Molecular and clinical pharmacology

Lamotrigine has complex pharmacological properties. It inhibits voltage-sensitive sodium channels, resulting in neuronal stabilisation, which probably underpins its anticonvulsant properties; however, the mechanisms underlying its mood-stabilising and antidepressant effects are uncertain. It blocks L-, N- and P-type calcium channels, weakly inhibits 5-HT₃ 5-hydroxytryptamine receptors, inhibits the release of glutamate in ventral striatum limbic areas and reduces gamma-aminobutyric acid A receptor-mediated neurotransmission, but it has minimal effects on adrenergic, dopaminergic (D₁ and D₂), cholinergic, histaminic (H₁), 5-HT₂ and N-methyl-d-aspartate glutamate receptors. It exerts antioxidant and neuroprotective effects and weakly inhibits the enzyme dihydrofolate reductase (which is important in folate metabolism), although this effect is not associated with changes in serum or erythrocyte folate concentrations.

The metabolism of lamotrigine shows first-order kinetics with a 29-h half-life. It undergoes rapid and complete absorption, with 98% absolute bioavailability that is unaffected by food. It is inactivated by hepatic glucuronidation, the major metabolite being inactive. Optimal therapeutic levels in patients with epilepsy have not been established, nor is there any established level–response relationship in patients with affective disorders. Its pharmacokinetics are influenced by genetic polymorphisms, weight and some concomitantly prescribed medicines. Clearance is increased in pregnant women and during concomitant treatment with oestrogen-containing oral contraceptive pills. ^{Reference Methaneethorn and Leelakanok1}

Lamotrigine treatment always requires careful monitoring. The best-known adverse effect is development of skin rashes: between 5% and 10% of patients will develop some form of rash. Severe cutaneous adverse reactions are relatively uncommon (the incidence has been variably estimated as 1 in 1000 to 1 in 2500) but include drug rash with eosinophilia and systemic symptoms, Stevens–Johnson syndrome (with involvement of <10% of skin) and toxic epidermal necrolysis (more than 30% of skin), usually occurring via a type IV hypersensitivity mechanism. Severe cutaneous adverse reactions with lamotrigine should be regarded as potentially life-threatening, necessitating admission to an intensive care or burns unit: the potential fatality rate with Stevens–Johnson syndrome is around 5–10%. Multiple risk factors for dermatological reactions have been identified; these include female sex, younger age, inappropriately rapid dosage titration and concomitant use of valproate-containing medicines.

Lamotrigine in patients with affective disorders or other conditions

Lamotrigine can be beneficial as part of overall clinical management in patients with bipolar disorder. A systematic review including 20 randomised controlled trials (RCTs) and 20 cohort studies (with a total of 12 307 lamotrigine-treated patients) examined its efficacy in acute treatment of bipolar depressive episodes, acute treatment of manic or hypomanic episodes, and maintenance treatment (‘prophylaxis’). ^{Reference Haenen, Kamperman, Prodan, Nolen, Boks and Wesseloo2} Studies were stratified by control (active medication or placebo) and strategy (monotherapy or add-on), and 24 studies were included in the meta-analysis (15 RCTs involving 950 lamotrigine-treated patients and nine cohort studies involving 812 lamotrigine-treated patients). For acute treatment of bipolar depressive episodes, there were insufficient data to establish superiority versus placebo for monotherapy; for add-on, there was superiority versus placebo but no difference versus comparators. For acute treatment of mania or hypomania, a single RCT found no difference compared with lithium as monotherapy. In maintenance treatment, monotherapy was superior to placebo but no different to lithium, and add-on therapy resulted in a longer median time to relapse (10 months) than placebo (3.5 months). Furthermore, five high-quality register-based studies found that lamotrigine was associated with lower hospital admission rates than other commonly used treatments. The authors emphasised its efficacy as an add-on treatment for bipolar depressive episodes and as maintenance treatment for prevention of relapse and recurrence. ^{Reference Haenen, Kamperman, Prodan, Nolen, Boks and Wesseloo2}

Lamotrigine treatment may be beneficial in management of comorbid substance use disorders in some bipolar patients. There are insufficient data to establish whether monotherapy is efficacious in acute treatment of patients with unipolar depression, and, although some have asserted that lamotrigine has broadly similar efficacy in unipolar and bipolar depressive episodes, the evidence for its efficacy as an add-on treatment is inconclusive. However, it is superior to placebo in reducing positive and negative symptoms as an augmentation therapy in patients with schizophrenia being treated with clozapine.

