Search

Ultrasound of the liver is not good to pick up mild steatosis. Controlled attenuation parameter (CAP) evaluated in transient elastography (FibroScan) is widely available in India. However, data regarding the diagnostic accuracy and optimal cut-off values of CAP for diagnosing hepatic steatosis are scarce in Indian population. MRI-PDFF is an accurate technique for quantifying hepatic steatosis. Thus, this study examined the diagnostic accuracy and optimal cut-off values of CAP for diagnosing steatosis with MRI-PDFF as reference standard. A total of 137 adults underwent CAP and MRI-PDFF measurements prospectively. A subset of participants (n = 23) underwent liver biopsy as part of liver transplantation evaluation. The optimal cut-off values, area under the receiver operating characteristic (AUROC) curves, sensitivity, and specificity for CAP in detecting MRI-PDFF ≥5% and ≥10% were assessed. The mean age and body mass index (BMI) were 44.2 ±10.4 years and 28.3 ±3.9 kg/m2, respectively. The mean hepatic steatosis was 13.0 ±7.7% by MRI-PDFF and 303 ±54 dB/m by CAP. The AUROC of CAP for detecting hepatic steatosis (MRI-PDFF ≥5%) was 0.93 (95% CI, 0.88–0.98) at the cut-off of 262 dB/m, and of MRI-PDFF ≥10% was 0.89 (95% CI, 0.84–0.94) at the cut-off of 295 dB/m. The CAP of 262 dB/m had 90% sensitivity and 91% specificity for detecting MRI-PDFF ≥5%, while the CAP of 295 dB/m had 86% sensitivity and 77% specificity for detecting MRI-PDFF ≥10%. The optimal cut-off of CAP for the presence of liver steatosis (MRI-PDFF ≥5%) was 262 dB/m in Indian individuals. This CAP cut-off was associated with good sensitivity and specificity to pick up mild steatosis. [ABSTRACT FROM AUTHOR]

Generally diagnosis of non-alcoholic fatty disease is made on imaging, however, mild steatosis is difficult to diagnose on imaging. Liver biopsy is the procedure of choice but is not carried out as it is an invasive procedure. We describe our experience of 157 liver biopsies in living liver donors with normal body mass index (BMI) <23 kg/M2 (lean). The study was conducted at a tertiary care center in north India. Data of lean living donors who underwent a liver biopsy before donation were analyzed. Data are presented as percentage, mean, or median (25–75 interquartile range). Of 718 donors who had a liver biopsy before donation, 157 (21.8%) donors were lean (BMI < 23 kg/M2). Seventy-eight percent of lean donors had no or only one metabolic risk factor. Fifty-three (33.7%) of lean donors had nonalcoholic fatty liver (NAFL) in liver biopsy. When donors with NAFL were compared to donors with normal histology, donors with NAFL had significantly higher aspartate transaminase (26.6 ± 7.5 versus 23.7 ± 5.4, p = 0.007), alanine transaminase (33.4 ± 11.7 versus 27.8 ± 10.7, p = 0.003), and gamma glutamyl transpeptidase [25 (16–40.5) versus 18 (14–23), p = 0.003]. Only triglycerides (TGs) were statistically different among metabolic factors in lean NAFL and normal histology groups, 97 (70–161) versus 86 (62.5–114.7), p = 0.043. A total of 30% donors in the lean NAFL group had TGs >150 mg/dl as compared with 12.5% in the normal histology group, p = 0.009. Other metabolic risk factors were not statistically different. One third of lean donors had NAFL. Among all metabolic risk factors, only higher TGs levels showed a significant association with NAFL. [ABSTRACT FROM AUTHOR]