BACKGROUND: Prucalopride is currently licensed in the European Union for treatment of chronic constipation for women (approved in 2009) and men (approved in 2015) in whom laxatives have been ineffective. Given prior safety experience with other 5-HT4 agonists, a multidatabase study is planned to evaluate cardiovascular safety of prucalopride.
OBJECTIVES: To describe baseline characteristics of patients who were newly prescribed prucalopride versus polyethylene glycol (PEG), focusing on cardiovascular risk factors.
METHODS: Adult new users of prucalopride were identified in the Clinical Practice Research Datalink (CPRD) and The Health Improvement Network {THIN), from April 2010 through May 2014, by following a common protocol in both databases. Aggregate data were pooled after resolution of duplicate practices. Prucalopride patients were matched to PEG patients (5: 1) by age, sex, calendar year of index prescription, and practice (CPRD only). Patient demographics, baseline comorbidities and comedications, and prescribing patterns for each cohort are described.
RESULTS: The pooled data included 1,037 new users of prucalopride and 5,867 new users of PEG; 95% of patients were female and 66% were aged 18-54 years. History of hospitalization for cardiovascular disease was similar for each cohort (acute myocardial infarction: 0.5% vs 0.6%; stroke: 0.3% vs 0.7%; ischemic heart disease: 3.3% vs 2.8%), but the prucalopride cohort had a higher proportion of patients taking antihypertensive medications (48.0% vs 42.4%). More patients using prucalopride had gastrointestinal-related visits than PEG patients: 55.8% vs 11.1% had greater than or equal to 2 outpatient visits for constipation, and 15.4% vs 6.3% had greater than or equal to 2 outpatient visits for irritable bowel syndrome.
CONCLUSIONS: Differences in prior history of gastrointestinal disease may be due to selective channeling, because prucalopride is indicated for patients in whom laxatives have been ineffective. Despite the similarity in measured cardiovascular risk factors observed between the cohorts, accounting for channeling may control for important unmeasured confounding and will be important in the analysis of cardiovascular outcomes.
