Lamotrigine: the legal and safety record

The boxed warning, verbatim, with its denominator

Lamotrigine carries a boxed warning — FDA’s most prominent — for serious skin rashes. Here is what it actually says [2]:

“LAMICTAL can cause serious rashes requiring hospitalization and discontinuation of treatment. The incidence of these rashes, which have included Stevens-Johnson syndrome, is approximately 0.3% to 0.8% in pediatric patients (aged 2 to 17 years) and 0.08% to 0.3% in adults receiving LAMICTAL.”— LAMICTAL prescribing information, Boxed Warning (rev. 10/2025) [2]

Read both halves of that. The rash is real: hospitalizations, Stevens-Johnson syndrome, and — in worldwide postmarketing experience — rare cases of toxic epidermal necrolysis and rash-related death, too few for FDA to estimate a rate. And the denominator is real: for an adult, roughly 1 to 3 in 1,000 by the label’s own figures — and the label states plainly that “the rate of serious rash is greater in pediatric patients than in adults” [2]. A page that quotes “SJS” without those numbers is running a scare; a page that rounds them to zero is selling reassurance the documents do not contain.

The label also carries its own mitigation, and it is not fine print. Nearly all life-threatening rashes occurred within 2 to 8 weeks of starting, and the named risk factors beyond age are concomitant valproate, exceeding the recommended starting dose, exceeding the recommended dose escalation, and the HLA-B*1502 allele [2]. That is why lamotrigine starts absurdly low and climbs over five weeks — the slow titration is the rash-risk management, which is also why the label sends anyone who has been off the drug more than five half-lives back to the initial schedule rather than their old dose [2]. Benign rashes happen too, and the label concedes it cannot reliably tell you which is which in advance — hence: any rash, prescriber, today.

The 2012 GSK resolution: where Lamictal actually appears

In July 2012, GlaxoSmithKline agreed to plead guilty and pay $3 billion — the largest health care fraud settlement in U.S. history at the time [1]. Most secondary sources flatten what that resolution was. Kept in DOJ’s own structure:

• The criminal plea did not involve Lamictal. GSK pleaded guilty to a three-count criminal information: two counts of introducing misbranded drugs — Paxil and Wellbutrin — into interstate commerce, and one count of failing to report safety data about Avandia to FDA, paying a $956,814,400 criminal fine plus $43,185,600 in forfeiture [1].
• Lamictal is in the civil allegations. The $2 billion civil settlement resolved alleged conduct including “promoting the drugs Paxil, Wellbutrin, Advair, Lamictal and Zofran for off-label, non-covered uses and paying kickbacks to physicians to prescribe those drugs.” Specifically, DOJ wrote that the settlement “resolves allegations that GSK promoted Lamictal, an anti-epileptic medication, for off-label, non-covered psychiatric uses, neuropathic pain and pain management.” GSK paid $1.043 billion for the off-label and kickback allegations as a group — there is no per-drug figure [1].
• The release’s own disclaimer applies. “Except to the extent that GSK has agreed to plead guilty to the three-count criminal information, the claims settled by these agreements are allegations only, and there has been no determination of liability” [1]. A civil settlement of allegations is not an admission, and this page will not write it as one.

One more precision, in the other direction: unlike gabapentin, whose psychiatric use has no approval at all, lamotrigine does have a psychiatric approval — maintenance treatment of bipolar I disorder to delay mood episodes [2]. A bipolar-maintenance prescription today is on-label. The civil allegations were about promotion beyond the label as it stood; they say nothing against the approved use, and the approved use does not launder the alleged conduct. Both facts, held apart.

Two FDA safety communications since

2018 — hemophagocytic lymphohistiocytosis (HLH). On 25 April 2018, FDA warned that lamotrigine can cause this rare, life-threatening immune over-activation and required a new warning in the prescribing information. The scale, in FDA’s own accounting: eight confirmed or suspected cases worldwide from the drug’s 1994 approval through September 2017 — all hospitalized, one death — with symptoms within 8 to 24 days of starting, and with FDA noting unreported cases likely exist [4]. It typically announces itself as a persistent fever above 101°F, often with rash — which folds into the same practical rule this page already gave you: fever or rash in the early weeks is an immediate prescriber conversation, and FDA’s advice in the same document was not to stop the medicine on your own [4]. The warning now lives in section 5.2 of the label [2].

2021 — heart rhythm. On 31 March 2021, FDA reported that laboratory testing it had required showed lamotrigine, at therapeutically relevant concentrations, can increase the risk of serious arrhythmias, which can be life-threatening, in patients with clinically important structural or functional heart disease — heart failure, valvular disease, congenital heart disease, conduction system disease, ventricular arrhythmias, channelopathies such as Brugada syndrome, significant ischemic disease, or multiple coronary risk factors [3]. The label records the mechanism as Class IB antiarrhythmic activity at therapeutically relevant concentrations [2]. Read the boundary as precisely as the finding: the concern is for hearts in those categories, and the documents assert nothing beyond them. FDA also ordered the same evaluation for ten other sodium channel blockers — this was a class question, not a lamotrigine pile-on [3]. And its patient advice repeated the rule this drug keeps generating: do not stop “without first talking to your prescriber because stopping lamotrigine can lead to uncontrolled seizures, or new or worsening mental health problems” [3]. The warning is now section 5.4 of the label [2].

