br />

Health Research Institute/University of Ottawa<br />

Purpose: A retrospective database analysis was conducted<br />

to a) determine the extent to which the American<br />

College of Chest Physicians (ACCP) guidelines for<br />

VTE prophylaxis are followed after total hip replacement<br />

(THR) and total knee replacement (TKR) and b)<br />

evaluate the incidence of VTE for patients receiving and<br />

not receiving prophylaxis according to ACCP guidelines<br />

(‘ACCP’ and ‘non-ACCP’, respectively).<br />

Method: A claims database associated with a large US<br />

health plan was linked to the Premier database, which<br />

provides details of in-patient medication use. Patients<br />

≥18 years undergoing TKR/THR and enrolled in<br />

the health plan 90 days before and 90 days following<br />

discharge from hospitalization (or until death) were<br />

included. Patients were considered to have received<br />

ACCP-guideline prophylaxis if they: a) received<br />

LMWH, fondaparinux, or VKA following surgery b)<br />

initiated prophylaxis within one day of surgery (for<br />

THR patients) and c) were prescribed prophylaxis<br />

for a minimum of ten days, or until the occurrence of<br />

major bleeding, VTE, or death. In addition, the number<br />

of DVTs and PEs occurring in ACCP and non-ACCP<br />

patients was recorded.<br />

Results: Of the 30,644 eligible patients from the health<br />

plan, 3,497 patients were linked to the in-patient database.<br />

Except for geographic indicators, there were no<br />

significant differences in demographics or baseline comorbidities<br />

between those included and excluded from<br />

the final study sample. Of the 3,497 linked patients,<br />

1,395 (40%) received ACCP prophylaxis. <strong>The</strong> number<br />

of DVTs occurring in the ACCP and non-ACCP groups<br />

were 28 (2.01%) and 79 (3.76%), suggesting that non-<br />

ACCP patients were almost twice as likely as ACCP<br />

patients to have a DVT (p=0.0521). <strong>The</strong> number of PEs<br />

occurring in the ACCP and non-ACCP groups were 2<br />

(0.14%) and 25 (1.19%), respectively, suggesting that<br />

non-ACCP patients were 8.5 times more likely than<br />

ACCP patients to experience a PE (p15 and/or flexion


COA/CORS/CORA 277<br />

163. MINIMALLY-INVASIVE TOTAL<br />

KNEE ARTHROPLASTY: POTENTIAL<br />

COMPLICATIONS BASED ON A LARGE<br />

SINGLE-SURGEON COHORT REVIEW<br />

Michael G. Zywiel, Peter M. Bonutti*, Slif D.<br />

Ulrich, Mike S. McGrath, Michael A. Mont<br />

Rubin Institute for Advanced Orthopaedics;<br />

*Bonutti Clinic<br />

Purpose: Minimally-invasive total knee arthroplasty has<br />

generated tremendous recent interest, but there have<br />

been reports of complications resulting from these new<br />

techniques. However, most studies have less than one<br />

year follow-up and are anecdotal in their results. <strong>The</strong><br />

purpose of this study was to retrospectively evaluate a<br />

series of 1,000 consecutive minimally invasive total knee<br />

arthroplasties, to describe the clinical and radiographic<br />

complications of this technique, to determine whether<br />

complication rates changed with experience, and finally<br />

to compare complication rates to a match group operated<br />

by the same surgeon.<br />

Method: Clinical and radiographic data were reviewed<br />

for the first 1,000 consecutive minimally-invasive knees<br />

(820 patients) performed by a single surgeon with a<br />

minimum 2 year post-operative period. All patients<br />

were operated using the mid-vastus approach, with a<br />

mean incision length of 10 cm (range, 8 to 13 cm). Nine<br />

patients were lost prior to 2 year follow-up, leaving 990<br />

knees (811 patients) in this report with annual followup.<br />

After determining overall results, data was stratified<br />

into 100-knee groups to determine whether complication<br />

rates changed over time. Finally, complication rates<br />

were compared to a matched group of 50 knees operated<br />

by the same surgeon.<br />

Results: <strong>The</strong>re were a total of 45 clinical complications<br />

(4.5%), including 20 (2.0%) manipulations under<br />

anesthesia, 12 (1.2%) arthroscopies for painful crepitus,<br />

4 (0.4%) component changes, and 3 (0.3%) spacer<br />

changes. <strong>The</strong>re were 3 (0.3%) impending radiographic<br />

failures. Stratification of results showed a complication<br />

rate of 6% over the first 200 knees, and 1% over<br />

the subsequent 800 knees. In comparison, the matched<br />

group had a complication rate of 4% over mean followup<br />

of 3 years (range, 2 to 4).<br />

Conclusion: This large series with minimum two year<br />

follow-up describes many of the long-term complications<br />

of this technique. Additionally, this large series<br />

suggests that high volume knee surgeons may require<br />

as many as 200 minimally invasive arthroplasties before<br />

achieving optimum proficiency with this technique.<br />

Nonetheless, comparison to a matched group suggests<br />

that even with this learning curve, complication rates<br />

need not increase markedly during this period if surgeons<br />

remain attentive to the potential complications<br />

associated with minimally invasive techniques.<br />

164. INCIDENCE OF SPINAL EPIMORPH<br />

RELATED COMPLICATIONS IN TOTAL<br />

JOINT ARTHROPLASTY<br />

Aaron Bigham, James J. Howard, Sugantha<br />

Ganapathy<br />

University of Western Ontario<br />

Purpose: Spinal epimorph is commonly used as part of<br />

multimodal analgesia for patients undergoing total joint<br />

arthroplasty. Patients who receive spinal epimorph are<br />

at risk for certain post- operative complications. <strong>The</strong><br />

purpose of this study was to determine the incidence of<br />

complications in patients undergoing total joint arthroplasty<br />

with administration of spinal epimorph compared<br />

to patients undergoing the same procedure who did not<br />

receive spinal epimorph as part of their analgesia.<br />

Method: A retrospective chart review of 72 patients in<br />

which two age, sex and procedure matched groups were<br />

compared for differences in known spinal epimorph complications.<br />

One group received spinal epimorph as part<br />

of their analgesia while the comparative group did not<br />

receive spinal epimorph but rather continuous infusion<br />

of local analgesia. Data extraction involved foley and<br />

oxygen usage, documented nausea/vomiting and puritis,<br />

associated risk factors and complications such as prostate<br />

disease and urinary tract infections, and secondary outcome<br />

measures such as Gravol and Benadryl usage.<br />

Results: Comparison of the two well-matched groups demonstrated<br />

that patients who received spinal epimorph had<br />

increased rates of foley insertion (p=0.0026), foley duration<br />

(p=0.015), oxygen usage (p=0.0053), documented<br />

puritis (p=0.0006) and Benadryl usage (p=0.0053). Trends<br />

towards increased nausea/vomiting (p=0.17), antiemetic<br />

use (p=0.16) and urinary tract infections (p=0.15) were<br />

seen in the spinal epimorph group, although these differences<br />

did not reach statistical significance.<br />

Conclusion: <strong>The</strong> use of spinal epimorph as part of an analgesia<br />

