Dr. Shah is a PGY-2 Psychiatry Resident, Department of Psychiatry and Behavioral Sciences, University of Louisville School of Medicine, Louisville, Kentucky. Dr. Elsayed is a Post-doctoral Research Fellow, Department of Psychiatry and Behavioral Sciences, University of Louisville School of Medicine, Louisville, Kentucky. Dr. El-Mallakh is Professor and Director, Mood Disorders Research Program, Department of Psychiatry and Behavioral Sciences, University of Louisville School of Medicine, Louisville, Kentucky.
Disclosures Dr. Shah and Dr. Elsayed report no financial relationships with any companies whose products are mentioned in this article, or with manufacturers of competing products. Dr. El-Mallakh is a speaker for Axsome, Intra-Cellular Therapies, Janssen, Lundbeck, Myriad, Noven, Otsuka, and Teva, and has received research grants/funding from Sunovion.
Bipolar disorder (BD) is a recurrent, life-long psychiatric illness affecting nearly 2% of the world population1,2 that is characterized by episodes of mania and depression interspersed among periods of relative mood stability.3 The illness causes an enormous health burden, which makes understanding its pathophysiology and treatment patterns a substantial priority.4 In the 1950s, lithium was found to be effective for treating acute manic episodes and preventing relapse in BD.5 Since then, valproate and carbamazepine also have been FDA-approved for treating mania.6,7 Antipsychotics have also shown evidence of efficacy in BD treatment,8,9 particularly for use in acute settings for more rapid effect or for a limited duration,10 which has led some to refer to them as “mood stabilizers.”11
In this article, we describe changes in trends of prescribing medications to treat BD, the role of ion dysregulation in the disorder, and how a better understanding of this dysregulation might impact the choice of treatment.
Changes in pharmacotherapy for bipolar disorder
From 1997 through 2016, the use of lithium for BD decreased from >30% of patients to 17.6% (with a nadir of 13.9% from 2009 to 2012).12 Over the same period, the use of nonlithium mood stabilizers decreased from 30.4% to approximately 4.8%, while second-generation antipsychotic (SGAs) use increased from 12.4% to 50.4%.12 Distressingly, antidepressant use increased from approximately 47% to 56.8%, and antidepressant use without concomitant mood stabilizers increased from 38% to 40.8%, although the rate of antidepressants without either a mood stabilizer or an antipsychotic remained relatively stable (14.9% to 16.8%).12 In randomized trials, when added to mood stabilizers, antidepressants have consistently failed to separate from placebo,13-15 but they can destabilize the illness, resulting in increases in mania, depression, and subsyndromal mixed symptoms.16-18
It is easy to understand clinicians’ attempts to address their patients’ distress due to depressive symptoms that do not resolve with mood stabilizers.19,20 Similarly, the increased use of antipsychotics is driven by evidence that antipsychotics are effective for treating bipolar depression and preventing the recurrence of manic and (for some antipsychotics) depressive episodes.21,22 However, long-term antipsychotic use causes brain volume change in patients with schizophrenia23 or major depressive disorder24 and in nonhuman primates25,26; metabolic abnormalities27-31; and cardiovascular adverse effects.32 Antipsychotics are believed to be associated with withdrawal psychosis.33,34 In the head-to-head Clinical Health Outcomes Initiative in Comparative Effectiveness for Bipolar Disorder (Bipolar CHOICE) study, quetiapine was as effective as lithium but associated with more adverse effects.35 Importantly, the estimated disability-adjusted life years of patients with BD increased by 54.4% from 6.02 million in 1990 to 9.29 million in 2017, which is greater than the increase in the incidence of BD (47.74%) over the same time.36 This means that despite the dramatic increase in treatment options for people with BD, functional outcomes have declined.
One major difference between antipsychotics and mood stabilizers is that antipsychotics do not alter the underlying abnormal pathology of BD.37 An ideal pharmacologic intervention is one that corrects a known pathophysiologic anomaly of the condition being treated. There are no demonstrated abnormalities in the dopamine or serotonin systems in individuals with BD, but long-term use of antipsychotics may create dopaminergic alterations.33 One of the most reproducible biomarkers associated with manic and bipolar depressed mood states is increased intracellular sodium38,39 and reduced ability to correct a sodium challenge.40-42 By normalizing intracellular sodium levels, lithium and the mood-stabilizing anticonvulsants uniquely and specifically counter known physiologic abnormalities in patients with BD.37,43
The role of ion dysregulation
The pathophysiology of BD remains elusive. A multitude of lines of evidence link BD to abnormal neuroimaging findings,22,44,45 oxidative stress,46 inflammation,47 and mitochondrial disease,48 but there is still no unifying understanding of these findings. Ion dysregulation appears to be central to understanding and treating BD.38,39
Despite extensive genetic studies, no genes have been identified that mediate >5% of the risk for BD. Nonetheless, 74% of all genes identified as mediating risk for BD code for proteins essential for the regulation of ion transport and membrane potential.49 The 2 genes that contribute the greatest risk are CACNA1C and ANK3, which code for a calcium channel and a cytoskeletal protein, respectively.50ANK3 codes for ankyrin G, which plays a role in proper coupling of the voltage-gated sodium channels to the cytoskeleton.51 An additional risk gene, TRANK1, contains multiple ankyrin-like repeat domains, which suggests some shared functions with ANK3.52 More importantly, the most reproducible pathophysiologic findings in BD are dysregulation of sodium, potassium, hydrogen, and calcium transport, with consequent alteration of depolarization potential, neuronal excitability, and calcium-mediated processes.38,39,53-56 For example, increased sodium and calcium within cells have been observed in both mania and bipolar depression, and these levels normalize during euthymia. All medications that are effective for treating BD may reduce intracellular sodium or calcium; traditional mood stabilizers do so directly by inhibiting voltage-sensitive sodium channels in an activity-dependent manner or displacing intracellular sodium,43,57 whereas antipsychotics do so indirectly by increasing sodium pump activity through inhibition of second messengers of the dopamine D2 family of receptors.37