SoA abstracts - JICS - The Intensive Care Society

State of the Art 2010 abstracts © The Intensive Care Society 2011
ICS medal and research abstract presentations
Research Gold Medal Presentations winner
Unravelling the paradox, how inflammation
leads to immuno-suppression and
secondary infection
A Conway Morris University of Edinburgh, Edinburgh, UK
Nosocomial infections occur in 25-35% of patients admitted to intensive
care, 1 similar to rates found in neutropenic patients. 2 Notably the
predominant organisms causing nosocomial infections differ from those
causing community-acquired infections in immuno-competent people. 3
The high rates of nosocomial infection, and the specific organisms
involved argue for the influence of host vulnerabilities. However, most
critically ill patients who acquire nosocomial infection are not classically
immuno-suppressed (eg HIV infection or neutropaenia), suggesting that
they suffer from a more subtle, as yet poorly defined, immune dysfunction.
Paradoxically, many illnesses that precipitate intensive care admission are
accompanied by immune hyper-activity. 4
Animal models of sepsis and trauma have implicated the endogenous,
pro-inflammatory molecule C5a in neutrophil dysfunction. 5,6 Due to the
rapid clearance (2-3 minute half life) of C5a from the circulation,
measurement of plasma concentrations gives an imprecise account of
neutrophil exposure to this molecule. 7 However, when C5a binds to its
predominant receptor CD88, the receptor is internalised and surface
expression decreased, 7 with a strong inverse correlation to exposure, so
acting as a proxy marker for C5a exposure. 8
I hypothesised that excessive release of C5a would act to inhibit
neutrophil function in humans, as seen in animal models. 5,6 Second, I
hypothesised that C5a drives a defect in phagocytosis by inhibiting RhoA,
the key mediator of actin polymerisation in complement-facilitated
phagocytosis and that similar defects would be seen in neutrophils from
critically ill patients. Third, I hypothesised that C5a-mediated neutrophil
dysfunction would predict those at greatest risk of nosocomial infection, and
act additively with other immune dysfunctions (monocyte deactivation 9 and
elevated immuno-suppressive regulatory T-cells 10 ) in this manner.
C5a mediates neutrophil dysfunction in ventilator-associated
pneumonia
The initial year’s work involved a cohort of 73 critically ill patients with
suspected ventilator-associated pneumonia (VAP). 8 I demonstrated that
these patients had neutrophils that were unable to effectively phagocytose
or produce reactive oxygen species, and these defects correlated
significantly with the marker of C5a exposure, CD88 (Figure 1a). This
relationship was specific to CD88, as there was no correlation with other
markers of neutrophil activation (eg CD11b). However, the neutrophils
retained capacity for tissue toxicity, 8 and patients with confirmed VAP
displayed an intense alveolar inflammation. 11,12
Exploring the mechanism of C5a-mediated neutrophil dysfunction
The apparent paradoxical finding of a pro-inflammatory molecule driving
dysfunction required exploration of potential underpinning mechanisms,
occurring in years 2 and 3. Recombinant human C5a (rhC5a) applied to
healthy neutrophils was able to impair phagocytosis, and reproduce the
correlation with CD88 and phagocytosis seen in the patients with VAP8.
C5a inhibited RhoA and actin polymerisation in response to a phagocytic
target, effects which could be prevented by blocking PI3K‰ with the
isoform specific small molecule inhibitor, IC87114 (Figure 1b). 13 Similarly
PI3Kδ inhibition prevented the inhibition of phagocytosis. Granulocytemacrophage
colony stimulating factor (GM-CSF) was able to reverse the
inhibition of RhoA and actin polymerisation and restore phagocytosis when
applied after C5a exposure. 14
The correlation of phagocytosis with CD88 was confirmed in
neutrophils obtained from a new cohort of critically ill patients (see
below), and similar inhibition of RhoA and actin polymerisation to that
seen in-vitro was demonstrated, alongside ex-vivo resurrection of
phagocytosis by GM-CSF. 14
C5a-mediated neutrophil dysfunction precedes and predicts
nosocomial infection
The VAP study recruited patients at time of clinical diagnosis, and only
included those with a single nosocomial infection. As such I could not
determine whether the neutrophil dysfunction preceded infection, or was
simply an epiphenomenon. In the final year I undertook a further study
examining the prevalence of neutrophil dysfunction in a group of 95
critically ill patients (patients requiring support of ≥1 organ systems for
≥48 hours), taking serial blood samples. C5a mediated neutrophil
dysfunction was determined by measuring surface CD88 expression using
flow cytometry, and occurred in 73% of the patients examined. It occurred
in those with both sterile and septic inflammation. 15 Infection was
determined by pre-defined criteria, in cases of dubiety an expert consensus
panel adjudicated. Patient data was censored from 48 hours before
nosocomial infection occurred to reduce the risk of infection itself causing
the effect seen.
Those patients who experienced C5a-mediated dysfunction had
increased relative risk and positive likelihood ratio (RR and +LR) acquiring
nosocomial infection (RR 5.4 (95% CI 1.4-21) +LR 1.5, p=0.01), a
relationship that remained significant when adjusted for time effects in a
survival analysis (Hazard ratio 4.1 95% CI 1.1-6.8, p=0.03) (Figure 1c).
C5a-mediated neutrophil dysfunction acts additively with other
immune defects
Peripheral blood monocytes have been previously shown to be ‘deactivated’
in sepsis and other systemic inflammatory conditions, 9 and this can be
determined by surface HLA-DR expression. 16 Additionally, I have
demonstrated that elevated numbers of regulatory T-cells (T-regs) are found
in critical illness 17 and that these also predict nosocomial infection. 18
The 95 critically ill patients from above also had monocyte function
determined (by HLA-DR expression) alongside quantification of T-reg
numbers. Those patients with low monocyte HLA-DR expression had a RR
of nosocomial infection of 2.9 (95%CI 1.2-6.8) +LR 1.5, while those with
elevated T-regs had RR of 2.3 (95%CI 1.3-4.1) +LR 2.2. Many patients had
dysfunction of more than one cell type, with a cumulative increase in risk
(Table 1). The combined immune dysfunctions’ association with
nosocomial infection remained significant when corrected for potential
confounders in a multiple logistic regression model (p

State of the Art 2010 abstracts
suitably equipped laboratory. Future work will focus on immune profile
guided therapeutic trials for immuno-dysfunction as well as an exploration
of whether the immune defects identified are simply additive or if there is
synergism between them.
Funding and acknowledgements: This work was funded by an academic
clinical training fellowship from the Chief Scientist Office (CAF/08/13), the
Sir Jules Thorn Charitable Trust and NHS Lothian Research and
Development.
Figures for this abstract can be accessed at www.ics.ac.uk/
Meetings_Seminars/main_meetings/gold_medal_ab
References
1. Vincent JL. Nosocomial infection in adult intensive-care units. Lancet 2003;361:2068-77.
2. EORTC International Antimicrobial Therapy Project Group. Three antibiotic regimens in
the treatment of infections in febrile granulocytopenic patients with cancer. J Infect Dis
1978;137:14-29.
3. Torres A, Rello J. Update in community-acquired and nosocomial pneumonia. 2009. Am
J Respir Crit Care Med 2010;181:782-87.
4. Adibconquy M, Cavaillon J. Stress molecules in sepsis and systemic inflammatory
response syndrome. FEBS Lett 2007;581:3723-33.
5. Huber-Lang MS, Younkin EM, Sarma JV et al. Complement-induced impairment of
innate immunity during sepsis. J Immunol 2002;169:3223-31.
6. Flierl MA, Perl M, Rittirsch D et al. The role of C5a in the innate immune response after
experimental blunt chest trauma. Shock 2007;29:25-31.
7. Oppermann M, Götze O. Plasma clearance of the human C5a anaphylatoxin by binding
to leucocyte C5a receptors. Immunology 1994;82:516-21.
8. Conway Morris A, Kefala K, Wilkinson T et al. C5a mediates peripheral blood
neutrophil dysfunction in critically ill patients. Am J Respir Crit Care Med 2009;180:
19-28.
9. Döcke WD, Randow F, Syrbe Uet al. Monocyte deactivation in septic patients:
restoration by IFN-gamma treatment. Nat Med 1997;3:678-81.
10. Venet F, Chung C-S, Kherouf H et al. Increased circulating regulatory T cells contribute
to lymphocyte anergy in septic shock patients. Intensive Care Med 2009;35:678-86.
11. Conway Morris A, Kefala K, Wilkinson TS et al. Diagnostic importance of pulmonary
interleukin-1 beta and interleukin-8 in ventilator-associated pneumonia. Thorax 2010;65:
201-07.
12. Conway Morris A, Kefala K, O’Kane C et al. Ventilator-associated pneumonia is
characterised by elevated neutrophil proteases. (Abstract (oral presentation), ICS State of
the Art Meeting 2009). JICS 2010;11:58-59.
13. Conway Morris A, Rossi AG, Walsh TS, Simpson AJ. C5a drives phagocytic dysfunction
via a PI3K delta-dependent inhibition of actin polymerization. (Abstract, ICS State of the
Art Meeting 2009). JICS 2010;11:61.
14. Conway Morris A, Rossi AG, Walsh TS, Simpson AJ. Inhibition of Rho-A by C5a leads
to defective neutrophil phagocytosis in critically ill patients. (Abstract (oral presentation),
ESICM Annual Congress 2010). Intensive Care Med in press.
15. Conway Morris A, Antonelli J, Brittan M et al. C5a mediated neutrophil dysfunction
precedes and predicts nosocomial infection in critical illness. (Abstract, ESICM Annual
Congress 2010). Intensive Care Med in press.
16. Caille V, Chiche J-D, Nciri N et al. Histocompatibility leukocyte antigen-D related
expression is specifically altered and predicts mortality in septic shock but not in other
causes of shock. Shock 2004;22:521-26.
17. Conway Morris A, Brittan M, Antonelli J et al. Immuno-suppressive regulatory
(CD4 + CD25 + FOXP3 + ) T-cells are elevated in septic and non-septic critical illness. Abstract
submitted for ICS State of the Art Meeting 2010.
18. Conway Morris A, Brittan M, Antonelli J et al. Combined immuno-phenotyping of
neutrophils, monocytes and T-cells enables accurate prediction of the risk of nosocomial
infection. Abstract submitted for ICS State of the Art Meeting 2010.
Research Gold Medal Presentations
Iron is an important factor in pulmonary
physiology
TG Smith, Nuffield Department of Anaesthetics, John Radcliffe Hospital,
Oxford, UK
Scientific background
The interaction between hypoxia and pulmonary physiology is central to
intensive care medicine. Physiological responses to hypoxia include an
increase in pulmonary arterial pressure caused by hypoxic pulmonary
vasoconstriction. This phenomenon is important in intensive care
medicine, such as in patients with acute respiratory distress syndrome
(ARDS), 1 as well as in other areas of anaesthesia and medicine. 2 However it
can become pathological – pulmonary hypertension frequently complicates
hypoxic lung disease and worsens patient survival. 3-5 The mechanisms
underlying pulmonary responses to hypoxia are poorly understood. This
body of work has investigated these mechanisms and their therapeutic
implications. Our original hypothesis was that the hypoxia-inducible factor
Figure 1. Effect of intravenous iron on systolic pulmonary artery pressure and
serum ferritin in healthy sea-level residents during altitude hypoxia at 4,340
metres. Baseline measurements (Day 0) were made at sea level prior to
ascent to high altitude. Dashed lines and open symbols indicate pre-infusion
measurements, while solid lines and closed symbols indicate post-infusion
measurements. Infusions were given immediately after measurements were
made on the morning of Day 3 (indicated by the dotted vertical line). Black
lines and symbols indicate the group that received an infusion of iron sucrose
200 mg (n=11), while grey lines and symbols indicate the group that received a
placebo infusion (n=11). Infusion with iron reversed much of the elevation in
pulmonary artery systolic pressure caused by hypoxia (p