protocol surrounding total joint arthroplasty is associated<br />

with increased complications and patient discomfort<br />

when compared to patients who received continuous local<br />

infiltrative analgesia. Further research is needed to determine<br />

if alternate modes of analgesia provide adequate pain<br />

relief when compared with spinal epimorph and thus provide<br />

suitable alternatives with less complications.<br />

COA PAPER SESSION 18:<br />

SPORTS LOWER EXTREMITY<br />

165. RETURN TO SPORT FOLLOWING<br />

SINGLE-BUNDLE ACL RECONSTRUCTION:<br />

PATELLAR TENDON AUTOGRAFT VERSUS<br />

HAMSTRING AUTOGRAFT<br />

Randy Mascarenhas*, Michael Tranovich † , Eric J.<br />

Kropf, James Irrgang, Freddie H. Fu, Christopher<br />

D. Harner<br />

*University of Manitoba; † Lake Erie College of<br />

Osteopathic Medicine; University of Pittsburgh<br />

Purpose: This study sought to examine return to sports<br />

in athletes younger than 25 following ACL reconstruction<br />

with either patellar tendon (PT) or hamstring (HS)<br />

autografts using a matched-pairs case-control experimental<br />

design.<br />

Method: Twenty-three matched pairs were obtained<br />

based on gender (56.5% Female), age (18.3±2.5yrs PT<br />

vs.17.6±2.6 HS), and length of follow-up (4.7±2.1yrs<br />

PT vs. 4.2±1.6 HS). All patients reported participating<br />

in very strenuous (soccer, basketball etc.) or strenuous<br />

(skiing, tennis etc.) sporting activity 4-7 times/ week<br />

prior to their knee injury. Outcomes other than return to<br />

play included the IKDC, SAS, ADLS, SF-36, knee range<br />

of motion, laxity, and hop/jump testing.<br />

Results: Most patients in both groups were able to participate<br />

in very strenuous or strenuous sporting activity<br />

at follow-up [18 (78.3%) PT vs. 19 (82.6%) HS]. However,<br />

only 13 (56.5%) of the patellar tendon subjects<br />

and 10 (43.5%) of the hamstrings patients were able to<br />

return to pre-injury activity levels in terms of frequency<br />

and type of sport (p=.63). Hamstrings patients showed<br />

higher ADLS (p


278 COA/CORS/CORA<br />

168. SOCCER AND ACL RECONSTRUCTION<br />

IN TEENAGE GIRLS<br />

Krish Maragh, Lauren Beaupré, Allyson Jones,<br />

David Otto<br />

University of Alberta<br />

Purpose: Females are at greater risk for anterior cruciate<br />

ligament (ACL) injury than males. Soccer may be a significant<br />

risk factor for ACL injury in adolescent females.<br />

ACL injury has significant consequences, including early<br />

onset of osteo-arthritis. <strong>The</strong> purpose of the study was<br />

to determine a) the number of ACL reconstruction surgeries<br />

performed on females between the ages of 13-<br />

18 inclusive in the Capital Health (CH) region from<br />

December 2000 to November 2005, and b) those due to<br />

soccer injuries. Secondly, we describe factors relating to<br />

the mechanism of injury.<br />

Method: Utilizing regional administrative data, we<br />

performed a standardized chart review and telephone<br />

interviews with female adolescents who underwent<br />

ACL reconstruction in the aforementioned time period.<br />

Information gathered included: a) Age at reconstruction<br />

procedure. b) Indoor versus outdoor soccer playing<br />

surface. c) Level of play and frequency of participation.<br />

<strong>The</strong> Alberta Soccer Association provided the number of<br />

registrants in indoor and outdoor seasons over the same<br />

time period.<br />

Results: 2,824 ACL reconstruction operations were performed<br />

between December 2000 and November 2005.<br />

Reconstructions in females took place at an earlier age<br />

than in males. <strong>The</strong>re were 266 ACL reconstructions in<br />

256 adolescent females, of which 253 charts were available<br />

for review. One hundred and eleven (44%) knees<br />

were injured during soccer play. Seven patients with<br />

eight ACL reconstructions who injured their ACL playing<br />

soccer could not be located. ACL injuries occurred<br />

during indoor soccer in 52 (51%) cases despite higher<br />

registration in outdoor soccer during the same timeframe.<br />

77 (74%) subjects played competitively and 81<br />

(79%) subjects played two or more times/week.<br />

Conclusion: Approximately 10% of ACL reconstructions<br />

were performed on adolescent females. Nearly<br />

50% of ACL injuries occurred during soccer play, with<br />

a similar number seen in indoor versus outdoor play.<br />

Study limitations include the use of administrative data<br />

to assess the number of ACL reconstructions rather than<br />

ACL injuries. An awareness of the propensity of knee<br />

injuries in female soccer players is important. With the<br />

increased participation of young females in soccer and<br />

the serious lifelong implications of ACL rupture, prevention<br />

and training should be improved to lower the<br />

incidence of injury.<br />

169. RELIABILITY OF TUNNEL ANGLE<br />

IN ACL RECONSTRUCTION: FREE HAND<br />

VERSUS MECHANICAL GUIDE TECHNIQUE<br />

Jeff Leiter, Nevin de Korompay, Lindsey<br />

MacDonald, Carling MacDonald, Warren G.<br />

Froese, Peter B MacDonald<br />

Pan Am Clinic<br />

Purpose: <strong>The</strong> increasing number of ACL reconstructions<br />

has led to the introduction of new techniques irrespective<br />

of the fact optimal tunnel angle placement has yet to<br />

be established. Improper tunnel angle placement is associated<br />

with a variety of complications including graft<br />

failure. <strong>The</strong> purpose of this retrospective study was to<br />

compare the reliability of tibial tunnel angles produced<br />

by two experienced surgeons using a free hand method<br />

or mechanical guide (HowellTM 65° Tibial Guide).<br />

Method: Tibial tunnel angles in the coronal and sagittal<br />

planes were determined from anteroposterior and lateral<br />

radiographs, respectively, taken at 2 to 6 months<br />

postoperatively. Fifty-two sets of digital radiographs<br />

were analyzed (free hand = 28, mechanical = 24) with<br />

the knee in full extension 100 cm from the beam source.<br />

Tunnel angle measurements were calculated using NIH<br />

ImageJ software. Each angle was measured by two<br />

investigators on three separate occasions with minimum<br />

7 days between each analysis.<br />

Results: <strong>The</strong>re was a significant difference (p 0.75) to<br />

moderate (0.75 –0.40), respectively.<br />

Conclusion: Tibial tunnel angles in the coronal plane<br />

produced with a mechanical guide are more accurate<br />

than those drilled free hand when the intended angle<br />

of placement is 65°. <strong>The</strong> method used to measure tibial<br />

angles in this study was reliable within and between<br />

investigators. Further research will be conducted to<br />

investigate the correlation between tunnel angle placement<br />

and patient outcome measures.<br />

170. THE OTTAWA TREATMENT PROTOCOL<br />

FOR SEPTIC ARTHRITIS FOLLOWING ACL<br />

RECONSTRUCTION<br />

Emilio Lopez-Vidriero, Olufemi R. Ayeni, Tracy<br />

Rupke, Ahmad Bin Nasser, Donnald Johnson<br />

University of Ottawa<br />

Purpose: To present our clinical and quality of life<br />

outcomes after one year treatment with our protocol<br />

including graft retention.<br />

Method: Seventeen of 1, 847 patients who underwent<br />

ACL reconstruction surgery were identified as infected<br />

following retrospective chart review in our institution<br />

(University of Ottawa) from 1995 to 2005. Mean age<br />

was 37 years old (range 18-56). Gender ratio was 14<br />

male/ three female. Laterality 13 left/ four right knees.<br />

<strong>The</strong> diagnosis was achieved by clinical suspicion and<br />

serum markers (ESR, CPR, WBC) followed with aspiration<br />

and culture of intraarticular liquid. After that,<br />

our treatment protocol included IV antibiotics (empiric<br />

and culture guided) and knee arthroscopy performing<br />

debridement and lavage with 12L of saline irrigation<br />

as well as graft retention when possible. Clinical and<br />

Radiographic data were collected at a minimum of<br />

one year follow-up (IKDC, KT-1000, Lysholm, SF-12,<br />

Tegner, Cybex Strength testing).<br />

Results: <strong>The</strong> Incidence of septic arthritis following ACL<br />

reconstruction was 0, 92%. Bacteriology results were as<br />

follows: seven staphylococcus aureus, four Propionibacterium<br />

acnes, one Klebsiella oxytoca, five no growth.<br />

<strong>The</strong> time from ACL reconstruction to first symptoms of<br />

infection was 37 days (range 4-63). <strong>The</strong> time from symptoms<br />

