Transient impaired glucose tolerance in South African Indians does not carry a risk for progression to NIDDM.

OBJECTIVE: To evaluate the significance of transient impaired glucose tolerance (IGT) in terms of the risk of progression to NIDDM and the serum insulin response during oral glucose tolerance test (OGTT) in a prospective study on the natural history of IGT in South African Indians. RESEARCH DESIGN AND METHODS: This is a report on 87 subjects who formed part of a 4-year prospective study in 128 subjects classified with IGT at baseline (year 0) using World Health Organization criteria for glucose tolerance. Subjects were reexamined at years 1 and 4. At year 1, based on OGTT results, the subjects were divided into three groups: transient IGT (normal glucose tolerance [trIGT], n = 40), persistent IGT (pIGT, n = 47), and diabetes (n = 41). Analysis was performed on the 87 subjects who were classified as IGT at year 0, but who had not progressed to NIDDM by year 1 of the study At baseline (year 0), a modified OGTT was performed; between years 1 and 4, the OGTT included timed midtest samples for plasma glucose and serum insulin. Analysis of predictive factors for progression to diabetes or reversion to normal glucose tolerance was undertaken using year 0 as baseline.
RESULTS: By year 4, 72 subjects (82.8%) completed the study Of the 32 subjects in the trIGT group, none (0%) had progressed to NIDDM, 11 (34.4%) had reverted to IGT (N-IGT), and 21 (65.6%) had persisted with normal glucose tolerance (N-N); of the 40 subjects in the pIGT group, 16 (40%) had progressed to NIDDM (IGT-D), 17 (42.5%) had persisted with IGT (IGT-IGT), and 7 (17.5%) had reverted to normal glucose tolerance (IGT-N). Significant predictive factors for reversion to normal glucose tolerance included absence of obesity (P = 0.0131, odds ratio [OR] 4.2, 95% CI 1.4-13.1) and 2-h plasma glucose level (P = 0.027, OR 2.4, 95% CI 1.11-5.13) at baseline (year 0). Intergroup (cross-sectional) analysis showed that the serum insulin response was higher in the pIGT than in the trIGT subgroup (fasting serum insulin: IGT-N vs. N-IGT and N-N, 16.9 +/- 1.9 vs. 6.8 +/- 2.1 and 6.1 +/- 2.4 microU/ml, respectively, P < 0.001; 2-h postload serum insulin: IGT-IGT vs. N-IGT, 116.8 +/- 2.2 vs. 60.3 +/- 1.7 microU/ml, P < 0.05). By contrast, the insulinogenic index was higher in the trIGT subgroups both at year 1 (90-min: N-N vs. IGT-D, 48.9 +/- 3.9 vs. 14.1 +/- 2.5; P < 0.05) and at year 4 (N-N vs. remaining four subgroups, P < 0.01 at 60 min and 90 min). Intragroup (prospective) comparisons showed that in the N-IGT subgroup, the mean 2-h insulinogenic index was lower at year 4 than at year 1 (19.9 +/- 1.7 vs. 66.0 +/- 2.7; P < 0.05).
CONCLUSIONS: In this 4-year prospective study in South African Indians, transient IGT carries no risk of progression to NIDDM. The significant predictive factors for reversion to normal glucose tolerance include lower baseline obesity prevalence and 2-h postload plasma glucose level. Moreover, in this group, beta-cell secretory function appeared to deteriorate with worsening of glucose tolerance.