Level 2: Mid-level evidence

The efficacy of acetaminophen (Tylenol and others) and hyoscine butylbromide alone and in combination was evaluated in a study of 1,935 patients aged 18-70 years who had recurrent crampy abdominal pain for at least two months (Aliment Pharmacol Ther. 2006;23:1741-1748). The single-blind trial had a one-week placebo run-in period; bowel habits and diagnostic criteria for irritable bowel syndrome were not factors for inclusion. Pain intensity was measured once daily in the evening and recorded on a visual analog scale (0-100 mm) for most severe pain of the day. Patients with satisfactory compliance measures and a pain score of at least 30 mm on the last day of the run-in period were randomized to one of four groups for three weeks (N = 1,644): hyoscine butylbromide 10 mg p.o. three times daily, acetaminophen 500 mg p.o. three times daily, combination hyoscine butylbromide 10 mg plus acetaminophen 500 mg p.o. three times daily, or placebo p.o. three times daily. Not included in the intention-to-treat analysis were 73 patients (4.4%) who had missing or irregular data or failed to take the study drug; another 448 patients (29%) were not included in the per-protocol analysis because of major protocol violations, noncompliance, or other reasons. The study was conducted in Germany and sponsored by the butylbromide manufacturer.

All active treatments significantly reduced mean pain intensity more than placebo (by 4-5 mm on 100-mm scale) and significantly reduced mean pain frequency more than placebo (by 0.2 points on 0-3 scale). Mean pain intensity score (0-100 mm scale) at baseline was about 50 mm in all four groups. The mean score during treatment with placebo was 31 mm (mean change 19 mm) compared with a mean score of 25-26 mm (mean change 23-24 mm) during treatment among the three active-treatment groups (P <.0001 compared with placebo). The baseline pain frequency score (0-3 scale) was about 1.7 in all four groups. The mean frequency score during treatment with placebo was 1.1 (mean change 0.5) compared with the mean score of 0.9-1.0 (mean change 0.7) among the three active-treatment groups (P <.0001 compared with placebo). A global efficacy assessment of at least “good” was reported by about 44% of the placebo group and about 60% of active-treatment groups (NNT 7), based on extrapolation from a figure in the full-text article. The rates of adverse events were 11% with placebo, 14% with acetaminophen (NNH 33), 16% with hyoscine (NNH 20), and 17% with combination therapy (NNH 16).