© Springer International Publishing Switzerland 2017
Gin S. Malhi, Marc Masson and Frank Bellivier (eds.)The Science and Practice of Lithium Therapy10.1007/978-3-319-45923-3_55. Neuroimaging and Lithium
(1)
AP-HP, Hôpitaux Universitaires Mondor, DHU PePsy, Université Paris Est, Pôle de Psychiatrie, Créteil, France
(2)
INSERM U955, IMRB, Team 15 ‘Psychiatrie Translationnelle’, 40 rue de Mesly, 94000 Créteil, France
(3)
Fondation Fondamental, Créteil, France
(4)
UNIACT, Neurospin, I2BM, CEA Saclay, Gif sur Yvette, France
Abstract
Neuroimaging holds the promise of being a means to improve our understanding of lithium’s mechanism of action. Structural brain imaging studies of individuals exposed to lithium, mainly using magnetic resonance imaging (MRI), have repeatedly evidenced an increased volume of grey matter in cross-sectional, but also in longitudinal, studies. They suggest that lithium has a neurotrophic and neuroprotective effect, particularly in the hippocampus. This effect seems to be supported by lithium-treated rodent models. Findings in white matter are more heterogeneous and sparse. We still do not know, however, whether the neurotrophic effects associated with lithium are a direct consequence of the element or are secondary to symptomatic improvement, although the effect does seem lithium specific. Finally, in the near future, neuroimaging of lithium may assist the clinician through the identification of biomarkers of response to lithium and through the direct measurement of lithium brain concentrations using MRI spectroscopy.
Keywords
NeuroimagingLithiumMRINeurotrophicSpectroscopyKey Points
Neuroimaging studies suggest that lithium has a neurotrophic effect.
This neurotrophic action of lithium mostly impacts grey matter.
A key challenge is the discovery of biomarkers of response to lithium.
MRI spectroscopy may provide a clearer understanding of lithium’s mechanism of action.
5.1 Introduction
Lithium has been used for more than 60 years and is the standard treatment for bipolar disorder. However, its mechanism of action is still unknown. In this chapter, we describe neuroimaging studies exploring lithium’s effects on brain structure and function. Most of these studies have been performed using a cross-sectional design and, along with a few longitudinal neuroimaging studies, bring evidence of a putative neuroprotective and neurotrophic effect of lithium. We will then discuss future developments of lithium imaging, such as magnetic resonance spectroscopy of lithium.
5.2 Lithium and Brain Volumes
Several cross-sectional studies have explored the associations between lithium medication and total brain, grey matter, white matter or regional volumes. Longitudinal neuroimaging studies of lithium’s effects offer the advantage of better control of confounding factors and biases, such as selection effect (e.g. patients already under lithium may have different clinical and neuroimaging characteristics than other patients at baseline). Therefore longitudinal neuroimaging studies probably lead to a more valid understanding of lithium’s mechanism of action than cross-sectional studies do.
Through both cross-sectional and longitudinal studies, we will focus successively on:
Lithium neurotrophic properties
Lithium specificity with respect to these properties
5.2.1 Neurotrophic Effects
The first study on lithium-induced increased volumes in the human brain was published in 2000 in The Lancet. This work used 3D magnetic resonance imaging (MRI) and brain segmentation to study pharmacologically induced increases in grey matter volume. It found a total increase of grey matter volume of about 3 % (24 cm3) after 4 weeks’ treatment with lithium in ten patients with bipolar disorder (bipolar I patients suffering from depressive relapse). However, the authors found no modification in cerebral white matter. This suggested that lithium has neurotrophic effects (Moore et al. 2000).
Several cross-sectional studies have been performed using magnetic resonance spectroscopy of N-acetylaspartate (NAA) in patients receiving lithium. NAA, which is synthesised in mitochondria, is a putative marker of neuronal viability, helps to maintain myelin and is involved in neuron-to-glia signalling. In a first study, nine older adults with bipolar I disorder treated with lithium had in vivo measurements of serum and brain lithium levels taken using magnetic resonance spectroscopy of lithium, NAA and myo-inositol. The positive relationship identified between higher brain lithium levels and elevated NAA levels in older patients with bipolar disorder may support lithium neuroprotective, neurotrophic and mitochondrial function-enhancing effects (Forester et al. 2008). Another study evidenced that, after 4 weeks of lithium, this treatment was associated with an increase in both brain NAA levels in healthy controls and in patients with bipolar disorder (Moore et al. 2002). These findings provide indirect support for lithium having neuroprotective effects and for there being negative effects of the illness burden on prefrontal NAA levels in patients with bipolar disorder (Hajek et al. 2012).
5.2.1.1 Specificity of Effect
Two questions arise concerning the neurotrophic effect of lithium:
Is it specific to lithium, or is this effect common across other medications for bipolar disorder?
Does the effect have regional specificity?
Is the Neurotrophic Effect Specific to Lithium?
An animal study by Vernon and colleagues (Vernon et al. 2012) strongly suggests that lithium has specific neurotrophic brain effects. For 8 weeks, rodents received either lithium or haloperidol. Treatment was then interrupted for the following 8 weeks, while measurements were taken using longitudinal in vivo structural MRI. Finally, confirmation was obtained with postmortem findings using unbiased stereology. Chronic haloperidol treatment induced decreases in whole brain volume (−4 %) and cortical grey matter (−6 %), along with hypertrophy of the striatum (+14 %). In contrast, chronic lithium treatment induced increases in whole brain volume (+5 %) and cortical grey matter (+3 %), without a significant effect on striatal volume. Following 8 weeks of drug withdrawal, haloperidol-induced changes in brain volumes normalised, whereas the lithium-treated animals retained significantly greater total brain volumes, as confirmed postmortem. This animal model provides strong evidence for lithium having direct and specific neurotrophic effects.