State of the Art 2010 abstracts
Figure 2. Effect of venesection on systolic pulmonary artery pressure and
serum ferritin in chronic mountain sickness patients during altitude hypoxia
at 4,340 metres. Eleven patients with chronic moutain sickness were studied.
Measurements shown in this figure are those made prior to any iron or
placebo infusions. The black bar indicates the period over which staged
isovolaemic venesection of two litres of blood occurred. Over the following
two weeks, a progressive reduction in ferritin (p

State of the Art 2010 abstracts
Extracellular myeloperoxidase and markers
of inflammation in the sepsis syndromes
NS MacCallum, National Heart and Lung Institute, Imperial College,
London, UK
Introduction
The systemic inflammatory response syndrome (SIRS) and its sequelae
represent a formidable problem, together constituting the leading cause of
morbidity and mortality in critically ill patients. 1-3 SIRS, which is clinically
indistinguishable from sepsis, is seen in association with a variety of noninfective
insults, 4 including surgery necessitating cardiopulmonary-bypass
(CPB). 5 The incidence is insult dependent, with 15-27% 6-7 of critically ill
patients suffering from severe sepsis, and up to 96% of ICU admissions 2,5
affected by SIRS.
Neutrophils are activated early in the host immune response, and are
central to the pathogenesis of sepsis/SIRS. 8 The neutrophil capacity for
bacterial killing lacks selectivity, despite stringent regulation; and thereby
carries the potential to inflict collateral damage to and destruction of host
tissue, 9 through generation of pro-oxidant mediators. 10
Myeloperoxidase (MPO) generated hypochlorous acid (HOCl) is the
most bactericidal oxidant produced by the neutrophil. HOCl and an array
of oxidant molecules generated by MPO/HOCl 10,11 are implicated in altered
cell signalling, growth arrest and tissue damage. 12,13 Caeruloplasmin (CP)
has been shown to have an affinity for MPO; when bound it compromises
the enzymatic activity of MPO. 14,15 CP therefore plays a potentially key role
in redox regulation, modulating pro-inflammatory responses related to
HOCl under certain defined circumstances. 16,19
Hypothesis
We hypothesised that excess release of extracellular (plasma) MPO,
produces a net increase in HOCl production, resulting in redox imbalance
in favour of pro-oxidant forces. A sub-optimal anti-oxidant response,
specifically CP, places patients at greater risks of developing organ
dysfunction. We investigated the following:
Aims
First, using in vitro experiments we explored the relationship between
MPO protein and activity, MPO-CP binding, and the effect of the latter on
MPO activity.
Second, two populations with SIRS (sepsis and post-CPB) were
recruited in order to elucidate the relationship between MPO exposure,
MPO activity, and CP; to place these within the framework of associated
inflammatory mediators; and to demonstrate their relationship to SIRSinduced
organ dysfunction.
Methods
Research ethical approval: Royal Brompton Hospital Research Ethics
Committee.
Populations: Severe sepsis, n=44; post-CPB, n=52; healthy controls, n=21;
non-cardiac surgical controls, n=8.
Morbidity: Defined within the sequential organ failure assessment score
(SOFA). 20
Biochemical techniques: Spectrophotometry (enzyme kinetics, spectral
absorbance); immunoprecipitation; ELISA; radial-immunodiffusion; gelchromatography;
Western blotting; nondenaturing electrophoresis; gaschromatography–mass-spectroscopy.
Results
Methodology
An assay was developed to determine MPO activity using the chromogen
tetramethylbenzidine, with validation by detection of chromogen oxidation
product by spectral absorbance, and the absence of activity following
immunoprecipitation of MPO from a purified protein solution. Hemeproteins
in the plasma peroxidase-pool 21 contribute to measured activity.
Plasma from healthy controls and patients with SIRS was therefore
subjected to immunoprecipitation of MPO. The MPO fraction, 35% of the
measured activity, correlated with total activity (Figure 1a). There was no
relationship between plasma MPO protein and activity.
MPO activity
Inhibited in protein solution by CP in a dose dependent manner (Figure 1b
and c); this was not reproducible in plasma or blood. MPO–CP binding
affinity was confirmed in a purified protein system by (a) gelchromatography;
(b) immunoprecipitation of MPO/CP by anti-CP/MPO
antibody respectively, followed by immuno-detection (ELISA, Western
blot); (c) and nondenaturing electrophoresis. Detection of the MPO–CP
complex in plasma was inconclusive.
MPO exposure
Patients with severe sepsis and those post-CPB were divided into groups
based on cumulative MPO exposure over the 72 hour study period.
Group 1: In severe sepsis greater MPO exposure was associated with an
increase and/or non-resolution in: leukocyte count, CRP, and IL-6. There
was no difference in MPO activity, CP, IL-8, TNF, and IL-10. MPO exposure
was associated with non-resolution of organ dysfunction and increased 3-
chlorotyrosine (a marker of HOCl-induced oxidative damage 22 ) Figure 2a
and b. By contrast, a relative increase in CP was associated with lower MPO
exposure and a decrease in organ dysfunction (1.4 SOFA points). MPO
activity was decreased compared to controls, and demonstrated no
relationship to MPO or CP protein.
Group 2: Post-CPB, greater MPO exposure was associated with an
increase and/or non-resolution in: leucocyte count, CRP, and IL-6, IL-8, and
IL-10; only the former differed between groups. MPO activity and CP
decreased significantly, with no difference between groups, nor discernable
relationship between indices. There was no difference in organ dysfunction
or 3-chlorotyrosine.
Figure 2a and b. Relationship of neutrophil count to plasma myeloperoxidase
protein and activity over time: MPO grouping. Mean –SEM of neutrophil count,
plasma MPO and plasma MPO activity for group A – higher (n=20), group B –
lower (n=20) plasma MPO AUC during the 72 hour study period. Group A had
a greater neutrophil response (p=0.051) and MPO concentration (p

