Although they are often considered as a class, anticonvulsants in fact have varied mechanisms and effects, says Andres M. Kanner, MD, a professor of neurological sciences and psychiatry at Rush Medical College, in Chicago. Valproate and lamotrigine tend to stabilize mood, for instance, while pregabalin (Lyrica)—which was not found to increase suicide risk in the study—has antianxiety effects, he says.
“These drugs definitely are being used more and more by psychiatrists to treat patients with a variety of psychiatric disorders, but primarily mood and anxiety disorders,” says Dr. Kanner.
Because having a mental illness such as bipolar disorder or depression is, by far, the biggest predictor of suicide, Dr. Patorno and her colleagues factored the patient’s diagnosis (and a range of other health variables) into their analysis. The results “argue that the risk was derived from the specific drug that the patient was taking and not their underlying conditions,” Dr. Patorno says.
If she were taking one of the riskier medications herself, Dr. Patorno adds, she “would probably re-evaluate the therapy” in consultation with her physician.
But the study has some important limitations, Dr. Kanner says. For one thing, he explains, topiramate—which is most commonly prescribed for migraine—is known to worsen some psychiatric conditions and many doctors won’t prescribe it to patients with a personal or family history of mental illness. Therefore, the comparison group may have already been at lower risk of suicidal thoughts or attempts, he says.
Dr. Kanner also notes that health conditions such as depression and epilepsy, which sometimes overlap, can combine to affect suicide risk in ways that may have eluded the researchers’ models. “You’re dealing with a very complex interaction,” he says. “To just attribute [the increased risk] to antiepileptic medication doesn’t explain the facts properly.”
Pinpointing whether anticonvulsants boost suicide risk in and of themselves will likely require a study that follows people over time, rather than one that looks back at previous treatment, as the current investigation did, Dr. Kanner adds.
According to Dr. Bazil, another limitation of the study is that Dr. Patorno and her team used standardized insurance-billing codes to determine a patient’s diagnosis. These codes can be imprecise, Dr. Bazil says, and doctors who suspect that a patient has (or is at risk for) depression may decide to prescribe one anticonvulsant versus another for that reason, without necessarily listing the billing code for depression in the patient’s file.
Dr. Kanner emphasizes that patients who are currently taking the drugs for any reason should not stop taking them without consulting their doctor, and that patients should also tell their doctor if they have a history of mental illness or if it runs in their family. “Very often clinicians will not inquire about these issues, and that’s where things get lost,” he says.
People taking the drugs should tell their doctor if they begin to feel depressed, or have thoughts of suicide, says Dr. Bazil, but patients should not be overly alarmed by the study.
“The main thing is to be cautious about the risks versus the benefits,” he says. “In my mind, you have to keep it in perspective. There may be an increased risk of suicidal ideation, [but] most of these patients have tremendous benefits from the drugs.”
The study was funded by the Harvard School of Public Health and HealthCore, a research subsidiary of WellPoint, a health benefits company that serves the Blue Cross Blue Shield network.
