PREVALENCE OF LIVER STEATOSIS AND FIBROSIS AND RELIABILITY OF ULTRASOUND IN DETECTING NAFLD IN OBESE PATIENTS

Background: Non-alcoholic fatty liver disease (NAFLD) is the most frequent chronic liver disease in Western countries, and in particular in obese patients the prevalence is higher than in general population. It has a large clinical spectrum, ranging from simple steatosis to steatohepatitis (NASH) and cirrhosis. Aim of the study: To evaluate in obese patients who underwent bariatric surgery in the last 3 years: 1) the histological prevalence and severity of liver steatosis (LS) and fibrosis; 2) the reliability of ultrasound (US) in diagnosing steatosis. Patients and methods: We reviewed preoperative liver US and intraoperative liver biopsy results in 57 bariatric surgery patients (age 42 ± 12.0 years, Body Mass Index (BMI) 43.5 ± 7.1), along with their clinical characteristics. Steatosis was diagnosed when more than 5 % of liver parenchyma was involved, the NAFLD activity score (NAS) and grade of fibrosis were determined according to Clinical Research Network system for scoring activity and fibrosis in non-alcoholic fatty liver disease (NASH CRN). US was performed after a 12-hour fast. US diagnosis of LS was evaluated with bright liver echopattern. Results: The prevalence of liver steatosis was 81 % (CI 95 %: 71-90), in patients with LS, the NAS Score was: 0-2 (not NASH) in 11 patients (24%; 13-37); 3-4 (borderline) in 20 patients (43 %; 30-58); ≥ 5 (NASH) in 15 patients, (33 %; CI 95 %: 21-46). Fibrosis was present in 40 patients (87 %; CI 95 %:75-95), bridging fibrosis (F= 3) in 3 patients, cirrhosis in 1 patients. The Sensitivity of US in steatosis diagnosis was 93% (CI 95 %:81-98), Specificity 42% (CI 95 %:9-75), Positive Predictive Value 89.3% (CI 95 %:83-93), Negative Predictive Value 50 % (CI 95 %:13-72), Accuracy 84.9% (CI 95 %:72- 93). Conclusions: In obese patients the prevalence of steatosis was higher than in the general population (82%), NASH and borderline-NASH were present in 24 and 43%, respectively. Finally, severe fibrosis or cirrhosis was present in 3 and 1 patient, respectively (in total 10 %); this low prevalence may be due to the young age of patients, confirming that earlier therapy could be associated with less irreversible liver damage. The US in diagnosing liver steatosis showed a good sensitivity, but insufficient specificity, confirming that in severe obesity the US reliability is low