RÉSUMÉ
AIM: The aim of this work was to examine the efficacy of oral metronidazole in reducing posthaemorrhoidectomy pain versus placebo. METHOD: Forty patients were randomized to either metronidazole and standard care or placebo and standard care (21 metronidazole, 19 placebo) in a double-blinded, randomized controlled trial. The main outcome measure was posthaemorrhoidectomy pain scores over 21 days, measured on a 10-point Likert scale. RESULTS: There were no significant differences between groups with regards to age, gender, smoking status, self-reported general health or quality of life, haemorrhoid-related pain, haemorrhoid-related impact on quality of life, reported satisfaction with surgery, experience of surgery, median overall pain score or likelihood of recommending surgery to others. For reported median worst pain scores and defaecation-related pain, a trend to significance was identified between groups on days 16 and 18-21, with the metronidazole group reporting less pain. However, these differences were not significant when prespecified Bonferroni correction criteria were used. Using multilevel mixed effects modelling, the impact of time on median worst pain score was identified to be highly significant (P < 0.0001) whereas treatment allocation (placebo versus metronidazole) did not significantly affect the improvement in patients' reported pain (P = 0.8837). CONCLUSION: Our data do not support the hypothesis that postoperative metronidazole has a clinically meaningful effect on posthaemorrhoidectomy pain. This study adds to the previous literature, and implies that it should not be routinely used as an adjunct to analgesia.