An Alternative Approach to the Diagnosis of ... - Diabetes Care

R E V I E W A R T I C L E
An Alternative Approach to
the Diagnosis of Diabetes
With a Review of the
Literature
MAYER B. DAVIDSON, MD
ANNE L. PETERS, MD
DAVID L. SCHRIGER, MD, MPH
T
wo thoughtful articles in the May
1994 issue of Diabetes Care argued
the case for screening for type II di-
Knowler (2) discussed the general principles
of screening and made the following
points that are pertinent for diabetes.
abetes (1, 2). Harris and Modan (1) First, hyperglycemia is a continuum, and
pointed out that 1) large population sur- therefore the choice of cutoff values for
veys repeatedly show that approximately diagnosis is somewhat arbitrary. Second,
one-half of people with type II diabetes screening for diabetes is unusual because,
are undiagnosed; 2) type II diabetes (i.e., depending on the test(s) chosen, the
hyperglycemia) is present for ~10 years screening test may also be the diagnostic
before the diagnosis is made; 3) compli- one. Third, as for all diseases, the screencations
are often present at the time of ing test should have high test-perfor-
diagnosis; 4) diabetes (of which type II mance characteristics, i.e., sensitivity,
accounts for 90% of cases) exacts a dread- specificity, and predictive value. Finally,
ful toll on the population (e.g., 15% of all in a nonresearch setting, relatively strin-
blindness, 35% of all end stage renal disgent criteria should be selected to maxiease,
50% of all nontraumatic lower exmize the chance that detected subjects
tremity amputations, >100 billion dol- will benefit from therapy. Interestingly,
lars annual cost); 5) it is well established Harris and Modan (1) concluded that the
that hyperglycemia is the proximate cause oral glucose tolerance test (OGTT) should
of the microvascular (retinopathy and ne- be used as the primary screening (and diphropathy)
and neuropathic complicaagnostic) method, whereas Knowler (2)
tions of diabetes; and 6) early interven- concluded that a glycated hemoglobin
tion (e.g., diet, exercise, medications) is level measurement should be the method
helpful in reducing hyperglycemia. of choice.
From the Department of Medicine (M.B.D., A.L.P.), Cedars-Sinai Research Institute, Cedars-
Sinai Medical Center, and the UCLA School of Medicine (M.B.D., A.L.P., D.L.S.), Los Angeles,
California.
Address correspondence and reprint requests to Mayer B. Davidson, MD, 2103 Overland
Ave., Los Angeles, CA 90025.
Received for publication 15 August 1994 and accepted in revised form 2 March 1995.
FPG, fasting plasma glucose; IGT, impaired glucose tolerance; NDDG, National Diabetes Data
Group; NGT, normal glucose tolerance; OGTT, oral glucose tolerance test; ULN, upper limit of
normal.
BACKGROUND FOR
DIAGNOSTIC CRITERIA —
Before 1979, there were many different
criteria for interpreting OGTT results (3)
and no consensus existed regarding the
best cutoff points for diagnosing diabetes.
Studies performed in the 1970s and
1980s (4-7) evaluated the association between
2-h post-glucose load values
(two studies [6,7] also assessed fasting
plasma glucose [FPG] concentrations,
and in one, glycated hemoglobin levels
were also compared [7]) and the subsequent
development of the microvascular
complications of diabetes. These
complications, especially diabetic retinopathy,
are considered to be specific
for diabetes. Two of these studies (4,5)
evaluated patients who had abnormal
2-h capillary blood glucose values after
administration of 50 g of oral glucose
(fasting values were not considered).
