Secondary Logo

The Six Antiepileptic Drugs with the Highest Risk for Adverse Skin Reactions

Lehmann, Christine

doi: 10.1097/01.NT.0000552574.19444.05
For Your Patients
Back to Top | Article Outline




Six antiepileptic drugs had the highest estimated risk of Stevens-Johnson syndrome and toxic epidermal necrolysis: lamotrigine, carbamazepine, phenytoin, zonisamide, rufinamide, and clorazepate, according to an analysis of reports in the FDA Adverse Event Reporting System. The findings underscore the need for neurologists to warn patients of the potential risks for severe skin complications with these drugs.

Antiepileptic drugs (AEDs) as a class have been associated with a significantly higher risk of Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN), both rare and potentially fatal skin reactions, than other classes of drugs. But some have a higher risk than others, according to a new analysis of data from the US Food and Drug Administration Adverse Event Reporting System (FAERS).

The study, published online November 5 in Epilepsia, identified six AEDs with the highest estimated risk of SJS/TEN: lamotrigine, carbamazepine, phenytoin, zonisamide, rufinamide, and clorazepate.



“As prescribing clinicians become aware of the risks these drugs pose to patients, they can educate them about the signs of SJS/TEN skin reactions including fever, rash, and blister-like lesions in the trunk and mucous membranes,” Aisling R. Caffrey, PhD, MS, senior author of the study published November 5 in Epilepsia, told Neurology Today. “This can lead to early identification and reduce the number and severity of these adverse events,” said Dr. Caffrey, associate professor of health outcomes at the College of Pharmacy at the University of Rhode Island.

Greater than 90 percent of SJS/TEN reactions associated with AEDs occur within the first two months of treatment initiation, the study authors said. In severe cases SJS/TEN present one of the greatest risk factors for mortality they noted. Approximately 5 percent of patients die from SJS, while about 30 percent of patients die from TEN, according to the paper.

Epileptologists who were not involved with the study pointed out that lamotrigine and carbamazepine have black-box warnings about potential complications. “But it is not the standard of care for most neurologists to warn patients taking these other relatively high-risk drugs about possible SJS/TEN reactions,” said Orrin Devinsky, MD, FAAN, professor of neurology and director of NYU Langone's Comprehensive Epilepsy Center. “Although these drugs are prescribed less frequently, we still need to warn people of the risk of SJS/TEN.”

Back to Top | Article Outline


Dr. Caffrey and her colleagues at several institutions in Rhode Island reviewed FAERS adverse event reports for SJS/TEN from July 2014 through December 2017. They reviewed AEDs with an FDA indication for epilepsy or seizures; reports using brand names were categorized under their generic names. They excluded follow up reports, duplicate reports, and reports missing event date, sex, and age.

The investigators calculated the risk estimates of SJS/TEN for the entire AED class, and individual medications compared with non-AEDs, using proportional reporting ratios, reporting odds ratios, and corresponding 95 percent confidence intervals.



The researchers found that AEDs had the highest number of SJS/TEN reports (198) and nearly nine times the estimated risk of SJS/TEN compared with other medication classes.

“We expected that AEDs would have a higher risk than other classes of drugs, but we were surprised by the magnitude of the difference,” said the lead study author Eric P. Borrelli, PharmD, and PhD student in the College of Pharmacy at the University of Rhode Island.

The AEDs with the highest number of total SJS/TEN reported adverse events were lamotrigine (106), carbamazepine (22), and phenytoin (14), which confirms previous research findings. [See “By the Numbers” for the specific risk estimates of the six highest-risk AEDs.]

Significant associations were also observed for valproic acid, eslicarbazepine, oxcarbazepine, clonazepam, and levetiracetam, according to the paper.

Dr. Caffrey noted that the wide variation in risk between AED medications and the confidence intervals was related to the number of reports submitted. For example, the risk estimates for rufinamide and clorazepate had wide confidence intervals due to only a single SJS/TEN reaction and a low incidence of total adverse reactions.

“However, the lower end of these wide intervals was still similar to the overall risk of AEDs as a class,” said Dr. Borrelli. “Even with only one report, these findings are still meaningful and statistically significant.”