to arthroscopic debridement was 5, 5 days (range<br />

0-33). Serology markers were as follows: ESR mean 69<br />

(range 23-128), CPR 136 (50-387), WBC 10.3 (6-15).<br />

<strong>The</strong> average follow up was 41 months (range 12-85).<br />

Sixteen of 17 grafts were retained after 1.53 procedures<br />

per patient. Antibiotic treatment lasted 5.2 weeks on<br />

average. <strong>The</strong> average clinical outcomes were: Lysholm<br />

77 (44-98), IKDC 74 (46-95), Quadriceps strength 71<br />

% of non operative side, and Hamstrings strength 70 %<br />

of non operative side, KT -1000 side to side difference<br />

1 mm (-3 to 4). Degenerative radiological changes were<br />

noted in three patients. Two later repeat reconstructions<br />

occurred due to instability (11,1%). No delayed recurrence<br />

of infection was noted.<br />

Conclusion: ACL reconstruction is a safe procedure<br />

being the incidence of septic arthritis in our series of 0,<br />

92%. Our protocol of treatment is clinically effective<br />

and allows for a stable knee in 88, 9% of the patients<br />

avoiding the appearance of a delayed infection.<br />

171. AUTOMATIC CLASSIFICATION OF 3D<br />

KINETIC DATA OF HEALTHY AND ACL<br />

DEFICIENT PARTICIPANTS<br />

Alexandre Fuentes*, Neila Mezghani, Nicola<br />

Hagemeister, Jacques A. de Guise<br />

*Université de Montreal; École de Technologie<br />

Superieure<br />

Purpose: Gait analysis has become an innovative<br />

approach to assess the biomechanical adaptations due to<br />

an ACL injury. However, interpreting the large amount<br />

of data collected often requires an expert. <strong>The</strong>refore,<br />

there is a need to develop an automatic method capable<br />

to distinguish kinetic pattern of an ACL deficient<br />

patients from an asymptomatic population.<br />

Method: 26 ACL deficient patients and 30 asymptomatic<br />

participants took part in a treadmill gait analysis.<br />

3D ground reaction forces (vertical, medio-lateral and<br />

anterior-posterior) were collected using the ADAL 3D<br />

treadmill. Features were extracted from the 3D ground<br />

reaction forces as a function of time and then classified<br />

by the nearest neighbour rule using a wavelet decomposition<br />

method. <strong>The</strong> classification method was tested on<br />

our data base of 56 participants.<br />

Results: <strong>The</strong> proposed classification method obtained<br />

an accuracy of 90%. <strong>The</strong> classification accuracy per<br />

class was higher for the ACL deficient group allowing<br />

classifying correctly 25 out of 26 ACL deficient patient.<br />

25 out of the 30 asymptomatic participants were properly<br />

classified.<br />

Conclusion: This study shows that an automatic objective<br />

computer method could be used in a clinical setting<br />

to help diagnose an anterior cruciate ligament injury<br />

during a gait analysis evaluation. Future studies should<br />

apply this method on a larger database including data<br />

from patients with other musculoskeletal pathologies to<br />

help diagnose other injuries.<br />

172. A SURVEY STUDY REGARDING THE<br />

NATURAL HISTORY AND TREATMENT OF<br />

ANTERIOR CRUCIATE LIGAMENT INJURY<br />

AMONG MEMBERS OF THE CANADIAN<br />

ORTHOPAEDIC ASSOCIATION<br />

Sheila McRae, Jaskarnip Chahal, Jeff Leiter, Peter<br />

B. MacDonald, Robert Marx*<br />

Pan Am Clinic; *Hospital for Special Surgery<br />

Purpose: To describe the current practices and opinions<br />

of members of the Canadian Orthopaedic Association<br />

(COA) pertaining to anterior cruciate ligament (ACL)<br />

reconstruction.<br />

Method: All orthopaedic surgeon members of the COA<br />

residing in Canada were sent an email invitation to take<br />

part in a survey via an internet-based survey manager.<br />

Expanding on a previously published survey (Marx<br />

et al., 2003), the current survey was comprised of 30<br />

questions regarding the natural history of ACL-injured<br />

knees, surgical and post-surgical treatment choices, and<br />

success of the surgery. Clinical agreement was defined<br />

as greater than 80% agreement in choice of response<br />

option.<br />

Results: Two hundred and eighty-three surgeons<br />

(49.3%) responded to the survey. Responses of the 143<br />

surgeons (50.5%) who indicated they performed ACL<br />

reconstruction in the last year are presented. Clinical<br />

agreement with respect to surgical technique was demonstrated<br />

with respect to only three practices – ipsilateral<br />

graft harvest (100%), single incision approach (86.1%),<br />

and manual graft tensioning (81.6%). In terms of natural<br />

history, the only area of agreement was that hamstring<br />

and quadriceps strength affects function in ACL<br />

deficient knees (92%). Although less than the clinical<br />

agreement threshold, a majority of surgeons indicated<br />

their preference for semitendinosis-gracilis autograft<br />

(73%), transtibial versus anteromedial portal for establishment<br />

of the femoral tunnel (65 versus 29%), notchplasty<br />

(65% only with impingement) and promotion of<br />

full weight-bearing and range of motion immediately<br />

post-surgery (72.1 and 74.8%, respectively). <strong>The</strong> most<br />

frequent surgeon-reported complications were tunnel<br />

widening (9.8%) and graft failure (4.4%). A greater<br />

proportion of high-volume surgeons used a manual tensioning<br />

device intra-operatively and permitted earlier<br />

return to sport (p


COA/CORS/CORA 279<br />

173. MANAGEMENT OF MULTIPLE<br />

LIGAMENT INJURED KNEE: A SURVEY OF<br />

CURRENT PRACTICE IN CANADA<br />

Robert G. McCormack, Rafael Martinez, Tom<br />

Herschmiller*, Kian Chung †<br />

UBC; *University of Sidney; † University of<br />

Singapore<br />

Purpose: Treatment of knee dislocations remains challenging<br />

and controversial. Several strategies for the management<br />

of multiple ligament knee injuries have been<br />

described and there are multiple unresolved questions.<br />

<strong>The</strong>se include the indications for surgery, repair versus<br />

reconstruction, surgical timing and graft selection. <strong>The</strong><br />

aim of this survey was to identify areas of agreement<br />

and controversy, to define the current standard of care<br />

and help generate research questions.<br />

Method: Using the standard techniques for survey development<br />

we presented six clinical scenarios of acute knee<br />

dislocations (at least three ligaments) to all orthopaedic<br />

surgeons affiliated with Canadian medical schools. <strong>The</strong><br />

scenarios were designed to cover the common combinations<br />

of knee dislocations in both a 25 year old active<br />

individual and a 50 year old sedentary individual. <strong>The</strong><br />

responses were divided into three groups. Group A<br />

consisted of those with fellowship training, or practices<br />

focused on, sports knee. Group B was comprised of surgeons<br />

with trauma fellowship or a sub-specialty trauma<br />

practice. Group C were the remaining surgeons without<br />

these subspecialty foci. We report on the responses of<br />

groups A and B.<br />

Results: An average of ten different treatment algorithms<br />

were reported for each scenario but there was<br />

agreement on the need for early surgical management<br />

in the young active individual with a knee dislocation<br />

and all dislocations involving a lateral sided injury.<br />

Conversely, there was a lack of consensus regarding the<br />

need for surgical treatment of the 50 year old individual<br />

with bi-cruciate injury and medial collateral ligament.<br />

<strong>The</strong> most common combinations of reconstruction and<br />

repair are reported for each clinical scenario. Surgeons<br />

strongly favor early surgical intervention, within three<br />

weeks of injury, except for the 50 year old sedentary<br />

individual with a bi-cruciate plus medial sided injury.<br />

Allograft was the most popular choice to reconstruct the<br />

PCL and lateral ligament. For the ACL graft there was<br />

a near equal distribution between hamstrings, patellar<br />

tendon and the use allograft tissue.<br />

Conclusion: In the absence of higher level evidence, the<br />

information from this survey helps define the standard<br />

of care in Canada and identifies areas of controversy<br />

which would be a priority for a multi-centre prospective<br />

trial.<br />

174. POSTEROMEDIAL CORNER INJURY<br />

PATTERNS IN TRAUMATIC KNEE<br />

DISLOCATIONS<br />

Jaskarndip Chahal, Dawn Pearce*, Tom<br />

McCarthy, Jeff Dawson, Anthea Liebenberg*,<br />

Daniel B. Whelan*<br />

University of Toronto; *St. Michael’s Hospital,<br />

University of Toronto<br />

Purpose: Traumatic knee dislocations are complex injuries.<br />

A thorough knowledge of the pattern of ligament<br />

damage is essential to plan for definitive treatment. Injuries<br />

to the structures of the posteromedial corner (posterior<br />

oblique ligament, the semi-membranosus tendon<br />

and its expansions, the meniscofemoral and meniscotibial<br />

ligaments, posterior horn of the medial meniscus and<br />

posterior reflections of the deep and superficial medial<br />

collateral ligament) can contribute to rotational laxity<br />

and have not been previously described in the setting of<br />

knee dislocations. We set out to evaluate the injury patterns<br />

of the posteromedial corner in a series of traumatic<br />

knee dislocations.<br />

Method: A radiographic (MRI) study was conducted on<br />

22 multi-ligament knee injuries in 22 patients, managed<br />

at a level one trauma centre by a single surgeon between<br />

July 2006 and May 2008.<br />

Results: <strong>The</strong>re were 14 male and 6 female patients with<br />

an average age of 29. <strong>The</strong> mechanism of injury was high<br />

velocity in 10 cases, and low velocity in 12 cases. One<br />

case (ACL/PCL/posterolateral corner injury) was associated<br />

with a vascular injury. While the incidence of MCL<br />

injury was 59%, injury specific to the PMC was evident<br />

on MRI in 81.8% (18/22) of cases. <strong>The</strong> deep and superficial<br />

MCL were involved in 13/18 and 12/18 PMC injuries,<br />

respectively. <strong>The</strong> posterior oblique ligament was<br />

injured in 11/18 cases and the semi-membranosus and<br />

its expansions were injured in 10/18 cases (all distal).<br />

Injuries to the posterior horn of the medial meniscus<br />

(9/18 cases) were associated with a tear of the meniscofemoral/meniscotibial<br />

ligaments in all cases.<br />

Conclusion: Injury to the PMC was common with highgrade<br />

multi-ligament knee injuries in this series. PMC<br />

injuries were more common than MCL injuries alone.<br />

Injury to the posterior horn of the medial meniscus was<br />

predictive of more extensive PMC injury in all cases.<br />

<strong>The</strong> posteromedial corner of the knee is an under recognized<br />

area where important injuries can occur in the setting<br />

of a dislocatable knee. Future research will involve<br />

the correlation of the injury patterns described in this<br />

series to clinical measures of knee instability and laxity.<br />

COA PAPER SESSION 19: TRAUMA UPPER<br />

EXTREMITY AND TRAUMA GENERAL<br />

175. EVALUATION OF PRIMARY CARE<br />

MANAGEMENT FOR ISOLATED LIMB<br />

INJURY: STUDY ON 166 CONSECUTIVE<br />

PATIENTS REFERRED TO ORTHOPAEDIC<br />

SURGERY IN A LEVEL 1 TRAUMA CENTER<br />

Dominique Rouleau, Stefan Parent, Debbie<br />

Feldman*, Valérie Deslauriers*<br />

Hôpital Sacré-Coeur de Montreal; *Université de<br />

Montreal<br />

Purpose: Musculoskeletal injuries affect up to 13% of<br />

adults annually. Despite this high incidence, quality of<br />

primary care, including analgesia, may be sub-optimal.<br />

<strong>The</strong> goal of this study is to describe the quality of primary<br />