The patients were classified as either diabetic
(2-h blood glucose ^200 mg/dl
[11.1 mmol/1]) or "borderline" diabetic
(2-h blood glucose 120-199 mg/dl
[6.7-11 mmol/1]) and were followed for
5-8 years with ophthalmoscopic examination
through dilated pupils to detect
diabetic retinopathy. Diabetic retinopathy
was defined in these studies as the
presence of vascular lesions, such as microaneurysms
alone (small red dots) or
obvious hemorrhage with or without
Table 1—5- to 8-year follow-up in subjects
with IGT or diabetes based on a 2-h
capillary blood glucose measurement after a
50-g glucose load
Subjects with diabetic retinopathy
and an initial 2-h blood glucose
measurement
120-199 mg/dl >200 mg/dl
Reference (6.7-11.1 mmol/1) (11.1 mmol/1)
5/245 *
0/112
20/108T
6/38
Three patients who had diabetic retinopathy at
baseline were excluded. TEight patients who had
diabetic retinopathy at baseline were excluded.
DIABETES CARE, VOLUME 18, NUMBER 7, JULY 1995 1065

Alternative approach to diagnosing diabetes
Table 2—Results of a 3-year follow-up in Pima Indians
Initial FPG
140 mg/dl (7.8 mmol/1)
2-h plasma glucose
200 mg/dl (11.1 mmol/1).
n
181
39
117
43
60
Retinopathy
2*
15
0
0
7t
Proteinuria
Data are n at risk at baseline. tThese two patients had a 2-h plasma glucose level ^200 mg/dl (11.1 mmol/1),
and one had an FPG between 130 and 139 mg/dl (7.2-7.7 mmol/1). TFive of these patients had FPG
concentrations S140 mg/dl (7.8 mmol/1). Data from Pettitt et al. (6).
exudates. Exudates alone did not constitute
diabetic retinopathy. The results
of these studies are summarized in Table
1.
Several of these studies were carried
out in Pima Indians (6,7). There is a
high prevalence of diabetes in this population,
and fasting and 2-h plasma glucose
concentrations and glycated hemoglobin
levels are found in a bimodal distribution,
which allows for the clear identification of
patients with diabetes. Both cross-sectional
and longitudinal data on diabetes
and its complications in this population
have been collected since 1965. At least
two of the studies published provide longitudinal
data on the development of retinopathy
and nephropathy in these patients
for several years compared with
baseline data on glucose tolerance (6,7).
Initially, glycated hemoglobin levels were
not available. A 3-year follow-up was provided
for patients who had a 75-g OGTT
with both FPG and 2-h plasma glucose
measured (Table 2), and 6-year follow up
was available for patients who had a measurement
of 2-h plasma glucose value
alone (Table 3). The presence or absence
of diabetic retinopathy (identified by microaneurysms,
hemorrhages, and neovascularization,
but not hard exudates because
they were considered less specific
for diabetes) was determined based on
ophthalmoscopic examination through
dilated pupils, and the presence of diabetic
nephropathy was defined as "heavy
proteinuria" (urine protein-to-creatinine
0
2
0
0
2
ratio >1.0). These studies suggested that
FPG values > 140 mg/dl (7.8 mmol/1) and
2-h plasma glucose values ^200 mg/dl
(11.1 mmol/1) defined a subset of Pima
Indians who were more likely to develop
diabetic retinopathy or nephropathy.
In a later study in Pima Indians,
941 patients without retinopathy or nephropathy
at baseline were followed for
4.5 (1.4-8.3) years (7). The development
of retinopathy or nephropathy (assessed
using the same criteria described above)
during follow-up was compared with
baseline fasting and 2-h glucose concentrations
and glycated hemoglobin levels.
The frequency distributions of these three
parameters of glycemia were analyzed by
a maximum-likelihood estimation using a
model of two overlapping distributions
(7). Antimodal cutpoints that provided
the minimum overlap between the two
components of the distribution were an
FPG concentration of 167 mg/dl (9.3
mmol/1), a 2-h plasma glucose concentration
of 227 mg/dl (12.6 mmol/1), an
HbAlc level of 7.8%, and an HbA: level of
Table 3—Results of a 6-year follow-up in Pima Indians
Initial 2-h blood glucose level
200 mg/dl (11.1 mmol/1)
n
335
111
113
9.4% (7). The ~5 year incidence of microvascular
complications in Pima Indians
without these complications at baseline
above and below the cutpoints is
shown in Table 4.
