Abstract
Bipolar disorder is an episodic, highly impairing mood disorder that is estimated to have a prevalence of 2–3% in the general population and is one of the leading causes of years lived with a disability.(1) Lithium is the gold standard for the treatment of bipolar disorder, and although it is a simple element, its effects on the brain are very complex.(2) Lithium’s potential neurotrophic and neuroprotective effects raise the intriguing possibility that it can potentially ameliorate abnormalities in brain structure and thus alter the disease trajectory.
17.1 Introduction
Bipolar disorder is an episodic, highly impairing mood disorder that is estimated to have a prevalence of 2–3% in the general population and is one of the leading causes of years lived with a disability.(1) Lithium is the gold standard for the treatment of bipolar disorder, and although it is a simple element, its effects on the brain are very complex.(2) Lithium’s potential neurotrophic and neuroprotective effects raise the intriguing possibility that it can potentially ameliorate abnormalities in brain structure and thus alter the disease trajectory.
Clinically, over 60% of patients experience long-term reduction of mood episodes and improved quality of life with lithium maintenance treatment; 20–30% are excellent responders to lithium monotherapy with full remission for at least five years.(3) The use of lithium in bipolar disorder has been surpassed, however, by the heavily marketed second-generation antipsychotic medications, especially for its acute treatment, although in adults their efficacy is similar.(4) Because treatments that are effective for acute episodes tend to be continued for maintenance therapy, fewer individuals are likely to be prescribed lithium for maintenance treatment. Lithium may exert its strongest beneficial effects including antisuicide properties during maintenance treatment.(5) Therefore, any evidence that lithium has beneficial effects on the brain and on clinical outcome could reassert lithium’s importance in the treatment armamentarium for bipolar disorder.
This chapter will critically review the results of studies that have used magnetic resonance (MR) imaging to examine the effects of lithium on brain structure in individuals with bipolar disorder. The focus will be on studies that were designed specifically to assess these effects using either cross-sectional or longitudinal designs. Thus, this selective review does not include studies that considered medication effects in ancillary or secondary analyses, given several literature reviews on this topic have already been conducted.(6) We also discuss the possibility that changes in brain volume assessed using MR imaging may be confounded by the properties of signal changes associated with properties of water osmosis.(7) We then discuss studies that have evaluated the effects of lithium on the brain in postmortem work to gain additional insight into lithium’s purported mechanism(s) of action.
Studies that correlate clinical response to lithium with neuroimaging findings such as volume changes suggest a potential mechanism of therapeutic action for lithium. We therefore discuss the role of brain-derived neurotrophic factor (BDNF) in brain development and specifically its role in hippocampal growth. A potential neurotrophic component of lithium’s therapeutic action would be supported if significant associations exist between changes in serum BDNF levels, hippocampus volume, and clinical response. Thus, evidence for a relationship among hippocampal structural changes associated with lithium treatment coupled with BDNF activity are discussed. Finally, we provide evidence that the effects of lithium may be most robust within the dentate gyrus of the hippocampus and discuss implications for the neurobiology of bipolar disorder. We conclude with directions for future research.
17.2 Magnetic Resonance Imaging Studies
Lithium treatment is most consistently associated with increases in hippocampus volume in multiple human cross-sectional MR imaging studies, and in the few longitudinal studies that have been published (see Table 17.1). Evidence for regionally specific findings has been inconsistent, however. Bearden and colleagues (8) reported that (mainly right) hippocampal volume in lithium-treated patients with bipolar disorder was 10.3% greater compared to non-medicated healthy controls and 13.9% greater compared to bipolar patients not treated with lithium. In that study, there were no significant associations between hippocampal volume and lithium dosage, blood level, or treatment duration. Similarly, Foland and colleagues (9) reported that patients with bipolar disorder treated with lithium demonstrated greater volume compared to bipolar patients not taking lithium including the (% in parentheses) left amygdala (3.72%), right amygdala (1.72%), left hippocampus (3.45%), and right hippocampus (2.73%). There were no significant associations between brain volume and illness duration, and prior number of manic or depressive episodes. Along these lines, Hajek et al. (10) reported that patients with bipolar disorder treated with lithium had significantly larger hippocampal volume compared to patients with bipolar disorder not treated with lithium and healthy controls. In addition, volumes were comparable among patients treated with lithium regardless of the number of prior mood episodes. In one of the few pediatric studies conducted to date, Baykara et al. (11) reported that right hippocampal volume was enlarged in patients with bipolar disorder treated with lithium, relative to untreated controls.
