0021-972X/90/7106-1447$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1990 by The Endocrine Society

Vol. 71, No. 6 Printed in U.S.A.

Association of Fasting Glucose Levels with a Delayed Secretion of Insulin after Oral Glucose in Subjects with Glucose Intolerance* RICHARD W. BERGSTROMf, PATRICIA W. WAHL, DONNA L. LEONETTI, AND WILFRED Y. FUJIMOTO Departments of Medicine, Biostatistics, and Anthropology, University of Washington, Seattle, Washington 98195

ABSTRACT. Two hundred and nineteen second generation Japanese-American men were classified with a 75-g oral glucose tolerance test: 77 with normal glucose tolerance, 74 with impaired glucose tolerance (IGT), and 68 with noninsulin-dependent diabetes mellitus (NIDDM). The peak insulin response to the oral glucose load was progressively delayed with each of the 3 glucose tolerance categories. A similar finding was observed with the peak C-peptide response to oral glucose, except for the absence of distinction between IGT and NIDDM. Variables measuring the initial rate of insulin or C-peptide secretion (030 min) after oral glucose also demonstrated a progressive diminution with increasing glucose intolerance. The relative incremental insulin response at 30 min and the relative incremental C-peptide response at 30 min were highly correlated with the fasting glucose levels (r = -0.61 and r = -0.62; P < 0.0001,

T

respectively). Variables measuring the 0-30 min secretory response had high variances, whereas the variance for fasting glucose was low. Twelve men who were initially classified as IGT subsequently developed NIDDM. These 12 men had significantly higher fasting glucose levels at baseline than the remaining men who did not develop diabetes, but the 30 min secretory parameters after oral glucose, although lower in those who subsequently developed diabetes, were not significantly different at baseline. However, if fasting glucose is used as a surrogate measure of secretory response, these 12 men appear to have had an impairment of oral glucose-stimulated insulin secretion antedating the development of NIDDM. The inability of the secretory parameters to detect the abnormality may be due to a type II statistical error, which may be resolved by a larger sample size. (J Clin Endocrinol Metab 7 1 : 1447-1453, 1990)

HE PEAK insulin response to an oral glucose load is delayed with noninsulin-dependent diabetes mellitus (NIDDM) and impaired glucose tolerance (IGT). This phenomenon has been described by many investigators (1-7) and has been termed the biochemical inertia of /3-cell response (2). The pattern is so characteristic of NIDDM that Kraft (6) has compiled and studied over 14,000 oral glucose tolerance tests (OGTT) with measurements of both glucose and insulin levels for 5 h and from these studies has proposed an alternative classification scheme for glucose intolerance and NIDDM based on the time of the peak insulin response. Seltzer and associates (2) have used the insulinogenic index at 30 min to describe the phenomenon. They have shown that

this index is inversely related to the severity of glucose intolerance (2). While this index and other measures of the initial rate of insulin secretion after oral glucose are likely to be related to postprandial glucose levels, the relationship of these measures to fasting glucose levels has not been defined. We examined the relationship between fasting glucose levels and several measures of the initial rate of insulin secretion after an oral glucose load in a study sample of Japanese-American men composed of an approximately equal distribution of men with normal glucose tolerance (NGT), IGT, and NIDDM. This study sample is derived from a population that is known to have a high prevalence of glucose intolerance.

Received December 19,1989. Address requests for reprints to: Dr. Wilfred Y. Fujimoto, Division of Metabolism, Endocrinology, and Nutrition, Department of Medicine RG-26, University of Washington, Seattle, Washington 98195. * This work was supported by USPHS Grant DK-31170 and the facilities and services of the Diabetes Endocrinology Research Center (DK-17047), the University of Washington Clinical Research Center (RR-37), and the Clinical Nutrition Research Unit (DK-35816). t Supported by a Juvenile Diabetes Foundation International Postdoctoral Fellowship, the Nordstrom Research Fund in Metabolism and Food, and Diabetes and Endocrine Training Grant DK-07247.

Materials and Methods The Japanese-American Community Diabetes Study is an epidemiological study designed to evaluate the factors associated with NIDDM and its attendant complications (8, 9). Among 229 Japanese-American Nisei men (American-born and -reared sons of immigrants from Japan) originally enrolled into the study, 77 had NGT, 74 had IGT, and 78 had NIDDM. Details pertaining to the recruitment, enrollment, and charac-

1447

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 11:21 For personal use only. No other uses without permission. . All rights reserved.

