PROPRANOLOL IN THE PREVENTION OF RECURRENT UPPER GASTROINTESTINAL-BLEEDING IN PATIENTS WITH CIRRHOSIS UNDERGOING ENDOSCOPIC SCLEROTHERAPY -A RANDOMIZED CONTROLLED TRIAL

The purpose of this study was to investigate the possible value of con tinuous administration of propranolol in the prevention of recurrent u pper gastrointestinal bleeding in patients with cirrhosis undergoing c hronic endoscopic sclerotherapy. Among 239 patients admitted for acute variceal bleeding, 85 with cirrhosis were randomized to receive scler otherapy either alone (40) or in combination with propranolol (45). Sc lerotherapy was carried out with an intravariceal injection of 5% etha nolamine oleate through a fiberoptic endoscope. The procedure was perf ormed every week, until the esophageal varices at the gastroesophageal junction were too small for any further injections. Varices were rein jected if they recurred. Propranolol was given orally twice a day unti l heart rate was reduced by 25% in the resting position. The mean foll ow-up period was 23.2 and 24.2 months for sclerotherapy and the sclero therapy plus propranolol groups, respectively. During this period a si gnificant (P = 0.001) reduction in the recurrence of esophageal varice s was observed in patients treated with the combination of sclerothera py plus propranolol compared with those treated with sclerotherapy alo ne. However, the time of rebleeding from any source or from esophageal varices did not differ significantly between the two groups. In the s clerotherapy group 21 patients rebled (35 bleeding episodes) compared with 14 (22 episodes) in the combination therapy group. Patients in th e sclerotherapy group were more prone to bleed from gastric varices an d congestive gastropathy than patients treated with the combination of sclerotherapy plus propranolol (P = 0.012). Twenty-five patients in t he endoscopic sclerotherapy group developed complications attributed t o sclerotherapy compared with 23 patients in the sclerotherapy plus pr opranolol group. Complications directly attributable to propanolol wer e observed in 11 patients. Three of these patients stopped taking the drug due to heart failure (1) and flapping tremor (2). Eight patients (17.8%) died in the latter group while the corresponding figure in the sclerotherapy group was nine (22.5%). It is concluded that the contin uous administration of propranolol may reduce incidences of recurrent upper gastrointestinal hemorrhage from gastric sources in patients wit h cirrhosis undergoing chronic sclerotherapy.