Advertisement

Accumulation of Lithium in the Hippocampus of Patients With Bipolar Disorder: A Lithium-7 Magnetic Resonance Imaging Study at 7 Tesla

  • Jacques Stout
    Affiliations
    NeuroSpin, Commissariat à l’Énergie Atomique, Paris-Saclay University, Gif-sur-Yvette, France
    Search for articles by this author
  • Franz Hozer
    Affiliations
    NeuroSpin, Commissariat à l’Énergie Atomique, Paris-Saclay University, Gif-sur-Yvette, France

    Department of Psychiatry, Corentin-Celton Hospital, Assistance Publique—Hôpitaux de Paris, Issy-les-Moulineaux, France

    Paris University, Paris, France

    Institut National de la Santé et de la Recherche Médicale U955 Team 15 “Translational Psychiatry,” Paris-East University, Créteil, France
    Search for articles by this author
  • Arthur Coste
    Affiliations
    NeuroSpin, Commissariat à l’Énergie Atomique, Paris-Saclay University, Gif-sur-Yvette, France
    Search for articles by this author
  • Franck Mauconduit
    Affiliations
    NeuroSpin, Commissariat à l’Énergie Atomique, Paris-Saclay University, Gif-sur-Yvette, France

    Siemens Healthineers, Saint-Denis, France
    Search for articles by this author
  • Nouzha Djebrani-Oussedik
    Affiliations
    Saint-Louis–Lariboisière–F. Widal Hospitals, Assistance Publique—Hôpitaux de Paris, Paris, France
    Search for articles by this author
  • Samuel Sarrazin
    Affiliations
    NeuroSpin, Commissariat à l’Énergie Atomique, Paris-Saclay University, Gif-sur-Yvette, France
    Search for articles by this author
  • Joel Poupon
    Affiliations
    Saint-Louis–Lariboisière–F. Widal Hospitals, Assistance Publique—Hôpitaux de Paris, Paris, France
    Search for articles by this author
  • Manon Meyrel
    Affiliations
    Paris University, Paris, France

    Saint-Louis–Lariboisière–F. Widal Hospitals, Assistance Publique—Hôpitaux de Paris, Paris, France

    Institut National de la Santé et de la Recherche Médicale UMRS-1144, Paris, France
    Search for articles by this author
  • Sandro Romanzetti
    Affiliations
    Neurology, Rheinisch-Westfälische Technische Hochschule Aachen University Hospital, Aachen, Germany
    Search for articles by this author
  • Bruno Etain
    Affiliations
    Paris University, Paris, France

    Saint-Louis–Lariboisière–F. Widal Hospitals, Assistance Publique—Hôpitaux de Paris, Paris, France

    Institut National de la Santé et de la Recherche Médicale UMRS-1144, Paris, France

    FondaMental Foundation, Créteil, France
    Search for articles by this author
  • Cécile Rabrait-Lerman
    Affiliations
    NeuroSpin, Commissariat à l’Énergie Atomique, Paris-Saclay University, Gif-sur-Yvette, France
    Search for articles by this author
  • Josselin Houenou
    Affiliations
    NeuroSpin, Commissariat à l’Énergie Atomique, Paris-Saclay University, Gif-sur-Yvette, France

    Institut National de la Santé et de la Recherche Médicale U955 Team 15 “Translational Psychiatry,” Paris-East University, Créteil, France

    Département Médico-Universitaire de Psychiatrie et d’Addictologie, Mondor University Hospital, Assistance Publique—Hôpitaux de Paris, Créteil, France

    FondaMental Foundation, Créteil, France
    Search for articles by this author
  • Frank Bellivier
    Affiliations
    Paris University, Paris, France

    Saint-Louis–Lariboisière–F. Widal Hospitals, Assistance Publique—Hôpitaux de Paris, Paris, France

    Institut National de la Santé et de la Recherche Médicale UMRS-1144, Paris, France

    FondaMental Foundation, Créteil, France
    Search for articles by this author
  • Edouard Duchesnay
    Affiliations
    NeuroSpin, Commissariat à l’Énergie Atomique, Paris-Saclay University, Gif-sur-Yvette, France
    Search for articles by this author
  • Fawzi Boumezbeur
    Correspondence
    Address correspondence to Fawzi Boumezbeur, Ph.D., NeuroSpin, Centre CEA de Saclay, Bâtiment 145, 91191 Gif-sur-Yvette Cedex, France.
    Affiliations
    NeuroSpin, Commissariat à l’Énergie Atomique, Paris-Saclay University, Gif-sur-Yvette, France
    Search for articles by this author
Open AccessPublished:March 05, 2020DOI:https://doi.org/10.1016/j.biopsych.2020.02.1181

      Abstract

      Background

      Lithium (Li) is a first-line treatment for bipolar disorder (BD). To study its cerebral distribution and association with plasma concentrations, we used 7Li magnetic resonance imaging at 7T in euthymic patients with BD treated with Li carbonate for at least 2 years.

