Abdominal Obesity as a Predictive Factor of Nonalcoholic Fatty Liver Disease Assessed by Ultrasonography and Transient Elastography in Polycystic Ovary Syndrome and Healthy Women

Abstract EN

Polycystic ovary syndrome (PCOS) and nonalcoholic fatty liver (NAFLD) share similar clinical presentations including obesity, insulin resistance (IR), and metabolic abnormality.

The predictive factors of NAFLD in women with PCOS and specifically in Asian women are not well established.

Associated factors for NAFLD assessed by ultrasound (US) among a group of PCOS and healthy women were determined and diagnostic accuracy between US and transient elastography (TE) for NAFLD was compared and correlated.

Sixty-three women with ages ranging from 20 to 40 years participated in the present cross-sectional study.

Forty-two women with PCOS as diagnosed by the Rotterdam criteria and 21 healthy women were recruited into the study.

Women with underlying hepatic diseases and history of alcohol consumption >20 g/day were excluded.

Biochemical and hormonal testing, anthropometrics, liver US, and TE were assessed.

Waist circumference (WC) greater than 80 cm was the only predictive factor for NAFLD as assessed by US in the whole group (adjusted odds ratio [aOR] 5.49, 95% confidence interval [CI]: 1.85–16.26, p <0.001).

The value of the TE-based controlled attenuation parameter (CAP) was significantly correlated with stage of steatosis as assessed by US (correlation coefficient = 0.696, p <0.001).

The diagnostic accuracies of dichotomized CAP ≥236 dB/m assessed for NAFLD using US as the gold standard were 84% and 78% sensitivity and specificity, respectively, with the area under the curve at 0.81 (p <0.001).

Abdominal obesity, rather than the presence of PCOS, was shown to be the independently associated factor for NAFLD.

WC could be used as the primary screening tool before performing complicated intervention for detection of steatosis.

TE is an alternative noninvasive detection tool in women with PCOS for NAFLD and hepatic fibrosis identification.