This recent study using Pima Indians
is the most complete in terms of
evaluating all of the glycemic tests for diagnosing
diabetes and the development
of complications. It was not available at
the time the National Diabetes Data
Group (NDDG) criteria for diagnosing diabetes
were developed (8) (see below),
and it uses somewhat higher glycemic
cutpoints for diagnosing diabetes than
had been used earlier. The conclusions
from all of these studies are based on the
premise that defining glycemic indicators
that predict the development of microvascular
complications (particularly retinopathy)
should be used to make the diagnosis
of diabetes.
In the studies from the 1970s, the
data defining glycemic cutpoints for diagnosing
diabetes are somewhat sketchy. In
two of the studies (4,5), FPG concentrations
were not available for the individuals
with 2-h plasma glucose concentrations
^200 mg/dl (11.1 mmol/1) who
subsequently developed microvascular
complications, making an analysis of FPG
versus 2-h plasma glucose concentrations
impossible. In the third study (6), FPG
concentrations were available for subjects
followed for 3 years (but not in those followed
for 6 years). Of the seven subjects
who developed retinopathy, all had a 2-h
value ^200 mg/dl (11.1 mmol/1) at baseline.
Five of these subjects also had an
FPG concentration ^140 mg/dl (7.8
mmol/1), one had an abnormal FPG con-
Retinopathy
3
3
23
Proteinuria
Data are n at risk at baseline. *Three of these subjects had possible causes for developing proteinuria other
than diabetes. Data from Pettitt et al. (6).
1066 DIABETES CARE, VOLUME 18, NUMBER 7, JULY 1995
4*
0
5

Table A—5-year incidence of retinopathy and nephropathy in Pima Indians
Retinopathy
Nephropathy
2-h plasma glucose
[mg/dl (mmol/1)]
227(12.6)
769 (0%) 158 (17.5%)
768 (1.2%) 173 (3.6%)
Data are n at risk at baseline (5-year incidence).
centration between 130-139 mg/dl (7.2-
7.7 mmol/1), and in the remaining subject
the FPG level was 200(ll.l)
=£200(11.1)
NDDG
Venous whole
blood glucose
< 100 (5.6)
200(ll.l)
—
=£200(11.1)
• Unequivocal elevation of plasma glucose
(>200 mg/dl [11.1 mmol/1]) and
classic symptoms of diabetes, including
polydipsia, polyuria, polyphagia, and
weight loss.
• FPG >140 mg/dl (7.8 mmol/1) on two
occasions.
WHO
Venous whole
blood glucose

Alternative approach to diagnosing diabetes
Table 6—Prevalence of nondiagnostic OGTT by NDDG criteria and diagnosis by
WHO criteria
Reference
12
13
14
15
Total
Data are n'
Total
OGTT
3704
543
2040
216
6503
Nondiagnostic
OGTT
536
179
405
54
1174(18)
• FPG < 140 mg/dl (7.8 mmol/l) and two
OGTTs with the 2-h plasma glucose
^200 mg/dl (11.1 mmol/l) and one intervening
value ^200 mg/dl (11.1
mmol/l) after a 75-g glucose load.
