Fourteen RCTs (n=7,984 patients) were included in the meta-analysis. Of these fourteen RCTs, four were abstracts and ten were full articles. The full articles met all the quality criteria. Funnel plot analysis did not show any evidence of publication bias
Abdominal pain and discomfort relief: 5-hydroxytryptamine antagonists were associated with statistically significant relief of abdominal pain and discomfort compared to control (RR 1.30, 95% CI 1.22 to 1.39; 10 RCTs). The calculated number-needed-to-treat was 7.7. There was no evidence of statistically significant heterogeneity.
Global improvement of IBS symptoms: Compared to patients receiving control, those receiving 5-hydroxytryptamine antagonists had a statistically significant increase in global improvement of IBS symptoms (RR 1.60, 95% CI 1.49 to 1.72; seven RCTs). The number-needed-to-treat was 4.2. There was also no evidence of statistically significant heterogeneity.
Constipation: Patients receiving 5-hydroxytryptamine antagonists were statistically significantly more likely to report constipation (RR 4.28, 95% CI 3.28 to 5.60, 14 RCTs) compared to those in the control group. The calculated overall number-needed-to-harm was 4.7. However, there was evidence of statistically significant heterogeneity (I2=65%). The authors stated that this heterogeneity was due to constipation being reported as adverse events by patients in some trials and identified from bowel habit diaries in other trials. Results of subgroup analyses were reported (see full article).