Second study shows that octreotide may prevent early rebleeding in cirrhosisBMJ 1998; 316 doi: https://doi.org/10.1136/bmj.316.7140.1320 (Published 25 April 1998) Cite this as: BMJ 1998;316:1320
- Gennaro D'Amico, Consultant in gastroenterology,
- Flavia Politi, Research fellow,
- Adele D'Antoni, Research fellow,
- Gandolfo Giannuoli, Senior registrar,
- Linda Pasta, Consultant in gastroenterology,
- Giovanni Vizziani, Senior registrar,
- Mario Traina, Senior registrar,
- Alberto Morabito, Associate professor of medical statistics,
- Luigi Pagliaro, Full professor of medicine
EDITOR—Jenkins et al have reported a randomised trial showing that long term subcutaneous octreotide together with sclerotherapy significantly reduces the risk of recurrent bleeding from oesophageal varices in liver cirrhosis.1
In a double blind placebo controlled pragmatic trial we have found that a 15 day course of subcutaneous octreotide was effective in preventing early rebleeding in cirrhosis. After acute bleeding of the upper digestive tract had been controlled, 262 consecutive patients with cirrhosis were randomised to receive octreotide 100 μg subcutaneously three times a day for 15 days (n=131) or placebo (n=131). β Blockers or sclerotherapy or both were added to the trial treatment as soon as possible, our hypothesis being that if octreotide reduced the risk of early rebleeding then β blockers or sclerotherapy, or both, could maintain that early beneficial effect. Separate randomisation and analysis were performed according to whether patients were eligible to receive β blockers or sclerotherapy or both (101 placebo and 97 octreotide) or not (30 placebo, 34 octreotide). Rebleeding from any source within 15 days was the primary measure of treatment efficacy; rebleeding within 42 days was assessed as a secondary measure. The table summarises the results.
Rebleeding within 15 days among the 198 patients randomised to receive β blockers or sclerotherapy, or both, was 27% (27/101) in the placebo group and 15% (15/97) in the octreotide group (P=0.05); among the 64 who did not receive these treatments they were 33% (10/30) and 47% (16/34) respectively (P=0.29). Corresponding figures in the whole series were 28% (37/131) and 24% (31/131) respectively (P=0.40). Among the patients who received β blockers or sclerotherapy, or both, those taking octreotide showed a significant reduction in rebleeding episodes (18 v 35, P=0.03), blood transfusions (50 v 75, P=0.04), and days in hospital (1190 v 1544, P<0.0001); this beneficial effect was maintained 42 days after randomisation. Mortality was not affected by octreotide in either subgroup of patients.
The 64 patients not randomised to receive β blockers or sclerotherapy or both had advanced liver disease (35 had Child-Pugh grade C disease; 30 died within 42 days). This might explain the lack of an effect of octreotide in these patients. Octreotide for early rebleeding failed to show any beneficial effect in another trial,2 in which 47% of patients had Child-Pugh grade C disease. By contrast, patients with grade C disease made up 28% of the patients in Jenkins et al's trial, 24% of the 198 patients benefiting from octreotide in our trial, and 25% of the patients in another positive trial.3
Thus we agree with Jenkins et al that octreotide may prevent early rebleeding in cirrhosis that is not advanced.