Housekeeping, because it keeps happening: the 2018 communication’s fda.gov URL returned HTTP 200 as recently as February 2026 and 404s today, so we cite the archived copy plus the label section FDA required — the same URL-rot pattern we documented on the sertraline record [4]. The 2021 communication is still live at fda.gov [3].

Recalls: thirteen, including two Class I — read them for what they are

Scope first: the openFDA enforcement endpoint is reliable from roughly 2012 onward, so everything here is a statement about the covered window, not the drug’s whole history [5]. Within it, lamotrigine has thirteen recall records — and unlike several drugs in this series, two are Class I, FDA’s most serious category. Both are terminated, and both are worth reading exactly [5]:

• 2019, Taro: 100 mg lamotrigine tablets cross-contaminated with enalapril maleate, a blood-pressure drug — a wrong-substance-in-the-bottle failure, recalled nationwide.
• 2016, Impax: 3,074 boxes labeled 200 mg that contained 100 mg orally disintegrating tablets. For most drugs a silent half-dose is an inconvenience; for an anti-seizure drug whose label warns against abrupt dose drops, it is presumably why this drew Class I.

The remaining eleven are Class II and III — manufacturing and quality problems [5]. None of the thirteen is a finding about the molecule as designed; all of them are findings about factories and packaging lines. Both statements are true, and a page that gives you only one of them is steering. The label, the FAERS reports and the caveats that go with them are on our lamotrigine page.

Common questions

Was Lamictal part of GSK’s $3 billion settlement in 2012?

Yes — but in the civil half, and the distinction matters. GSK pleaded guilty to a three-count criminal information covering the misbranding of Paxil and Wellbutrin and the failure to report safety data about Avandia; Lamictal is not in the plea. Lamictal appears in the civil settlement, which resolved allegations that GSK promoted it for off-label, non-covered psychiatric uses, neuropathic pain and pain management, and paid kickbacks to physicians. The release itself states that, except for the guilty plea, the claims settled are allegations only, with no determination of liability. Pages that write "GSK pleaded guilty over Lamictal" are wrong in a checkable way.

Does lamotrigine have a black box warning?

Yes. The boxed warning covers serious skin rashes requiring hospitalization, including Stevens-Johnson syndrome, with rash-related deaths reported. The label’s own incidence figures: approximately 0.3% to 0.8% in pediatric patients aged 2 to 17, and 0.08% to 0.3% in adults. Nearly all life-threatening rashes occurred within 2 to 8 weeks of starting. Benign rashes are also caused by lamotrigine, and the label says it is not possible to reliably predict which rashes will prove serious — which is why any rash goes to the prescriber immediately.

What actually raises the rash risk?

The label names four factors beyond age: taking valproate at the same time, exceeding the recommended starting dose, exceeding the recommended dose escalation, and carrying the HLA-B*1502 allele. The first three are why the slow titration schedule exists — it is the label’s own mitigation, not caution theater. The label also notes serious rash has occurred without any of these factors, so titration lowers the risk; it does not zero it.

What should I do if I get a rash on lamotrigine?

Contact your prescriber immediately — same day, not next appointment. The label says lamotrigine should ordinarily be discontinued at the first sign of rash unless the rash is clearly not drug related, and that is the prescriber’s call to make quickly, because discontinuing may not stop a rash from progressing. Do not silently stop on your own and tell no one: abrupt discontinuation carries seizure risk, and the label allows faster-than-taper withdrawal exactly when safety requires it — which is a decision to make with the person who can weigh both.

Can I just stop taking lamotrigine?

Not abruptly, and not alone. The label states lamotrigine should not be abruptly discontinued — in epilepsy because seizure frequency can increase, and even in the bipolar trials two patients had seizures shortly after abrupt withdrawal. Unless safety concerns require faster, the label calls for a taper over at least two weeks, roughly 50% dose reduction per week. One more label rule worth knowing: if you have been off it for more than five half-lives, restarting means going back to the initial titration schedule — restarting at your old dose resets the rash risk the slow start exists to manage.

Has lamotrigine had serious recalls?

Yes — and unlike several drugs in this series, that includes FDA’s most serious category. The openFDA enforcement database returns thirteen lamotrigine recall records (the endpoint is reliable only from roughly 2012 onward, so this is the covered window, not all history): two Class I, seven Class II, four Class III, all now terminated. The two Class I events: 100 mg tablets cross-contaminated with enalapril maleate (Taro, 2019), and boxes labeled 200 mg that contained 100 mg tablets (Impax, 2016). Both are manufacturing and packaging failures, not findings about the molecule — though for this drug specifically, a silent dose drop is not harmless, which is presumably why the mislabel drew Class I.

Sources

Last verified 2026-09-08. Corrections change this date.