State of the Art 2010 abstracts
Discussion
In the validated plasma MPO activity assay, the activity due to MPO and the
total measured activity correlated well. The proportion of measured activity
attributable to MPO was similar in health and SIRS, despite differing
plasma concentrations of MPO protein. This may reflect a functional
capacity of MPO, supporting a role during normal physiological states. 23-29
CP inhibited MPO activity in solution but not plasma; possibly due to the
presence of endogenous CP, dysfunctional exogenous CP, 30 or plasma
factors. CP binding affinity to MPO was confirmed in protein solution, with
MPO-CP complex visualisation for the first time.
In the SIRS populations studied, MPO was associated with an increase
of inflammatory indices. Greater plasma MPO appears to be detrimental in
terms of non-resolution of organ dysfunction and CRP in severe sepsis, but
exerts no impact on inflammatory indices or morbidity post-CPB. MPO
activity was decreased in both populations, despite leukocyte response and
elevated MPO protein. Post-CPB the activity did not return to pre-operative
levels, a phenomenon not previously documented.
CP appears to neither to modulate MPO activity in SIRS post-CPB, nor
in severe sepsis. Thus, despite in vitro data confirming the ability if CP to
inhibit MPO activity, no such evidence could be demonstrated in patient
plasma. The data presented here did not support a role for the modulation
of MPO by CP within 72 hours of onset of the syndromes studied. A rise in
CP was however associated with a decrease in MPO protein in sepsis,
perhaps, in conjunction with its binding affinity, conferring a regulatory
role of the former in the elimination of extracellular MPO protein.
The reduced enzymatic capacity of MPO during SIRS may be due to
diminished functional capacity of MPO (eg molecular damage); 31-34
reduction of substrate or co-factor availably; 35 or inhibition. 14,15,33,36-38 MPO
activity may be subject to regulation. However, the evidence of chlorination
indicates that this population may also be either more susceptible to, or
subject to, elevated HOCl-induced oxidative damage. 39 This paradox
possibly relates to substrate availability, 40 a mechanism that would support
the increased production of HOCl.
The data highlight further evidence for a regulatory role of MPO
activity. MPO activity levels on resolution of organ dysfunction and
systemic inflammation are similar to, and indeed return to those levels seen
pre-operatively and in health. The ‘excess’ MPO detectable in plasma is
enzymatically dysfunctional, retaining an intact epitope detectable
immunologically. Increased release of MPO into plasma may be
physiological, compensating for the limited enzymatic lifespan and
associated molecular damage sustained. The mechanism of MPO clearance
remains undetermined. Dysfunctional MPO protein may thus remain in
plasma for a prolonged time period until cleared.
Conclusion
The relationship between plasma MPO and its measured activity remains to
be further elucidated. What seems clear from the data presented here is that
plasma MPO protein is not representative of activity in severe sepsis or post-
CPB; and that maintenance of a constant activity, or return to a ‘normal’
activity following resolution of disease process seems to be desirable.
Acknowledgements: This research was generously funded by the British
Heart Foundation through Clinical Research Fellowship. NSM conducted
all experiments over a three year period with PhD supervision by Dr GJ
Quinlan and Professor TW Evans. The Royal Marsden Hospital and Dr MB
Hacking for access to surgical control patients. The staff from the Unit of
Critical Care research group and the Royal Brompton Hospital.
Figures 1a, b and c and 2c can be accessed at www.ics.ac.uk/
Meetings_Seminars/main_meetings/gold_medal_ab
References
1. Angus DC, Linde-Zwirble WT, Lidicker J et al. Epidemiology of severe sepsis in the
United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med
2001;29:1303-10.
2. Brun-Buisson C. The epidemiology of the systemic inflammatory response. Intensive Care
Med 2000;26 Suppl 1:S64-74.
3. Martin GS, Mannino DM, Eaton S et al. The epidemiology of sepsis in the United States
from 1979 through 2000. N Engl J Med 2003;348:1546-54.
4. Bone RC, Balk RA, Cerra FB et al. Definitions for sepsis and organ failure and guidelines
for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference
Committee. American College of Chest Physicians/Society of Critical Care Medicine.
Chest 1992;101:1644-55.
5. MacCallum NS, Gordon SE, Quinlan GJ et al. Abstract 23: The Systemic Inflammatory
Response Syndrome Following Cardiac Surgery. Circulation 2007;116
(16_MeetingAbstracts):II_928-.
6. Brun-Buisson C, Meshaka P, Pinton P et al. EPISEPSIS: a reappraisal of the epidemiology
and outcome of severe sepsis in French intensive care units. Intensive Care Med
2004;30:580-88.
7. Padkin A, Goldfrad C, Brady AR et al. Epidemiology of severe sepsis occurring in the
first 24 hrs in intensive care units in England, Wales, and Northern Ireland. Crit Care
Med 2003;31:2332-38.
8. Das UN. Critical advances in septicemia and septic shock. Crit Care 2000;4:290-96.
9. Marshall JC. Neutrophils in the pathogenesis of sepsis. Crit Care Med 2005;33(12
Suppl):S502-5.
10. MacCallum NS, Quinlan GJ, Evans TW. The role of neutrophil-derived myeloperoxidase
in organ dysfunction and sepsis. In: Vincent JL, ed. Yearbook of Intensive Care and
Emergency Medicine: Springer: 2007;173-78.
11. Weiss SJ, Lampert MB, Test ST. Long-lived oxidants generated by human neutrophils:
characterization and bioactivity. Science 1983;222:625-28.
12. Pullar JM, Vissers MC, Winterbourn CC. Living with a killer: the effects of
hypochlorous acid on mammalian cells. IUBMB Life 2000;50:259-66.
13. Vissers MC, Pullar JM, Hampton MB. Hypochlorous acid causes caspase activation and
apoptosis or growth arrest in human endothelial cells. Biochem J 1999;344 Pt 2:443-49.
14. Griffin SV, Chapman PT, Lianos EA et al. The inhibition of myeloperoxidase by
ceruloplasmin can be reversed by anti-myeloperoxidase antibodies. Kidney Int
1999;55:917-25.
15. Segelmark M, Persson B, Hellmark T et al. Binding and inhibition of myeloperoxidase
(MPO): a major function of ceruloplasmin Clin Exp Immunol 1997;108:167-74.
16. Baker CS, Evans TW, Haslam PL. Measurement of ceruloplasmin in the lungs of patients
with acute respiratory distress syndrome: is plasma or local production the major source
Respiration 2000;67:533-38.
17. Festa M, Mumby S, Nadel S et al. Antioxidant protection against iron in children with
meningococcal sepsis. Crit Care Med 2002;30:1623-29.
18. Gutteridge JM, Quinlan GJ, Mumby S et al. Primary plasma antioxidants in adult
respiratory distress syndrome patients: changes in iron-oxidizing, iron-binding, and free
radical-scavenging proteins. J Lab Clin Med 1994;124:263-73.
19. Pepper JR, Mumby S, Gutteridge JM. Sequential oxidative damage, and changes in ironbinding
and iron-oxidising plasma antioxidants during cardiopulmonary bypass surgery.
Free Radic Res 1994;21:377-85.
20. Vincent JL, Moreno R, Takala J et al. The SOFA (Sepsis-related Organ Failure
Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group
on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive
Care Med 1996;22:707-10.
21. Peroxidase activity of hemoproteins released into plasma as sources of oxidative stress in
sepsis: a preliminary report. A60. Annual Congress Society of Critical Care Medicine;
2007; Orlando, FL, USA. SCCM.
22. Winterbourn CC, Kettle AJ. Biomarkers of myeloperoxidase-derived hypochlorous acid.
Free Radic Biol Med 2000;29:403-09.
23. Claesson R, Karlsson M, Zhang Y, et al. Relative role of chloramines, hypochlorous acid,
and proteases in the activation of human polymorphonuclear leukocyte collagenase. J
Leukoc Biol 1996;60:598-602.
24. Arnhold J. Properties, functions, and secretion of human myeloperoxidase. Biochemistry
(Mosc) 2004;69:4-9.
25. Eiserich JP, Baldus S, Brennan ML et al. Myeloperoxidase, a leukocyte-derived vascular
NO oxidase. Science 2002;296:2391-94.
26. Midwinter RG, Vissers MC, Winterbourn CC. Hypochlorous acid stimulation of the
mitogenactivated protein kinase pathway enhances cell survival. Arch Biochem Biophys
2001;394:13-20.
27. Schoonbroodt S, Legrand-Poels S, Best-Belpomme M et al. Activation of the NF-kappaB
transcription factor in a T-lymphocytic cell line by hypochlorous acid. Biochem J
1997;321(Pt 3):777-85.
28. Schildknecht S, Ullrich V. Peroxynitrite as regulator of vascular prostanoid synthesis.
Arch Biochem Biophys 2009;484:183-89.
29. Evans TJ, Buttery LD, Carpenter A et al. Cytokine-treated human neutrophils contain
inducible nitric oxide synthase that produces nitration of ingested bacteria. Proc Natl
Acad Sci USA 1996;93:9553-58.
30. Sengelov H, Kjeldsen L, Borregaard N. Control of exocytosis in early neutrophil
activation. J Immunol 1993;150:1535-43.
31. Marchetti C, Patriarca P, Solero GP et al. Genetic characterization of myeloperoxidase
deficiency in Italy. Hum Mutat 2004;23:496-505.
32. Nauseef WM, Root RK, Malech HL. Biochemical and immunologic analysis of hereditary
myeloperoxidase deficiency. J Clin Invest 1983;71:1297-307.
33. Galijasevic S, Maitra D, Lu T et al. Myeloperoxidase interaction with peroxynitrite:
chloride deficiency and heme depletion. Free Radic Biol Med 2009; In Press, Corrected
Proof.
34. Chapman AL, Hampton MB, Senthilmohan R et al. Chlorination of bacterial and
neutrophil proteins during phagocytosis and killing of Staphylococcus aureus. J Biol
Chem 2002;277:9757-62.
35. Zhang R, Shen Z, Nauseef WM et al. Defects in leukocyte-mediated initiation of lipid
peroxidation in plasma as studied in myeloperoxidase-deficient subjects: systematic
identification of multiple endogenous diffusible substrates for myeloperoxidase in
plasma. Blood 2002;99:1802-10.
36. Park YS, Suzuki K, Mumby S et al. Antioxidant binding of caeruloplasmin to
myeloperoxidase: myeloperoxidase is inhibited, but oxidase, peroxidase and
immunoreactive properties of caeruloplasmin remain intact. Free Radic Res 2000;33:
261-65.
37. Qian M, Eaton JW, Wolff SP. Cyanate-mediated inhibition of neutrophil myeloperoxidase
activity. Biochem J 1997;326(Pt 1):159-66.
60
Volume 12, Number 1, January 2011 JICS

State of the Art 2010 abstracts
38. Andrews PC, Krinsky NI. A kinetic analysis of the interaction of human
myeloperoxidase with hydrogen peroxide, chloride ions, and protons. J Biol Chem
1982;257:13240-45.
39. Winterbourn CC. Biological reactivity and biomarkers of the neutrophil oxidant,
hypochlorous acid. Toxicology 2002;181-82:223-27.
40. Pascual C, Karzai W, Meier-Hellmann A et al. A controlled study of leukocyte activation
in septic patients. Intensive Care Med 1997;23:743-48.
What determines poor psychological
outcomes after admission to the intensive
care unit
D Wade*, R Raine*, J Weinman † , R Hardy*, R Tupprasoot ‡ , M Mythen ‡ ,
DCJ Howell ‡
*University College London, UK. † Institute of Psychiatry, King’s College
London, UK. ‡ University College Hospital NHS Foundation Trust, London,
UK
Background
It is well documented that the mental health and quality of life of ICU
survivors may be poor. 1,2 Extreme psychological reactions such as
terror, panic and confusion can occur during ICU admission. 3 Subsequently,
patients may suffer from depression, anxiety and posttraumatic stress
disorder (PTSD). A national guideline 4 requires that intensive care patients
should be assessed for risk of psychological morbidity and offered
rehabilitation. To provide support for patients and improve outcomes, more
evidence is needed about the psychological impact of intensive care.
Therefore the aims of this research were to assess the prevalence of
psychological morbidity three months after a level 3 intensive care
admission and to identify clinical, psychological and socio-economic
predictors of poor outcome. An additional aim was to investigate the
formation of intrusive memories of intensive care as a key process in the
development of PTSD. The work was carried out as an MRC-funded PhD
by DW between September 2007 and September 2010.
Methods
Initially a systematic review was conducted of studies of psychological
morbidity and quality of life following intensive care admission.
Subsequently, a prospective cohort study of 157 consecutive ICU patients,
who received level 3 care, was performed. At ICU discharge, patients were
assessed for mood, stress, delirium and memory distortions. Data about
illness, treatment and drugs in the ICU were collected.
At three months, PTSD, depression, anxiety, quality of life
and socioeconomic circumstances (SEC) were assessed. A total of
100 patients (64%) participated in the follow-up assessment, ensuring that
the study was fully-powered. Statistical analysis included the detection of
univariable associations between predictors and outcomes, multiple
regression to identify independent risk factors and mediation analysis. 5
Finally seventeen cohort participants who developed intrusive memories
of the ICU were interviewed at 4-5 months for a qualitative study.
Results
The systematic review of 46 studies found few high quality studies and
estimates of prevalence of PTSD ranged from 0% to 62%. Depression was
found in 3%-47%, and anxiety in 5%-43% of patients. Few risk factors were
reported and evidence for clinical predictors was inconsistent.
The cohort study showed that during level 3 ICU admissions the
incidence of mood disturbance, delirium and physical stress was 47%, 66%
and 56% respectively. At three months, the prevalence of PTSD was 27.1%
(CIs: 18.3, 35.9%), of depression 46.3% (CIs: 36.5-56.1%) and of anxiety
44.4% (CIs: 34.6%-54.2%). PTSD was predicted by Therapeutic
Intervention Scoring System (TISS) score, number of organs supported,
number of drug groups received in ICU and sepsis markers. C-Reactive
Protein (CRP) at admission (r=0.248, p=0.014), and noradrenaline use
(mean difference=5.1, p=0.033) both predicted PTSD, as did highest CRP
and white cell count (WCC) during admission.
Table 1 shows that the strongest clinical predictor of PTSD was days of
sedation (r=0.268, p=0.008). Benzodiazepine usage was the strongest
clinical predictor of depression (mean difference: 7.44 (CIs: 1.81, 13.07),
p=0.010) and inotrope usage predicted worse anxiety (mean difference:
7.63 (CIs:1.90,13.37), p=0.01). Steroids predicted better physical quality of
life (5.57 (9.96,1.18), p=0.029), as did anaesthetics (4.45(0.04,8.94)).
There was a trend for opioid analgesia to be associated with better
depression, anxiety and mental health scores (7-8 points difference).
Psychological risk factors for PTSD included extreme ICU reactions
such as mood disturbance (r=0.495, p

