The painkillers the patients were taking included naproxen (Aleve), ibuprofen (Advil, Motrin), diclofenac (Voltaren, Cataflam), celecoxib (Celebrex), etoricoxib (Arcoxia), rofecoxib (Vioxx), lumiracoxib (Prexige) or placebo.
Overall, the number of heart events among patients taking NSAIDs was low, the researchers found. In 29 trials, 554 heart attacks occurred. In 26 trials, 377 strokes were reported, and in 28 trials, 676 people died.
Compared with patients taking placebo, those taking rofecoxib and lumiracoxib had twice the risk of heart attack, and those taking ibuprofen had more than three times the risk of stroke. The highest risks for cardiac death were associated with etoricoxib and diclofenac, where the risk was around four times greater than for placebo, the researchers found.
Naproxen appeared to be the least harmful medication, they noted.
"Our study provides the best available evidence on the safety of this class of drugs," the researchers wrote. "Although uncertainty remains, little evidence exists to suggest that any of the investigated drugs are safe in cardiovascular terms. Cardiovascular risk needs to be taken into account when prescribing any non-steroidal anti-inflammatory drug," they concluded.
Commenting on the study, Dr. Eric J. Topol, director of the Scripps Translational Science Institute and one of the experts who uncovered the risks associated with Vioxx, said a meta-analysis can leave many questions unanswered but is of value nonetheless.
"Pooling large data sets like this winds up with ambiguity as it homogenizes differences in patient population characteristics, dose of drugs, how endpoints were ascertained and when, etc.," he said. "I am not sure if the conclusions reflect or agree with other meta-analysis results," Topol noted.
"Despite the limitations, this study has made many excellent contributions
All rights reserved