A new prognostic algorithm based on stage of cirrhosis and HVPG to improve risk-stratification after variceal bleeding

Background & Aims HVPG decrease ≥20% or ≤12mmHg (“responders”) indicates good prognosis during propranolol/nadolol treatment but requires two HVPG measurements. We aimed at simplifying risk‐stratification after variceal bleeding using clinical data and HVPG. Methods 193 cirrhotic patients (62% with ascites and/or hepatic encephalopathy, HE) included within 7‐days of bleeding had HVPG measured before and at 1‐3 months of treatment with propranolol/nadolol plus endoscopic band ligation. End‐points: Rebleeding and rebleeding/transplantation‐free survival for 4‐years. Another cohort (n=231) served as validation set. Results During follow‐up 45 patients had variceal bleeding and 61 died. HVPG‐responders (n=71) had lower rebleeding‐risk (10% vs 34%, p=0.001) and better survival than 122 non‐responders (61% vs 39%, p=0.001). Patients with/HE (n=120) had lower survival than patients without (40% vs 63%, p=0.005). Among patients with ascites/HE, those with baseline HVPG≤16mmHg (n=16) had low rebleeding‐risk (13%). By contrast, among patients with ascites/HE and baseline HVPG>16mmHg, only HVPG‐responders (n=32) had good prognosis, with lower rebleeding‐risk and better survival than non‐responders (n=72) (respective proportions: 7% vs 39%,p=0.018; 56% vs 30% p=0.010). These findings allowed developing a new algorithm for risk‐stratification in which HVPG‐response was only measured in patients with ascites and/or HE and baseline HVPG>16mmHg. This algorithm reduced the grey‐zone (high‐risk patients not dying on follow‐up) from 46% to 35% and decreased by 42% the HVPG measurements required. The validation cohort confirmed these results. Conclusion Restricting HVPG measurements to patients with ascites/HE and measuring HVPG‐response only if baseline HVPG>16mmHg improves detection of high‐risk patients while markedly reducing the number of HVPG measurements required.