Oral Disposition Index Predicts the Development of Future Diabetes Above and Beyond Fasting and 2-h Glucose Levels

Cardiovascular and Metabolic Risk

¹Division of Metabolism, Endocrinology and Nutrition, Department of Medicine, Veterans Affairs Puget Sound Health Care System, Seattle, Washington
²Department of Medicine, University of Washington, Seattle, Washington
³Geriatric Research Education and Clinical Center, Baltimore Veterans Affairs Medical Center, Baltimore, Maryland
⁴Division of Gerontology, University of Maryland School of Medicine, Baltimore, Maryland
⁵Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, University of Washington, Seattle, Washington
⁶Epidemiologic Research and Information Center, Veterans Affairs Puget Sound Health Care System, Seattle, Washington
⁷Department of Anthropology, University of Washington, Seattle, Washington

Diabetes Care 2009 Feb; 32(2): 335-341. https://doi.org/10.2337/dc08-1478

Article Figures & Tables

Download figure
Open in new tab
Download powerpoint
Figure 1—
The computed slopes (——) and 95% CIs for the slopes (– – –) for ln(ΔI_0–30/ΔG_0–30) versus ln(1/fasting insulin) are plotted for subjects with NGT (slope −0.87 [95% CI −1.13 to −0.61]) (A), IGM (−0.84 [−1.05 to −0.63]) (B), and diabetes (DM) (−0.76 [−1.16 to −0.35]) (C). D: The hyperbolic curves for NGT, IGM, and diabetes, assuming a slope of −1, are plotted for ΔI_0–30/ΔG_0–30 versus 1/fasting insulin.
Download figure
Open in new tab
Download powerpoint
Figure 2—
A: The DI_O (ΔI_0–30/ΔG_0–30 × 1/fasting insulin) decreases from NGT to IGM to diabetes (DM) (median [interquartile range]). B: The logarithmic means for baseline ΔI_0–30/ΔG_0–30 and 1/fasting insulin values are plotted for subjects with NGT (▪ and □) and IGM (• and ○) as nonprogressors (▪ and •) and progressors (□ and ○) relative to the hyperbolic curves. Progressors had lower β-cell function at baseline. C: The number of subjects who developed diabetes over the 10-year follow-up period by quintiles of baseline DI_O. D: ROC curves comparing ability of baseline ΔI_0–30/ΔG_0–30 (dotted line), 1/fasting insulin (dashed line), and DI_O (solid line) to predict cumulative diabetes at 10 years.

Table 1—

Subject characteristics by glucose tolerance category

	NGT	iIFG	iIGT	IFG + IGT	IGM	Diabetes
n	244	60	118	76	254	115
Age (years) (mean ± SD)	48.6 ± 11.9	56.4 ± 10.7	54.0 ± 11.5	59.1 ± 9.5§	56.1 ± 10.9*	61.1 ± 7.4*
Sex (female/male)	120/124	20/40	79/39†	23/53§	122/132	49/66
BMI (kg/m²)	23.5 ± 0.2	24.7 ± 0.3	23.7 ± 0.3	25.8 ± 0.4§	24.6 ± 0.2*	25.7 ± 0.3*
Abdominal circumference (cm)	84.0 ± 0.51	87.7 ± 0.8	85.5 ± 0.9	90.0 ± 0.9§	87.6 ± 0.5*	91.0 ± 0.8*
FPG (mmol/l)	4.89 (0.58)	5.72 (0.39)	5.06 (0.50)†	5.89 (0.39)§	5.56 (0.72)*	7.56 (3.33)*
2-h plasma glucose (mmol/l)	6.22 (1.44)	6.94 (1.08)	8.69 (1.11)†	9.19 (1.78)‡§	8.47 (1.56)*	15.17 (8.56)*
Fasting insulin (pmol/l)	66 (36)	81 (48)	69 (48)	78 (51)	78 (54)*	102 (96)*
HOMA-S	81.9 (45.2)	63.9 (36.4)	79.2 (50.2)	65.8 (38.8)	70.7 (46.9)*	47.0 (44.7)*
ΔI_0–30/ΔG_0–30 (pmol/mmol)	105.7 (102.5)	100.2 (101.9)	81.7 (92.6)	77.9 (65.0)	80.9 (86.5)*	29.9 (41.4)*
incAUC_ins/glu (pmol/mmol)	148.6 (138.6)	138.5 (146.5)	104.3 (83.8)†	105.6 (77.8)‡	113.3 (90.8)*	33.8 (49.0)*

Data are reported as mean ± SEM or median (interquartile range) unless otherwise specified.
*
↵* P < 0.001 NGT versus IGM versus diabetes;
†
↵† iIFG versus iIGT,
‡
↵‡ iIFG versus IFG + IGT,
§
↵§ iIGT versus IFG + IGT, P < 0.05.