BACKGROUND: It is unknown which is the best therapy to treat haemodynamic non-responders to pharmacological therapy after variceal bleeding. AIM: To evaluate the efficacy of adding banding ligation to drugs to prevent variceal rebleeding in haemodynamic non-responders to drugs. METHODS: Fifty-three cirrhotic patients with variceal bleeding underwent a hepatic venous pressure gradient (HVPG) measurement 5 days after the episode. Nadolol and nitrates were then titrated to maximum tolerated doses. A second HVPG was taken 14 days later. Responders (HVPG ≤12 mm Hg or ≥20% decrease from baseline) were maintained on drugs and non-responders had banding ligation added to drugs. RESULTS: Mean follow-up was 28 months. In 5 patients the second HVPG could not be performed because of early rebleeding. The remaining 48 patients were classified as responders (n=24) and non-responders (n=24), who had banding added. No baseline differences were observed between groups. Variceal rebleeding occurred in 12% of the 48 patients whose haemodynamic response was assessed. Responders on drug therapy presented a 16% rebleeding rate, whilst non-responders rescued with banding showed an 8% rebleeding rate. Rebleeding-related mortality was not different between groups. CONCLUSION: In a HVPG-guided strategy, adding banding ligation to drugs is an effective rescue strategy to prevent rebleeding in haemodynamic non-responders to drug therapy.
BACKGROUND: It is unknown which is the best therapy to treat haemodynamic non-responders to pharmacological therapy after variceal bleeding. AIM: To evaluate the efficacy of adding banding ligation to drugs to prevent variceal rebleeding in haemodynamic non-responders to drugs. METHODS: Fifty-three cirrhoticpatients with variceal bleeding underwent a hepatic venous pressure gradient (HVPG) measurement 5 days after the episode. Nadolol and nitrates were then titrated to maximum tolerated doses. A second HVPG was taken 14 days later. Responders (HVPG ≤12 mm Hg or ≥20% decrease from baseline) were maintained on drugs and non-responders had banding ligation added to drugs. RESULTS: Mean follow-up was 28 months. In 5 patients the second HVPG could not be performed because of early rebleeding. The remaining 48 patients were classified as responders (n=24) and non-responders (n=24), who had banding added. No baseline differences were observed between groups. Variceal rebleeding occurred in 12% of the 48 patients whose haemodynamic response was assessed. Responders on drug therapy presented a 16% rebleeding rate, whilst non-responders rescued with banding showed an 8% rebleeding rate. Rebleeding-related mortality was not different between groups. CONCLUSION: In a HVPG-guided strategy, adding banding ligation to drugs is an effective rescue strategy to prevent rebleeding in haemodynamic non-responders to drug therapy.