REDUCED SAMPLING PROTOCOLS IN ESTIMATION OF INSULIN SENSITIVITY AND GLUCOSE EFFECTIVENESS USING THE MINIMAL MODEL IN NIDDM

Recent work in healthy subjects, the aged, and subjects with gestation al diabetes or drug-induced insulin resistance using minimal model ana lysis of the tolbutamide-modified frequently sampled intravenous gluco se tolerance test suggested that a reduced sampling regimen of 12 time points produced unbiased and generally acceptable estimates of insuli n sensitivity (S(I)) and glucose effectiveness (S(G)) compared with a full sampling schedule of 30 time points. We have used data from 26 in sulin-modified frequently sampled intravenous glucose tolerance tests in 21 subjects with NIDDM to derive and compare estimates of S(I) and S(G) from the full sampling schedule (S(I(30)), S(G(30)) with those es timated from the suggested 12 time points (S(I(12)), S(G(12)) and thos e estimated with the addition of a 25-min time point (S(I(13)), S(G(13 ))). Percentage relative errors were calculated relative to the corres ponding 30 time-point values. A statistically significant bias of 15% (97% confidence interval from 7.4 to 25.6%, interquartile range 25%) w as introduced by the estimation of S(I(12)) but not S(I(13)) (1%, 97% confidence interval from -9.4 to 9.3%, interquartile range 21%). Resul ts for S(G(12)) (-12%, 97% confidence interval from -46.7 to 1.2%, int erquartile range 49%) and S(G(13)) (-5%, 97% confidence interval from -27.8 to 6.8%, interquartile range 37%) were statistically equivocal. The precision of estimation of S(I(12)), S(G(12)), and S(G(13)) measur ed by the interquartile range of the percentage relative errors was po or. The precision of determination measured by the median minimal mode l coefficient of variation was 18, 29, and 27% for S(I(30)), S(I(12)), and S(I(13)) and 9, 11, and 11% for S(G(30)), S(G(12)), and S(G(13)), respectively. Thus, the application of minimal model analysis to the 12 time-point protocol of the insulin-modified IVGTT for the estimatio n of S(I) and S(G) in NIDDM may necessitate an inordinately large numb er of subjects. Although the 13 time-point protocol may be more accept able for the assessment of S(I) in population studies, we recommend re tention of the full sampling schedule where feasible.