The Hemoglobin Glycation Index Is Not an Independent Predictor of the Risk of Microvascular Complications in the Diabetes Control and Complications Trial
Article Figures & Tables
Figures
- FIG. 1.
Scatter plot of the A1C versus HGI for all visits of all DCCT subjects, with the estimated regression line; intercept 8.2, slope 1.08 A1C % increase per unit increase in the HGI, and correlation of 0.77.
- FIG. 2.
Scatter plots of the observed A1C versus the observed within-profile MBG at all visits during the DCCT. A: For all participants. B–D: For those participants in the high, moderate, and low HGI groups, respectively. The regression line shown in each panel was derived from the simple regression of A1C on the MBG from the population. The horizontal line is the mean A1C in each panel.
- FIG. 3.
Cumulative incidence of retinopathy progression within the high, moderate, and low HGI groups obtained from a Cox proportional hazards regression model adjusted for the effects of age, diabetes duration, sex, treatment group, cohort, and MBG as a time-dependent covariate. The adjusted relative risk (RR) for moderate versus low HGI and for high versus low HGI with P value and the overall 2-df Wald test, χ2, and P values are shown. A: Without adjustment for A1C. B: With adjustment for the A1C at eligibility and the time-dependent current updated mean A1C.
- FIG. 4.
Cumulative incidence of nephropathy progression (advanced microalbuminuria) within the high, moderate, and low HGI groups, obtained from a Cox proportional hazards regression model adjusted for the effects of age, diabetes duration, sex, treatment group, cohort, and MBG as a time-dependent covariate. The adjusted RR for moderate versus low HGI and for high versus low HGI with P value and the overall 2-df Wald test, χ2, and P values are shown. A: Without adjustment for A1C. B: With adjustment for the A1C at eligibility and the time-dependent current updated mean A1C.
- FIG. 5.
Cumulative incidence of retinopathy progression within the high, moderate, and low HGI groups for those subjects within the low MBG group, obtained from a Cox proportional hazards regression model adjusted for the effects of age, diabetes duration, sex, treatment group, cohort, and MBG as a time-dependent covariate. The adjusted RR for moderate versus low HGI and for high versus low HGI with P value and the overall 2-df Wald test, χ2, and P values are shown. A: Without adjustment for A1C. B: With adjustment for the A1C at eligibility and the time-dependent current updated mean A1C.
- FIG. 6.
Cumulative incidence of retinopathy progression within the high, moderate, and low HGI groups for those subjects within the high MBG group, obtained from a Cox proportional hazards regression model adjusted for the effects of age, diabetes duration, sex, treatment group, cohort, and MBG as a time-dependent covariate. The adjusted RR for moderate versus low HGI and for high versus low HGI with P value and the overall 2-df Wald test, χ2, and P values are shown. A: Without adjustment for A1C. B: With adjustment for the A1C at eligibility and the time-dependent current updated mean A1C.
Tables
- TABLE 1
DCCT baseline characteristics by treatment group
Conventional Intensive n 729 710 Cohort Secondary intervention 377 ± 51.7 348 ± 49.0 Primary prevention 352 ± 48.3 362 ± 51.0 Age (years) 26.5 ± 7.1 27.1 ± 7.1 Race (Caucasian) 703 ± 96.4 686 ± 96.6 Sex (male) 394 ± 54.0 365 ± 51.4 Duration of type 1 diabetes (years) 5.7 ± 4.1 6.0 ± 4.2 A1C at eligibility (%) 9.05 ± 1.63 9.08 ± 1.59 MBG at DCCT baseline (mmol/l) 12.8 ± 4.4 13.0 ± 4.6 Data are n (%) and means ± SD. P > 0.10 for all comparisons.
- TABLE 2
MBG, HGI, and A1C among all visits for subjects classified by low, moderate, and high levels of the HGI
HGI group n (subjects) n (Obs) MBG HGI A1C Low (<−0.425) 467 11,338 10.00 (3.75) −0.91 (0.78) 7.25 (1.17) Moderate (−0.425 to 0.249) 470 11,351 10.17 (3.99) −0.09 (0.75) 7.85 (1.29) High (>0.25) 502 11,350 12.20 (4.68) 0.92 (1.12) 9.34 (1.66) Data are means (SD). Obs, quarterly visit values.
In this Issue
