BACKGROUND

The possible influence exerted by mechanical factors and/or compressive phenomena on myocardial strain parameters in healthy individuals with opposite obesity phenotypes (android vs gynoid) has never been previously investigated. Accordingly, we aimed at evaluating the relationship between anthropometrics, such as the waist-to-hip ratio (WHR), modified Haller index (MHI, the ratio of chest transverse diameter over the distance between sternum and spine), and epicardial adipose tissue (EAT), and left ventricular (LV)-global longitudinal strain (GLS), in healthy women with opposite obesity phenotypes (android vs gynoid).

METHODS

Forty healthy women with obesity (body mass index (BMI) ≥30 Kg/m and WHR ≥0.85 ("android group") (52.5±13.2 yrs), 40 age- and BMI-matched healthy women with obesityand WHR <0.78 ("gynoid group") (49.8±13.4 yrs) and 40 age-matched healthy women without obesity (BMI <30 Kg/m) (controls) (50.3±12.5 yrs) were retrospectively analyzed. All women underwent transthoracic echocardiography implemented with echocardiographic strain analysis of all cardiac chambers. Correlation between LV-GLS and anthropometrics (WHR, MHI, and EAT) was assessed in both groups of obese women. Age, WHR, homeostasis model assessment for insulin resistance (HOMA-IR), and left ventricular mass index (LVMi) were included in the logistic regression analysis performed for evaluating the independent predictors of reduced LV-GLS magnitude (less negative than -20%) in women with android obesity.

RESULTS

Compared to the other groups of women, those with android obesity were found with significantly greater LVMi, higher LV filling pressures, and lower biventricular and biatrial deformation indices. A strong inverse correlation between LV-GLS and all anthropometrics (WHR, MHI, and EAT) was demonstrated in both groups of women with obesity. Univariate logistic regression analysis revealed that WHR (OR 1.58, 95%CI 1.22-2.03, p<0.001) and LVMi (OR 1.09, 95%CI 1.02-1.16, p=0.006) were independently correlated with LV-GLS impairment in women with android obesity. On multivariate logistic regression analysis, the WHR maintained a statistically significant association with the above-mentioned outcome (OR 1.68, 95%CI 1.14-2.48, p=0.009). Receiver operating characteristic (ROC) curve analysis showed that a WHR value ≥1.01 had 93% sensitivity and 100% specificity for detecting LV-GLS impairment in women with android obesity (AUC=0.98; 95%CI 0.96-1.00).

CONCLUSIONS

Anthropometrics may strongly influence cardiac mechanics in healthy women with obesity. The WHR is associated with reduced LV-GLS magnitude in healthy women with android obesity, independent of age, glycometabolic status, and LV size.