Association of Anthropometric Indices With the Development of Diabetes Among Hypertensive Patients in China: A Cohort Study

BACKGROUND: Patients with comorbidity of hypertension and diabetes are associated with higher morbidity and mortality of cardiovascular disease than those with hypertension or diabetes alone. The present study aimed to identify anthropometric risk factors for diabetes among hypertensive patients who were included in a retrospective cohort study. METHODS: Hypertensive adults without diabetes were recruited in China. Demographic, clinical, biochemical, and anthropometric indices were collected at baseline and during the follow-up. Anthropometric measures included BMI, waist circumference, waist-to-height ratio (WHtR), and waist-to-hip ratio, and several novel indices. To estimate the effect of baseline and dynamic changes of each anthropometric index on risk of new-onset diabetes (defined as self-reported physician-diagnosed diabetes and/or use of hypoglycemic medication, or new-onset FPG≥7.0 mmol/L during follow-up), Cox regression models were used. RESULTS: A total of 3852 hypertensive patients were studied, of whom 1167 developed diabetes during follow-up. Multivariate Cox regression analyses showed that there was a graded increased risk of incident diabetes with successively increasing anthropometric indices mentioned above (all P<0.05). Regardless of the baseline general obesity status, elevated WHtR was both related to higher risk of diabetes; the HRs (95%CI) of baseline BMI<24 kg/m(2) & WHtR≥0.5 group and BMI≥24 kg/m(2) & WHtR≥0.5 group were 1.34 (1.05, 1.72), 1.85 (1.48, 2.31), respectively. Moreover, the dynamic changes of WHtR could sensitively reflect diabetes risk. Diabetes risk significantly increased when patients with baseline WHtR<0.5 progressed to WHtR≥0.5 during the follow-up (HR=1.63; 95%CI, 1.11, 2.40). There was also a decreasing trend towards the risk of incident diabetes when baseline abnormal WHtR reversed to normal at follow-up (HR=1.93; 95%CI, 1.36, 2.72) compared with those whose WHtR remained abnormal at follow-up (HR=2.04; 95%CI, 1.54, 2.71). CONCLUSIONS: Central obesity is an independent and modifiable risk factor for the development of diabetes among hypertensive patients. Measuring indices of central obesity in addition to BMI in clinics could provide incremental benefits in the discrimination of diabetes among Chinese hypertensive patients. Dynamic changes of WHtR could sensitively reflect changes in the risk of diabetes. Therefore, long-term monitoring of hypertensive patients using non-invasive anthropometric measures and timely lifestyle intervention could effectively reduce the development of diabetes.