Study and patient characteristics
Study descriptions are provided in Table 1. Study endpoint details for each study are stated in the Supplementary Data. The quality of studies evaluation was assessed, using version 2 of the Cochrane risk-of-bias tool for randomised trials (RoB 2), into two categories: some concerns and low risk of bias (Supplemental Figures 4.6, 5.6, 6.6, 7.6, 8.6 and 9.6). Seventeen RCT compared prokinetics to placebo29–45. Seven compared peripheral µ-opioid receptor antagonists to placebo22–28. Four compared non-steroidal anti-inflammatory drugs to placebo11–14. One study compared glucocorticoids (dexamethasone) to placebo17. Two compared erythromycin to placebo18,19. Two studies compared gastrograffin to placebo20,21. Three compared probiotics to placebo46–48. One study compared colloid infusion to placebo16 and one study compared oral carbohydrates to placebo15. Erythromycin has a dual antibiotic and prokinetic effect, which is why this treatment is analysed separately 49. We have classified erythromycin in a category outside the prokinetic treatments with regard to its reference therapeutic class. Indeed, erythromycin being a macrolide antibiotic, it has a potential effect on the balance of the microbiota.
The 38 included studies covered 6,371 patients: 2,091 patients for prokinetic studies (including 860 receiving placebo), 3,118 patients for µ-opioid receptor antagonist studies (including 1151 receiving placebo), 276 patients for non-steroidal anti-inflammatory drug studies (including 137 receiving placebo), 302 patients for glucocorticoid studies (including 151 receiving placebo), 156 patients for erythromycin studies (including 80 receiving placebo), 129 patients for gastrografin studies (including 65 receiving placebo), 169 patients for probiotics studies (including 85 receiving placebo), 80 patients for colloid infusion studies (including 42 receiving placebo) and 50 patients for oral carbohydrate studies (including 25 receiving placebo). Pooled baseline characteristics are displayed in Table 2.