Adipopenia correlates with higher portal pressure in patients with cirrhosis

Abstract Background & Aims In cirrhosis, hepatic venous pressure gradient (HVPG) and imaging body composition assessment can influence prognosis. We assessed whether adipose and skeletal muscle tissues reflect the severity of portal hypertension (PH), and whether they improve non‐invasive prediction of decompensation and death. Methods We included 84 cirrhosis patients with HVPG and computed tomography (CT) within 12 weeks of HVPG at a single centre. L3 vertebra CT images were used for body composition indexes (cm 2 /m 2 ): total adipose tissue index (TATI), visceral adipose tissue index (VATI), subcutaneous adipose tissue index (SATI), intramuscular adipose tissue index (IMATI), skeletal muscle index (SMI) and psoas muscle index (PMI). Correlations were calculated between indexes, HVPG and standard non‐invasive tests for PH. Twelve‐month decompensation and death predictors were determined. Results The following were the characteristics for the patients included in the study: male 61%; BMI 28 ± 5 kg/m 2 ; alcoholic liver disease in 51%, non‐alcoholic steatohepatitis in 24%; HVPG 14 ± 6 mm Hg; 45% compensated. The median follow‐up was 11 (4‐17) months. HVPG correlated with SATI ( r = −0.282, P = 0.01), TATI ( r = −0.220, P = 0.045) and SATI/VATI index ( r = −0.240, P = 0.03). In compensated patients, lower VATI (HR 0.94 (0.90‐0.99), P = 0.01) was associated with 12‐month decompensation. Combining TATI and liver stiffness × spleen size over‐platelet count risk score added discriminative capacity for 12‐month decompensation (AUROC 0.91 vs 0.87). IMATI was independently associated with mortality in decompensated patients. MELD‐Na combined with IMATI discriminated excellently for mortality (AUROC 0.94; P < 0.001). Conclusions Hepatic venous pressure gradient inversely correlates with imaging markers of adipose tissue, while markers of sarcopenia were unrelated to PH. In compensated patients, TATI improves non‐invasive prediction of decompensation. In decompensated patients, IMATI independently predicted mortality.