Using FAERS had several limitations, the researchers acknowledged: Because the total number of patients using these medications is unknown, they could not calculate incidence and prevalence, and they suspected underreporting of adverse events due to voluntary reporting to FAERS and a bias for reporting more adverse events in the first two years after a drug comes on the market. Also, drugs with frequent side effects can increase the number of overall adverse effects, and therefore decrease reporting odds ratios and proportional rating ratios.



“We suspect there are significant associations between SJS/TEN and the AEDs we identified,” said Dr. Caffrey, “but we can't infer these associations without incidence data. This study gives us a starting point of having an estimate for potential risk. Future studies are needed to evaluate incidence and risk estimates so we know the actual risk of SJS/TEN facing people treated with these drugs.”

Back to Top | Article Outline


“The significance of this paper is that the authors examined a large data set of adverse event reports and confirmed that lamotrigine has a relatively high risk of SJS/TEN,” said Dr. Devinsky. “They also found that three other drugs have an even higher relative risk: zonisamide, rufinamide, and clorazepate, and that two other medications, phenytoin and carbamazepine, have half the risk.”

The data from FAERS provide a good estimate of the overall US population, Dr. Devinsky noted, but lack the power or sufficient sample size to enable a data breakdown by ethnicity. “It would be interesting to know, for example, if the relative risk for zonisamide is higher in one population than another.”

Although the researchers did not analyze the data by age, neurologists who regularly prescribe AEDs for children said the analysis would factor into their treatment decisions. “Based on my clinical experience, I would have reason to be concerned about the antiepileptic drugs with higher relative risks found in this study because all the risks seen in adults are seen in children,” said Douglas R. Nordli Jr., MD, professor of pediatrics and chief of pediatric neurology at the University of Chicago Medical Center.

When child neurologists/epileptologists prescribe any AED, they tell parents there is a risk of an idiosyncratic reaction, said Dr. Nordli. “The problem is that a rash can start off looking innocent and turn into something more serious. We always tell patients that if you see a new rash, you must hold the medicine and alert us so we can examine it and investigate further.”

Because previous studies have indicated that SJS/TEN reactions may be associated with the starting dose and rate of titration for some AEDs, Dr. Nordli generally likes to start with the lowest efficacious dose and then titrate up slowly. A similar approach to Dr. Nordli's has also been used with lamotrigine, which the study authors recommend.



“We also need to make sure that the medication is justified by verifying that the patient has epilepsy, and that the risks of the medication outweigh the risk of recurring seizures,” said Dr. Nordli.

In this issue, this story and another on page 10 discuss the FDA Adverse Event Reporting System. Both point to its potential for providing transparency about possible drug-related complications. But they both point to limitations, as well, in interpreting outcomes.

Back to Top | Article Outline


The six AEDs with the highest risk estimates (>20) for SJS/TEN were:

  • zonisamide (ROR 70.2, 95% CI 33.1-148.7; PRR 68.7, 95% CI 32.9-143.5)
  • rufinamide (ROR 60.0, 95% CI 8.3-433.5; PRR 58.9, 95% CI 8.4-411.5)
  • clorazepate (ROR 56.0, 95% CI 7.8-404.1; PRR 55.1, 95% CI 7.8-385.0)
  • lamotrigine (ROR 53.0, 95% CI 43.2-64.9; PRR 52.2, 95% CI 42.7-63.7)
  • phenytoin (ROR 26.3, 95% CI 15.5-44.7; PRR 26.1, 95% CI 15.4-44.2)
  • carbamazepine (ROR 24.5, 95% CI 16.0-37.5; PRR 24.3, 95% CI 16.0-37.1)
Back to Top | Article Outline


Dr. Borrelli reported no disclosures. Dr. Nordli is a consultant to Eisai Pharmaceuticals. Dr. Devinsky has served on the medical or scientific advisory board and received compensation from Privateer Holdings/Tilray, Egg Rock/Papa & Barkley, Empatica, Receptor Life Sciences, RettCo, Pairnomix, Tevard Biosciences, and Engage PEO. He has also received consulting fees from GW Pharmaceuticals, and Zogenix.

Back to Top | Article Outline


•. Borrelli EP, Lee EY, Descoteaux AM, et al Stevens Johnson syndrome and toxic epidermal necrolysis with antiepileptic drugs: An analysis of the US Food and Drug Administration Adverse Event Reporting System Epilepsia 2018; Epub 2018 Nov 5.
    © 2018 American Academy of Neurology