Effect of fenfluramine-derivative diet pills on cardiac valves: A meta-analysis of observational studies

Molly Sachdev, William C. Miller, Thomas Ryan, James G. Jollis

Research output: Contribution to journalArticlepeer-review

101 Scopus citations


Background: Fenfluramine-derivative diet pills were withdrawn from the market in 1997 because of an association with valvular regurgitation, but subsequent estimates of the prevalence of this condition have varied widely. We systematically reviewed evidence regarding the prevalence of valvular disease after fenfluramine exposure. Methods: We searched multiple databases with multiple search terms. Conference proceedings from 1997 onward were searched by index. Authors of eligible studies were contacted to identify unpublished works. Selection criteria were liberally determined. Ten of the identified 11 articles met these criteria. Reviewers assessed the studies' methodologic quality by use of a standard form to evaluate selection, attrition, performance, and detection bias. The studies were analyzed in 2 groups on the basis of length of exposure (<90 days or >90 days). The Mantel-Haenszel method was used to summarize data. Quantitative and qualitative tests for heterogeneity were performed. Tests for publication bias were also done. Results: Tests for heterogeneity were nonsignificant after removing 1 outlier trial. The pooled prevalence of valvular regurgitation meeting Food and Drug Administration criteria (at least mild aortic regurgitation or at least moderate mitral regurgitation) among patients treated for >90 days was 12.0% compared with 5.9% for the unexposed group (prevalence odds ratio 2.2, 95% Cl 1.7-2.7). The combined analyses also identified a small but statistically significant increase in mitral regurgitation not previously identified by individual studies (exposed 3.5%, unexposed 1.8%, prevalence odds ratio 1.6, 95% Cl 1.05-2.3). Among patients exposed for <90 days, a trend toward more regurgitation was not statistically significant by either combined Food and Drug Administration criteria (exposed 6.8%, unexposed 5.8%, prevalence odds ratio 1.4, 95% Cl 0.8-2.4) or by individual valve. Conclusions: These data indicate that fenfluramine-associated valvular regurgitation is less common than initially reported, but still present in 1 of 8 patients treated for >90 days.

Original languageEnglish
Pages (from-to)1065-1073
Number of pages9
JournalAmerican heart journal
Issue number6
StatePublished - Dec 1 2002


Dive into the research topics of 'Effect of fenfluramine-derivative diet pills on cardiac valves: A meta-analysis of observational studies'. Together they form a unique fingerprint.

Cite this