State of the Art 2010 abstracts
Strengths of the cohort study include the representativeness of the
sample, achieved by approaching every level 3 patient admitted to ICU over
a ten month period. Participants were assessed at ICU discharge, reducing
risk of recall bias. A comprehensive group of risk factors was explored,
robust methods of outcome assessment were employed and a rigorous
follow-up system ensured a good response rate. Limitations were that the
sample size was not large and many statistical associations were found,
increasing the risk of type 1 errors.
The finding that different clinical factors correlate with post-ICU PTSD,
depression and anxiety is of particular interest and the biological
mechanisms behind these observations should be investigated further. This
research also suggests that modifications in clinical practice to reduce the
intensity of ICU interventions, to use inotropes and sedatives judiciously
and prescribe appropriate opiate analgesia, may improve psychological
outcome. In the meantime patients should be routinely assessed for
psychological distress and offered support in ICU and post discharge. A
clinical and psychological intervention based on these findings should be
designed and evaluated in a randomised controlled trial.
References
1. Dowdy DW, Eid MP, Sedrakyan A et al. Quality of life in adult survivors of critical
illness: a systematic review of the literature. Intensive Care Med 2005;31:611-20.
2. Jackson JC, Hart RP, Gordon SM et al. Post-traumatic stress disorder and post-traumatic
stress symptoms following critical illness in medical intensive care unit patients: assessing
the magnitude of the problem. Crit Care 2007;11:R27.
3. Granberg A, Engberg IB, Lundberg D. Patients’ experience of being critically ill or
severely injured and cared for in an intensive care unit in relation to the ICU syndrome.
Part 1. Intensive Crit Care Nursing 1998;14:294-307.
4. National Institute for Health and Clinical Excellence. Clinical Guideline 83:
Rehabilitation after critical illness. NICE 2009.
5. Baron RM, Kenny DA. The moderator-mediator variable distinction in social
psychological research. J Personality Social Psychol 1986;51:1173-82.
6. Ware JE, Sherbourne CD. The MOS 36-item short-form health survey(SF-36). 1.
Conceptual framework and item selection. Medical Care 1992;6:473-83.
Developing novel renoprotective agents for
the prevention of post cardiac surgery
acute kidney injury
NN Patel*, T Toth † , C Jones*, H Lin*, P Ray ‡ , SJ George*, G Welsh † ,
SC Satchell † , GD Angelini*, GJ Murphy*
*Bristol Royal Infirmary, Bristol, UK. † Southmead Hospital, Bristol, UK.
‡
Weston General Hospital, Weston-Super-Mare, UK
Background
Acute kidney injury (AKI) following cardiac surgery, defined as a >25%
reduction in glomerular filtration rate, is associated with a four-fold
increase in postoperative mortality. Despite the importance of this clinical
problem, our understanding of the underlying processes is poor and there
is no effective treatment. This is a reflection of the poor homology between
rodent models, the mainstay of research into kidney injury, and that which
occurs in humans, and it has been suggested that large animal models of
AKI with closer homology to humans are required if clinical progress is to
be achieved.
The aim of this programme of research was:
1. To undertake a systematic review of the evidence from randomised
controlled clinical trials that have evaluated pharmacological agents for
the prevention of AKI post cardiac surgery.
2. To characterise post-cardiopulmonary bypass (CPB) AKI in a novel
porcine recovery model with potentially greater homology to cardiac
surgery patients.
3. To determine the effect of endothelin-A (ET-A) receptor blockade on
post CPB AKI in swine.
4. To determine the effect of phosphodiesterase-5 (PDE-5) inhibition on
post CPB AKI in swine.
Methods
Systematic review
We searched PubMed, Embase and the Cochrane Central Register of
Controlled Trials for randomised controlled trials comparing renoprotective
pharmacological interventions with control in adult patients undergoing
cardiac surgery with cardiopulmonary bypass. We extracted data for
mortality, need for renal replacement therapy (RRT), incidence of AKI, and
creatinine clearance at 24-48 hours. The meta-analysis was performed in
line with recommendations from the Cochrane Collaboration and the
Quality of Reporting of Meta-analyses guidelines with standard software.
Porcine recovery model of post-CPB AKI
Plan of investigation
Pigs (n=8 per group) were randomised to the following groups:
Group 1. Sham operation. Pigs underwent a neck dissection under
general anaesthesia.
Group 2. CPB only. Pigs underwent 2.5 hours of CPB.
Group 3. CPB plus sitaxsentan sodium (ET-A antagonist). Pigs
underwent 2.5 hours of CPB. Each pig received an infusion of sitaxsentan
sodium 0.7 mg/kg over 30 minutes, at commencement of CPB.
Group 4. CPB plus sildenafil citrate (PDE-5 inhibitor). Pigs underwent
2.5 hours of CPB. Each pig received an infusion of sildenafil 10 mg over 30
minutes, at commencement of CPB.
Experimental methods
Thirty-two adult female Large White Landrace crossbred pigs weighing 50-
70 kg were used. Minimally invasive CPB was achieved via Smart
Cannulae ® (Smartcanula LLC, Lausanne, Switzerland) placed in the aorta
and right atrium via the right internal carotid artery and external jugular
vein respectively. Heparinisation, priming, temperature, perfusion pressure,
pump flows and acid base balance were managed according to standard
protocols. Total CPB time was 2.5 hours. Sham procedure animals
underwent similar general anaesthesia, surgical dissection and
heparinisation as CPB operated animals. Urine output was measured via a
urethral catheter. Animals were recovered, re-anaesthetised and reevaluated
at 24 hours.
Collection of serum and urine samples and measurement of renal and
endothelial function, oxygenation, and perfusion occured at four time
points: Baseline, the end of CPB, 1.5 hours post CPB and 24 hours post
CPB. Organ harvest was performed prior to euthanasia.
Outcomes
1. Renal function: Creatinine clearance (primary outcome), free water
clearance and fractional sodium excretion were calculated from urine
volumes and serum samples taken at the four specified time points
using standard formulae.
2. Renal injury: H&E stained 5 µm formalin fixed, paraffin embedded
sections were scored for tubular injury by a renal histopathologist
blinded to treatment allocation. Urinary IL-18, albumin and protein
were measured in the urine.
3. Renal endothelial function: Renal blood flow was recorded using a
Transonic flow probe placed on the renal artery via a mini-laparotomy.
Endothelial dysfunction was determined by the change in renal blood
flow in response to a supra-renal aortic infusion of acetylcholine
(0.1-10 µg/kg/min). Cortical perfusion and medullary oxygenation was
measured by cortical surface probes and O 2 sensors connected to a dual
channel tissue monitoring system (Oxylite, Oxford Optronix, Oxford
UK).
4. Mechanisms:
Endothelial injury – immunocytochemistry (ICC) for lectin Dolichos
biflorus agglutinin, Pecam-1 and ve-Cadherin. Nitric oxide (NO)
bioavailability was assessed by measuring urinary nitrate/nitrite
concentration using the Greiss reaction.
Endothelial activation – ICC for endothelin-1, eNOS and iNOS,
Inflammatory cell infiltrate – ICC for MAC-387.
High energy phosphates – will be measured using HPLC of snap frozen
tissue,
Apoptosis – assessed using In-Situ End Labelling (ISEL).
Results
Systematic review
Forty-nine randomised controlled trials involving 4,605 patients were
included. Pharmacological interventions included dopamine, fenoldopam,
62
Volume 12, Number 1, January 2011 JICS

State of the Art 2010 abstracts
Mean (SEM) Sham CPB alone CPB + CPB + P value P value
sitaxsentan sildenafil (sitax) (sildenafil)
Creatinine +22.4 -27.6 +39.7 +19.3 0.010 0.009
clearance Δ 24 hr (13.4) (15.0)* (11.0)** (7.8)**
(mL/min)
Proteinuria Δ 24 +5.36 +29.26 -7.68 -0.2

State of the Art 2010 abstracts
3. To investigate the effects of flow targeted haemodynamic therapies on
tissue microvascular flow and oxygenation and plasma markers of
inflammation in patients following major abdominal surgery.
4. To investigate changes in sublingual microvascular flow in patients with
sepsis and severe sepsis within six hours of hospital admission and
compare these data to that of healthy volunteers.
5. To investigate the dose related effects of noradrenaline on tissue
microvascular flow and oxygenation in patients with septic shock.
Methods and results
All studies received relevant approvals from the local audit committee,
Research ethics committees and the Medical and Healthcare products
Regulatory Agency as appropriate. In each of the clinical investigations
cutaneous PtO 2 was measured continuously using a Clark electrode
(TCM400, Radiometer, Denmark). Microvascular flow was measured
intermittently using cutaneous laser Doppler flowmetry (MoorLab,
Moor Instruments, UK) and visualised using sublingual sidestream
darkfield imaging (Microscan, Microvision Medical Netherlands). Cardiac
output and DO 2 I were measured continuously using lithium indicator
dilution and pulse power analysis (LiDCOplus, LiDCO Ltd, UK) in
studies 2, 3 and 5 and supra sternal Doppler (USCOM Ltd, Australia) in
study 4.
Study 1: Characterisation of the high-risk surgical population in a
large NHS Trust
Data describing in-patient non-cardiac surgical procedures performed in
our NHS Trust between April 2002 and March 2005 were obtained from
hospital and critical care databases. Healthcare Resource Groups (HRG)
codes were extracted and then ranked according to mortality rates. HRGs
with ≥5% mortality were prospectively defined as high-risk. The high-risk
surgical population accounted for 75.4% of deaths in hospital but only
9.3% of admissions. Although the high-risk population accounted for less
than 10% of cases they utilised 23% of in-patient bed days. Only 35.3% of
the high-risk patients were admitted to a critical care unit at any stage after
surgery. Of 294 high-risk patients who died, only 144 (49.0%) were
admitted to critical care at any time. Mortality rates were high amongst
patients admitted to critical care following initial management on a
standard ward (29.9%).
Study 2: Observational study of peri-operative changes in tissue
microvascular flow and oxygenation
Measurements were made prior to and at 0, 2, 4, 6 and 8 hours after major
abdominal surgery. Patients received usual clinical care. Of 25 patients
recruited, two died (8%) and 14 (56%) developed post-operative
complications. In patients who developed complications, sublingual
microvascular flow index (MFI) (small vessels