Lamotrigine in the period of reproductive potential

Awareness of the damaging effects of valproate-containing medicines in pregnancy and the resulting regulatory restrictions on their use in clinical practice have led to increased attention being paid to the use of lamotrigine in women of reproductive potential. The ability of lamotrigine to prevent depressive episodes in women of reproductive age with bipolar disorder can be of enormous benefit. It is recognised that women with bipolar disorder are at increased risk of relapse in the postpartum period, particularly if they have not taken medication during pregnancy. A meta-analysis of four randomised placebo-controlled maintenance studies – in which lamotrigine was administered at a dosage of 150–400 mg per day for up to 76 weeks to women aged 18–45 years – found that the median time to intervention for a mood episode (the primary outcome) was 323 days for lamotrigine, compared with 127 days for placebo (for all episodes). Much of the benefit of lamotrigine was because of its superior efficacy in preventing depressive episodes, as it was not superior to placebo in prevention of manic episodes. Lamotrigine was also generally well tolerated. During double-blind treatment, there was a similar incidence of adverse events and adverse events leading to withdrawal, the most common adverse effects being headache (16%, compared with 14% with placebo) and nausea (13%, compared with 11% with placebo). ^{Reference Vieta, Ghorpade, Biswas, Sarkar, Phansalkar and Cooper3}

When considering possible prescription of any medication to a patient of reproductive potential, it is important to be aware of potential teratogenicity. An early retrospective study suggested that first trimester monotherapy (in 791 women) might increase the risk of oral clefts (lip or palate), although a population-based case–control study of infants with congenital malformations (226 806 pregnancies) found the excess risk of oral clefts to be less than 1 in 550 exposed babies, with no significant increase in risk compared with controls. ^{Reference Dolk, Wang, Loane, Morris, Garne and Addor4} Systematic reviews have found that lamotrigine exposure during pregnancy was not associated with increased risks of either birth complications or congenital malformations, compared with disease-matched controls and non-exposed controls. Prospective cohort studies have provided further reassurance: an international (47 countries) study including 10 121 pregnancies in individuals exposed to anticonvulsant monotherapy found a lower rate of major congenital malformations (MCMs) with lamotrigine (3.1%, 95% CI 2.5–3.7) than with most other anti-epileptics (whereas there were dose-dependent effects for carbamazepine, phenobarbital and valproate ^{Reference Battino, Tomson, Bonizzoni, Craig, Perucca and Sabers5} ); and a population-based cohort study (in five Nordic countries) of 8339 pregnancies in individuals exposed to first-trimester lamotrigine monotherapy found no evidence of a higher risk of MCMs compared with pregnancies in those not exposed to anti-seizure medication (ASM), but there were higher risks of MCMs with valproate and topiramate compared with lamotrigine. ^{Reference Cohen, Alvestad, Cesta, Bjørk, Leinonen and Nørgaard6}

There is no well-characterised lamotrigine withdrawal syndrome in adults, and neonatal withdrawal symptoms appear to be uncommon. There have been highly variable estimated of the correlation between maternal serum level and breast milk level (range: 0.6–90.3%). Potential effects of lamotrigine on breastfed infants, described in case reports, include breathing difficulties and somnolence; however, clinical advice is that lamotrigine treatment is not a reason to discontinue breastfeeding.

Neurodevelopment and mental disorders in childhood and adolescence

Increasing awareness of the hazards of valproate-containing medicines when administered in utero has understandably focused attention on possible adverse effects of lamotrigine, including effects on neurodevelopment and on the emergence of mental health problems in children and adolescents. Most data are derived from studies of the offspring of mothers with epilepsy and may not necessarily be fully applicable to considerations of developmental conditions among children or mothers prescribed lamotrigine for affective disorders.

