Objective: To determine the efficacy of lansoprazole 30 mg given in the morning compared with high-dose ranitidine 300 mg twice daily in the treatment of patients with oesophageal strictures.
Design: A multicentre, outpatient, double-blind, parallel group, prospectively randomized clinical trial.
Patients: One hundred and fifty-eight patients (lansoprazole 30 mg n = 78, ranitidine 600 mg n = 80) were enrolled from 19 centres in the UK over 23 months.
Interventions: Patients with an oesophageal stricture were randomized to receive either lansoprazole 30 mg once daily or high-dose ranitidine 300 mg twice daily for 12 months. Dilatation was performed at entry and repeat endoscopies were scheduled at 6 and 12 months and additionally at other times if there was symptomatic relapse. Redilatation was performed as required and according to a predefined scale. The patient's assessment of dysphagia over the previous 7 days was recorded by the investigator at 1, 3, 6, 9 and 12 months. Safety was assessed by laboratory tests, physical examination and all adverse events.
Main outcome measures: Efficacy was assessed primarily by the time to redilatation, the proportion of patients requiring at least one redilatation, and the number of redilatations over 12 months. The relief of dysphagia and reduction in stricture grade were secondary efficacy measures.
Results: The time to redilatation was longer and the probability of no redilatation were higher in the lansoprazole group than in the ranitidine group; for all patients randomized (intention to treat principle), this difference was of borderline significance (life table, P = 0.053). The proportions of patients requiring at least one redilatation during the 12-month treatment period were 30.8% (24/78) with lansoprazole and 43.8% (35/80) with ranitidine (all patients randomized, chi 2 test, P = 0.092). Compared to ranitidine, patients receiving lansoprazole reported significantly lower dysphagia grades at 6 months (stratified Wilcoxon test, P = 0.0086) but not at 12 months (stratified Wilcoxon test, P = 0.074). A greater proportion of patients in the ranitidine group-33.8% (27/80)-withdrew prematurely compared to the lansoprazole group (26.9%, 21/78). The most frequent reasons for premature withdrawal were adverse events and protocol violations. There were no clinically significant differences in incidence or severity of adverse events between the two groups. The mean increase in gastrin levels after 12 months' treatment was significantly greater for patients in the lansoprazole group (124.2 pg/ml, P = 0.0056) than those in the ranitidine group (31.9 pg/ml). No significant changes in gastric mucosal histology were detected for patients in either group.
Conclusion: It is concluded that lansoprazole 30 mg once daily is superior to ranitidine 300 mg twice daily in relieving dysphagia, and at least as effective in reducing the need for a repeat dilatation.