Non-invasive tests for clinically significant portal hypertension after HCV cure.

BACKGROUND & AIMS

Non-invasive tests (NIT)s for clinically significant portal hypertension (CSPH; hepatic venous pressure gradient [HVPG] ≥10mmHg) have predominantly been studied in patients with active HCV-infection. Investigations after HCV-cure are limited and yielded conflicting results. We conducted a pooled analysis to determine the diagnostic/prognostic utility of liver stiffness-measurement (LSM)/platelet count (PLT) in this setting.

METHODS

418 patients with pre-treatment HVPG≥6mmHg who achieved sustained virological response (SVR) and underwent post-treatment-HVPG-measurement were assessed, of which 324 (HVPG/NIT-cohort) also had paired data on pre-/post-treatment-LSM/-PLT. The derived LSM/PLT-criteria were then validated against the direct endpoint decompensation in 755 compensated advanced chronic liver disease (cACLD) patients with SVR (cACLD-validation-cohort).

RESULTS

HVPG/NIT-cohort: Among cACLD patients, the pre-/post-treatment prevalence of CSPH was 80%/54%. The correlation between LSM/HVPG increased from pre- to post-treatment (r=0.45 vs. 0.60), while that of PLT/HVPG remained unchanged. For given LSM/PLT-values, HVPG tended to be lower post- vs. pre-treatment, indicating the need for dedicated algorithms. Combining post-treatment-LSM/-PLT yielded a high diagnostic accuracy for post-treatment-CSPH in cACLD (AUC: 0.884 [95%CI: 0.843-0.926]). Post-treatment-LSM<12kPa & PLT>150G/L excluded CSPH (sensitivity: 99.2%), while LSM≥25kPa was highly specific for CSPH (93.6%).

CACLD-VALIDATION-COHORT

The LSM<12kPa & PLT>150G/L-criterion was achieved in 42.5% of patients and their 3-year decompensation risk was 0%. In patients with post-treatment-LSM≥25kPa (prevalence: 16.8%), 3-year decompensation risk was 9.6%, while it was 1.3% in those meeting none of the above criteria (prevalence: 40.7%).

CONCLUSIONS

NITs can estimate the probability of CSPH after HCV-cure and predict clinical outcomes. cACLD patients with LSM<12kPa & PLT>150G/L (CSPH-excluded; no decompensation risk) may be discharged from portal hypertension surveillance (NITs and/or endoscopy), if no co-factors are present, while patients with LSM≥25kPa require surveillance/treatment (CSPH-ruled-in; increased decompensation risk).

LAY SUMMARY

Measurement of liver stiffness by a specific ultrasound device and platelet count (a simple blood test) are broadly used for the non-invasive diagnosis of increased blood pressure in the veins leading to the liver, which drives the development of complications in patients with advanced liver disease. The results of our pooled analysis refute previous concerns that these tests are less accurate after the cure of hepatitis C virus (HCV) infection. We have developed diagnostic criteria that facilitate the personalized management after HCV-cure and allow for a de-escalation of care in a high proportion of patients, thereby decreasing disease burden.