State of the Art 2010 abstracts
6. Pearse R, Dawson D, Fawcett J et al. Early goal-directed therapy after major surgery
reduces complications and duration of hospital stay. A randomised, controlled trial
[ISRCTN38797445]. Crit Care 2005;9:R687-693.
7. Wilson J, Woods I, Fawcett J et al. Reducing the risk of major elective surgery:
randomised controlled trial of preoperative optimisation of oxygen delivery. BMJ 1999;
318:1099-103.
8. Boyd O, Grounds RM, Bennett ED. A randomized clinical trial of the effect of deliberate
perioperative increase of oxygen delivery on mortality in high-risk surgical patients.
JAMA 1993;270:2699-707.
9. De Backer D, Creteur J, Preiser JC et al. Microvascular blood flow is altered in patients
with sepsis. Am J Respir Crit Care Med 2002;166:98-104.
10. Rivers E, Nguyen B, Havstad S et al. Early goal-directed therapy in the treatment of
severe sepsis and septic shock. N Engl J Med 2001;345:1368-77.
11. Landry DW, Oliver JA. The pathogenesis of vasodilatory shock. N Engl J Med 2001;
345:588-95.
12. Dellinger RP, Levy MM, Carlet JM et al. Surviving Sepsis Campaign: international
guidelines for management of severe sepsis and septic shock: 2008. Crit Care Med 2008;
36:296-27.
Muscle dysfunction in severe sepsis
P S Zolfaghari, University, College London, London, UK
Introduction
Recent evidence suggests mitochondrial dysfunction and subsequent
cellular metabolic shutdown may be important pathophysiological
mechanisms of multi-organ failure. 1-4 This may underlie the profound
muscle dysfunction seen in these critically ill patients. 5
Mitochondria supply adenosine triphosphate (ATP) to meet cell
metabolic demands. By oxidising substrates, the mitochondrial respiratory
chain generates a proton gradient across the inner mitochondrial
membrane (the mitochondrial membrane potential – Δψ m ) that provides
energy to drive ATP synthesis. 6 Cycling of phosphate to form ATP is vital
for muscle function as its accumulation causes a reduction in the force of
contraction. 7 The respiratory chain is inhibited/damaged by reactive
species 8 produced excessively in sepsis. 1,9,10 Mitochondrial dysfunction can
also result from lack of substrate, decreased protein turnover, and increased
uncoupled (non-ATP generating) respiration – ‘proton leak’.
Previous studies have examined mitochondrial function ex-vivo, using
isolated mitochondria or homogenised tissue samples. I have extended
this scope by applying new techniques to investigate temporal effects of
sepsis on:
1. Whole animal metabolism and muscle function.
2. Transmembrane potential difference (Δψ m ) in freshly isolated living
skeletal muscle (diaphragm)
3. Mitochondrial proton leak and phosphorylation kinetics.
Methods
Experiments were performed on 30-week male C57-black mice under
Home Office guidelines. Sepsis was induced by intraperitoneal (i.p.)
injection of 20 mL/kg caecal slurry. Two groups of sham animals (fed and
starved) received i.p. 0.9% saline. All animals received subcutaneous 0.82%
saline/5% dextrose (50 mL/kg) at 6 hr and 18 hr. Mice were housed in
metabolic cages to measure whole body oxygen consumption (VO 2 ) and
carbon dioxide production (VCO 2 ).
Echocardiography was performed at 0 hr, 6 hr and 24 hr post-induction
of sepsis under brief (but equi-depth) isoflurane anaesthesia. Core
temperature was also measured.
At 24h, sepsis severity was scored using a panel of clinical variables
based on individual appearance, behaviour and biochemical data. 3 Animals
were then sacrificed, diaphragm muscle removed intact and 1.5 mm wide
strips of muscle dissected and attached to an ergometer force transducer
continuously superfused with oxygenated Krebs’ solution. Maximal tetanic
isometric force and the power generated while imposing length changes
(simulating in-vivo functional movement of the diaphragm) were
measured. 11 Fatigue testing involved measuring power output during one
minute of repetitive length changes, with 2Hz electrical stimulation
corresponding to the mouse’s respiratory rate.
The role of the ryanodine receptor in muscle dysfunction was explored
by measuring isometric force generation following exposure to 30 mM
caffeine.
Δψ m was measured in freshly isolated oxygenated diaphragm strips by
multiphoton confocal microscopy using a voltage-dependent fluorescent
probe, tetramethylrhodamine methylester (TMRM) 100nM. 12 Increased dye
fluorescence is indicative of a higher Δψ m .
Changes in Δψ m were explored using modular kinetic analysis of proton
leak and phosphorylation in mitochondria isolated from skeletal muscle. 13
This involved energisation of mitochondria by the Complex II substrate
succinate (together with the Complex I inhibitor, rotenone) in a chamber
allowing simultaneous measurement of oxygen consumption and Δψ m .
Changes in oxygen consumption were measured in the presence (State III)
and absence (State IV respiration) of ADP during alterations in Δψ m
achieved by sequential inhibition of the respiratory chain.
Results
Profound (60-70%) decreases in VO 2 were observed in severely septic mice
within hours, corresponding with development of hypothermia. At 24 hr,
core temperature was 37.5±0.2°C [sham] vs 31.5±1.2°C [severe septic]
(p

State of the Art 2010 abstracts
References
1. Brealey D, Brand M, Hargreaves I et al. Association between mitochondrial dysfunction
and severity and outcome of septic shock. Lancet 2002;360:219-23.
2. Crouser ED. Mitochondrial dysfunction in septic shock and multiple organ dysfunction
syndrome. Mitochondrion 2004;4:729-41.
3. Brealey D, Karyampudi S, Jacques TS et al. Mitochondrial dysfunction in a long-term
rodent model of sepsis and organ failure. Am J Physiol Regul Integr Comp Physiol 2004;
286:R491-497.
4. Callahan LA, Supinski GS. Sepsis induces diaphragm electron transport chain
dysfunction and protein depletion. Am J Respir Crit Care Med 2005;172:861-68.
5. Lanone S, Taillé C, Boczkowski J et al. Diaphragmatic fatigue during sepsis and septic
shock. Intensive Care Med 2005; 31:1611-17.
6. Duchen MR. Mitochondria in health and disease: perspectives on a new mitochondrial
biology. Mol Aspects Med 2004;25:365-451.
7. Martyn DA, Gordon AM. Force and stiffness in glycreated rabbit psoas fibres. Effect of
calcium and elevated phosphate. J Gen Phsyiol 1992;99:795-816.
8. Brown GC. Regulation of mitochondrial respiration by nitric oxide inhibition of
cytochrome c oxidase. Biochim Biophys Acta 2001;1504:46-57.
9. Haden DW, Suliman HB, Carraway MS et al. Mitochondrial biogenesis restores oxidative
metabolism during Staphylococcus aureus sepsis. Am J Respir Crit Care Med 2007;176:
768-77.
10. Boczkowski J, Lisdero CL, Lanone S et al. Endogenous peroxynitrite mediates
mitochondrial dysfunction in rat diaphragm during endotoxemia. FASEB J 1999;13:1637-
46.
11. Stevens ED, Faulkner JA. The capacity of mdx mouse diaphragm muscle to do
oscillatory work. J Physiol 2000;522:457-66.
12. Duchen MR, Surin A, Jacobson J. Imaging mitochondrial function in intact cells. Meth
Enzymol 2003;361:353-89.
13. Parker N, Affourtit C, Vidal-Puig A et al. Energization-dependant endogenous activation
of proton conductance in skeletal muscle mitochondria. Biochem J 2008;412:131-39.
14. Kreymann, G, Grosser S, Buggisch P et al. Oxygen consumption and resting metabolic
rate in sepsis, sepsis syndrome and septic shock. Critical Care Med 1993;21:1012-19.
15. Peres Bota D, Lopes Ferreira F, Melot C et al. Body temperature alterations in the
critically ill. Intensive Care Med 2004;30:811-16.
16. Callahan LA, Stofan DA, Szweda LI et al. Free radicals alter maximal diaphragmatic
mitochondrial oxygen consumption in endotoxin-induced sepsis. Free Radic Biol Med
2001;30:129-38.
17. Friedrich O. Critical illness myopathy: what is happening Curr Opin Clin Nutr Metab
Care 2006;9:403-09.
18. Zink W, Kaess M, Hofer S et al. Alterations in intracellular Ca2+-homeostasis of skeletal
muscle fibers during sepsis*. Critical Care Med 2008;36:1559-63.
Research Free Paper Presentations Winner
Combined immuno-phenotyping of
neutrophils, monocytes and T-cells enables
accurate prediction of the risk of
nosocomial infection
A Conway Morris*, M Brittan*, J Antonelli † , C McCulloch † , IF
Laurenson † , DG Swann † , A Hay † , TS Walsh † , AJ Simpson*
*MRC Centre for Inflammation Research, University of Edinburgh,
Edinburgh, UK. † Royal Infirmary of Edinburgh, Edinburgh, UK
Nosocomial infection is a significant problem in intensive care units and is
thought to be linked to critical illness-associated immune dysfunction. This
study aimed to examine the relationship between three immune cell types
(neutrophils, monocytes and regulatory T-cells) and acquisition of
nosocomial infection.
Ninety-five critically ill patients (patients requiring support of ≥1 organ
systems for ≥48 hours) were recruited within 48 hours of admission to
intensive care. Serial blood samples were taken, and measures made of
markers of neutrophil dysfunction (low CD88 expression 1 ), monocyte
deactivation (low HLA-DR expression 2 ) and levels of immunosuppressive
regulatory T-cells (T-regs). Infection was determined by pre-defined criteria.
Patient data were censored from 48 hours before nosocomial infection
occurred to reduce the risk of infection itself causing the effect seen.
Thirty-three patients developed nosocomial infection. Neutrophil
dysfunction was associated with a relative risk (RR) and Likelihood ratio
(LR) of nosocomial infection of 5.4 (95%CI 1.4-21, LR1.5) compared to
patients without dysfunction, monocyte deactivation an RR of 2.9 (95%CI
1.2-6.8, LR 1.5) and elevated T-regs with an RR of 2.3 (95%CI 1.3-4.1, LR
2.2). Many patients had dysfunction of more than one cell type, with a
cumulative increase in risk (Table). The relationship was significant when
adjusted for potential confounders in a logistic regression analysis (p8 units, and coagulopathy and a greater ICU mortality, but did not
directly compare transfused with non-transfused patients.
In order to make this comparison Kaplan-Meier survival estimates were
plotted. Patients transferred/discharged or in ICU over 30 days (the cut-off
in ISOC data collection) were censored. The curves were compared using
the log rank test, and a Hazard Ratio (HR) was calculated using Cox
regression, with RBCT as the only covariate. To adjust for potential
confounding a Cox model adjusted for RBCT, Apache II score, and the
presence of coagulopathy was fitted. The HR for RBCT in the unadjusted
and adjusted models was compared.
The Kaplan-Meier Plot suggested patients receiving RBCT had better
survival. Figure 1 shows the curves separating early and diverging until day
15, after which the numbers at risk (and precision of estimates) was low.
66
Volume 12, Number 1, January 2011 JICS

State of the Art 2010 abstracts
The unadjusted HR suggested that receiving RBCT reduced the relative risk
of all cause mortality by 29% (95% CI 11 to 43). After adjusting for
APACHE II score and coagulopathy this association was strengthened
(unadjusted HR 0.71 (95% CI: 0.57 to 0.89); adjusted HR 0.54 (0.43 to
0.68). Our data suggest an association between receiving RBCT and greater
ICU survival. These finding could be explained by residual confounding,
however they are consistent with the results of recent studies that used
more comprehensive statistical adjustment. 4 It is likely that the
combination of these findings and the study cited supports the need for
further large pragmatic trials of RBCT in the critically ill.
Numbers at risk:
Non RBCT 1290 189 67 25 9 4 0
RBCT 633 313 201 130 81 58 6
Log rank x 2 =8.82, p=0.03
Hazard ratio = 0.71 (95% CI 0.57 to 0.89)
Figure 1. Kaplan-Meier survival estimates in patients receiving RBCT and
those not receiving RBCT.
Acknowledgements: This work was done as part of the first author’s
Masters Dissertation. Thanks is owed to the individuals involved in the
original ISOC study, and to Professor Robin Prescott for extracting the data
from the original ISOC dataset.
References
1. Hebert PC, Wells G, Blajchman MA et al. A multicentre, randomized, controlled clinical
trial of transfusion requirements in critical care. N Engl J Med 1999;340:409-17.
2. Vincent JL, Baron JF, Reinhart K et al. Anemia and blood transfusion in critically ill
patients. JAMA 2002;288:1499-507.
3. Corwin HL, Gettinger A, Pearl RG et al. The CRIT Study: Anemia and blood transfusion
in the critically ill – Current clinical practice in the United States. Crit Care Med
2004;32:39-52.
4. Sakr Y, Lobo S, Knuepfer S et al. Anemia and blood transfusion in a surgical intensive
care unit. Crit Care 2010;14:R92.
5. Walsh TS, Stanworth SJ, Prescott RJ et al. Prevalence, management and outcomes of
critically ill patients with prothrombin time prolongation in United Kingdom intensive
care units. Crit Care Med 2010; in press.
Long-term effects of organ failure in the
critically ill
NI Lone, TS Walsh
Centre for Population Health Sciences, University of Edinburgh,
Edinburgh, UK
The Sequential Organ Failure Assessment (SOFA) score has been shown to
be associated with ICU mortality. 1 However, few studies have investigated
the relationship between SOFA score and long-term mortality. The aims of
this study were to report the relationship between organ failure and longterm
mortality in ICU patients, and to investigate if the total burden of
organ failure had an ongoing effect on mortality.
We undertook a secondary analysis of a dataset collected prospectively
for a multicentre cohort study. We excluded those with a neurological
diagnosis and those under 16. Daily SOFA score was calculated
prospectively for each patient and the worst SOFA score during the ICU
stay for each of the five organs was recorded in the dataset. A measure of
total organ failure burden was derived as the sum of the worst SOFA scores
at any time point during the stay. Individual organ SOFA scores were
classified into three categories: 0, 1-2, and 3-4. Total organ failure burden
was categorised into mild (0-5), moderate (6-11) and severe (≥12). Death
status was retrieved by linking the original study dataset to the Scottish
Death Registry database. Short and long-term mortality were defined as
death at three months and five years from ICU admission. A logistic
regression model was used to adjust for the following confounders: age,
sex, and APACHE II score on admission to ICU.
Of 1023 admissions during the study period, 151 (15%) were excluded
(neurological diagnosis n=141). The mean age was 59, 380 (44%) were
female, and the mean APACHE II score on admission was 20.3. ICU
mortality was 26%. Short-term mortality was 40%, and long-term mortality
was 58%. In multivariable analyses (n=745 due to missing data),
cardiovascular (p

