Prevalence of coronary risk factors in non-insulin dependent (type 2) diabetics

A cross sectional study was conducted to find the prevalence of coronary risk factors in non-insulin dependent diabetic (NIDDM) patients and to compare and co-relate these risk factors in type II diabetics with and without electrocardiographic and/or symptomatic evidence of coronary heart disease (CHD).

One hundred sixty-seven consecutive NIDDM patients (77 males, and 90 females) attending the diabetic clinic at Dr. RML Hospital, New Delhi were studied. Only known NIDDM cases, already on treatment and without any history of ketosis or congestive heart failure were included. Coronary risk factors comprising of age, gender, duration and treatment for diabetes, smoking, physical activity, hypertension, truncal obesity, lipids, microalbuminuria (semiquantitative) and glycemic control have been particularly ascertained in all the cases. The data was analysed using 'Epi Info version 6.0'.

The mean age of patients was 53.12 year and 8.86 year was the mean duration of diabetes. 28.6% of the diabetic men were found to be currently smoking and/or consuming alcohol, 82% were involved in sedentary physical activity and 20.4% had family history of CHD. Central obesity was observed in 46.7% of the cases; more so in females. 31.74% of cases were hypertensive; more females than males had hypertension (33.8% vs 30%). Poor glycemic control (HbA1c > = 9.5%) was seen in 16.8% of the cases. In about 52.5% of the total group hypertriglyceridemia was noted. Microalbuminuria could be found in 35.93%. CHD was diagnosed in 15.57% of cases in this study.

The present study revealed that high levels of serum cholesterol (p = 0.000004), LDL (p = 0.00003), HbA1c (p = 0.002), microalbuminuria (p = 0.000006) and hypertension (p = 0.00006) are significant associates of CHD in NIDDM (both the sexes). Among the female NIDDM cases, in addition BMI (p = 0.01), Waist-hip ratio (WHR) (p = 0.003) and low HDL level (p = 0.008) are important correlates of CHD. Multiple logistic regression analysis was used to allow for confounding between variables. Microalbuminuria alone entered the 'best' model for CHD prediction. Other risk factors, though significant, provided inadequate models for CHD prediction.