A Cochrane review included 5 studies with a total of 537 subjects. Lamotrigine was given as an adjuvant only to antipsychotic agents. Two studies required people to be treatment resistant to past drug trials. The other three studies required persisting positive symptoms without mention of resistance to treatment. Most people in the studies were moderately to severely ill.
Equal proportions of people allocated to lamotrigine or placebo had no global response (RR 1.06, CI 0.73 to 1.54, n=208, 1 study). There was no significant difference between groups in the proportions of people whose mental state did not improve (RR 1.26, CI 0.81 to 1.97, n=297, 3 studies). In lamotrigine groups there was a significant reduction in the PANSS total scores (WMD -16.88, CI -8.57 to -25.18, p=0.0001, n=67, 2 studies), positive symptom sub-scale scores (WMD -5.10, CI -8.86 to -1.34, n=65, 2 studies) and negative symptom sub-scale scores (WMD -5.25, CI -7.07 to -3.43, n=67, 2 studies). Most cognitive measures showed no differences. The proportion of participants leaving studies was about 25% at 12 weeks (RR 0.96, CI 0.71 to 1.29, n=537, 5 studies). The lamotrigine group did experience the outcome of any adverse effects significantly more frequent than the placebo group (RR 1.19, CI 1.02 to 1.38, CI 5 to 90 NNH 10, n=429, 2 studies). Nausea was found to be more common (9%) in the lamotrigine group compared to placebo (RR 2.26, CI 1.05 to 4.88, n=465, 3 studies).
Comment: The quality of evidence is downgraded by limitations in study quality (e.g., inadequate follow up, poor reporting, omission of many outcomes important for clinical practice) and by indirectness (selected population, i.e. hospital patients with well-defined schizophrenia or schizoaffective disorder without co-morbidities). There were no usable data on service outcomes, general functioning, behaviour, engagement with services, satisfaction with treatment or economic outcomes.