State of the Art 2010 abstracts
Mean (SEM) Sham (n=7) CPB alone CPB + Sildenafil P value
(n=7) (n=7)
Creatinine clearance +22.2 (11.6) -29.1 (13.1)* +19.3 (7.8)** 0.009
Δ 24 hr (mL/min)
Urinary IL-18 post 27.5 (13.5) 236.9 (37.0)* 0.02 (0.003)**

State of the Art 2010 abstracts
presence of sepsis, admission haemoglobin, length of ICU stay and
study centre.
Binary logistic regression was used. Two regression models were fitted,
the first included variables present at admission to ICU and the second
included all ten variables. Additionally, a unifactorial association of each of
the predictor variables with RBCT was estimated to aid interpretation of the
multifactorial models.
Results are shown in Table 1, on the previous page. For baseline data
age, admission Hb, Apache II score and study centre were independent
predictors of RBCT in ICU. The second model upheld these predictors and
additionally suggested ICU stay over two days, bleeding in ICU and ICU
coagulopathy as further independent predictors of RBCT. Gender, weight
and sepsis were not significant predictors.
Illness severity, age, admission haemoglobin, presence of coagulopathy
and length of ICU stay have been described previously as independent
predictors of RBCT and/or are biologically plausible. 4,5 The apparent
association with study centre was interesting and could have resulted from
case mix. An alternative explanation is variation in clinical practice
between centers.
Acknowledgements: This work was done as part of the first author’s
Masters Dissertation. Thanks is owed to the individuals involved in the
original ISOC study, and to Professor Robin Prescott for extracting the data
from the original ISOC dataset.
References
1. Vincent JL, Sakr Y, Sprung C et al. Are blood transfusions associated with greater
mortality rates Anesthesiology 2008;108:31-39.
2. Sakr Y, Lobo S, Knuepfer S et al. Anemia and blood transfusion in a surgical intensive
care unit. Crit Care 2010;14:R92.
3. Walsh TS, Stanworth SJ, Prescott RJ et al. Prevalence, management and outcomes of
critically ill patients with prothrombin time prolongation in United Kingdom intensive
care units. Crit Care Med 2010; In Press.
4. Walsh TS, Saleh E-E-D. Anaemia during critical illness. Bri J Anaesthesia 2006;97:
278-91.
5. Chant C, Wilson G, Friedrich JO. Anemia, transfusion and phlebotomy practices in
critically ill patients with prolonged ICU length of stay: a cohort study. Crit Care
2006;10:R140.
Long-term survival of alcoholic liver
disease following critical care admission: a
national cohort study
AM Shaik-Dawood*, NI Lone † , TS Walsh*
*Edinburgh Royal Infirmary, Edinburgh, UK. † Centre for Population
Health Sciences, University of Edinburgh, Edinburgh, UK
The number of patients with alcoholic liver disease (ALD) admitted to
critical care is increasing in line with an increase in alcohol consumption
in the United Kingdom. 1 There is little published literature on long-term
outcomes for ALD patients following intensive care unit (ICU) admission
at a national level. The aim of this study was to report long-term
outcomes and resource utilisation for ALD patients requiring ICU
admission in Scotland.
The Scottish Intensive Care Society Audit Group (SICSAG)
prospectively collects data on all ICU admissions to all participating ICUs
in Scotland. Data were extracted for all patients admitted during a threeyear
period (1/1/2005 to 31/12/2007) meeting at least one of the
following diagnostic codes: ALD, alcohol dependence (AD), upper gastrointestinal
bleed (GIB), cirrhosis, hepatic encephalopathy, portal
hypertension and liver transplant. We selected the study cohort by
including only patients with a diagnostic code of ALD. In addition, those
with a diagnostic code of alcohol dependence combined with a code
indicating cirrhosis ± complications were included. Sepsis-related and
upper GIB diagnoses were defined by a process of consensus. Survival
analyses were undertaken using the long-rank test to compare Kaplan-
Meier plots. Death status was retrieved by linking the dataset to the
Scottish Death Registry database. The shortest period of follow up was
19-months, the longest 57-months. Therefore 18-month mortality data
was available for all cases.
Of 2872 admissions in the extract, 1699 (59.2%) were excluded
(transplants, isolated AD, unspecified GIB and non-alcoholic liver diseases).
The remaining cohort (n=1173) comprised the ALD study cohort, of whom
16 (1.4%) were missing outcome data. The mean age was 52.0, 759 (65%)
were male and the mean APACHE II score was 23.3. The prevalence of
mechanical ventilation, cardiovascular-support and haemofiltration was
85%, 60% and 19% respectively and 17% received all three. The prevalence
of a sepsis-related diagnosis was 48.3% and GIB 25.1%. The mean length of
stay was 6.2 (ICU) and 18.3 (hospital) days. Short-term mortality was 42.9%
(ICU), 55.2% (hospital) and long-term mortality was 64.4% (one year),
67.1% (18 months). The short-term mortality was higher than the whole
Scottish ICU and hospital mortality between 2005 and 2007.2,3 Eighteenmonth
mortality increased with number of organs supported (chi-square
test-for-trend p

State of the Art 2010 abstracts
annually. 1 Further costs to the criminal justice system, social care and the
economy bring the total ARH expenditure to approximately £20 billion.
These harms are not gender or age specific and as a result form part of the
caseload for many NHS staff. 2
The North West has been identified as having the second highest rates
of admissions to hospital in the UK. 1 This audit was conducted to assess
the impact that ARH have upon an ICU within this area.
Data was collected during an eight week prospective audit from patients
identified by staff as having an ARH. Patients with a significant history of
alcohol excess were also included in this study as it has been shown that
these patients respond less well to treatment. 3 Information was collected
regarding the reason for admission, maximum number of organs supported,
age, diagnosis, gender, postcode, date admitted, whether the ICU admission
was planned or related to trauma, whether self harm was involved, whether
the patient was post operative and the patient outcome.
Of the 200 admissions, 21 were alcohol related; one of which was
planned. The mean cost per admission was £5,007.84 for non alcohol
related and £13,457.46 for alcohol related admissions. The mean age of
alcohol related admissions was 50 compared to 62 for non alcohol related.
Of the alcohol related admissions, 38% died compared to 10% of non
alcohol related. There were 15 admissions related to trauma, seven were
alcohol related. Fourteen percent of non alcohol related admissions were
admitted on a weekend compared to 62% of alcohol related.
Alcohol played a significant role in the ICU workload over the 8 weeks
studied. ARH patients were younger, were more likely to die during the
admission, be unplanned and be as a result of trauma. Alcohol related
admissions cost considerably more. Further study is required as the sample
size was small.
To reduce the impact that ARH have upon the ICU, further emphasis
needs to be put upon primary prevention such as a minimum price for
alcohol and tighter regulations for sponsorship to decrease both
affordability and awareness. There is no fast acting and simple solution to
this culture of excessive drinking, but it is clear that these significant
monetary and human costs caused by alcohol related harms cannot
continue.
References
1. NHS Confederation and Royal College of Physicians. Too much of the hard stuff: what
alcohol costs the NHS. [Online]. 2010 [Cited 11/07/2010] http://www.nhsconfed.org/
Publications/Documents/Briefing_193_Alcohol_costs_the_NHS.pdf
2. House of Commons Health Committee. Alcohol. Report of first session 2009-2010.
[Online]. 2010 [Cited 11/07/2010] http://www.publications.parliament.uk/pa/
cm200910/cmselect/cmhealth/151/151i.pdf
3. Uusaro A, Parviainen I, Tenhunen JJ, Ruokonen E. The proportion of intensive care unit
admissions related to alcohol use: a prospective cohort study. Acta Anaesthesiol Scand
2005;49:1236-40..
Research Poster Presentations winner
Interactions between allogenic red cell
transfusion and cardiopulmonary bypass in
post cardiac surgery acute kidney injury
NN Patel*, H Lin*, T Toth † , C Jones*, P Ray ‡ , SC Satchell † , GI Welsh † , R
Cardigan § , SJ George*, GD Angelini*, GJ Murphy*
*Bristol Royal Infirmary, Bristol, UK. † Southmead Hospital, Bristol, UK.
‡
Weston General Hospital, Weston-Super-Mare, UK. § National Health
Service Blood and Transplant
Acute kidney injury (AKI) post cardiac surgery is associated with mortality
rates approaching 20%. Clinical studies have demonstrated associations
between allogenic red blood cell (RBC) transfusion and AKI in patients
undergoing cardiac surgery, however causality has not been established.
The aim of this study was to determine whether transfusion has a causal
effect on the development of AKI in a large animal experimental model of
post cardiopulmonary bypass (CPB) AKI with significant homology to that
which occurs in cardiac surgery patients.
Adult pigs were randomised to undergo Sham procedure, CPB, Sham +
RBC Transfusion or CPB + RBC Transfusion in a factorial model design
with recovery and reassessment at 24 hours.
CPB reduced renal perfusion pressure and the circulating haematocrit
resulting in endothelial injury and activation, increased endothelin-1 and
cortical adenosine expression, global endothelial dysfunction and
vasoconstriction, medullary hypoxia, glomerular inflammatory cell
infiltration and sequestration of activated platelets, proteinuria, tubular
epithelial cell stress and a 47% reduction in creatinine clearance (CrCl), or
a mean reduction of 64.9 mL/min (95% CI 20.2-109.6) at 24 hours.
Transfusion did not cause significant endothelial activation and injury
however it did result in glomerular inflammation and platelet
sequestration, a reduction in cortical microvascular flow, increased
endothelin-1 expression, medullary hypoxia and a reduction of CrCl –
30.53 mL/min (-58.47 to -2.59), p=0.034, at 24 hours. Transfusion in CPB
pigs increased renal perfusion pressure, reduced endothelial injury and
expression of cortical vasoconstrictors attributed to CPB and improved
medullary oxygenation at 24 hours. This was associated with a
reduction in glomerular inflammation, proteinuria and markers of tubular
epithelial stress, although the improvement in CrCl relative to CPB was
not significant (mean difference 33.5 mL/min, 95% CI -13.7 to 80.77,
p=0.141)
In this model RBC transfusion and CPB cause acute kidney injury
via distinct mechanisms. Transfusion reverses some of the pathophysiological
processes attributable to CPB induced oxygen supply
dependency. These findings highlight the complexity of acute kidney
injury and suggest novel therapeutic targets: endothelial dysfunction and
platelet activation.
Research Poster Presentations
Applicability of palliative quality measures
to end-of-life care in ICUs in the UK and
Israel
Up to 90% of patient deaths in ICU involve ‘end-of-life’ decision-making.
Previous studies have demonstrated that there are differences in intensive
care units between and within countries in: patient and family members’
preference for involvement in decisions about end of life care; end-of-life
care practices and the role of health professionals in shared decision
making. 1-4 The Department of Health strategy for end-of-life care places
emphasis on conducting timely discussions to allow patients’ wishes to be
met; studies comparing end of life care across countries identify ICUs in
Israel as having higher rates of withholding treatment at end-of-life (the
ETHICUS study 3,4 ). The original study team acknowledge that these
differences may be due to religious affiliation of physicians; however, these
data do suggest that there are lessons that can be learnt from colleagues in
Israel regarding the conduct of timely discussions with carers and, where
possible, patients.
Studies have shown that the quality of ICU end-of-life care needs to be
improved. In order to improve the quality of care, some method must be
developed to measure it. An extensive body of work in the United States
(US) has identified a series of Palliative Quality Measures (PQM) for end of
life care in intensive care units. This study examined whether the palliative
quality measures identified in the US apply in Israel and the UK.
The study was conducted in three phases:
Phase 1: Medical Record Review (n=84 records) to examine validity,
reliability of the PQM tool in Israel and UK
Phase 2: Focus Group Interviews with ICU Nurses (n=48-54
participants) to examine what is considered as a ‘good death’ in ICUs in
Israel and the UK
Phase 3: Observation of nurse-to-nurse handover (n=120 observations
across six ICUs) to examine verbal communication of PQM.
Data collection is Israel was undertaken in April/May 2010; UK data
collection commenced in November 2010.
References
1. Azoulay E, Pochard F, Chevret S, Adrie C et al. Half the family members of intensive
care unit patients do not want to share in the decision-making process: a study in 78
70
Volume 12, Number 1, January 2011 JICS

