most thoroughly studied of the novel anticonvulsant agents, is FDA-indicated to
delay relapse into mood episodes in bipolar disorder type I. It is not FDA-indicated for type II bipolar
disorder (as many seem to believe). It
has been proven ineffective for every other aspect of bipolar disorder (acute
depression, acute mania/mixed episodes, rapid-cycling).
Its biochemical effect involves inhibiting the presynaptic release of excitatory amino acid neurotransmitters such as glutamate and aspartate; this effect may or may not explain its psychotropic properties. Lamotrigine is metabolized by the liver and is moderately (over 50%) protein-bound. Its half-life is 25 hours, which allows for simple once daily dosing.
Since it can be somewhat stimulating, and its effect is on glutamate, which is not normally part of the biology of sleep, PL editors have come to the conclusion that lamotrigine is best given in the morning, not in the evening, as is commonly the case. This will prevent impairment of the normal stages of sleep.
Divalproex competes with lamotrigine for hepatic glucuronidation, inhibiting lamotrigine’s metabolism and increasing its half-life to 60 hours, while carbamazepine, phenytoin, and primidone enhance its metabolism, decreasing the half-life to 15 hours. When used with valproate, the dose of lamotrigine should be halved due to the markedly longer half-life.
Dosing for bipolar disorder is discussed in detail below, but efficacy has been shown in the 50-200 mg/d range, though the maximal dose can go up to 500 mg/d. It should never be dosed more quickly than 25 mg/d per week due to serious rash risk (see below).
The current PL recommendation, and a common conservative practice, is to increase lamotrigine 25 mg every two weeks in the average patient (not every week as had been common practice by us previously). Thus, achieving a target dose of 100–200 mg per day may take up to 2-3 months. This slow titration is usually not as much of a problem in the outpatient treatment of depression and prophylaxis of mood episodes.
PL recommends that clinicians do not dose lamotrigine above 200 mg/d, mainly because there is no evidence of more benefit at higher doses (400 mg/d was similar to 200 mg/d in the only randomized data on this matter in bipolar illness). Further, the risk of rash is highest as long as one is increasing the dose, so the higher one goes, the longer the period of risk. Lastly, side effects of anxiety and cognitive impairment increase with doses above 200 mg/d.
Most side effects with lamotrigine are rare and mild. These include headache, tremor, somnolence, and dizziness; in clinical trials, only 2% of bipolar disorder patients discontinued lamotrigine due to adverse events. However, about 10-20% of patients develop a common but non-serious rash. The FDA requires discontinuation of lamotrigine should rash occur due to the risk of progression to the rare, but potentially fatal, Stevens-Johnson syndrome.
Stevens-Johnson syndrome is a serious rash in which patients can experience symptoms equivalent to a severe burn. A large proportion of patients die due to bacterial superinfection; those who survive can be disfigured. While obviously severe, Stevens-Johnson syndrome is rare, and seems most associated with rapidity of titration of lamotrigine. In the early 1990s, when lamotrigine was first studied in large-scale clinical trials, Stevens Johnson syndrome was observed in 1/1000 adult patients and 4/1000 children and adolescents. Consequently, lamotrigine is not allowed for use below age 15 for non-epileptic indications. The above rates were observed with relatively rapid titration of the medication.
Combined therapy with valproate and lamotrigine increases the non-serious rash rate, and potentially the serious risk of rash.
Besides speed of dosing titration, the most important risk factor for rash, in the PL view, is other drug allergies, particularly antibiotics. Data on file with the manufacturer indicate that the risk of rash with lamotrigine increases 4-5 fold in persons who have antibiotic allergies. My forensic experience with cases of SJS with lamotrigine has also been that antibiotic allergies tend to be present in persons who develop SJS with lamotrigine. Other risk factgors, suggesting immunologic reactivity, are asthma, autoimmune disorders, hayfever, allergic rhinitis, and food allergies. In such patients, PL recommends either avoiding lamotrigine, or automatically institute a 12.5 mg every 2 week titration. Clinicians who are afraid of using this agent due to its potentially serious medical risks should feel much more comfortable using it with a conservative titration such as this one outlined here. Patients’ fears can also be allayed by pointing out the marked lowering of risk of serious rash with a slow titration. Also, the serious rash risk appears to be highest in the first months of treatment. Once patients are taking a stable dose of lamotrigine for long-term prophylaxis, they no longer appear to possess significant risk of serious rash.
Since lamotrigine, like lithium, has a great deal of evidence supporting its efficacy, it is important that clinicians learn to manage its side effects and allay their patients’ fears so that this effective medication can be used in patients who would benefit from it. PL suggests summarizing the situation to my patients in this manner: This is a very effective medication, with very few short or long-term side effects, except for a potentially serious rash risk, which can be lessened by a slow titration. The non-serious rash variant occurs in 10-20% of patients. Otherwise, this medication is very tolerable.
