Major depression recurrence is associated with differences in obesity-related traits in women, but not in men.

BACKGROUND

Obesity-related cardiometabolic comorbidity is common in major depressive disorder (MDD). However, sex differences and MDD recurrence may modify the MDD-obesity-link.

METHODS

Sex-specific associations of MDD recurrence (single [MDD] or recurrent episodes [MDD]) and obesity-related traits were analyzed in 4.100 adults (51.6% women) from a cross-sectional population-based cohort in Germany (SHIP-Trend-0). DSM-IV-based lifetime MDD diagnoses and MDD recurrence status were obtained through diagnostic interviews. Obesity-related outcomes included anthropometrics (weight, body mass index, waist- and hip-circumference, waist-to-hip ratio, waist-to-height ratio), bioelectrical impedance analysis of body fat mass and fat-free mass, and subcutaneous (SAT) and visceral adipose tissue (VAT) from abdominal magnetic resonance imaging. Sex-stratified linear regression models predicting obesity-related traits from MDD recurrence status were adjusted for age, education, and current depressive symptoms.

RESULTS

790 participants (19.3%) fulfilled lifetime MDD criteria (23.8% women vs. 14.5% men, p<0.001). In women, MDD was inversely associated with anthropometric indicators of general and central obesity, while MDD was positively associated with all obesity-related traits, except waist-to-hip ratio and fat-free mass. In women, MDD versus MDD was associated with higher levels of obesity across all outcomes except fat-free mass. In men, MDD was positively associated with SAT regardless of MDD recurrence. Additionally, lifetime MDD was positively associated with VAT in men. Results remained significant in sensitivity analyses after exclusion of participants with current use of antidepressants.

CONCLUSIONS

The MDD-obesity association is modified by MDD recurrence and sex independent of current depressive symptoms. Accounting for sex and MDD recurrence may identify individuals with MDD at increased cardiometabolic risk.