Assessment of portal hypertension by transient elastography in patients with compensated cirrhosis and potentially resectable liver tumors.

Share

Abstract

Background & Aims Patients with cirrhosis and small hepatocellular carcinoma with normal bilirubin and hepatic venous pressure gradient (HVPG) <10mmHg have >70% 5-year survival after hepatic resection. On the contrary, patients with HVPG ⩾10mm Hg (clinically significant portal hypertension, CSPH) frequently develop decompensation following surgery, with around 50% 5-year survival. Liver stiffness (LS) evaluation by transient elastography might non-invasively identify CSPH. We investigated the usefulness of LS predicting CSPH in patients with compensated cirrhosis and potentially resectable liver tumors. Methods Ninety-seven consecutive Child-Pugh A patients with potentially resectable liver tumors referred for HVPG measurement were prospectively evaluated. In fasting conditions LS was measured before the hemodynamic study. Results HVPG could be measured in all patients, whereas LS could not be measured in 18 (18.5%) obese patients. In the 79 patients with valid LS, 32 (40.5%) had CSPH; mean HVPG was 8.8±4.7mmHg. Mean LS was 18.4±12.3kPa. LS showed a moderate correlation with HVPG (r=0.552; p<0.001). LS<13.6kPa had high sensitivity (91%) but low specificity (57%) excluding CSPH. Conversely, LS>21kPa had low sensitivity (53%) and high specificity (91%) predicting CSPH. 35% of patients had LS between 13.6 and 21kPa (“grey zone”). Conclusions These data suggest that in real-life scenarios half of patients with potentially resectable liver nodules can be non-invasively classified as having or not CSPH by LS. However, in the remaining half, LS is either not applicable or inaccurate. In this last population HVPG is still a non replaceable method to detect CSPH.