State of the Art 2010 abstracts
French intensive care units. Crit Care Med 2004;32:1832-38.
2. Pochard F, Azoulay E, Chevret S, Lemaire F et al. Symptoms of anxiety and depression
in family members of intensive care unit patients: ethical hypothesis regarding decisionmaking
capacity. Crit Care Med 2001;29:1983-97.
3. Benbenishty J, Ganz FD, Lippert A et al. Nurse involvement in end-of-life decision
making: the ETHICUS study. Intensive Care Med 2006;32:129-32.
4. Sprung CL, Woodcock T, Sjokvist P et al. Reason, considerations, difficulties and
documentation of end-of-life decisions in European intensive care units: the ETHICUS
study. Intensive Care Med 2008;34:271-77.
Sodium bicarbonate is as effective as
hypertonic saline for reduction of raised
intracranial pressure after traumatic
brain injury
CP Bourdeaux, JM Brown
Frenchay Hospital, Bristol, UK
In a previous study we demonstrated that 8.4% sodium bicarbonate (HSB)
was effective at reducing raised intracranial pressure (ICP) after severe
traumatic brain injury (TBI). 1 In this randomised controlled trial we
compared the effects of an equiosmolar dose of 8.4% sodium bicarbonate
with our standard treatment of 100mL of 5% sodium chloride (HS) for the
treatment of raised ICP after TBI.
Ethical approval was granted by the research ethics committee for
Wales and the medical and healthcare products regulatory agency approved
the use of 8.4% sodium bicarbonate. Patents were recruited to the trial if
they had suffered severe TBI (GCS 20 mm Hg for >5 minutes) patients were randomised to receive
either HS (100 mL 5% saline) or HSB (85 mL 8.4% sodium bicarbonate).
Patients were studied for six hours after each dose, monitoring ICP,
mean arterial pressure (MAP) and cerebral perfusion pressure (CPP).
Patients could be re-randomised, treated and studied for further episodes
every 24 hours. The primary outcome measure was change in ICP
after treatment.
Twenty episodes of raised ICP were studied in 11 patients. Ten episodes
were treated with HS and 10 episodes were treated with HSB. Six patients
had suffered a road traffic collision and five patients had fallen from a
height. Baseline variables were not different between the treatment groups.
Intracranial pressure (mm Hg)
25
20
15
10
Figure 1. Mean intracranial pressure vs time.
Hypertonic saline
Sodium bicarbonate
0 100 200 300 400
Time (minutes)
All patients responded to initial therapy (ICP 20 mm Hg. No patients required a second dose of HSB.
Analysis of the data using a 2 way ANOVA indicates that there was a
statistically significant fall in ICP from baseline at all time points, p

State of the Art 2010 abstracts
This work confirms the finding of elevated T-regs in human sepsis. It
extends this finding to patients with non-septic critical illness, suggesting
that this is part of a stereotyped counter-regulatory response to systemic
inflammation. It contributes to the growing understanding of immune
suppression among the critically ill.
Funding: This work was funded by the Chief Scientist Office.
Reference
1. Venet F, Chung C-S, Kherouf H et al. Increased circulating regulatory T cells contribute
to lymphocyte anergy in septic shock patients. Intensive Care Med 2009;35:678-86.
Neutrophil microparticles and their
contents as potential novel biomarkers in
sepsis
J Dalli*, J Radhakrishnan* † , X Yin ‡ , M Mayr ‡ , JC Knight † , C Hinds* † ,
M Perretti*
William Harvey Research Institute, Barts and the London School of
Medicine and Dentistry, London, UK. † Wellcome Trust Centre for Human
Genetics, University of Oxford, Oxford, UK. ‡ King’s College British Heart
Foundation Centre and Centre for Bioinformatics, School of Physical
Sciences and Engineering, London, UK
Microparticles were originally described as platelet dust, but it is now
recognised that they are more than inert by-products of cellular
activation. 1,2 Studies conducted in a number of inflammatory conditions
have shown that plasma levels of neutrophil derived microparticles are
elevated in these patients, suggesting that they might be playing an
important pathogenic role. To date, very little is known about the
contents of neutrophil-derived microparticles. We therefore analysed the
protein content of microparticles derived from activated human
neutrophils.
First neutrophils were stimulated with 1 mmol fLMP under two
distinct conditions, either in suspension or post adhesion to human
umbilical endothelial cells. Microparticles were extracted from
supernatant using ultracentrifugation and then subjected to proteomic
analysis (liquid chromatography followed by mass tandem spectrometry).
Among the many candidates (shared or unique for each population) two
proteins differentially expressed between the two groups and two that
were equally expressed were identified by Western blotting and flow
cytometry. We then proceeded to analyse protein expression in plasma
microparticles obtained from healthy volunteers and patients suffering
from sepsis recruited to the Genomic Advances in Sepsis (GAinS) study.
The initial proteomic analysis showed that there were over 600
different proteins present in these microstructures of which more then
200 were significantly enriched between the two populations of
stimulated neutrophils. Western blotting and flow cytometry confirmed
the presence of ceruloplasmin and alpha-2-macroglobulin in plasmaderived
microparticles and demonstrated differential microparticle
associated protein expression between healthy volunteers and patients
with sepsis (n=10; P

State of the Art 2010 abstracts
TACE activity — a potential indicator of
monocyte inflammatory status during
sepsis
DJP O’Callaghan, KP O’Dea, AJ Scott, M Takata, AC Gordon
Imperial College, London, UK
The immune response during sepsis is comprised of both a systemic
inflammatory response syndrome (SIRS) and a compensatory antiinflammatory
response syndrome (CARS). Monocytes can display an
altered response status, eg a state of priming or tolerance, when challenged
with an inflammatory stimulus. These states, as measured using in-vitro
assays, may be considered representative of those occurring in SIRS and
CARS respectively.
Tumour necrosis factor-α (TNF) converting enzyme (TACE) is
responsible for the shedding of membrane proteins, including TNF and its
receptors, and is expressed on monocytes. 1 Selective substrate shedding
can have opposing effects, potentially allowing TACE to dictate monocyte
inflammatory balance.
To determine TACE activity within a two-hit lipopolysaccharide (LPS)
model, human monocytes were isolated from whole blood via density
gradient centrifugation and magnetic activated cell sorting utilising CD14
positive bead selection. Cells were re-suspended and placed in nonadhesive
conditions, with variable doses of LPS (low dose 1 ng/mL, high
dose 1 µg/mL, and nil as control) for 16 hours.
n=5 for all groups, data display as mean ± standard error. For each group:
grey bar represents baseline activity afer 16 hour pre-treatment, black bar
represents activity after LPS 1μg/mL for one hour.
Monocytes were washed and exposed to a further LPS stimulus
(1 µg/mL for one hour) before having their TACE activity quantified using
a cell based fluorometric catalytic activity assay. 2 TACE expression levels
were determined via flow cytometry. A t-test was used for analysis.
LPS stimulation induced up-regulation of TACE activity in controls and
in the low-dose LPS group. In the high-dose LPS group baseline TACE
activity remained elevated and did not increase on further stimulation.
There were no significant differences in TACE expression after pretreatment
between groups.
These findings suggest that TACE has a memory of previous LPS
exposure at two levels: activity is higher and cannot be re-induced.
Furthermore monocyte inflammatory status can be assessed at the level
of altered cellular protein processing activity, suggesting TACE activity
may be a novel and useful indicator of such status during the course
of sepsis.
References
1. Black R, Rauch C, Kozlosky C et al. A metalloproteinase disintegrin that releases
tumour-necrosis factor-alpha from cells. Nature 1997;385:729-33.
2. Alvarez-Iglesias A, Wayne G, O’Dea KP et al. Continuous real-time measurement of
tumour necrosis factor-a converting enzyme activity on live cells. Lab Invest
2005;85:1440-48.
Prevention of post-cardiopulmonary
bypass acute kidney injury and endothelial
dysfunction using sitaxsentan sodium, an
endothelin-a receptor antagonist
NN Patel*, T Toth † , C Jones*, H Lin*, P Ray ‡ , SJ George*, G Welsh † , SC
Satchell † , GD Angelini*, GJ Murphy*
*Bristol Royal Infirmary, Bristol, UK. † Southmead Hospital, Bristol, UK.
‡
Weston General Hospital, Weston-Super-Mare, UK
Acute kidney injury (AKI) post cardiac surgery is associated with mortality
rates approaching 20%. Our objective was to characterise post cardiopulmonary
bypass (CPB) AKI in an animal model with significant
homology to cardiac surgery patients and to assess the effect of Sitaxsentan
Sodium, an endothelin-A receptor antagonist on these changes.
Adult White-Landrace pigs (50-70 kg, n=21) were randomised to
undergo either: a) sham procedure, b) 2.5 hours of CPB, or c) 2.5 hours of
CPB + sitaxsentan sodium (0.7 mg/kg). Perfusion pressure and hydration
were standardised. Endpoints included serial functional and biochemical
measures of AKI. All pigs were recovered for 24 hours prior to in-vivo
measurement of renal endothelial function, nephrectomy and histological
assessment.
Mean (±95% CI) Sham CPB alone CPB + Sitaxsentan p value
(n=7) (n=7) (n=7)
Creatinine clearance +22.4 -27.6 +39.7 0.010
Δ 24 hrs (mL/min) (-10.4, +55.3) (-64.4, +9.1)* (+9.3, +70.2)**
Proteinuria Δ +5.36 +29.26 *-7.68

