Study designs of evaluations included in the review
Articles not reporting original data (e.g. reviews, editorials, news articles, letters, case reports) were excluded. Studies where the total number of patients could not be determined, was less than 20, or did not equal the sum of the patients in subgroups mentioned were excluded.
Specific interventions included in the review
Only studies of the diagnostic accuracy of MA-HBS and/or C-HBS were eligible for inclusion in the review. For MA-HBS a positive test had to be defined as non-visualisation of the gallbladder 20-30 min after morphine administration. For MA-HBS studies morphine augmentation had to be generally commenced 60 min after the start of the study if there was non-visualisation of the gallbladder.
Studies were excluded if the criteria for positive and negative test results were not described or if outdated radiopharmaceuticals were used (e.g. 131I Rose Bengal).
Reference standard test against which the new test was compared
Included studies had to confirm diagnosis using an acceptable 'gold' standard - surgery with confirmation by pathology, autopsy, or clinical follow-up with establishment of an alternative diagnosis.
Participants included in the review
The included studies were of patients with suspected acute cholecystitis (may have included some patients with severe illness) who had fasted for at least two hours before the study.
Studies were excluded if they were of atypical patients (e.g. jaundiced patients only or pregnant patients only). Studies that selected patients after surgery had established the diagnosis were also excluded.
In addition, articles that focused exclusively on the following populations were excluded: patients with liver dysfunction; patients with chronic cholecystitis; patients with late development of symptoms of acute cholecystitis after another illness; patients who used total parenteral nutrition; patients fasting longer than 24 hrs without cholecystokinin pre-treatment.
In C-HBS studies, data from patients whose gallbladders were visualised within 60 min, and who were therefore not candidates for MA-HBS, were excluded.
Outcomes assessed in the review
Studies were excluded if sensitivity and specificity could not be calculated from the data presented. Studies where information on acute cholecystitis could not be differentiated from other diagnoses involving the biliary tract were also excluded.
How were decisions on the relevance of primary studies made?
Non-physician research staff reviewed candidate titles. Two independent readers reviewed the remaining studies using pre-determined exclusion criteria relating to patient selection, diagnostic method, gold standard and study design. Reviewers were asked not to consider the authors, institution and publication source.