Effects of Promotility Agents on Gastric Emptying and Symptoms: A Systematic Review and Meta-analysis

Abstract

Studies have reported a lack of association between improvements in gastric emptying (GE) and upper gastrointestinal (UGI) symptoms with promotility drugs. However, GE test methods were suboptimal in some studies. We assessed improvements in GE and UGI symptoms in patients given promotility agents in studies with optimal or moderate test methods (scintigraphy or breath test, solid meal, >2 hours duration) compared to studies with suboptimal GE test methods. With an expert librarian, we completed an extensive search of publications in the Ovid MEDLINE (1946 to present), EMBASE (1988 to January 2018), and EBM Reviews Cochrane Central Register of Controlled Trials, without restrictions on language or year. Two independent reviewers evaluated the following inclusion criteria: randomized, blinded, parallel, or crossover trials of 5HT4 agonists, D2 receptor antagonist, or ghrelin agonists; trials that measured change in GE (T1/2) or composite UGI symptoms; trials of patients with functional dyspepsia and gastroparesis; and trials of GE test methods. Standardized mean differences (units expressed as SD) were used to standardize symptom assessments that were not uniform across studies. Random effects model was used to analyze data and meta-regression was used to evaluate the association between change in GE and UGI symptoms. Of 899 studies considered, 22 studies assessed change in GE; 23 evaluated UGI symptoms; and 14 evaluated GE and UGI symptoms. Promotility agents significantly accelerated GE (T1/2) in all studies (mean reduction in T1/2, 16.3 minutes; 95% confidence interval, -22.1 to -10.6 minutes) and in studies that used optimal GE test methods (mean reduction in T1/2, 23.6 minutes; 95% confidence interval, -32.3 to -14.9 minutes). Promotility agents also significantly reduced UGI symptoms (mean reduction, 0.25 SD; 95% confidence interval, -0.37 to -0.13 SD). Meta-regression found no significant association between change in GE and UGI symptoms. However, when only studies with optimal GE test methods were evaluated, there was a significant positive association between improvement in GE and UGI symptoms (P= .02). In a meta-analysis of published trials, we found promotility agents to significantly accelerate GE (when optimal test methods were used) and to produce significant improvements in UGI symptoms.