Study designs of evaluations included in the review
Randomised controlled trials (RCTs) were eligible. Studies that did not provide standard deviations or standard errors were excluded from the meta-analysis.
Specific interventions included in the review
Comparisons of H2RAs with a concurrent placebo control were eligible. Studies that used only nasogastric tube suction as a control intervention were excluded. Only comparisons of parenteral cimetidine (1,200 to 2,000 mg/day, usually in divided doses every 4 to 6 hours) with placebo or fasting were included. The cointerventions, used in both groups as needed, included: nil by mouth, intravenous hydration, analgesia, antibiotics, and nasogastric tube.
Participants included in the review
The inclusion criteria were not defined in terms of the participants. Patients with alcoholic, biliary, idiopathic, and miscellaneous-cause pancreatitis were included. Most of the participants were male (40 to 96% across the studies), with a mean age ranging from 40 to 55 years across the studies. The mean baseline initial serum amylase was 2.5 to 5.6 times higher than the normal upper limit.
Outcomes assessed in the review
Studies that reported preplanned outcomes measures or the number of patients who died, developed complications or dropped out, were eligible. The duration of abdominal pain was also assessed. The authors intended to assess fever, leucocytosis and the dose of analgesia used, but insufficient data were reported in primary studies.
How were decisions on the relevance of primary studies made?
Two investigators independently screened the titles and abstracts of the identified studies, and any disagreements were resolved by a third investigator. The agreement between the reviewers for the study selection process was 100%.