Female urinary incontinence and obesity assessed by anthropometry and dual-energy X-ray absorptiometry: Analysis from the 2008-09 Korean National Health and Nutrition Examination Survey.

Objective In the present study we evaluated the association between obesity, assessed by dual energy X-ray absorptiometry (DEXA), and urinary incontinence (UI). Methods The study was performed on 5792 women who had taken part in the Korean National Health and Nutrition Examination Survey. UI was deemed to be present if a woman answered "yes" to the question "Do you have current UI?". Obesity was assessed using anthropometry and DEXA. Data were analyzed using Chi-squared tests, t-tests, receiver operating characteristic curves, and logistic regression analysis. Results The UI group had significantly higher mean (+/- SD) waist circumference (78.5 +/- 10.0 vs, 82.4 +/- 9.1 kg) and body mass index (23.3 +/- 3.4 vs. 24.2 +/- 3.1 kg/m(2)) than the non-UI group. In addition, total fat mass (18.5 +/- 5.3 vs. 19.4 +/- 4.9 kg), trunk fat mass (9.3 +/- 3.4 vs. 10.1 +/- 3.2 kg), the trunk fat/leg fat (mass) ratio (1.58 +/- 0.54 vs. 1.73 +/- 0.50), total body fat percentage (32.3 +/- 5.4% vs. 33.0 +/- 5.0%), and trunk fat percentage (32.4 +/- 7.3% vs. 33.9 +/- 6.6%) were significantly higher in the UI group. Of these parameters, the trunk fat/leg fat ratio showed highest sensitivity (83.6%), with a cut-off value of 1.272. Before and after adjustment, trunk fat/leg fat ratio >1.272 was significantly related to UI and had the highest odds ratio (OR) among all DEXA parameters (adjusted OR 1.807; 95% confidence interval 1.343-2.431). Conclusion Obesity parameters obtained using DEXA are closely related to UI. Of these parameters, the trunk fat/leg fat ratio is the strongest in predicting the presence of UI. In addition, the present study has found a novel trunk fat/leg fat ratio cut-off value for defining obesity related to the UI.