Five RCTs were included in the review (n=702 patients, sample size 45 to 238). Three trials were high quality and scored 3 on the modified Jadad scale. Two trials scored 2.
Complication rate: The overall complication rate was 31%. Early endoscopic retrograde cholangiopancreatography (ERCP) was associated with a statistically significant reduction in complication rate compared with conservative management (pooled RD -8.7%, 95% CI -15.8 to -1.5; NNT=12 patients; five RCTs). Subgroup analysis revealed that, whilst early ERCP was associated with a statistically significant reduction in complication rate compared with conservative management in patients with severe pancreatitis (RD -38.5%, 95% CI -53 to -23.9; NNT=3 patients, three RCTs), there was no statistically significant difference in patients with mild pancreatitis (RD 1.8%, 95% CI -5.6 to 9.3; three RCTs).
Mortality rate: The overall mortality rate was 6%. There was no statistically significant difference in the mortality rate of patients who had undergone early ERCP compared with those who had received conservative management. Subgroup analysis showed that there was no statistically significant difference in mortality rate of patients with severe pancreatitis and those with mild pancreatitis.
Hospital stay: In patients with predicted mild pancreatitis, there was no difference in length of hospital stay in patients who had undergone early ERCP compared with those who had received conservative management (unclear if this was based on one or two RCTs). Two trials showed that patients with predicted severe pancreatitis who underwent early ERCP (9.5 days, range 6 to 36; 15.4 ± 3 days; p< 0.03) had a statistically significant shorter stay in hospital compared with those who received conservative management (17.0 days, range 4 to 74; 63 ± 4.8 days; p<0.05).
There was no evidence of statistical heterogeneity for any of the analyses.
The number of null trials needed for a statistically non significant result was three.