On the other hand, clinicians should avoid simply prescribing lamotrigine without much thought, given to the small but real risk of fatality. Patients also need to be cautioned explicitly and carefully, to never increase lamotrigine dosing on their own. Sometimes patients are used to other drugs (like amphetamines) which have immediate dose-related effects; they need to be educated that messing with lamotrigine dosing is literally a matter of life and death.
When lamotrigine was first indicated for bipolar disorder, many clinicians, including PL board members, were impressed by its benefits. After some time, though, it became clear that negative studies on many medications are not published, or are slowly released, so as to create an overly positive image of a drug’s efficacy. This appears also to have been the case with lamotrigine. For a short time, under legal injunction, its manufacturer was forced to post all these negative data on its website (www.gsk.com); these data were then removed, but before their removal, the PL editor downloaded the data and summarized it in a review paper. There the reader can find the following evidence:
Almost all of those studies were either not published or were only published in partial form as summary results combined with other studies. A few studies were published as positive, based on secondary analyses, despite negative primary outcomes. What this means is that the drug was the same as placebo in its main analysis, though sometimes later analyses suggested some benefits in a particular subgroup. The latter subgroup benefits were not replicated in follow-up studies, however (e.g., possible benefit in a type II rapid cycling subgroup in one study was not found in another study).
Overall, although PL thinks lamotrigine is a useful drug and very helpful to many patients, the glossing over of negative data, combined with the effects of marketing its real positive benefits, has resulted in clinicians forming an overly favorable impression of the extent of efficacy of this drug.
To be clear: PL is not saying it does not work at all; it has maintenance preventive benefits. But the data strongly indicate that it does not have acute mood benefits and that it does not improve rapid cycling bipolar disorder (which is not surprising: nothing does, except antidepressant discontinuation). Some would argue otherwise, suggesting for instance that its acute mood benefits are hard to demonstrate in 2 month studies of acute depression due to its slow titration; that may be the case, but it remains the case that it has not been proven effective for acute depression (or mania or rapid cycling).
That is the bad news. Now let’s examine the good news that this agent seems to have preventive benefits, more so for depression than for mania prevention. While this is the case based on its two maintenance studies, there is often the misconception that those same studies prove that lamotrigine is more effective than lithium in prevention of depression (and vice versa, lithium better prevented mania than lamotrigine). This may or may not be the case. Keep in mind that those studies were “enriched”: they only included patients who had initially responded to lamotrigine before they entered the randomized maintenance study. Thus, it was not a fair comparison with lithium (to do so, half of the enrolled patients would have needed to be chosen based on initially responding to lithium before the study). Thus, one can say that, in lamotrigine responders for acute mood symptoms, lamotrigine is more effective than lithium in prevention of depression. However, one cannot say, in general, that lamotrigine is more effective than lithium in prevention of depression. In contrast, the fact that lithium was more effective in mania prevention, despite the initial preselection of patients as lamotrigine responders, does demonstrate that lithium is clearly more effective in prevention of mania.
Thus, PL thinks it is misleading to call lithium a mood stabilizer “from above” (more antimanic than antidepressant) and lamotrigine a mood stabilizer “from below” (more antidepressant than antimanic). In fact, lithium is likely as effective in prevention of depression - in "fair" non-enriched trials - as lamotrigine. Since lithium is also clearly more effective than lamotrigine in prevention of mania, the PL conclusion is that lithium is overall a more effective agent than lamotrigine.
In sum, lamotrigine is a useful drug, but as with most new drugs, the hype may have been larger than the reality. It should be used where it is effective, not where it is not.
Legal claims regarding SJS with lamotrigine have begun to occur, leading PL to offer some basic forensic advice to clinicians. The usual issues are inadequate warning regarding the risk of SJS, too rapid dosing, or inappropriate indication of lamotrigine. First, with any patient given lamotrigine, the clinician should document: “warned regarding SJS.” Further detail can also be provided, such as warning regarding the risk of disfigurement, not only death, and spelling out of the dosing titration. Second, patients should clearly be warned to never increase the dose on their own, and clinicians should not dose the drug faster than 25 mg/week (and preferably 25 mg every 2 weeks). PL is aware that the PDR dosing instructions and the sample pack differ with these recommendations (the FDA labeling recommends 25 mg/d for 2 weeks, then 50 mg/d for 2 weeks, then 100 mg/d for one month, then 200 mg/d). This dosing titration, developed for epilepsy, is unnecessarily fast for the long-term prevention of mood episodes in bipolar disorder. Jumping from 100 to 200 mg/d overnight seems much too rapid. One will never faulted for dosing this drug too slow, since it has no acute efficacy. Third, this drug has only one proven indication: the prevention of mood episodes in bipolar I disorder. It is not indicated or proven in type II bipolar disorder, or acute major depression in that setting, for instance. Clinicians should not turn to lamotrigine as their first-line drug for type II bipolar depression, since it has not been shown effective in that setting, and since it has real medical risks. If SJS should occur, that kind of indication would increase a clinician’s legal risk. If lamotrigine is used, discussion of other standard mood stabilizers should occur and be documented and the rationale for use of lamotrigine should be given.