Adverse Effects of Cyclooxygenase 2 Inhibitors on Renal and Arrhythmia Events: Meta-Analysis of Randomized Trials
Jingjing Zhang,* Eric L. Ding,†‡ and Yiqing Song†
(JAMA, 296:1619-1632, 2006)
*Renal Division and †Division of Preventive Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, MA; and ‡Departments of Epidemiology and Nutrition, Harvard School of Public Health, Boston, MA.
Recent evidence indicates that the cyclooxygenase 2 (COX-2) inhibitors may have undesirable cardiovascular effects. The nephrotoxic potential of these drugs has been recognized, but results of various studies are inconsistent; thus, the renal safety profile of selective COX-2 inhibitors is not yet firmly established. This comprehensive class-wide meta-analysis examined randomized, double-blinded trials of the COX-2 inhibitors (rofecoxib, celecoxib, valdecoxib, parecoxib, etoricoxib, and lumiracoxib) to assess the consistency and robustness of evidence for adverse effects on renal and arrhythmia events.
After a systematic search of EMBASE and MEDLINE databases, along with the Cochrane Controlled Trials Register, the CRISP database of the National Institutes of Health, and relevant references of retrieved articles, the Food and Drug Administration reports, and contacts with investigators, 114 informative reports were included in the analysis. Information on year of publication, characteristics of participants, duration of the trial, drug, control, dose, and events was extracted according to a standardized method.
The 114 reports included 127 trial populations (40rofecoxib, 37 celecoxib, 29 valdecoxib/parecoxib, 15 etoricoxib, 6 lumiracoxib trials), with a total of 116,094 participants. From a total of 6394 composite renal events, 2670 were peripheral edema, 3489 hypertension, and 235 renal dysfunction. A total of 286 arrhythmia events occurred. Use of the different COX-2 inhibitors indicated substantial heterogeneity, and thus, no class effect. Patients taking rofecoxib had increased risks of composite renal events compared with control patients; rofecoxib increased the risk of peripheral edema, hypertension, and renal dysfunction. Other COX-2 inhibitors, overall, were not associated with increased renal events. Celecoxib seemed to have a lower risk for both renal dysfunction and hypertension compared with controls. For arrhythmia events, only patients taking rofecoxib were at increased risk compared with controls. Celecoxib showed no effect on arrhythmia; valdecoxib plus parecoxib had a marginally lower arrhythmia risk. The limited number of trials could not provide conclusive results for etoricoxib and lumiracoxib. Under meta-regression stratified analyses, increased risk of renal events was consistently shown for rofecoxib regardless of comparison to placebo, nonselective nonsteroidal anti-inflammatory drugs (NSAIDs), or mixed comparison of other controls. Both higher dose and longer trial duration (>25 mg/d, ≥6 months) further increased the risk of renal events. Stratified results for the other COX-2 inhibitors did not reveal notable effect modifications. Time-cumulative analyses showed that, by the end of 2000, rofecoxib was associated with overall adverse renal events as well as the specific events of hypertension and peripheral edema. The evidence for adverse effects on renal dysfunction was apparent by 2005. The adverse effects of rofecoxib on arrhythmia were first apparent by the end of 2004.
Although a class effect was not evident for renal or arrhythmia events for all COX-2 inhibitors, rofecoxib increased the risk of these types of adverse effects. Further safety monitoring of current and emerging treatments is needed. Additionally, an active surveillance protocol may be beneficial.