Correlação entre ultrassonografia com Doppler e biópsia na gradação da esteatose hepática não alcoólica

Abstract

Nonalcoholic fatty liver disease (NAFLD) is common and widespread diagnosed by ultrasound (US). Steatosis grading by conventional US is subjective and presents great interobserver and intraobserver variability. Liver biopsy has been considered the gold standard for the diagnosis of NAFLD, but there are controversies about its indication. A non-invasive, objective and reproducible quantification of steatosis is necessary for clinical practice. The aim of this study was to assess the correlation between sonographic hepatorenal ratio (HRR), portal vein pulsatility index (VPI) and right hepatic vein pattern with liver biopsy for gradation of nonalcoholic steatosis. Ultrasound and Doppler US were performed in 82 subjects, among which 42, with NAFLD, also underwent liver biopsy. Forty normal volunteers were included as control group. The echogenicity of the hepatic parenchyma and right kidney cortex were measured using the histogram echo intensity and the hepatorenal ratio (HRR) was derived. The VPI was calculated as maximum velocity at minimum velocity divided by maximum velocity. The waveform of right hepatic vein was classified in monophasic, biphasic or triphasic. The specimens obtained through needle biopsy was stained by hematoxylin-eosin and Masson s Trichrome and classified as mild (up to 33% of the hepatocytes infiltrated by fat droplets), moderate (from 33 to 66%) and severe (over 66% of fatty infiltration in hepatocytes). The inflammatory infiltrate was discriminate as mild, moderate and severe intensity. The fibrosis was described according to location in the hepatic acinus, aspect and intensity. A significant correlation was found between histologic steatosis and HRR (r=0.80, P <0.01). The cutoff point of HRR presented sensitivity of 92.7%, specificity of 92.5% and accuracy of 92.6%. The mean and standard deviation in subgroups were: control 1.09±0.13; mild 1.46±0.24; moderate 1.52 ±0.27 and severe 2.04±0.31. The means of HRR were different among themselves, except between the mild and moderate subgroups. The VPI presented inverse and significant correlation with steatosis degree at biopsy (r= -0.74 P <0.01). The mean and standard deviation of VPI in control group was 0.34±0.08 and in steatosis group was 0.20±0.07 (P <0.01), dropping progressively according to the increase in steatosis degree. However, the VPI did not discriminate the steatosis subgroups. The Doppler waveform pattern of the right hepatic vein was predominantly triphasic in control and mild subgroup while the monophasic pattern was more frequent in severe steatosis (P< 0.01). However, the distribution of Doppler waveform pattern of the right hepatic vein did not show significant differences among mild, moderate and severe steatosis groups. In this study, Hepatorenal sonographic ratio was the best sonographic parameter for gradation of steatosis in patients with NAFLD.