care for ambulatory patients with isolated limb<br />

injury and to identify related factors.<br />

Method: A cross sectional study was undertaken on 166<br />

consecutive ambulatory adult patients with isolated limb<br />

injury who presented to orthopedics service in a Level<br />

one Trauma Centre. Quality of care was assessed by<br />

evaluating analgesia, walking aids, immobilization, and<br />

quality of referral diagnosis according to actual expert<br />

recommendations. Patient satisfaction was assessed by<br />

Visit Satisfaction Questionnaire.<br />

Results: This study revealed low quality of primary care<br />

for more than 50% of injured patients. More than half<br />

the patients had pain level over 5/10 and more than a<br />

quarter had insufficient/absent analgesia prescriptions.<br />

A third had unacceptable immobilization and 36% of<br />

patients with a lower limb injury did not receive a walking<br />

aid prescription. A total of 37% had an absent or<br />

inadequate referral diagnosis. Factors associated with<br />

lower quality depended on the specific quality indicator<br />

and included: living further away from the hospital,<br />

younger age, initially consulting at another health care<br />

center, having a fracture, and being a smoker.<br />

Conclusion: <strong>The</strong> high frequency of low quality of care<br />

underlines the necessity for orthopedic surgeons to be<br />

involved in primary care education. Identifying factors<br />

associated with lower quality of care will orient efforts<br />

to improve medical care of patients with isolated traumatic<br />

injury.<br />

176. PATIENT SATISFACTION FOLLOWING<br />

CLAVICLE FRACTURE FIXATION:<br />

HORIZONTAL VERSUS VERTICAL INCISION<br />

John Choi, Elton Edwards<br />

<strong>The</strong> Alfred Hospital<br />

Purpose: To document outcomes and patient satisfaction<br />

in relation to the incision used following clavicle<br />

fracture fixation. In literature, the incidence of incisional<br />

numbness following operative fixation of clavicle<br />

fractures is reported to be between 7-29%. Such wound<br />

related problems contribute significantly to the dissatisfaction<br />

of patients with operatively treated clavicle fractures.<br />

Wound related problems can be bothersome and<br />

disabling and this is poorly documented.<br />

Method: All primary clavicle fractures treated with<br />

plating at the Alfred Hospital between 01/06/2003 and<br />

01/06/2006 were included in the study. Patients were<br />

asked to complete paper-based questionnaires assessing<br />

satisfaction, pain, scar satisfaction, presence of numbness<br />

and the degree of disability following clavicle fixation.<br />

<strong>The</strong>ir clinical notes and X-rays were reviewed for<br />

evaluation. <strong>The</strong> study sample was then divided into two<br />

groups; horizontal incision versus vertical incision then<br />

the data was analysed.<br />

Results: <strong>The</strong> response rate was 65% (35/54). 74% of<br />

patients reported as having “good” or better outcome<br />

following their clavicle fracture fixation. <strong>The</strong>re was no<br />

statistically significant difference in pain scores. However,<br />

there were statistically significant differences observed in<br />

the presence of numbness (vertical 21% versus horizontal<br />

62%) and the disability from the numbness between the<br />

two incision types. Overall satisfaction between the two<br />

groups was also significantly different.<br />

Conclusion: This study confirms that scar-related problems<br />

significantly affect the satisfaction following plating<br />

of clavicle fractures and numbness appears to be one<br />

of the most significant factors. Vertical incisions appear<br />

to reduce the incidence of numbness and lead to better<br />

patient satisfaction. Our results suggest that vertical<br />

incision is an attractive alternative approach in clavicle<br />

fracture fixation.<br />

177. LOCKED PLATE FIXATION VERSUS<br />

NON-OPERATIVE TREATMENT FOR<br />

DISPLACED, EXTRA-ARTICULAR PROXIMAL<br />

HUMERUS FRACTURES: ARE FUNCTIONAL<br />

AND QUALITY OF LIFE OUTCOMES<br />

BETTER?<br />

Gerard Slobogean, Mohit Bhandari*, Peter J.<br />

O’Brien<br />

UBC; *McMaster University<br />

Purpose: To compare the functional outcome and quality-of-life<br />

following a displaced extra-articular proximal<br />

humerus fracture treated with open reduction and locking<br />

plate fixation versus non-operative management. To<br />

provide preliminary data for a subsequent prospective<br />

clinical trial.<br />

Method: Eligible subjects were identified through<br />

retrospective searches of a large emergency department<br />

admission database and the orthopaedic trauma<br />

database. All subjects ages ³ 55 treated for a proximal<br />

humerus fracture between 2002 to 2005 were invited to<br />

participate. <strong>The</strong> Disabilities of Arm, Shoulder, and Hand<br />

(DASH), Health Utilities Index Mark 3 (HUI), Euroqol-<br />

5D (EQ-5D), and the SF-36 questionnaires were mailed<br />

to all eligible subjects. Initial radiographs were reviewed<br />

using the AO/OTA classification system. Only patients<br />

with A3, B1, B2, or B3 fractures were included.<br />

Results: Thiry-four subjects were included: 15 were<br />

treated with sling immobilization and 19 with locked<br />

plate ORIF. <strong>The</strong> non-operative group was approximately<br />

seven years older (mean age 74 versus 67, p = 0.046).<br />

DASH scores were similar between the groups: ORIF<br />

26.6 ± 24 and Sling 26.5 ± 20. <strong>The</strong> 95% CI surrounding<br />

the 0.01 point difference (-16.0 to 15.9) slightly exceeds<br />

the 13 point cutoff for the instrument’s measurement<br />

error (minimal detectable change). Using univariable<br />

analysis, no statistically significant differences in health<br />

state values were detected. <strong>The</strong> mean HUI value for the<br />

ORIF group was 0.68 versus 0.75 for the sling (p=0.48).<br />

Mean EQ-5D values were 0.77 for the ORIF group<br />

and 0.80 for the sling group (p=0.73). <strong>The</strong> SF-36 PCS<br />

scores were also similar between the two groups: ORIF<br />

41.1 versus Sling 39.8 (p=0.77). When controlling for<br />

age and pre-injury function, a 0.09 point difference in<br />

HUI values was detected favouring the sling treatment<br />

(p=0.036). No differences in DASH, EQ-5D, or SF-36<br />

PCS scores were detected using regression models.<br />

Conclusion: <strong>The</strong> results of this small cohort suggest, for<br />

extra-articular fractures, the functional and quality of<br />

life outcomes may be similar between the two interven-<br />

J BONE JOINT SURG [BR] 2011; 93-B:SUPP III


280 COA/CORS/CORA<br />

tions. No trial comparing locked plate fixation and nonoperative<br />

management has been reported. A total of 96<br />

subjects will be needed for a prospective clinical trial<br />

comparing the two treatments (DASH difference 15,<br />

80% power, 0.05 two-sided alpha).<br />

178. TYPE 2D MONTEGGIA ELBOW<br />

FRACTURE-DISLOCATIONS: PATTERN<br />

OF INJURY, SURGICAL TECHNIQUE OF<br />

FIXATION AND OUTCOME<br />

Darius Viskontas*, Daphne M. Beingessner, Sean<br />

Nork, Julie Agel<br />

* UBC; University of Washington<br />

Purpose: To describe the pattern of injury, surgical technique<br />

and outcomes of Monteggia type IID fracture<br />

dislocations.<br />

Method: Design: Retrospective review of prospectively<br />

collected clinical and radiographic patient data in<br />

orthopaedic trauma database with prospectively collected<br />

outcome scores. Setting: Level 1 university based<br />

trauma center. Patients / Participants: All patients with<br />

Monteggia type IID fracture dislocations admitted from<br />

January 2000 to July 2005. Intervention: Review of<br />

patient demographics, fracture pattern, method of fixation,<br />

complications, additional surgical procedures, and<br />

clinical and radiographic outcome measures. Main Outcome<br />

Measurements: Clinical outcomes: elbow range<br />

of motion, QuickDASH (Quick Disabilities of the Arm,<br />

Shoulder and Hand), PREE (Patient Rated Elbow Evaluation),<br />

complications. Radiographic outcomes: quality<br />

of fracture reduction, healing time, degenerative change<br />