IGT
• FPG < 140 mg/dl (7.8 mmol/l) and 2-h
plasma glucose ^140 (7.8 mmol/l) and

Table 7—Results of OGTTs in individuals with glycated hemoglobin levels >\% above
the ULN
Reference
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Total
Total Diabetes 1GT NGT Undetermined
22
22
23
14
25
18
38
57
43
22
49
12
87
14
12
8
466
21
22
23
13
24
17
37
54
41
22
46
12
84
14
6
8
444
1
—
—
—
1
1
1
1
2
—
3
—
3
—
3
—
16
—
—
—
—
—
2
3
5 1
Data are n. For those undetermined, information not available to determine if individual was normal or had
IGT. Positive predictive value = true positive X 100/true positives + false positives = 95.3%.
physicians revealed that over two-thirds
never considered ordering an OGTT to
diagnose diabetes (22). In fact,

Alternative approach to diagnosing diabetes
stitute a diagnosis but would simply alert
these individuals (and their physicians) to
the potential benefits of altering their lifestyles
to minimize these risks. Regardless
of the results if an OGTT were performed
in subjects with glycated hemoglobin levels
120% of desirable body weight); 3)
Native American, African American, or
Hispanic ethnicity; 4) previously identified
IGT; 5) hypertension or dyslipidemia;
or 6) history of gestational diabetes
or delivery of a baby weighing >9 lb.
Since diabetes is so common in the elderly,
consideration should also be given to
measuring FPG concentrations in those
>65 years of age regardless of risk factors.
References
1. Harris MI, Modan M: Screening for
NIDDM: why is there no national program?
Diabetes Care 17:440-444, 1994
2. Knowler WC: Screening for NIDDM: opportunities
for detection, treatment, and
prevention. Diabetes Care 17:445-450,
1994
3. Valleron AJ, Eschwege E, Papoz L, Rosselin
GE: Agreement and discrepancy in the
evaluation of normal and diabetic oral
glucose tolerance test. Diabetes 24:585-
596,1975
4. Jarrett RJ, Keen H: Hyperglycaemia and
diabetes mellitus. Lancet ii: 1009-1012,
1976
5. Sayegh HA, Jarrett RJ: Oral glucosetolerance
tests and the diagnosis of diabetes:
results of a prospective study based
on the Whitehall survey. Lancet ii:431-
433,1979
6. Pettitt DJ, Knowler WC, Lisse JR, Bennett
PH: Development of retinopathy and proteinuria
in relation to plasma-glucose
concentrations in Pima Indians. Lancet ii:
1050-1052, 1980
7. McCance DR, Hanson RL, Charles MA,
Jacobsson LTH, Pettitt DJ, Bennett PH,
Knowler WC: Comparison of tests for glycated
haemoglobin and fasting and two
hour plasma glucose concentrations as diagnostic
methods for diabetes. Br Med J
308:1323-1328, 1994
8. National Diabetes Data Group: Classification
and diagnosis of diabetes mellitus
and other categories of glucose intolerance.
Diabetes 28:1039-1057, 1979
9. American Diabetes Association Position
Statement: Office guide to diagnosis and
classification of diabetes mellitus and
other categories of glucose intolerance.
Diabetes Care 16 (Suppl. 2):4, 1993
10. World Health Organization: Diabetes Mellitus:
Report of a WHO Study Group. Geneva,
World Health Org., 1985 (Tech.
Rep. Ser.,no. 727)
11. American Diabetes Association Committee
on Statistics: Standardization of the
oral glucose tolerance test. Diabetes 18:
299-310, 1969
12. Harris MI, Hadden WC, Knowler WC,
Bennett PH: International criteria for the
diagnosis of diabetes and impaired glucose
tolerance. Diabetes Care 8:562-567,
1985
13. Massari V, Eschwege E, Valleron AH: Imprecision
of new criteria for the oral glucose
tolerance test. Diabetologia 24:100-
106,1983
14. Modan M, Halkin H, Karasik A, Lusky A:
Effectiveness of glycosylated hemoglobin,
fasting plasma glucose, and a single post
load plasma glucose level in population
screening for glucose intolerance. Am J
Epidemiol 119:431-444, 1984
15. Forrest RD, Jackson CA, Yudkin JS: The
abbreviated glucose tolerance test in
screening for diabetes: The Islington Diabetes
Survey. Diabetic Med 5:557-561,
1988
16. Modan M, Harris MI, Halkin H: Evaluation
of WHO and NDDG criteria for impaired
glucose tolerance: results from two
national samples. Diabetes 38:1630-
1635, 1989
17. Olefsky JM, Reaven GM: Insulin and glucose
responses to identical oral glucose
tolerance tests performed forty-eight
hours apart. Diabetes 23:449-453, 1974
18. Kosaka K, Mizuno Y, Kuzuga T: Reproducibility
of the oral glucose tolerance test
and the rice-meal test in mild diabetes.