In humans, a longitudinal study conducted at Washington University included 13 medication-free patients with bipolar disorder. An MRI exam was performed on all patients using a clinical 1.5 T whole-body scanner. They then received lithium medication for 3 months. Then a second MRI was performed. Results revealed lithium-induced increases in grey matter volume (2.56 %, 17.6 cm3 increase). No white matter volume modification was observed. In contrast, valproate-treated patients with bipolar disorder did not show grey matter volume changes over time (Lyoo et al. 2010).
Similarly, according to an updated review of the effects of medication on neuroimaging findings in bipolar disorder, lithium use is associated with increased volumes—particularly in areas subserving emotion processing and mood regulation (the amygdala, hippocampus, anterior and subgenual cingulate cortex)—while the use of antipsychotic agents and anticonvulsants is generally not (Hafeman et al. 2012).
Is Lithium’s Neurotrophic Effect Regionally Specific?
The current neurobiological models of emotion regulation describe two steps:
An early pre-attentional step (tens of milliseconds after the stimulus). This automatic regulation step of emotions involves the classical deep limbic structures (the amygdala and hippocampus).
A later step (several hundreds of milliseconds after the stimulus) of voluntary emotion regulation. It involves cognitive structures such as the prefrontal cortex and cingulate dorsal gyrus. These structures modulate the activity of the amygdala and hippocampus.
Compared to healthy controls, during emotional processing, patients with bipolar disorder exhibit an overactive ventral-limbic network (including regions such as the amygdala, the parahippocampal gyrus, the anterior and subgenual cingulate) (Malhi et al. 2015). This ventral-limbic network is poorly modulated by the ventral prefrontal cortices due to defective prefrontal-limbic connectivity in bipolar patients. This altered balance between limbic and prefrontal networks is thought to underlie the abnormal emotional reactivity present in bipolar disorder and also mood switches (Houenou et al. 2012; Phillips et al. 2008).
5.2.1.2 Effects on the Hippocampus
In two areas of the adult human brain (the olfactory bulb and the dentate gyrus of the hippocampus), new neurons are generated throughout life and form an integral part of the normal functional circuitry (Lledo et al. 2006). These areas are natural targets for the neurotrophic effect of lithium. Therefore, Baykara and colleagues focused on the hippocampus (located in the medial part of the temporal lobe, which is also a key component of the emotion regulation network) (Baykara et al. 2012). Seventeen adolescents aged between 13 and 19 with type I bipolar disorder were included in the study. Six of them took lithium for 1–14 months. After this, their right hippocampal volumes were found to be significantly larger than before treatment. The authors did not find any change in hippocampus size in groups taking valproate or second-generation antipsychotics (SGA). However, this finding must be considered with caution, because this was a secondary aim of the study. The main finding was that there was no significant difference between the right and left hippocampus volumes of patients with bipolar disorder and controls.
A longitudinal controlled study of bilateral hippocampal volume increase in patients with bipolar disorder who took long-term lithium treatment is more convincing. The 12 included patients were medication naive, having never received any psychopharmacological treatment for psychiatric illness before entry into the study. Lithium was introduced simultaneously at the time of the first MRI scan (one brain MRI exam at the beginning of the study, another one after 2 years and the last one after 4 years). Results showed that bilateral hippocampal volumes were larger in the lithium (+4–5 %) group than in the unmedicated group. This increase happened mostly during the first 2 years of MRI monitoring. There was no brain volume difference at the beginning of the study between bipolar disorder subjects and control subjects. This suggests that lithium did not restore a decreased hippocampal volume due to bipolar disorder but that this medication has neurotrophic properties (Yucel et al. 2007).
Another longitudinal study reached a similar conclusion (Hajek et al. 2012). Whereas the comparison of hippocampal volumes between the non-lithium patients and controls showed marked differences (p < 0.05), there were no significant differences between young bipolar disorder patients and controls or between the lithium-treated bipolar disorder and control participants (0.50 and 0.30, respectively). These results are also in line with previous investigations, which have shown a lack of hippocampal volume differences amongst lithium-treated patients or patients at the beginning of illness and controls. The smaller hippocampal volumes in bipolar disorder patients selected because of minimal lifetime exposure to lithium may be secondary to the burden of illness (mean 25.6 ± 9.8 years of illness and 10.5 ± 5.1 episodes) (Hajek et al. 2012).
A recent study about familial contributions to hippocampal morphology in bipolar disorder found that lithium-treated bipolar I patients had significantly larger global hippocampal volume (p = 0.03) compared to healthy controls (9 %), non-bipolar co-twins (12 %) and non-lithium-treated bipolar I patients (8 %). In contrast, hippocampal volumes in non-lithium-treated bipolar I patients did not differ from those of non-bipolar co-twins and control twins. This result supports the hypothesis of neurotrophic effects of lithium in the hippocampus (Lyoo et al. 2010).
5.2.1.3 Effects on Fronto-limbic Structure
The fronto-limbic networks play a critical role in emotion regulation and in bipolar disorder. A study comparing hippocampal and amygdala volumes in older bipolar patients with controls using high-resolution MRI scans reported smaller hippocampal volumes in bipolar patients (p = 0.001) and also a smaller right amygdala (p = 0.01). Total hippocampal volume was negatively associated with the duration of depressive (p = 0.035) and manic (p = 0.027) episodes but not lithium use. It suggests that the neuroprogressive course was related to the severity of the disorder. Amygdala volumes were not associated with the duration of mood episodes (Wijeratne et al. 2013).
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
