Four RCTs were included in the review (n=856 patients; range 144 to 318). All of the RCTs scored 5 points on the Jadad scale.
Prophylactic nitroglycerin use was significantly associated with a reduction of overall post-endoscopic retrograde cholangiopancreatography pancreatitis (RR 0.60, 95% CI 0.39 to 0.92; 856 patients). The incidence of moderate or severe post-endoscopic retrograde cholangiopancreatography pancreatitis was not significantly reduced by nitroglycerin. There was no significant heterogeneity in the analyses.
Sensitivity analysis, excluding a trial in which diagnostic criteria were not proposed by Cotton et al, found that there was no longer any significant difference between treatment and placebo. The sensitivity analysis for severity of post-endoscopic retrograde cholangiopancreatography pancreatitis was not changed as there were no moderate or severe patients in the excluded trial.
Subgroup analyses of single centre trials (RR 0.58, 95% CI 0.35 to 0.97; 648 patients) or trials with low risk of post-endoscopic retrograde cholangiopancreatography pancreatitis (RR 0.38, 95% CI 0.19 to 0.75; 330 patients) supported the prophylactic effect of nitroglycerin on overall post-endoscopic retrograde cholangiopancreatography pancreatitis. Multi-centre trials or trials with moderate to high risk of post-endoscopic retrograde cholangiopancreatography pancreatitis did not demonstrate a significant difference between nitroglycerin and placebo. There was no significant effect of nitroglycerin in the prevention of moderate or severe post-endoscopic retrograde cholangiopancreatography pancreatitis in any of the subgroup analyses.
Hypotension and headache were the main adverse effects.