State of the Art 2010 abstracts
Pharmacological therapies for the
prevention of acute kidney injury following
cardiac surgery: a systematic review
NN Patel, CA Rogers, GD Angelini, GJ Murphy
Bristol Royal Infirmary, Bristol, UK
Post cardiac surgery acute kidney injury (AKI) is common, is associated
with a significant increase in morbidity and mortality. We aimed to
systematically review randomised trials that assessed the renoprotective
utility of pharmacological agents in patients undergoing cardiac surgery.
A systematic review of all randomised trials comparing renoprotective
pharmacological interventions with control in patients undergoing cardiac
surgery with cardiopulmonary bypass was undertaken.
47 randomised controlled trials involving 3942 patients were included.
Pharmacological interventions included dopamine, fenoldopam, calcium
channel antagonists, natriuretic peptides, diuretics, and N-acetylcysteine.
Most trials were of poor quality, with small sample sizes, under reporting
of randomisation procedure, allocation concealment and method of
blinding. There was marked variation in interventions being compared,
populations receiving the intervention and outcomes being measured.
No pharmacological intervention was associated with a significant
reduction in mortality or incidence of AKI. Fenoldopam and Atrial
Natriuretic Peptide (ANP) were associated with a 70% (NNT 21, 95% CI
11.3, 83.0) and 73% (NNT 14, 95% CI 8.6, 81.0) reduction, respectively,
in the need for RRT. Fenoldopam and sodium nitroprusside (SNP) were
associated with a significant increase of 26.04 mL/min (95% CI 18.94,
33.15) and 6.60 mL/min (95% CI 2.41, 10.79), respectively in creatinine
clearance at 24-48 hr, whereas dopamine was associated with a significant
reduction (-4.26 mL/min, 95% CI -7.14, -1.39).
Vasodilatory agents, such as fenoldopam and ANP show evidence of
renoprotection in patients undergoing cardiac surgery. Evidence does
not support the use of dopamine, diuretics or N-acetylcysteine. Further
studies evaluating the effect of novel renoprotective strategies on
clinical outcomes in adequately powered randomised controlled trials
are required.
Genetic polymorphism of the beta-2
adrenoreceptor may affect alveolar fluid
clearance and mortality in the acute
respiratory distress syndrome
W Backman*, Z Puthucheary † , J Skipworth † , J Rawal † , J Palmen † , M Hill † , S
Humphries † , R Chambers † , H Montgomery* †
*University College London Medical School, London, UK. † University
College London, London, UK
Acute respiratory distress syndrome (ARDS) is defined by the acute
onset of non-cardiogenic pulmonary oedema and consequent hypoxaemia.
Impaired alveolar fluid clearance (AFC) has been postulated to play
a key pathogenenic role. AFC is mediated by apical membrane epithelial
sodium channels (ENaC) and basal membrane Na + /K + ATPase,
upregulated by the intracellular signalling molecules cAMP. cAMP levels
rise with β2 adrenoceptor (β2R)and type 2A adenosine receptor
(A 2A R) stimulation.
Adenosine is catabolised by adenosine deaminase (ADA). Single
nucleotide polymorphisms (SNPs) exist in the genes for A2AR, ADA, and
β2R. The β2R rs1042717 SNP (G→A) is associated with β2R
desensitisation. We aimed to identify a role for AFC in human ARDS by
seeking association of genetic variation in β2R, A 2A R and ADA with the
development of ARDS or mortality in ARDS cases.
DNA samples were collected from 111 ARDS patients and 72 ICU
control patients and genotyped using a PCR sequencing detection system
with custom-made probes. No association was found between ADA SNPs
rs244076 (p=0.28) and rs11086932 (p=0.188), the A2AR SNP rs5751876
(p=0.698) or the β2R SNP rs1042717 (p=0.68), and the presence of ARDS
nor (using a dominant allele model) with ARDS-related mortality.
However, a trend for a mortality-association was noted with the β2R SNP
(p=0.075) (Figure 1). Such an association of a marker of adrenoceptor
desensitization with improved outcome would be consistent with recent
data from the BALTI trial investigators – where beta-adrenoceptor
activation may be associate with an excess ARDS mortality.
% mortality
45
40
35
30
25
20
15
10
5
0
GG
AG/AA
Expected
Observed
Figure 1. AG/AA genotype of β2R tends towards decreased mortality
compared to GG. Expected (overall) mortality: 31%. GG mortality: 38.6%.
AG/AA mortality: 18.5%. p=0.075 by Chi-square test for independence.
No specific therapies (other than those which are supportive) are yet
available which influence the development of ARDS or outcome from it.
Our results show a trend in mortality with β2R desensitisation. Further
studies with greater numbers (likely necessitating international
collaborations) are necessary to confirm the importance of genetic control
of alveolar fluid clearance in ARDS.
The relationship between agitation and
pain during physiotherapy in intensive care
L Salisbury*, K Everingham † , P Lapinlampi ‡ , I Cornwall † , F Frame § ,
T Walsh †
*University of Edinburgh, Edinburgh, UK. † NHS Lothian, Edinburgh, UK.
‡
GE Healthcare, Helsinki, Finland. § University of Leicester, Leicester, UK
The occurrence and causes of distress in critically ill patients are not well
described or understood. Identified stressors include physiotherapy, 1,2 but
no standards for management of sedation or analgesia during
physiotherapy exist. Recent evidence supporting routine ICU management
at ‘lighter’ levels of sedation potentially increases the importance of
interventions to avoid distress during interventions such as physiotherapy.
The aim of this study was to explore the relationships between agitation
and pain during physiotherapy.
A prospective observational study was undertaken. Patients admitted to
the ICU requiring mechanical ventilation and physiotherapy were recruited
into the study. A single session of physiotherapy was observed by an
independent assessor. Agitation was measured using the Richmond
Agitation Sedation Scale (RASS) and pain using the Behavioural Pain Scale
(BPS). The RASS and BPS were collected immediately prior to the
commencement of physiotherapy (Baseline) and after each treatment
intervention undertaken during the single physiotherapy session.
Spearman’s rank correlation coefficients were undertaken to explore the
relationships between baseline RASS and BPS scores and the change in and
worse scores of RASS and BPS during the physiotherapy session.
Forty-nine patients were recruited into the study (32 males; age
57.4±13.9 years; APACHE II 22.5±6.2; physiotherapy duration
14.9±4.7 mins; sedation received for 24 hours prior to physiotherapy
74
Volume 12, Number 1, January 2011 JICS

State of the Art 2010 abstracts
128.7±126.5 mL; opiates received for 24 hours prior to physiotherapy
39.1±37.3 mL; ICU admission to study 100.9±97.6 hours). Table 1
summarises the Spearman Rank correlation coefficients.
Table 1. Spearman Rank correlation coefficients
Weak associations were found between baseline RASS and BPS scores
and subsequent RASS and BPS scores during a single physiotherapy
session, with the exception of the association between baseline RASS and
worse RASS score during physiotherapy.
In patients who are mechanically ventilated and receiving
physiotherapy their baseline RASS was moderately associated with their
worse RASS score during physiotherapy. This indicates that a patient’s
conscious level at baseline is likely to increase during a single session of
physiotherapy. As a result physiotherapists should consider whether
patients at ‘lighter’ levels of sedation may require a bolus prior to
commencing physiotherapy to prevent agitation during the treatment
session. More work is required exploring the measurement and
management of pain during physiotherapy in intensive care.
References
1. Novaes et al. Intensive Care Med 1999;25:1421-26.
2. van de Leur et al. Crit Care 2004;8:R467-R473.
Treating delayed gastric emptying in
critical illness: metoclopramide,
erythromycin and bedside (Cortrak)
nasointestinal tube placement
S Taylor, A Manara, J Brown
Frenchay Hospital, Bristol, UK
Baseline RASS
Baseline BPS
Change in RASS during physiotherapy 0.27 0.24
Worse RASS during physiotherapy 0.65 0.17
Change in BPS during physiotherapy 0.35 0.10
Worse BPS during physiotherapy 0.39 0.26
Nasointestinal feeding can overcome delayed gastric emptying common to
critical illness. This improves outcome associated with early enteral
nutrition but current methods of tube placement have failed to gain
widespread success or use. 1 We determined whether the
electromagnetically-guided nasointestinal tube (EGNT-Cortrak) system
was effective in improving enteral feed retention compared to nasogastric
feeding plus prokinetics.
In ICU patients with gastric residual volumes >200 mL we compared
two treatment cohorts: a) IV metoclopramide±erythromycin; or b)
Nasointestinal feeding when EGNT became available. We retrospectively
determined placement success, feed retention and gastric residual volumes.
Of 69 EGNT placements in 62 patients, 87% reached the small
intestine. Enteral fluid input rose in all groups from ~700 mL pretreatment
to >1600 mL d10. In contrast, more feed was retained by the
EGNT group (metoclopramide: d1-3, 5-7; p=0.0001-0.02; erythromycin:
d1-2, 6-8; p=0.006-0.041) despite higher gastric residual volumes
(metoclopramide: pre-treatment day 2+1 and d2-3, 5; p=0.0001-0.03;
erythromycin: pre-treatment d2+1 and d2-4; p0.0002-0.04).
The median percentage of the enteral nutrition goal retained d1-10
was higher following EGNT (80-100%) than metoclopramide (n=58,
40-87%: days 1-2, 5-7, p†0.018) or erythromycin (n=38, 48-98%; days 1
and 5, p

T he Intensive Car e Society
State of the Art 2010 abstracts
Pre-alarm ward round Post-alarm ward round
intervention
intervention
Total alarms Alarms/bed/hour Total alarms Alarms/bed/hour
Median
Median
(interquartile
(interquartile
range)
range)
Total alarms 1013 7.0* (2.0–15.0) 604 3.5* (1.0–11.0)
Nuisance alarms 878 6.0** (1.0–13.5) 502 2.0** (1.0–9.0)
Relevant alarms 135 1.0 † (0.0–2.0) 102 1.0† (0.0–2.0)
* Difference significant at p=0.0025, ** Difference significant at p=0.0022, † p = 0.41
and total alarms is presented in the table. ‘Total alarms’ is the sum of
alarms over 15 hours.
Our results show that introducing an alarm ward round significantly
reduces the amount of alarms occurring at night on a general Intensive
Care Unit. Of importance is that the alarm round decreases the frequency
of nuisance alarms without decreasing the frequency of relevant alarms.
References
1. Imhoff M, Kuhls S. Alarm algorithms in critical care monitoring. Anesth Analg
2006;102:1525-37.
2. Meredith C, Edworthy J. Are there too many alarms in the intensive care unit An
overview of the problems. J Adv Nurs 1995;21:15-20.
3. Balogh D, Kittinger E, Benzer A, Hackl JM. Noise in the ICU. Intensive Care Med
1993;19:343-46.
Become a member and...
...in addition to becoming part of a growing network of medical professionals working within
of up to £1,000,000 in the event of death or
disability during a patient transfer.
informative journal.
at the Society’s meetings and events.
to keep members informed of recent
developments in Intensive Care as well as forthcoming meetings, events and
opportunities such as grants and prizes and other news.
– a two way communication channel from the
Society to its members and vice versa.
grade of membership.
at a
further details .
76
Volume 12, Number 1, January 2011 JICS