Response and Adverse Event Rates With Placebo in Gastroparesis: A Systematic Review and Meta-analysis

Multiple drugs have been used to treat gastroparesis symptoms, yet their therapeutic benefits are poorly understood partly due to lack of insight into response and adverse event rates with placebo in randomized controlled trials (RCTs). We evaluated these issues systematically in drug trials for gastroparesis.

We searched the medical literature through August 2, 2022 to identify RCTs comparing active drug with placebo in patients with gastroparesis. We assessed placebo response rates according to at least one of the following endpoints: improvement according to a composite outcome, nausea, vomiting, abdominal pain, bloating, or fullness, as well as total adverse events, and adverse events leading to withdrawal. We extracted data as intention-to-treat analyses with dropouts assumed to be treatment failures. We pooled placebo response and adverse event rates using a random effects model and expressed as proportions with 95% confidence intervals (CIs).

Thirty-five studies were eligible. Among 23 trials reporting a composite endpoint of improvement, the pooled placebo response rate was 29.3% (95% CI, 23.7%-35.2%). Pooled placebo response rates were higher in idiopathic compared with diabetic gastroparesis (34.2% vs 28.1%), among trials that did not use validated symptom questionnaires (31.2% vs 27.4%), and in RCTs of shorter duration (<4 weeks, 32.6% vs ≥9 weeks, 23.2%). Adverse events occurred in 33.8% (95% CI, 26.4%-41.8%) of patients with placebo, in 27 trials, and were less common in idiopathic compared with diabetic gastroparesis (17.9% vs 43.4%), trials of shorter duration (<4 weeks, 33.7% vs ≥9 weeks, 40.7%), and trials with lower randomization ratios of active drug to placebo (1:1, 26.7% vs 3:1, 50.5%).

This meta-analysis assessed placebo response and adverse event rates in gastroparesis. To accurately assess therapeutic gain, future trials should be a minimum of 8 weeks duration, use validated questionnaires, and distinguish gastroparesis subtypes.