and heterotopic ossification.<br />

Results: Sixteen patients were included in the study.<br />

All fractures united. <strong>The</strong>re were seven complications<br />

in 6 patients including 3 contractures with associated<br />

heterotopic ossification, 1 pronator syndrome and late<br />

radial nerve palsy, 1 radial head collapse and a DVT in<br />

the same patient and 1 with prominent hardware. Outcome<br />

scores were obtained on 11 patients at an average<br />

of 49 months (range 25 – 82 months) post-operatively.<br />

<strong>The</strong> average Quickdash score was 11 (range 0-43) and<br />

the average PREE score was 13 (range 0-34).<br />

Conclusion: Monteggia IID fracture dislocations are<br />

complex injuries with a recurring pattern. Rigid anatomic<br />

fixation, early range of motion and avoidance of<br />

complications leads to a good outcome.<br />

179. ELBOW HEMIARTHROPLASTY FOR<br />

DISTAL HUMERAL FRACTURES<br />

Alex Malone*, Peter Zarkadas † , Stuart Jansen,<br />

Jeff Hughes<br />

*University of Otago, Christchurch; † Lions Gate<br />

Hospital; Royal North Shore, Sydney<br />

Purpose: This study reviews the early results of elbow<br />

hemiarthroplasty for distal humeral fractures.<br />

Method: Elbow hemiarthroplasty was performed on<br />

30 patients (mean 65 years; 29-91) for unreconstructable<br />

fractures of the distal humerus or salvage of failed<br />

internal fixation. A ‘triceps on’ approach was used in<br />

six and an olecranon osteotomy in 24. A Sorbie Questor<br />

prosthesis (Wright Medical Technology) was used in 14<br />

patients and a Lattitude (Tornier) in 16. Clinical review<br />

at a mean of 25 months (3-88) included the American<br />

Shoulder and Elbow Surgeons elbow outcomes instrument<br />

(ASES), Mayo Elbow Performance Index (MEPI)<br />

and radiological assessment.<br />

Results: At follow up of 28 patients mean flexion deformity<br />

was 25 degrees, flexion 128 degrees, range of pronosupination<br />

165 degrees, mean ASES 83, MEPI 77 and<br />

satisfaction 8/10. Acute cases scored better than salvage<br />

cases. Re-operation was required in 16 patients (53%);<br />

two revisions to a linked prosthesis for periprosthetic<br />

fracture and aseptic loosening at 53 and 16 months, 12<br />

metalwork removals and four ulnar nerve procedures.<br />

Posterolateral rotatory instability was present in one<br />

elbow, four had laxity and pain on loading (two with<br />

prosthesis or pin loosening), four had laxity associated<br />

with column fractures (two symptomatic) and ten had<br />

asymptomatic laxity only. <strong>The</strong> triceps on approach had<br />

worse laxity and clinical scores. Uncomplicated union<br />

occurred in all olecranon osteotomies and 86% of<br />

column fractures. One elbow had an incomplete cement<br />

mantle and seven had lucencies >1 mm; one was loose<br />

but acceptable. Five prostheses were in slight varus and<br />

two were flexed. Two elbows had early degenerative<br />

changes and 15 developed an osteophytic lip on the<br />

medial trochlea.<br />

Conclusion: Early results of elbow hemiarthroplasty<br />

show good outcomes after complex distal humeral fractures,<br />

despite a technically demanding procedure, metalware<br />

removal in 40%, symptomatic laxity in 12% and<br />

column non-union in 8%. Better results are obtained for<br />

treatment in the acute setting and with use of an olecranon<br />

osteotomy.<br />

180. COMPARTMENT SYNDROME CAUSES A<br />

SYSTEMIC INFLAMMATORY RESPONSE AND<br />

REMOTE ORGAN INJURY<br />

Abdel-Rahman Lawendy, Gregory McGarr, Joel<br />

Phillips, David W. Sanders, Aurelia Bihari, Amit<br />

Badhwar<br />

University of Western Ontario<br />

Purpose: Severe compartment syndrome is associated<br />

with renal failure, end organ damage, and systemic<br />

inflammatory response syndrome (SIRS). Intravital<br />

videomicroscopy (IVVM) is a useful tool to study capillary<br />

perfusion and inflammation in end organs such<br />

as the liver and lungs. In this study, the systemic effect<br />

of hindlimb compartment syndrome was studied using<br />

hepatic IVVM. <strong>The</strong> purpose was to measure the effect<br />

of increased hindlimb intracompartmental pressure on<br />

hepatocyte viability, inflammation, and blood flow in a<br />

rodent model.<br />

Method: Ten Wistar rats were randomised into control<br />

(C) and Compartment Syndrome (CS) groups. Animals<br />

were anaesthetized with 5 % isoflurane. Mean<br />

arterial pressure was monitored using a carotid artery<br />

catheter. Elevated intracompartmental pressure (EICP)<br />

was induced by saline infusion into the anterior compartment<br />

of the hind limb and maintained for 2 hours<br />

between 30–40mmHg in the CS group. Two hours following<br />

fasciotomy, the liver was analyzed using IVVM<br />

to quantify capillary perfusion as a measure of microvascular<br />

dysfunction. <strong>The</strong> numbers of adherent and<br />

rolling leukocytes in venules and sinusoids were quantified<br />

to measure the inflammatory response. Irreversible<br />

hepatocyte injury was measured using a fluorescent vital<br />

dye which labels the nuclei of severely injured cells.<br />

Results: Hepatocellular injury was significantly higher<br />

in the CS group (325±103 PI labeled cells/10-1 mm2)<br />

compared to controls (30±12 PI labeled cells/10-1<br />

mm2)(p=0.0087). <strong>The</strong> number of adherent venular<br />

white blood cells (WBC) was significantly higher for the<br />

CS group (5±2/hpf) than controls (0.2±0.2)(p=0.0099).<br />

Volumetric blood flow was not significantly different<br />

between CS and controls.<br />

Conclusion: After only 2 hours of compartment syndrome<br />

in this animal model, the number of activated white<br />

blood cells increased 25-fold and liver cellular injury<br />

increased 10-fold compared to controls. Marked systemic<br />

inflammation and hepatocellular damage was detected in<br />

response to isolated limb compartment syndrome. Compartment<br />

syndrome is a low-flow ischemia/reperfusion<br />

injury with a profound inflammatory response. Further<br />

research into the severe end-organ damage associated<br />

with compartment syndrome is required.<br />

181. HYPOTHERMIA IN COMPARTMENT<br />

SYNDROME<br />

David W. Sanders, Gladys Chan, Amit Badhwar<br />

University of Western Ontario<br />

Purpose: Compartment syndrome is a limb-threatening<br />

condition. Treatment is urgent decompression by fasciotomy.<br />

However, orthopedic surgeons are often confronted<br />

by a limb at risk for compartment syndrome,<br />

in which treatments to preserve tissue might be considered.<br />

Hypothermia has shown promise as a technique<br />

of maintaining tissue viability in transplant surgery,<br />

replant surgery and soft tissue injury. Cooling reduces<br />

microvascular dysfunction, inflammation and edema.<br />

This study was designed to determine whether tissue<br />

cooling might reduce muscle damage in the setting of<br />

elevated intracompartmental pressure. Purpose This<br />

study investigated the effect of hypothermia on tissue<br />

perfusion, viability and the inflammatory response in an<br />

animal model of elevated intracompartmental pressure.<br />

We hypothesize that hypothermia will preserve muscle<br />

tissue viability in an animal model of elevated intracompartmental<br />

pressure.<br />

Method: Twenty Wistar rats were randomized. Five<br />

animals had elevated intracompartmental pressure for<br />

2 hours (CS). Five had elevated pressure and hindlimb<br />

cooling to 25oC (CS-HY). Five had hindlimb cooling<br />

to 25oC (HY) and 5 were control animals (C). All animals<br />

were anaesthesized for study. Core temperature was<br />

maintained over 30oC. Elevated ICP was maintained<br />

(30mmHg) using a saline infusion technique (groups CS<br />

and CS-HY). After 2 hours, fasciotomies were completed<br />

and intravital microscopy was used to measure tissue<br />

viability, microvascular perfusion and inflammation.<br />

Results: <strong>The</strong> use of hypothermia reduced tissue damage<br />

by approximately 50% in the CS-HY group (8.2%<br />

injured cells) compared with the CS group (16.5%<br />

injured cells). <strong>The</strong>re was no difference in capillary perfusion<br />

