Association of Type 2 Diabetes Mellitus with Biometric Variables: A study in Sir Ganga Ram Hospital, Lahore

Abstract

Introduction:  Diabetes mellitus is closely linked to obesity as measured by body mass index (BMI) and is especially related to abnormal fat distribution as indicated by an increased waist/hip ratio and waist/thigh ratio. Central obesity is known to be associated with diabetes mellitus type 2; however limb circumferences may also have a strong association.

Objective:  To compare various biometric variables in type 2 diabetics and non-diabetics.

Design:  Cross-sectional analytical study.

Place and Time:  Once a week, Outpatient Clinic, Department of Internal Medicine, Sir Ganga Ram Hospital, Lahore, from June 2007 to June 2008.

Statistical Tests:  SPSS 12 (Statistical Package for Social Sciences) used. Includes student’s t-test and logistic regression analysis.

Methodology:  One hundred and sixty type 2 diabetics out of which eighty eight are women are compared with one hundred and fifty non-diabetics matched for age, sex and socioeconomic status. Diabetics in this study are known diabetics and those with 2 hour post-prandial blood glucose levels in excess of 140mg/dl. Anthropometric measurements, including waist, hip, thigh, calf and upper arm circumferences, BMI, waist/hip and waist/thigh ratios, of all subjects were taken and the student’s t-test was applied to find out if they were significant differences between the measurements in diabetics and non-diabetics. Logistic regression analysis was used to show association of these biometric variables/anthropometric measurements with blood glucose levels.

Results:  The mean blood glucose level in the diabetics group is 251mg/dl, with mean duration of disease more than five years. Type 2 diabetes mellitus is associated with thinner thighs and calves in both men and women (p value < 0.05). On regression analyses a negative association of thigh and calf circumference is demonstrated with high blood glucose. Thigh and calf measurements show an inverse linear correlation (r = - 0.45) and (r = - 0.68) respectively. The coefficient of determi-nation (r² = 0.2 and r² = 0.5) suggests that thigh and calf circumferences have a 20% and 50% dependence on blood sugar respectively. Diabetic biometric measurements were found to be: mean waist circumference of 102 ± 16 cm in men and 92 ± 16 cm in women, mean BMI of 28 ± 5 kg/m² in men and 28.7 ± 7 kg/m² in women and mean waist/hip ratio of 0.99 and 0.93 in men and women respectively. However there was no significant difference in waist, hip circumference, body mass index, waist/thigh and waist/hip ratio between diabetics and non-diabetics.

Discussion:  Our focus is on how the biometric measurements of diabetics differ from non-diabetics in the South Asian population. The significant inverse relationship between thigh and calf circumference and high blood glucose level can be used as an observational clinical screening tool to quickly identify patients with poorly controlled blood sugar in a typical out-patient setting. Further study is required to determine whether these biometric changes are due to the diabetes itself; or whether having a certain body shape predisposes to diabetes mellitus; and also whether these changes can be reversed with good glycemic control.

Conclusion:  In our study sample type 2 diabetes mellitus is associated with significantly thinner thighs and calves than non-diabetes whereas central adiposity is prevalent in both diabetics and non-diabetics.

Key Words:  Type 2 diabetes mellitus, biometric variables, anthropometric measurements, BMI, thigh, calf, waist, waist/hip ratio. waist/thigh ratio.