Publisher Health

This post was originally published on this site

LOS ANGELES — The FDA’s warning about arrhythmia risk with lamotrigine (Lamictal) based on in vitro data wasn’t borne out in observational cohorts of people with epilepsy, although the risk did appear significant in people with bipolar disease.

The difference in findings presented at the American Epilepsy Society (AES) annual meeting appeared to be due to comparator drugs used between the two conditions.

In a study looking at Medicare and U.K. CALIBER datasets for new-onset epilepsy or seizure disorder cases, initial treatment with lamotrigine wasn’t associated with any higher risk of serious ventricular arrhythmias or death compared with levetiracetam as a “presumably cardiac ‘inert'” control, reported Samuel Terman, MD, of the University of Michigan Department of Neurology in Ann Arbor.

In a larger study of Medicare and commercial claims data, lamotrigine didn’t hold any greater ventricular tachycardia (VT) risk compared with other sodium channel blockers or other drugs for the same seizure indications.

However, lamotrigine used for bipolar disorder in that study did have higher VT risk than commonly used antipsychotics (HR 1.243, P=0.02), with an absolute risk difference of about 0.09% at 1 year, reported Przemyslaw Radwanski, PharmD, PhD, of the Ohio State University College of Pharmacy in Columbus.

Neither study showed a significantly greater risk for lamotrigine in patients with comorbid heart disease.

In 2021, the FDA warned about potential increased arrhythmia risk in people with cardiac disease who are taking lamotrigine based on reports of abnormal ECG findings and cases of chest pain, loss of consciousness, and cardiac arrest. Suggesting it could be a class effect, the agency required in vitro safety studies on all the sodium channel blockers, although that data has yet to be made publicly available.

The new observational data add to the AES treatment committee’s prior conclusion that “there’s nothing to this” based on an extensive review, said AES board of directors member David Vossler, MD, of the University of Washington School of Medicine in Seattle.

“When you look at this compared to other antiepileptic drugs, we’re not seeing a clear difference,” he summarized. However, he added, “We will know more when the manufacturers complete their additional in vitro studies.”

“There’s no need to take patients with epilepsy off lamotrigine,” Vossler concluded. “But if they have other significant heart conditions, including structural heart disease or known cardiac arrhythmias, getting an EKG test and perhaps even consulting with a cardiologist would be prudent. But for someone that’s otherwise healthy, I don’t believe you need an EKG or even need to see a cardiologist. And from both of these studies, there’s no evidence of increased risk.”

For the bipolar patients, “the question is maybe lithium, quetiapine [Seroquel], and risperidone [Risperdal] are less problematic than lamotrigine,” he noted. “That’s an observation that needs to be explained.”

With lamotrigine so commonly used for bipolar disorder, even a small increase in risk, if real, would mean serious risk for a meaningful number of patients over the course of a year in the United States, Radwanski argued.

Radwanski’s study used the Merative MarketScan Commercial Claims and Medicare Supplemental Database to find 2.6 million adults with at least one prescription filled for lamotrigine or controls comprising other medications for the same indication. For partial seizure, 10,275 persons on lamotrigine were matched with the same number on carbamazepine, levetiracetam, oxcarbazepine, or eslicarbazepine. For generalized tonic-clonic seizure, 5,855 pairs were matched for lamotrigine versus controls on valproate, levetiracetam, carbamazepine, or zonisamide (Zonegran). For bipolar disorder, 148,645 matched pairs were on lamotrigine versus comparators of lithium, quetiapine, valproate, or risperidone. Individuals with preexisting ventricular or atrial arrhythmia during the 6 months prior to enrollment were excluded.

Onset of VT while covered with one of these medications for at least 7 days during 365 days of follow-up was not significantly more common with lamotrigine in partial seizure patients (HR 1.477, 95% CI 0.886-2.462, P=0.14) or generalized tonic-clonic seizure patients (HR 1.427, 95% CI 0.622-3.271, P=0.40).

Although the statistical significance is limited, “the positive association is ubiquitous across epileptic conditions,” Radwanski concluded, suggesting that the smaller cohorts in those indications might have been underpowered to find small differences.

Terman’s study included a total of 53,652 patients in a Medicare 20% random sample or the U.K. CALIBER study from primary and secondary care practices. Enrollment was limited to adults with epilepsy or seizure diagnoses who were starting lamotrigine or levetiracetam without any antiseizure medication use in the prior year.

After adjusting for a wide range of demographics, comorbidities, and co-medications, 2-year cumulative incidence of VT or ventricular fibrillation in the Medicare group was 0.6 percentage points (95% CI -1.2 to 0.0) less common with lamotrigine than levetiracetam in the intent-to-treat analysis and 0.1 percentage points (95% CI -0.5% to 0.7%) more common with lamotrigine in the per-protocol analysis. In CALIBER, those figures were -0.1 percentage points (95% CI -0.3 to 0.1) and -0.1% (95% CI -0.4 to 0.2), respectively.

Adjusted 2-year cumulative incidence of death was also lower in the lamotrigine groups in each study, by an absolute -7.6% (95% CI -9.4 to -4.8) in the Medicare group and -3.1% (95% CI -2.2 to -4.0) in the CALIBER group.

“Our data support continued use of lamotrigine amongst patients starting a new [antiseizure medication], even amongst those at higher cardiac risk or using concomitant sodium channel blockers,” Terman’s group concluded.

Both studies were limited by their comparators and by the observational nature of the data that could not determine causality or eliminate the potential for residual confounding.