Table 17.1 Studies examining the effects of lithium on brain structure in bipolar disorder
| Study and type | Participants | MRI methods/design | Findings |
|---|---|---|---|
|
|
|
|
| Healthy controls vs. Bipolar I or II patients in any mood state. Patients treated with lithium for ≥ 2 weeks or unmedicated (no psychotropic medications for ≥ 2 weeks and at least 1 month off lithium)
|
| Hippocampal volume (mostly on right side) in lithium-treated bipolar patients was as follows:
No significant relationships seen between hippocampal volume and lithium dosage, lithium blood level, or duration of lithium treatment. |
|
|
| Lithium (+) patients had
Lithium (−) patients had similar gray matter volumes as controls White matter volumes did not significantly differ between lithium (+), lithium (−), and control subjects |
|
|
| Global scaling factors were not significantly different among groups Lithium (+) patients showed greater total amygdala and hippocampal volumes compared to lithium (−) patients, especially on the left side:
No significant correlations were found between volumes and illness duration, prior number of manic episodes or prior number of depressive episodes |
|
|
| Lithium (+) treated patients had
|
| No healthy controls. Patients with bipolar I, meeting criteria for remission for 6 months
| MRI scans were conducted using a 1.5 T GE system. Each MR image was normalized and segmented into gray matter, white matter, and CSF using unified segmentation in SPM5 using voxel-based morphometry | Lithium (+) treated patients had
Gray matter and total intracranial volumes were comparable between lithium (+) patients and those in other treatment groups White matter volumes were lowest in valproate patients, highest in carbamazepine patients CSF volumes lowest in antipsychotic patients, highest in valproate patients Valproate (+) patients had less white matter and higher CSF fractional volumes compared to all other groups |
| Healthy controls vs. Bipolar I or II patients. Lithium (+) group must have had adequate Li treatment for ≥2 years with Li levels 0.5 to 1.2 mmol/L on every blood test taken at least twice a year Lithium (−) group must have <3 months of lithium exposure, with no lithium at least 24 months prior to scan
|
| Lithium (+) patients had
Among lithium (+) patients, volumes were similar regardless of the number of mood episodes while on lithium treatment Left hippocampus was larger than right in all groups |
| Healthy controls vs. bipolar spectrum patients
|
| Lithium (+) patients had
Lithium (−) patients had
|
| Healthy controls vs. euthymic bipolar I patients
| MRI scans acquired on a 1.5 T Phillips scanner and segmented to generate volumetric measures of cortical and subcortical brain areas, ventricles, and global brain, using Freesurfer | Lithium (+) patients compared to untreated patients had
Compared to controls, lithium (+) patients had
|
| Healthy controls vs. bipolar patients randomly assigned to either lithium or valproate (VPA) treatment for 16 weeks
|
| Lithium (+) patients had
|
| 13 Healthy volunteers with no DSM-IV diagnosis were given lithium over 4 weeks, titrated to lithium level of at least 0.6 mEq/L. No controls. MRI scans performed at baseline and at the end of 4 weeks Lithium doses 600–1,500 mg/day, mean dose 1,281 mg/day Mean age = 25.9, SD = 10 Mean lithium levels:
|
| At the end of 4 weeks of lithium administration:
|
|
|
| Compared to lithium (−) treated patients, lithium (+) patients had
|
|
|
| Lithium (+) treated patients had
|
| Healthy controls vs. lithium (+) bipolar patients vs. untreated bipolar patients off all psychotropic meds for at least 2 weeks (usually due to noncompliance)
|
| Lithium (+) patients had
|
| Healthy controls vs. bipolar I patients in various states (euthymic, depressed, manic, hypomanic) Participants were required to be drug-free for at least 2 weeks prior to study
|
| Lithium (+) who were clinical responders had |
A few cross-sectional studies investigated the effects of lithium on hippocampal volume in patients with bipolar disorder who had either very short- or long-term treatment. Hajek et al. (12) studied 17 patients with bipolar disorder who had at least 2 years of regularly monitored lithium treatment, 12 bipolar patients with less than 3 months of total lifetime lithium treatment and no lithium treatment prior to 2 years before an MR imaging scan and 11 healthy controls. Voxel-based morphometry indicated that the non-lithium treatment group had smaller left hippocampal volume compared to controls with a trend for lower volumes than the lithium-treated group who did not differ from controls, consistent with meta-analysis (13). At the other end of the spectrum, Yucel et al. (14) compared hippocampal volume among three groups of patients with bipolar disorder including those treated with lithium between one and eight weeks, patients who were unmedicated at the time of scan, and patients treated with either valproic acid or lamotrigine. Results indicated a bilateral increase in hippocampal volume that was evident in the head of the hippocampus among patients treated with lithium, even after a brief period of treatment. These findings suggest that the effects of lithium on the brain may be evident even within weeks of treatment initiation.
Several studies examined the relationship between lithium treatment and hippocampal subfield volumes using MR imaging. In one of the largest cross-sectional studies to date Giakoumatos et al. (15) reported that 51 patients with psychotic bipolar disorder treated with lithium had thicker cortical volume and greater hippocampal subfield volumes compared to 135 patients with psychotic bipolar disorder not being treated with lithium and 342 healthy controls. Patients being treated with lithium had comparable hippocampal subregion volumes as healthy controls. In a study by Hartberg et al. (16), investigation of hippocampal subfield volumes revealed smaller total hippocampal volume, including the right CA1 and subiculum subfields, and bilateral CA2/3, CA4/DG subfields among the patients not being treated with lithium compared to healthy controls. Interestingly, in that study there was a significant positive association between lithium treatment duration and larger amygdala volume.