A meta-analysis (including a total of 18 studies) of neurodevelopmental outcomes after epilepsy-indicated monotherapy in pregnancy found no influence of lamotrigine on overall neurodevelopmental outcomes (five studies; odds ratio 0.84; 95% CI: 0.66–1.06), language delay or disorders (seven studies; odds ratio 1.16; 95% CI: 0.67–2.00), diagnosis or risk of autism spectrum disorder (five studies; odds ratio 0.97; 95% CI: 0.61–1.53) or the diagnosis or risk of attention-deficit hyperactivity disorder (ADHD) (four studies; odds ratio 1.14; 95% CI: 0.75–1.72). However, lamotrigine exposure was associated with psychodevelopmental delay or disorders (four studies; odds ratio 2.68; 95% CI: 1.29–5.56) and with cognitive development delay in children less than 3 years old (two studies; odds ratio 3.42; 95% CI: 1.17–9.98). ^{Reference Peron, Picot, Jurek, Nourredine, Ripoche and Ajiji7}

A cohort study of patients with epilepsy involving over 4 million pregnancies evaluated those exposed to anti-epileptic drugs between pregnancy week 19 and delivery, with linkage to health records of children before 8 years of age. The cumulative incidence of autism spectrum disorder (after exclusion of children with chromosomal abnormalities) was 1.9% in the full population of children with no exposure to medication (4.2 million children), 4.2% in those with mothers with epilepsy but no exposure to medication (8815 children), 6.2% in those exposed to topiramate (1030 children), 10.5% in those with epilepsy exposed to valproate (800 children), and 4.1% in those with epilepsy exposed to lamotrigine (4205 children). The risk of autism spectrum disorder was higher among children exposed to anticonvulsant medication, although this risk was attenuated for lamotrigine (and topiramate, another anticonvulsant) after adjustment for confounders. ^{Reference Hernández-Díaz, Straub, Bateman, Zhu, Mogun and Wisner8}

A prospective population-based cohort study (in five Nordic countries, involving over 4.5 million singleton children) included 38 661 children of mothers with epilepsy and examined prenatal exposure to ASM at any time between 30 days before the last menstrual period and birth to ascertain the cumulative risks of diagnosis of any child or adolescent psychiatric disorder at 18 years of age (from patient registers). These were estimated as follows: ASM-unexposed children, 31.3% (95% CI: 28.9–33.6); children exposed to any ASM, 30.8% (95% CI: 29.2–32.3) (adjusted hazard ratio (aHR) 1.17; 95% CI: 1.09–1.25); valproate-exposed children, 42.1% (aHR 1.80; 95% CI: 1.60–2.03) (mainly neurodevelopmental disorders); and lamotrigine-exposed children, 24.1% (aHR 0.91; 95% CI: 0.82–1.02; i.e. no significant increase). However, there were associations between topiramate and ADHD, and between levetiracetam (another anticonvulsant) and both ADHD and anxiety disorders. ^{Reference Dreier, Bjørk, Alvestad, Gissler, Igland and Leinonen9}

Guidance on lamotrigine prescribing

When making treatment decisions, it makes sense to rely on the highest-level evidence (i.e. meta-analyses and prospective population studies) rather than findings from isolated case reports or limited case series. The possible importance of effects of lamotrigine on dihydrofolate reductase and hence on folate metabolism are worth remembering, as folate deficiencies are associated with an increased risk of congenital malformations. National Health Service advice is for high-dose folic acid to be taken throughout pregnancy (5 mg rather than 400 µg). Similarly, clinicians should be aware that lamotrigine pharmacokinetics are altered during pregnancy, and levels should be monitored and the dose adjusted.

High-level evidence indicates that lamotrigine is effective in the prevention of depression in patients with bipolar disorder. When prescribing lamotrigine, great caution is needed when titrating the dosage to reduce the likelihood of adverse dermatological reactions. Lamotrigine does not appear to be associated with congenital malformations. High-level evidence has identified no increased risk of autism spectrum disorder, ADHD, or other child and adolescent mental health disorders; however, the possible effects of in utero lamotrigine exposure on neurodevelopment need further investigation. It should be used cautiously in breastfeeding mothers.

All conclusions relating to safety of medicines during the perinatal period are necessarily tentative. Treatment decisions for managing mental health problems in patients of reproductive potential are complex, owing to the need to consider both parent and child and to balance the potential risks of untreated illness and of exposure to medication, while recognising that such treatment decisions should not be avoided in clinical practice.