Diabetes 15:901-904, 1966
19. Wilkerson HLC, Butler FK, Francis JO'S:
The effect of prior carbohydrate intake on
the oral glucose tolerance test. Diabetes
9:386-391, 1960
20. Aparicio NJ, Puchulu FE, Gagliardino JJ,
Ruiz M, Llorens JM, Ruiz J, Lamas A, De
Miguel R: Circadian variation of the blood
glucose, plasma insulin and human
growth hormone levels in response to an
oral glucose load in normal subjects. Diabetes
23:132-137, 1974
1070 DIABETES CARE, VOLUME 18, NUMBER 7, JULY 1995

21. Rushforth NB, Miller M, Bennett PH:
Fasting and two-hour post-load glucose
levels for the diagnosis of diabetes. Diabetologia
16:373-379, 1979
22. Orchard TJ: From diagnosis and classification
to complications and therapy. Diabetes
Care 17:326-338, 1994
23. Melton LJ, Palumbo PJ, Chu C-P: Incidence
of diabetes mellitus by clinical type.
Diabetes Care 6:75-86, 1983
24. The Diabetes Control and Complications
Trial Research Group: The effect of intensive
treatment of diabetes on the development
and progression of long-term complications
in insulin-dependent diabetes
meilitus. N Engl J Med 329:977-986,
1993
25. Boucher BJ, Welch SG, Beer MS: Glycosylated
haemoglobins in the diagnosis of diabetes
meilitus and impaired glucose tolerance
for the assessment of chronic
hyperglycemia. Diabetologia 21:34-36,
1981
26. Svendsen PA,JorgensenJ, NerupJ: HbAlc
and the diagnosis of diabetes meilitus.
Ada Med Scand 210:313-316,1981
27. Clipson KL, Kansal PC, Poon M-C,
Boshell BR: Hemoglobin Ax in the diagnosis
of chemical diabetes meilitus. Horm
MetabRes 13:129-131, 1982
28. Kesson CM, Youn RE, Talwar D,
WhitelawJW, Robb DA: Glycosylated hemoglobin
in the diagnosis of non-insulindependent
diabetes meilitus. Diabetes
Care 5:395-398, 1982
29. Verillo A, De Teresa A, Golia R, Nunziata
V: The relationship between glycosylated
haemoglobin levels and various degrees
of glucose intolerance. Diabetologia 24:
391-393, 1983
30. Hall PM, John, MS, Cook G, Sheldon J,
Rutherford SM, Gould BJ: Glycosylated
hemoglobins and glycosylated plasma
proteins in the diagnosis of diabetes meilitus
and impaired glucose tolerance. Diabetes
Care 7:147-150, 1984
31. CederholmJ, Ronquist G, Wibell L: Comparison
of glycosylated hemoglobin with
the oral glucose tolerance test: a study in
subjects with normoglycemia, glucose intolerance
and non-insulin-dependent diabetes
meilitus. Diabete & Metab 10:224-
229,1984
32. Albutt EC, Nattrass M, Northam BE: Glucose
tolerance test and glycosylated haemoglobin
measurement for diagnosis of
diabetes mellitus:an assessment of the criteria
of the WHO expert committee on
diabetes meilitus 1980. Ann Clin Biochem
22:67-73, 1985
33. Lester E, Frazer AD, Shepherd CA, Woddroffe
FJ: Glycosylated haemoglobin as an
alternative to the glucose tolerance test for
the diagnosis of diabetes meilitus. Ann
Clin Biochem 22:74-78, 1985
34. Jeppsson J-O, Jerentorp P, Sundkvist G,
Englund H, Nylund V: Measurement of
hemoglobin Alc by a new liquid-chromatographic
assay: methodology, clinical
utility, and relation to glucose tolerance
evaluated. Clin Chem 32:1867-1872,1986
35. John WG, Richardson RW: Glycosylated
haemoglobin levels in patients referred
for oral glucose tolerance tests. Diabetic
Med 3:46-48, 1986
36. Forrest RD, Jackson CA, Yudkin JS: The
glycohaemoglobin assay as a screen or
screening test for diabetes meilitus: The
Islington Diabetes Survey. Diabetic Med
4:254-259, 1987
37. Little RR, England JD, Wiedmeyer H-M,
McKenzie EM, Pettitt DJ, Knowler WC,
Goldstein DE: Relationship of glycosylated
hemoglobin to oral glucose tolerance.
