Three RCTs were included (n=462). Studies scored 2 or 3 out of 5 for quality. Two RCTs used adequate randomisation methods; none were double-blind. All studies had complete follow-up.
There was no statistically significant difference between ERCP and conservative treatment in morbidity or mortality; both analyses were based on three studies. Significant heterogeneity was found for both analyses (I2 65% and 54%). Results were similar for patients with mild and severe pancreatitis.
After exclusion of one study with relatively low quality and enrolment of patients within 72 hours of admission (the other two studies were in patients within 72 hours of symptom onset), studies were homogeneous.
There was no statistically significant difference between treatments for morbidity. There was a nonsignificant trend favouring conservative treatment for mortality. Both analyses were based on two studies.