Review of: Chen PH et al, Br J Psychiatry 2018;1–6. doi:10.1192/bjp.2018.203

Study Type: Case-crossover study

People living with bipolar disorder already have a decreased life expectancy of 10+ years compared to the general population (Crump C et al, JAMA Psychiatry 2013;70:931–939). Most of these years of lost life have been attributed to cardiovascular and cerebrovascular causes. Bipolar disorder itself affects the vascular system, as do some of the medications that treat it. A new study parses out which of the mood stabilizers are the worst offenders when it comes to the risk of stroke.

The authors examined medical records of the Taiwan National Health Insurance Research Database to identify people between ages 15 and 65 who had been diagnosed with bipolar disorder and then later had their first stroke. Among the 19,433 cases of bipolar disorder, 609 had a stroke during the study period and were included in the analysis. Both ischemic and hemorrhagic strokes were included. This was a case-crossover design, in which the study subjects served as their own controls. In this case, the medications they took in the 2 weeks before the stroke were compared to four other 2-week periods in the year preceding the stroke. The particular medications examined were carbamazepine, valproic acid, lithium, and lamotrigine.

Only two of those mood stabilizers increased the risk of stroke. The worst offender was carbamazepine, which increased the risk of any kind of stroke (adjusted risk ratio [ARR] 1.68; p = 0.018). Valproic acid only increased the risk of hemorrhagic stroke (ARR 1.76; p = 0.022). Lithium and lamotrigine had no significant effect on the occurrence of stroke in this analysis.

One weakness of the study was that it only examined the acute effects of treatment, and not the risks of long-term exposure to mood stabilizers. The study also did not examine severity of illness and what burden that may have on stroke risk; presumably, less severe patients may have been prescribed lithium or lamotrigine. A strength was that the study controlled for concomitant use of other medications that might influence the risk of stroke, such as antipsychotics, cardiovascular drugs, and diabetic medications.

TCPR’S TAKE
This study should make us think a little more when choosing a mood stabilizer. The risk of stroke with carbamazepine has previously been reported in epilepsy (Chuang YC et al, Epilepsia 2012;53:120–128). Although atypical antipsychotics were not included in this study, their well-known metabolic risks would give us pause. If people already have risk factors for stroke, consider treatment with lithium or lamotrigine as first-line agents. Although not generally thought of as heart-friendly, lithium has numerous cardioprotective effects and lowers the risk of myocardial infarctions (Chen PH et al, Prog N Biol Psych 2019;88:208–214).