BERGSTROM ET AL.

1448

JCE & M • 1990 Vol71»No6

terization of the study group have been reported previously (8, NIDDM groups were examined by analysis of variance, and 9). Each of the subjects was studied at the Clinical Research pairwise contrasts were evaluated by the Student-NewmanCenter, University of Washington Medical Center. Glucose Keuls range test. Comparisons between the two subgroups of tolerance status was classified as NGT, IGT, and NIDDM men initially classified as IGT were evaluated by the Wilcoxon based upon medical information obtained from the subjects rank sum test. Spearman's correlation coefficient was used to and their physicians and the results of a 75-g OGTT. Those 10 determine linear associations between variables and was tested subjects with NIDDM who were receiving insulin therapy were for significance using the t test for correlations. All values are excluded from the present analysis, resulting in a study sample expressed as the mean ± SE. of 219 men. After a mean follow-up period of 30 months (range, 28-34 months), 146 of the nondiabetic men (NGT, 75; IGT, Results 71) participated in a follow-up examination, which included an OGTT. Fifteen (NGT, 3; IGT, 12) of the 146 Nisei men Increasing glucose intolerance was associated with a subsequently developed NIDDM (7). The protocol for this progressive delay in the peak insulin response after a 75research was reviewed and approved by the Human Subjects g oral glucose load (Fig. 1). Fifty-four percent of the men Review Committee at the University of Washington. Written with NGT had a peak insulin response that occurred at informed consent was obtained from each subject. 60 min, 39% of the men with IGT had a peak insulin response that occurred at 90 min, and 49% of the men Variables with NIDDM had a peak insulin response that occurred at 120 min. The peak C-peptide response to the 75-g oral Blood samples were drawn after a minimum 10-h fast and glucose load was also delayed with increasing glucose 30, 60, 90, 120, and 180 min after administration of 75 g oral glucose to measure serum glucose and plasma insulin and Cintolerance; however, there was no significant separation peptide levels. The incremental response at 30 min for these between IGT and NIDDM in the peak C-peptide rethree variables was calculated by subtracting the fasting value. sponse (Fig. 2). The relative incremental insulin or C-peptide response at 30 The insulinogenic index at 30 min, the incremental min was the incremental response for each of these two peptides insulin response at 30 min (numerator of the insulinodivided by the corresponding fasting peptide level. The insuligenic index at 30 min), and the relative incremental nogenic index at 30 min or the incremental (insulin/glucose) insulin response at 30 min were all inversely related to at 30 min is the quotient of the incremental insulin response the severity of glucose intolerance (Table 1). All interat 30 min divided by the incremental glucose response at 30 group differences (NGT vs. IGT, IGT vs. NIDDM, and min. Similarly, the incremental (C-peptide/glucose) at 30 min NGT vs. NIDDM) were significant (P < 0.05) by Stuwas the quotient of the incremental C-peptide response at 30 min divided by the incremental glucose response at 30 min. dent-Newman-Keuls test. Similarly, the corresponding The serum glucose was measured by an automated glucose C-peptide measures were all inversely related to the oxidase method (Department of Laboratory Medicine, Universeverity of glucose intolerance, and all intergroup differsity of Washington Medical Center). Both plasma insulin and ences were significant, except for the comparison of NGT C-peptide were measured by RIAs, performed at the Radioand IGT for incremental C-peptide response (Table 1). immunoassay Core of the Diabetes Endocrinology Research The relationships between these insulin secretory reCenter. The intra- and interassay coefficients of variation were, respectively, 5% and 8% for insulin and 1% and 11% for Cpeptide. 50

Statistical analysis The peak insulin or C-peptide response was the time point in the OGTT when the peak insulin or C-peptide response occurred for each individual. In those cases in which there was no single peak response (occurring at more than one time point), the peaks were determined by dividing equally among the time points for a total score of 1; e.g. if the highest values were at 60, 90, and 120 min and equal to one another, each time point was given a third of the score. The frequency distribution of the peak insulin or C-peptide for the three glucose tolerance categories was examined using x2 analysis. The variables with a skewed distribution were transformed by a square root transformation: incremental insulin response at 30 min, incremental (insulin/glucose) at 30 min, and relative incremental insulin and C-peptide responses at 30 min. All comparisons of measured variables among NGT, IGT, and