Other cross-sectional studies provide evidence that the effects of lithium may be more widespread in the brain. For example, Bearden et al. (17) reported that patients with bipolar disorder treated with lithium (n = 20) had greater volume in the frontal (9.8%), temporal (6.6%), and parietal (10.3%) lobes compared to eight patients with bipolar disorder of whom only one was taking psychotropic medication (i.e., citalopram). In addition, patients with bipolar disorder not treated with lithium had gray matter volumes comparable to healthy controls (n = 28). Also, in that study no differences in white matter volume were observed between patients treated with lithium and patients not receiving lithium. Germana et al. (18) reported that patients with bipolar disorder treated with lithium had more gray matter in the right subgenual anterior cingulate, left postcentral gyrus, hippocampus/amygdala complex, and left insula. In a study by Lopez-Jaramillo et al. (19), patients treated with lithium had significantly larger bilateral amygdala and thalamic volume (but no differences in hippocampal volume) compared to patients with bipolar disorder not treated with lithium. Moreover, compared to controls, patients treated with lithium had significantly larger bilateral amygdala, bilateral thalamus, and left hippocampus volume. Lastly, Sassi et al. (20) reported that patients with bipolar disorder treated with lithium had greater total brain gray matter volume compared to patients not currently receiving psychotropic treatment and healthy controls.
Results from longitudinal neuroimaging studies in patients with bipolar disorder treated with lithium provide stronger support compared to cross-sectional studies for the hypothesis that enhancement of neuroplasticity is a component of lithium’s therapeutic mechanism. Such studies have demonstrated brain changes in both the gray and white matter. In a single-blind randomized controlled clinical trial, Berk and colleagues (21) reported that lithium was more effective than quetiapine in slowing progression of white matter volume loss within the left internal capsule between baseline and twelve months without associated gray matter changes. In a study by Yucel et al. (22), patients were rescanned twice after baseline: approximately two years and then again four years following initiation of lithium maintenance therapy. They found increases of 4–5% in hippocampal volume after two years and these increases were maintained at the four-year scan and associated with improvements in cognitive functioning.
Several controlled trials investigated the effects of lithium on brain imaging measures prior to and then following controlled treatment. In a longitudinal study of twenty-four patients with bipolar disorder, Selek et al. (23) conducted MR imaging scans at baseline and then again following four weeks of lithium treatment. Participants were required to be drug-free for at least two weeks prior to study entry. Patients with bipolar disorder categorized as clinical responders to lithium had increased left prefrontal volume (and in particular the dorsolateral prefrontal cortex) following treatment. In contrast, patients with bipolar disorder categorized as clinical nonresponders to lithium had decreased left hippocampal and right anterior cingulate cortex volume following treatment. Lyoo et al. (24) conducted longitudinal MR imaging and evaluated clinical response to treatment in twenty-two patients with bipolar disorder who were psychotropic drug naive to mood stabilizers and antipsychotics. Patients were randomly assigned to receive either valproic acid or lithium and followed for sixteen weeks. Patients treated with lithium had greater gray matter volume (corresponding to an increase of 2.56%), which peaked at approximately ten to twelve weeks of treatment compared to both patients treated with valproic acid and healthy controls.
Two studies by Moore et al. (25) (26) examined the effects of lithium treatment on MR imaging measures in the context of a controlled clinical trial. In both studies, patients had a two-week medication washout period and were scanned prior to and then again following four weeks of blinded lithium treatment. In the first study (25), eight of the ten patients treated with lithium demonstrated an increase of approximately 3% in total gray matter volume without associated changes in white matter volume or cerebral water content. In the second study (26), these investigators reported an increase in total gray matter volume in twenty patients, which included both treatment responders and nonresponders. There was a trend for a correlation between clinical improvement and change in gray matter volume and increases were most prominent in the prefrontal cortex. Notably, there was an 8% increase in the left subgenual prefrontal cortex gray matter volume without any observed changes in total brain or white matter volume among patients treated with lithium.
One longitudinal study should be noted given that it investigated the effects of lithium on the brain in thirteen right-handed healthy volunteers (27). These individuals received MR imaging exams prior to and then following four weeks of lithium treatment at therapeutically relevant dosages. Using optimized voxel-based morphometry results indicated that both right and left dorsolateral prefrontal cortex and left anterior cingulate gray matter volume increased following lithium treatment. In addition, total white matter volume increased in contrast to total brain volume and total gray matter volume, which demonstrated no changes after lithium treatment. These data thus highlight brain changes associated with lithium treatment in the absence of a psychiatric illness confound.
Related posts:
Chapter 1 – Brain Imaging Methods in Mood Disorders
Chapter 2 – Neuroanatomical Findings in Unipolar Depression and the Role of the Hippocampus
Chapter 7 – Functional Connectome in Bipolar Disorder
Chapter 8 – Magnetic Resonance Spectroscopy Investigations of Bioenergy and Mitochondrial Function in Mood Disorders
Chapter 14 – Electrophysiological Biomarkers for Mood Disorders
Chapter 15 – Magnetoencephalography Studies in Mood Disorders
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