Diabetes 37:60-64, 1988
38. Guillauseau PJ, Charles MA, Paolaggi F,
Timsit J, Chanson P, Peynet J, Godard
V, Eschwege E, Rousselet F, Lubetzki J:
Comparison of HbAx and fructosamine
in diagnosis of glucose-tolerance abnormalities.
Diabetes Care 13:898-900,
1990
39. Motala AA, Omar MAK: The value of glycosylated
haemoglobin as a substitute for
the oral glucose tolerance test in the detection
of impaired glucose tolerance
(IGT). Diab Res Clin Pract 17:199-207,
1992
Davidson, Peters, and Schriger
40. Santiago JV, DavisJE, Fisher F: Hemoglobin
Alc levels in a diabetes detection program.
J Clin Endocrinol Metab 47:578-
580.1978
41. Tsuji I, Nakamoto K, Hasegawa T,
Hisashige A, Inawashiro H, Fukao A, Hismichi
S: Receiver operating characteristic
analysis on fasting plasma glucose,
HbAlc, and fructosamine on diabetes
screening. Diabetes Care 14:1075-1077,
1991
42. Larsen ML: The utility of glycated hemoglobin
in identification of impaired glucose
tolerance. Diabetes Res 12:67-70,
1989
43. Bolli G, Compagnucci P, Cartechini MG,
Santeusanio F, Cirotto C, Scionti L, Brunetti
P: HbAL in subjects with abnormal
glucose tolerance but normal fasting
plasma glucose. Diabetes 29:272-277,
1980
44. Flock EV, Bennett PH, Savage PJ, Webner
CJ, Howard BV, Rushforth NB, Miller M:
Bimodality of glycosylated hemoglobin
distribution in Pima Indians. Diabetes 28:
984-989,1979
45. Dix D, Cohen P, Kingsley S, Senkbeil J,
Sexton K: Glycohemoglobin and glucose
tolerance tests compared as indicators of
borderline diabetes. Clin Chem 25:877-
879.1979
46. Lev-Ran A, VanderLaan WP: Glycohemoglobins
and glucose tolerance. JAMA 241:
912-914, 1979
47. Orchard TJ, Daneman D, Becker DJ,
Kuller LH, LaPorte RE, Drash AL, Wagener
D: Glycosylated hemoglobin: a
screening test for diabetes meilitus? Preventive
Med 11:595-601, 1982
48. Shima K, Abe F, Chikakiyo H, Ito N: The
relative value of glycated albumin, hemoglobin
Alc and fructosamine when
screening for diabetes meilitus. Diab Res
Clin Pract 7:243-250, 1989
49. Charles MA, Fontbonne A, Thibult N,
Warnet J-M, Rosselin GE, Eschwege E:
Risk factors for NIDDM in white population.
Diabetes 40:796-799, 1991
DIABETES CARE, VOLUME 18, NUMBER 7, JULY 1995 1071