40 30 n

20 10 0 0

30

III

60

90

• 120 180

Time post oral glucose load (min)

FlG. 1. Frequency distribution by percentage of peak insulin response after an oral glucose load. The study sample was subdivided by glucose tolerance status: D, NGT; W, IGT; • NIDDM. x 2 (NGT-IGT-NIDDM) = 60.3, P < 0.0001; x 2 (NGT-IGT) = 22.6, P < 0.0001; x 2 (NGTNIDDM) = 57.2, P < 0.0001; x2 (IGT-NIDDM) = 15.7, P < 0.001.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 11:21 For personal use only. No other uses without permission. . All rights reserved.

DELAYED INSULIN RESPONSE AND FASTING GLUCOSE 50403020-

1

10-

0

30

L

60

1 1 1 90

1 1

1 _.E11 120 180

Time post oral glucose load (min)

FlG. 2. Frequency distribution by percentage of peak C-peptide response after an oral glucose load. The study sample was subdivided by glucose tolerance status: D, NGT; M, IGT; • . NIDDM. x 2 (NGT-IGTNIDDM) = 74.7, P < 0.0001; X2 (NGT-IGT) = 39.5, P < 0.0001; x 2 (NGT-NIDDM) = 54.0, P < 0.0001; x 2 (IGT-NIDDM) = 6.9, P = 0.08.

sponses at 30 min and the glucose levels at fasting and the five other time points of the OGTT were examined. Since the incremental (insulin/glucose) at 30 min (insulinogenic index) and the corresponding C-peptide measures are calculated using the fasting and 30 min glucose values, correlations of glucose values with these two indices were not performed. The incremental glucose responses at 30 min were not significantly different in the men with NGT and the men with IGT, but the difference was significant for both IGT vs. NIDDM and NGT vs. NIDDM (NGT, 3.95 ± 0.17 mM; IGT, 4.33 ± 0.17 mM; NIDDM, 5.36 ± 0.17 mM). Thus, the insulinogenic index and the incremental insulin response at 30 min are similar for IGT and NGT. However, for those with NIDDM, the difference in the incremental insulin response at 30 min is greater than that in the insulinogenic index. The relationships of the four secretory indices as well as the fasting insulin and C-peptide values to the six glucose values were determined for all of the men combined as well as separately for the three glucose

1449

tolerance categories (Table 2). The relationships of both the relative incremental insulin and the relative incremental C-peptide response at 30 min with the six glucose values were stronger than the corresponding relationships of either the fasting insulin and C-peptide values or the absolute incremental values with the six glucose values. When the three glucose tolerance categories were examined individually, the relationships between the secretory indices and glucose values was especially strong in the diabetic men. The relationship between one of these secretory indices, the relative incremental C-peptide response at 30 min, and fasting glucose is shown for the entire study population in Fig. 3. Two variables that were associated with the delay in the peak insulin response (the relative incremental insulin and C-peptide responses at 30 min) as well as fasting glucose were compared between the 12 men who were initially classified as IGT but developed diabetes 30 months later and the 59 men who were initially classified as IGT and remained nondiabetic. The fasting glucose value was significantly elevated in the men progressing to diabetes at both the baseline and follow-up period, while all of the insulin secretory variables were only significant at the follow-up (Table 3). Discussion The delay in the peak insulin secretory response to oral glucose as observed in this study is a common observation with glucose intolerance (1-3, 6, 10). The association of a delayed peak insulin response with glucose intolerance, occurring at 60 min for NGT, 90 min for IGT, and 120 min for NIDDM, was highly significant (x2 = 60.3; P < 0.0001). The pattern of a delayed peak C-peptide response to oral glucose was distinctive for NGT compared to either IGT or NIDDM, but was not for IGT compared to NIDDM. This difference may be related in part to the differences in clearance between C-

TABLE 1. The initial insulin secretory responses in the three glucose tolerance groups (mean ± SE)

Incremental (insulin/glucose) at 30 min

NGT (n = 77)

IGT (n = 74)

NIDDM (n = 67)

P value0

135 ± 12

96 ± 8

37 ± 4