comparing the CS and CS-HY groups (p>0.05).<br />

<strong>The</strong> number of adherent inflammatory cells was fewer<br />

comparing the CS-HY with the CS groups, but this did<br />

not reach statistical significance with the numbers available<br />

for study.<br />

Conclusion: Hypothermia preserved tissue viability in an<br />

animal model of elevated intracompartmental pressure.<br />

Fasciotomy remains the gold standard treatment for<br />

established compartment syndrome. However cooling<br />

may be useful to preserve tissue viability in extremities<br />

that are at risk of developing compartment syndrome.<br />

<strong>The</strong> clinical utility of hypothermia for compartment<br />

syndrome requires further study.<br />

182. A COMPARISON OF COMPLICATION<br />

RATES AND OUTCOMES FOLLOWING LIMB-<br />

LENGTHENING FOR POST-TRAUMATIC<br />

VERSUS CONGENITAL/DEVELOPMENTAL<br />

DEFORMITY<br />

Jeffrey M. Potter*, Peter O’Brien † , Piotr Blachut † ,<br />

Emil H. Schemitsch, Michael McKee<br />

* Queen’s University, Department of Medicine;<br />

†<br />

UBC; St. Michael’s Hospital, University of<br />

Toronto<br />

Purpose: To conduct a study to identify differences in<br />

complication rates and outcomes between previously<br />

recognized sub-groups commonly treated for limb<br />

length discrepancies (LLD).<br />

Method: Forty-two males and 13 females were treated<br />

for LLD at two level-one trauma centres. Mean LLD<br />

was 4.4 cm (range 1.8 to 18cm). <strong>The</strong>re were 44 femoral<br />

segments (in 41 patients) and 14 tibia segments lengthened.<br />

Forty were post-traumatic, and 18 congenital/<br />

developmental. Objective data regarding complications,<br />

length achieved, and lengthening duration was collected<br />

from patient records. Two groups were compared for<br />

differences: Developmental (congenital and developmental<br />

etiology combined; LLD occurred prior to<br />

skeletal maturity and treatment involved creating new<br />

length) versus post-traumatic (restoration of previously<br />

existing length), and tibia versus femoral lengthening.<br />

Results: A mean of 4.4 cm of length was achieved over a<br />

mean duration of 83 days, for a mean lengthening index<br />

of 18.9 days/cm. Superficial pin tract infections were the<br />

most common complication, occurring in 33 segments<br />

(56%). Deep infection occurred in six segments (10%).<br />

Three of these six had a history of open fracture, and<br />

a fourth had a history of infection during initial fracture<br />

management. All were successfully treated with<br />

irrigation and debridement, and exchange nailing. <strong>The</strong><br />

developmental group had significantly greater incidence<br />

of flexion contracture (13% versus 78%, p


COA/CORS/CORA 281<br />

(45% versus 16%, p=0.04). Tibia segments had a significantly<br />

greater lengthening index (29 d/cm versus 18<br />

d/cm, p=0.03).<br />

Conclusion: Limb lengthening is an involved process<br />

with potential for serious complications. Patients who<br />

had limb-lengthening for congenital/ developmental discrepancies<br />

had a higher rate of adjacent joint contracture<br />

and subsequent requirement for surgical release.<br />

Patients with post-traumatic lengthening had a higher<br />

rate of hardware removal, and the lengthening index<br />

was greater for tibiae than femora. Deep infection<br />

remains a significant concern. This study provides information<br />

for physicians and patients on the rate and type<br />

of complications that can be expected both overall, and<br />

within specific LLD treatment groups.<br />

183. LOCKED PLATING OF OPEN DISTAL<br />

FEMUR FRACTURES<br />

David Barei, Craig Greene*, Daphne M.<br />

Beingessner<br />

Harborview Medical Center; *Baton Rouge<br />

Orthopaedic Clinic<br />

Purpose: Non-union and secondary reduction loss complicate<br />

open distal femur fractures with bone loss. We<br />

hypothesize that locking plates decrease subsequent<br />

bone grafting in these injuries, yet maintain alignment;<br />

immediate post-fixation radiographic features predict<br />

primary union.<br />

Method: From 2001 to 2004 inclusive, 34 adults with<br />

36 open AO/OTA C-type distal femur fractures were<br />

reviewed. All were treated with locking plates and<br />

3-month minimum follow-up. Union required radiographic<br />

bridging callus on 2/4 cortices combined with<br />

lack of symptoms. Alignment was assessed on initial and<br />

united radiographs. Antibiotic beads within a metaphyseal<br />

defect defined clinically important bone loss.<br />

Results: Eleven of 20 fractures with bone loss (55%)<br />

underwent staged bone grafting to achieve union, versus<br />

two of 16 fractures without bone loss (13%). <strong>The</strong> presence<br />

of antibiotic beads was significantly associated<br />

with staged bone grafting (p


282 COA/CORS/CORA<br />

operatively and the mean percentage of correction, as<br />

well as loss of correction determined. We also analyzed<br />

the length of surgery (hours), length of hospitalization<br />

and complications.<br />

Results: <strong>The</strong> mean preoperative major curve was 52.5<br />

degrees in Group 1, 52. degrees in Group 2, and 48.8<br />

degrees in Group 3. <strong>The</strong> mean percent postoperative<br />

correction (POC%) at 1 month was 67.2, 65.2, and 63.4<br />

% in the 3 groups, respectively (p=0.531). <strong>The</strong> mean<br />

percent post-operative correction (POC%) at 24 month<br />

was 55.6% for Group 1, 56.6% for Group 2 and 51.5%<br />

for Group 3 (p=0.478). <strong>The</strong> mean percent loss of correction<br />

