HbA1c measured in stored erythrocytes and mortality rate among middle-aged and older women

Abstract

Aims/hypothesis: Diabetes is known to increase mortality rate, but the degree to which mild hyperglycaemia may be associated with the risk of death is uncertain. We examined the association between HbA1c measured in stored erythrocytes and mortality rate in women with and without diabetes.

Methods: We conducted a cohort study of 27,210 women>or=45 years old with no history of cardiovascular disease or cancer who participated in the Women's Health Study, a randomised trial of vitamin E and aspirin.

Results: Over a median of 10 years of follow-up, 706 women died. Proportional hazards models adjusted for age, smoking, hypertension, blood lipids, exercise, postmenopausal hormone use, multivitamin use and C-reactive protein were used to estimate the relative risk of mortality. Among women without a diagnosis of diabetes and HbA1c<5.60%, those in the top quintile (HbA1c 5.19-5.59%) had a relative risk of mortality of 1.28 (95% CI 0.98-1.69, p value for linear trend=0.14) compared with those with HbA1c 2.27-4.79%. Women with HbA1c 5.60-5.99% and no diagnosis of diabetes had a 54% increased risk of mortality (95% CI 1-136%) compared with those with HbA1c 2.27-4.79%. HbA1c was significantly associated with mortality across the range 4.50-7.00% (p value for linear trend=0.02); a test of deviation from linearity was not statistically significant (p=0.67). Diabetic women had more than twice the mortality risk of non-diabetic women.

Conclusions/interpretation: This study provides further evidence that chronic mild hyperglycaemia, even in the absence of diagnosed diabetes, is associated with increased risk of mortality. ClinicalTrials.gov ID no.: NCT00000479.

Figures

Fig. 1

Association of HbA1c measured in stored erythrocytes and risk of mortality among women with no self-reported history of diabetes. The solid line represents the RR of mortality associated with HbA1c calculated from Cox proportional hazards models. Penalised cubic splines were used to flexibly model the shape of the association. Dashed lines represent the 95% confidence interval. The models were adjusted for age at baseline in 5 year categories, strenuous exercise (rarely/never, <1 time/week, 1–3 times/week, ≥4 times/week), post-menopausal hormone use (never, past, current), multivitamin use (never, past, current), smoking status (never, past, current), BMI (<21, 21–22.9, 23–24.9, 25–26.9, 27–28.9, 29–30.9, ≥31 kg/m2), selfreported history of hypertension (yes or no), and quintile of triacylglycerol, HDL-cholesterol, LDL-cholesterol and hsCRP