(LOC%) at 24 months was 3.1%, 2.25%, and<br />

2.9% respectively(p=0.648). One-way ANOVA demonstrated<br />

no significant differences in patient age, number<br />

of levels fused, length of surgery, or length of hospital<br />

stay. <strong>The</strong>re were no implant related complications.<br />

Conclusion: Hybrid constructs with one or more pairs<br />

of pedicle screws and all-hook constructs for distal foundations<br />

in SSI provided similar major curve correction,<br />

percentage postoperative correction, and are associated<br />

with a minimal loss of correction postoperatively. <strong>The</strong>re<br />

was no significant difference between length of surgery,<br />

hospital stay or complications.<br />

188. THORACOSCOPIC ANTERIOR<br />

INSTRUMENTATION AND FUSION<br />

AS TREATMENT FOR ADOLESCENT<br />

IDIOPATHIC SCOLIOSIS: A SYSTEMATIC<br />

REVIEW OF THE LITERATURE<br />

Ron El-Hawary, David Russell, Alex M.<br />

Soroceanu, Colleen O’Connell<br />

Dalhousie University<br />

Purpose: Traditionally, the accepted treatments for<br />

adolescent idiopathic scoliosis (AIS) have included<br />

open anterior thoracotomy with instrumentation and<br />

posterior spinal fusion and instrumentation. Thoracoscopic<br />

instrumentation is a newer technique, whose role<br />

remains controversial. This systematic review of the literature<br />

aims to better understand thoracoscopic instrumentation<br />

as a treatment for AIS and to discuss it in the<br />

context of the alternative techniques currently used.<br />

Method: <strong>The</strong> most commonly used medical databases<br />

(PUBMED, Medline, EMBASE, Cinahl, and the<br />

Cochrane library) were searched up to April 2008 using<br />

the search terms “VATS”, “thoracoscopic scoliosis”<br />

and “thoracoscopic scoliosis instrumentation”. Two<br />

reviewers independently performed the literature evaluation.<br />

<strong>The</strong>re were no language restrictions. Because the<br />

number of randomized controlled trials was anticipared<br />

to be small, we included relevant non-randomized trials,<br />

observational studies, and uncontrolled studies.<br />

Results: Eleven studies met the strict inclusion criteria<br />

for the systematic review, of which the majority were<br />

level III and IV evidence. Four hundred and forty-five<br />

cases have been reported, 80% of them female, with<br />

the vast majority having a diagnosis of AIS. Similar<br />

surgical techniques were used and had a mean operative<br />

time of 355 minutes, mean blood loss of 444 ml,<br />

and mean hospital stay of 5.1 days. Mean pre-operative<br />

curve magnitude was 47.9o; post-operative curve magnitude<br />

was 16.3o, with a correction of 62%. Number of<br />

levels instrumented was 6.3, pulmonary function testes<br />

returned to pre-operative values by 2-years post-operative,<br />

and complication rate was 21.6%, including a pulmonary<br />

complication rate of 9.2%. SRS questionnaires<br />

revealed that patients were satisfied.<br />

Conclusion: <strong>The</strong> major drawbacks of the thoracoscopic<br />

approach are the operative time and incidence<br />

of early pulmonary complications. Advantages include:<br />

minimally invasive, less blood loss, short hospital stay,<br />

excellent curve correction, few levels fused, good patient<br />

satisfaction, and no long term effect on pulmonary function.<br />

With appropriate surgeon training, careful patient<br />

selection, and precise surgical technique, this technique<br />

can offer an acceptable alternative to the more traditional<br />

procedures.<br />

189. PROSPECTIVE COHORT ANALYSIS OF<br />

PRIMARY PYOGENIC INFECTION OF THE<br />

SPINE IN INTRAVENOUS DRUG USERS<br />

John Street, Brian Lenehan, Michael Boyd, Marcel<br />

Dvorak, Brian K. Kwon, Scott Paquette, Charles<br />

G. Fisher<br />

UBC<br />

Purpose: To evaluate the demographics, presentation,<br />

treatment and outcomes of spinal infection in a population<br />

of Intravenous Drug Users.<br />

Method: Data on all patients with pyogenic spinal<br />

infection presenting to a quaternary referral center was<br />

obtained from a prospectively maintain database.<br />

Results: Over the five-year study period, there were<br />

102 patients treated for Primary Pyogenic Infection of<br />

the Spine of which 51 were Intravenous Drug Users<br />

(IVDU). Of this IVDU group there were 34 males.<br />

Mean age was 43 years (range 25 – 57). Twenty-three<br />

had HIV, 43 Hepatitis C and 13 Hepatitis B. All were<br />

using cocaine, 26 were also using Heroin and 44 more<br />

than three recreational drugs. Thirty patients presented<br />

with axial pain with a mean duration of 51 days (range<br />

3-120). Thirty-one were ASIA D or worse with eight<br />

ASIA A. Mean Motor Score of patients with deficit was<br />

58.6. Most common ASIA Motor Levels were C4 and<br />

C5. Mean duration of neurological symptoms was seven<br />

days (range 1-60). Blood parameters on admission were<br />

in keeping with sepsis in immunocompromised patients.<br />

None had previous surgery for spinal infection. Twentysex<br />

were receiving IV antibiotics for known spinal<br />

infection. 44 patients were treated surgically. 32 had<br />

infection of the cervical spine, 9 Thoracic and 3 Lumbar.<br />

22 had a posterior approach alone, 13 had anterior only<br />

while 9 required combined. Mean operative time was<br />

263 mins (range 62 – 742). 13 required tracheostomy.<br />

7 required early revision for hardware failure and 2 for<br />

surgical wound infection. Mean duration of antibiotic<br />

treatment was 49 days (range 28-116). 26 patients had<br />

single agent therapy. 17 had MSSA and 17 MRSA. At<br />

discharge 28 patients had neurological improvement<br />

(mean 20 ASIA points, range 1-55), 11 had deterioration<br />

(mean 13, range 1-50) and 5 were unchanged.<br />

<strong>The</strong>re were no in-hospital deaths. At 2 years after index<br />

admission 13 patients were dead and none were attending<br />

the unit for follow-up.<br />

Conclusion: Primary pyogenic spinal infection in<br />

IVDU’s typically presents with sepsis and acute cervical<br />

quadriplegia. Surgical management must be prompt and<br />

aggressive with significant neurological improvement<br />

expected in the majority of patients.<br />

190. COMPARATIVE STABILITY OF TWO C1-<br />

C2 TRANSARTICULAR SCREW SALVAGING<br />

FIXATION TECHNIQUES<br />

Hossein K. Elgafy, Tejaswy Potluri, Ahmad<br />

Faizan, Scot Foster, Nikhil Kulkarni, Amanesh<br />

Goyal, Vijay Goel<br />

University of Toledo<br />

Purpose: <strong>The</strong> current gold-standard for atlanto-axial<br />

fixation is C1-C2 Transarticular Screw (TS) fixation. In<br />

certain cases, the complicated nature of vertebral artery<br />

injury could make the application of bilateral transarticular<br />

screws impossible. This study biomechanically<br />

compares three atlantoaxial transarticular salvaging<br />

fixation techniques.<br />

Method: Nine Fresh ligamentous human cervical spine<br />

specimens (C0-C4) were thawed and the tissue surrounding<br />

the spine, except the ligaments and discs,<br />

was carefully removed. Pure moments were applied to<br />

skull in increments of 0.5 Nm from 0 Nm to 2.0 Nm<br />

with the help of loading arms, nylon strings and pulleys.<br />

<strong>The</strong> specimens were tested in extension (EXT),<br />

flexion (FLEX), left lateral bending (LB), right lateral<br />

bending (RB), left axial rotation (LR) and right axial<br />

rotation (RR) for all the cases. <strong>The</strong> positions of the<br />

LEDs were recorded using an Optotrak Motion Measurement<br />

System (Northern Digital, Waterloo, Ontario,<br />

Canada) and was converted into three rotations (flexion/extension,<br />

lateral bending and axial rotation) using<br />

rigid body kinematic principles in relation to the fixed<br />

base. <strong>The</strong> specimens were tested intact and after type II<br />

odontoid fracture, were instrumented and tested with<br />

three fixation constructs: 1) C1-C2 TS on right side and<br />

C1LMS-C2PS on contralateral side 2) C1-C2 TS on<br />

right side and C1LMS-C2IL on the contralateral side<br />

and 3) C1-C2 TS on right side with sublaminar wire.<br />

Results: All of the three instrumented cases significantly<br />

reduced motion across C1-C2 segment in all the modes<br />

when compared to intact (P


COA/CORS/CORA 283<br />

192. THE PREVALENCE OF SPINAL<br />

MAGNETIC RESONANCE IMAGING<br />

CHANGES IN SURGICAL VERSUS NON-<br />

SURGICAL PATIENTS? A RETROSPECTIVE<br />

STUDY<br />

Frederick Cheng*, Oma Persaud, Raja Y.<br />

Rampersaud<br />

*University of Toronto, Toronto; University of<br />

Toronto, University Health Network, Krembil<br />

Neuroscience Program, Toronto<br />

Purpose: Magnetic resonance imaging (MRI) is the diagnostic<br />

imaging modality of choice for spinal disorders.<br />

<strong>The</strong> high prevalence of “abnormal” MRI findings within<br />

asymptomatic individuals is well established, however,<br />

referrals to spine surgeons are often based on symptomatically<br />

discordant or non-surgical MRI ‘abnormalities’.<br />

<strong>The</strong> purpose of this study was to determine the<br />

prevalence of typical spine MRI abnormalities among<br />

symptomatic surgical candidates (SC) and non-surgical<br />

(NS)patients.<br />

Method: A retrospective cohort study was conducted on<br />

1,585 patients (with a lumbar MRI) seen by a spine surgeon.<br />

<strong>The</strong> cohorts were compared in terms of the total<br />

amount, type, severity and number of levels of abnormalities<br />

on MRI. All patients were prospectively stratified<br />

regarding surgical candidacy.<br />

Results: <strong>The</strong>re was no difference between the cohorts<br />

(n=722-SC / 863-NS) in terms of the total amount of<br />

structural abnormalities present (p=0.26). <strong>The</strong>re was no<br />

difference in the prevalence of DDD, disc-herniation or<br />

previous surgery (p> 0.2). However, there was a higher<br />

prevalence of spinal stenosis (0.513 vs. 0.394) and spondylolisthesis<br />

(0.263 vs. 0.112) within the SC (p≤0.01).<br />

Logistic-regression showed that patients with disc-herniation,<br />

stenosis or spondylolisthesis were 1.49, 1.61 and<br />

2.84 times more likely to be SC respectively (p≤0.001).<br />

Subjects with a report of severe/large herniation or stenosis<br />

were 3.30 and 2.25 times (p


284 COA/CORS/CORA<br />

Decreased ankle power and persistant internal rotation<br />

are more frequently seen in feet that have undergone surgery<br />

despite initial nonoperative treatment, compared to<br />

those treated only by either the Ponseti protocol or the<br />

French physical therapy program.<br />

197. LONG TERM RESULTS OF ILIZAROV<br />

TREATMENT IN RELAPSED CLUB FEET- A<br />

COMPARISON OF SCORING SYSTEMS WITH<br />

A PATIENT’S PERSPECTIVE OF OUTCOME<br />

Amit Datta, Shakir Syed, Curtis Robb, Chris<br />

Bradish<br />

Royal Orthopaedic Hospital Birmingham, UK<br />

Purpose: <strong>The</strong> Ponseti method of clubfoot treatment has<br />

revolutionised the management of this condition. Prior<br />

to the introduction of the Ponseti regime to the UK in<br />

the late 1990’s children were frequently treated by open<br />

surgical releases. <strong>The</strong> aim of our study is to compare<br />

the patient’s perspective of outcome following Ilizarov<br />

treatment against the long-term outcome generated by<br />

the formal scoring systems.<br />

Method: We identified nine patients and 14 feet from<br />

the theatre logbooks, treated by the senior author (CB),<br />

with recurrent deformity of idiopathic clubfeet, using an<br />

ilizarov external fixator between 1994 and 1996. A variety<br />

of objective and subjective scoring systems were used<br />

to compare the results following Ilizarov treatment.<br />

Results: International Clubfoot Study Group (ICFSG)<br />

scores on six patients gave two excellent feet, one good<br />

foot, four fair feet and one poor foot. Giving an excellent/<br />

good rate of only 37.5% with a mean follow up of 13.5<br />

years. <strong>The</strong> Reinker & Carpenter scoring system resulted<br />

in five feet graded as excellent, one as good and two<br />

were rated poor. Giving an excellent/good rate of 75%.<br />

Functional questioning was also undertaken, six of seven<br />

(85%) patients deemed their treatment a success and<br />

were glad to have undergone treatment with an ilizarov<br />

frame. All but one patient is in higher education pursuing<br />

a vocational career or are in full time employment.<br />

Conclusion: Our results show that 85% of our patients<br />

who were treated with an Ilizarov frame for correction<br />

of a relapsed clubfoot were happy with their long term<br />

outcome. Thus the patient’s perspective of the long term<br />

results of Ilizarov treatment for relapsed club foot are very<br />

encouraging. <strong>The</strong>se results do not appear to correlate well<br />

with the International Clubfoot Study Group scores.<br />

198. SUPRACONDYLAR HUMERUS<br />

FRACTURES IN OLDER CHILDREN:<br />

TREATMENT MODALITIES AND<br />

OUTCOMES<br />

Brent G. Mollon, W. S. McGuffin † , Jamie A.<br />

Seabrook ‡ , K. Kellie Leitch †<br />

<strong>The</strong> University of Western Ontario, London, ON;<br />

†<br />

Schulich School of Medicine & Dentistry, <strong>The</strong><br />

University of Western Ontario, London, ON;<br />

‡<br />

Department of Paediatrics, Children’s Health<br />

Research Institute, Department of Sociology,<br />

London Health Sciences Centre, London, ON<br />

Purpose: <strong>The</strong> treatment algorithm for supracondylar<br />

humerus fractures in children under age seven is wellestablished.<br />

However, the best treatment option for these<br />

fractures in older children (8-14 year olds) is debated.<br />

<strong>The</strong> purpose of this study was to assess the efficacy of<br />

closed versus open fixation methods of this fracture type<br />

in older children. We hypothesize that closed reduction<br />

and percutaneous pinning (CRPP) is as effective as open<br />

reduction and internal fixation (ORIF).<br />

Method: A retrospective chart review was completed of all<br />

patients 8-14 years old treated for supracondylar humerus<br />

fractures at one centre from 2000-2007. IRB approval was<br />

obtained for this study. Demographics, treatment methods,<br />

pre- and post-operative complications, functional and<br />

radiographic outcomes were reviewed. Values are reported<br />

as mean ± standard deviation.<br />

Results: Seventy-eight eligible patients were identified:<br />

60 (76.9%) were treated with CRPP, and 18 (23.1%)<br />

were treated with ORIF. Demographics and fracture<br />

characteristics were similar between the CRPP and<br />

ORIF groups, although patients treated with ORIF were<br />

older (p


COA/CORS/CORA 285<br />

202. INTRATHECAL MORPHINE ANALGESIA<br />

IN IDIOPATHIC SCOLIOSIS SURGERY: DOES<br />

GENDER OR RACIAL GROUP AFFECT<br />

OPTIMAL DOSING?<br />

Jochen P. Son-Hing, Connie Poe-Kochert, Paul A.<br />

Tripi, Jennifer Potzman, George H. Thompson<br />

Rainbow Babies and Children’s Hospital<br />

Purpose: Do children from different gender or racial<br />

groups receive different analgesic doses for the same<br />

acute pain condition? We previously reported on intrathecal<br />

morphine for preemptive analgesia in children<br />

undergoing posterior spinal fusion (PSF) and segmental<br />

spinal instrumentation (SSI) for idiopathic scoliosis<br />

(IS). We determined the optimal dose range to maximize<br />

analgesia while minimizing adverse effects. <strong>The</strong> purpose<br />

is to ensure this adopted protocol is equally effective<br />

across gender and racial groups.<br />

Method: We studied 407 intrathecal morphine patients.<br />

Those given a moderate dose of 9-19 mcg/kg (n=293)<br />

had the most effective and safe postoperative pain relief.<br />

This group consisted of 246 female and 47 male patients.<br />

<strong>The</strong>re were 224 Caucasian (CA) and 63 African-American<br />

(AA) patients. Other ethnicities were excluded. Factors<br />

analyzed included postoperative Wong-Baker visual<br />

analog pain scores (VAS), time to first opioid rescue<br />

dose, total morphine dose over the first 48 hours, and<br />

postoperative complications.<br />

Results: For female and male gender, mean VAS pain<br />

scores in post-anesthesia care unit (PACU) were 0.48<br />

and 0.56, mean times to first opioid rescue dose were<br />

999.1 and 1003.3 minutes, and total morphine over the<br />

first 48 hours were 1.5mg/kg in both groups, respectively.<br />

Respiratory depression and PICU admission<br />

occurred in 2 (4.2.%) and 4 (1.6%) patients, respectively.<br />

For CA and AA patients, mean VAS pain scores<br />

in PACU were 0.48 and 0.46, mean times to first opioid<br />

rescue dose were 991.7 and 1031.9 minutes, and total<br />

morphine over the first 48 hours were 1.5mg/kg and<br />

1.3mg/kg, respectively. Respiratory depression occurred<br />

in 5 (2.2.%) and 2 (3.2%) patients and PICU admission<br />

occurred in 4 (1.8.%) and 4 (6.3%) patients, respectively.<br />

StudentÂ’s t-test and Fisher exact test demonstrated<br />

no significant differences between genders for all<br />

variables, and no significant differences between races<br />

except less total morphine for AA patients over the first<br />

48 hours (p=0.0024).<br />

Conclusion: An optimal intrathecal morphine dose range of<br />

9-19 mcg/kg provides effective and safe postoperative pain<br />

relief in children undergoing PSF and SSI for IS, regardless<br />

of gender or race. Intrathecal morphine can be given with<br />

the assurance that it does not discriminate against gender<br />

or provide less optimal analgesia to AA patients.<br />

203. APICAL VERTEBRA PEDICLE<br />

MORPHOLOGY IN SCOLIOSIS<br />

Denise Mackey, Firoz Miyanji, Renjit Varghese,<br />

Davor Saravanja, Christopher W. Reilly<br />

UBC<br />

Purpose: <strong>The</strong>re is scant literature with respect to reproducibility<br />

in radiological measurements of vertebral<br />

morphology. <strong>The</strong> purpose was to determine the reliability<br />

of measurement of various parameters of vertebral<br />

morphology in idiopathic scoliosis.<br />

Method: Ten patients with AIS were investigated with<br />

standardised low dose multi-slice helical CT. Axial<br />

reconstructions in the plane of the T8 (apical) vertebra<br />

were performed prone, as per Jamieson et al (2008).<br />

Antero-posterior (AP) canal diameter, left and right<br />

pedicle width, canal width, left and right mid-point to<br />

medial pedicle length, left and right pedicle length, and<br />

cord length, left and right transverse angles, and left and<br />

right canal area were measured by our spine surgeons<br />

and spine surgery fellow. Statistical analysis for intraclass<br />

coefficients (ICC) for intra and inter observer reliability<br />

was then performed.<br />

Results: Intra-observer reliability was excellent, with a<br />

mean ICC score of 0.930 (range 0.608-0.996), across all<br />

fourteen variables. Inter-observer reliability was very good<br />

with a mean ICC score of 0.890 (range 0.360-0.987),<br />

across all variables. <strong>The</strong>re was poor inter-observer reliability<br />

for measurement of the transverse pedicle angles<br />

(0.360 – 0.446). <strong>The</strong> intra-observer reliability for transverse<br />

pedicle angles, whilst good (0.608- 0.861), was<br />

worse than any of the other intra-observer reliabilities.<br />

Conclusion: We demonstrate excellent intra, and inter<br />

observer reliability for measurement of apical vertebrae<br />

morphology in AIS. This tool can be utilized in<br />

the further study of pedicle dysplasia. Measurement of<br />

transverse pedicle angle was less reliable than any of the<br />

other measurement variables. A standardised measurement<br />

of the morphology of vertebral canal, pedicles<br />

and vertebral body morphology is reliable both within<br />

individual observers, and across a group of observers. A<br />

standardised measure for further investigation has been<br />

validated which will enable study of the evolution of pedicle<br />

dysplasia over time. This will lead to a better understanding<br />

of the etiology of pedicle dysplasia in scoliosis.<br />

204. CONTINUOUS INTRAVENOUS<br />

MORPHINE INFUSION FOR POSTOPERATIVE<br />

ANALGESIA FOLLOWING POSTERIOR<br />

SPINAL FUSION FOR IDIOPATHIC SCOLIOSIS<br />

Connie Poe-Kochert, Paul A. Tripi, Jennifer<br />

Potzman, Jochen P. Son-Hing, George H.<br />

Thompson<br />

Rainbow Babies and Children’s Hospital<br />

Purpose: Postoperative pain is common following posterior<br />

spinal fusion (PSF) and segmental spinal instrumentation<br />

(SSI) for idiopathic scoliosis (IS). It is often<br />

treated with intravenous morphine patient controlled<br />

analgesia (PCA), but no studies have examined continuous<br />

morphine infusion. <strong>The</strong> purpose of this study was to<br />

identify the safety and efficacy of continuous morphine<br />

infusion without PCA for post-operative pain management<br />

in these patients.<br />

Method: We retrospectively reviewed 338 consecutive<br />

patients from 1992 to 2006 who received continuous<br />

morphine infusion. Following induction of general anesthesia<br />

and prior to surgical incision, patients received<br />

intrathecal morphine for preemptive analgesia. Anesthesia<br />

was maintained with 50% nitrous oxide and up to<br />

0.6% isoflurane, with minimal or no intravenous opioids.<br />

Following surgery, pre-ordered morphine infusion<br />

(0.01 mg/kg/hr) began when patients first reported pain.<br />

<strong>The</strong> infusion rate was titrated using a strict protocol<br />

based on frequent assessment of vital signs, Wong-Baker<br />

visual analog pain scores (VAS), and clinical status. <strong>The</strong><br />

infusion continued until patients were able to take oral<br />

analgesics at postoperative day 2-3. Factors analyzed<br />

included patient demographics, intrathecal morphine<br />

dosage, intraoperative intravenous opioid dosage, pain<br />

scores through the third postoperative day, interval to<br />

start of morphine infusion, total morphine requirement<br />

in the first 48 postoperative hours, and postoperative<br />

complications.<br />

Results: Mean intrathecal morphine dose was 15.45<br />

mcg/kg and mean interval to start of morphine infusion<br />

was 15:45 hours. Mean VAS pain scores were<br />

3.05, 4.48, 4.48, and 4.60 at 12 hours, 1, 2, and 3 days<br />

postoperatively. <strong>The</strong> total mean dosage of morphine in<br />

the first 48 hours postoperatively was 0.03 mg/kg/hr.<br />

Nausea/vomiting, pruritis, respiratory depression, and<br />

PICU admissions related to the morphine drip occurred<br />

in 13.3%, 4.1%, 0%, and 0% of the patients during the<br />

same time period.<br />

Conclusion: A low frequency of adverse events and a<br />

mean postoperative pain score of 5 or less demonstrates<br />

that continuous postoperative morphine infusion is a<br />

safe and effective method of pain management in children<br />

following PSF and SSI for IS. Continuous morphine<br />

infusion without PCA is a safe, alternative method of<br />

pain control for postoperative patients with IS.<br />

J BONE JOINT SURG [BR] 2011; 93-B:SUPP III

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!