      Methods

      Three-dimensional 7Li magnetic resonance imaging scans (N = 21) were acquired with an ultra-short echo-time sequence using a non-Cartesian k-space sampling scheme. Lithium concentrations ([Li]) were estimated using a phantom replacement approach accounting for differential T1 and T2 relaxation effects. In addition to the determination of mean regional [Li] from 7 broad anatomical areas, voxel- and parcellation-based group analyses were conducted for the first time for 7Li magnetic resonance imaging.

      Results

      Using unprecedented spatial sensitivity and specificity, we were able to confirm the heterogeneity of the brain Li distribution and its interindividual variability, as well as the strong correlation between plasma and average brain [Li] ([Li]B ≈ 0.40 × [Li]P, R = .74). Remarkably, our statistical analysis led to the identification of a well-defined and significant cluster corresponding closely to the left hippocampus for which high Li content was displayed consistently across our cohort.

      Conclusions

      This observation could be of interest considering 1) the major role of the hippocampus in emotion processing and regulation, 2) the consistent atrophy of the hippocampus in untreated patients with BD, and 3) the normalization effect of Li on gray matter volumes. This study paves the way for the elucidation of the relationship between Li cerebral distribution and its therapeutic response, notably in newly diagnosed patients with BD.

      Keywords

      Bipolar disorder (BD) is a chronic affective disorder characterized by recurrence of manic and depressive episodes and affects 1%–3% of the adult population worldwide (
      • Merikangas K.R.
      • Jin R.
      • He J.P.
      • Kessler R.C.
      • Lee S.
      • Sampson N.A.
      • et al.
      Prevalence and correlates of bipolar spectrum disorder in the world mental health survey initiative.
      ). It is a severe and debilitating illness leading to major disruptions of daily life and higher mortality, particularly by increasing the rate of suicides in affected individuals (
      • Grande I.
      • Berk M.
      • Birmaher B.
      • Vieta E.
      Bipolar disorder.
      ).
      For more than half a century, lithium (Li) salts have been used to treat BD (
      • Shorter E.
      The history of lithium therapy.
      ). Although other types of medication have appeared since, Li has remained one of the first-line drugs to treat BD (
      • Yatham L.N.
      • Kennedy S.H.
      • Parikh S.V.
      • Schaffer A.
      • Bond D.J.
      • Frey B.N.
      • et al.
      Canadian Network for Mood and Anxiety Treatments (CANMAT) and International Society for Bipolar Disorders (ISBD) 2018 guidelines for the management of patients with bipolar disorder.
      ), preventing acute manic episodes (
      • Bauer M.S.
      • Mitchner L.
      What is a “mood stabilizer”? An evidence-based response.
      ) and lowering the suicide rate (
      • Baldessarini R.J.
      • Tondo L.
      • Davis P.
      • Pompili M.
      • Goodwin F.K.
      • Hennen J.
      Decreased risk of suicides and attempts during long-term lithium treatment: A meta-analytic review.
      ). Despite decades of therapeutic use, mechanisms underlying Li effects remain largely unknown. One prevalent theory is that Li might exert neurotrophic and/or neuroprotective effects on gray matter (GM), preventing GM atrophy related to BD through inhibition of proapoptotic pathways, such as glycogen synthase kinase-3β (
      • Rowe M.K.
      • Wiest C.
      • Chuang D.M.
      GSK-3 is a viable potential target for therapeutic intervention in bipolar disorder.
      ). These effects seem to impact specific subcortical regions, which have been widely highlighted in the literature, particularly the hippocampus, amygdala, and thalamus (
      • Hibar D.P.
      • Westlye L.T.
      • van Erp T.G.
      • Rasmussen J.
      • Leonardo C.D.
      • Faskowitz J.
      • et al.
      Subcortical volumetric abnormalities in bipolar disorder.
      ,
      • Hallahan B.
      • Newell J.
      • Soares J.C.
      • Brambilla P.
      • Strakowski S.M.
      • Fleck D.E.
      • et al.
      Structural magnetic resonance imaging in bipolar disorder: An international collaborative mega-analysis of individual adult patient data.
      ,
      • Hajek T.
      • Bauer M.
      • Simhandl C.
      • Rybakowski J.
      • O’Donovan C.
      • Pfennig A.
      Neuroprotective effect of lithium on hippocampal volumes in bipolar disorder independent of long-term treatment response.
      ,
      • López-Jaramillo C.
      • Vargas C.
      • Díaz-Zuluaga A.M.
      • Palacio J.D.
      • Castrillón G.
      • Vieta E.
      • et al.
      Increased hippocampal, thalamus and amygdala volume in long-term lithium-treated bipolar I disorder patients compared with unmedicated patients and healthy subjects.
      ). It remains unclear, however, why Li would have localized effects on specific GM volumes. One hypothesis is that the distribution of Li throughout the brain could be heterogeneous owing to a discrepancy in the density of its available transporters, especially sodium channels (
      • Luo H.
      • Gauthier M.
      • Tan X.
      • Landry C.
      • Cisternino S.
      • Declèves X.
      Sodium transporters are involved in lithium influx in brain endothelial cells.
      ,
      • Jakobsson E.
      • Argüello-Miranda O.
      • Chiu S.W.
      • Fazal Z.
      • Kruczek J.
      • Pritchet L.
      • et al.
      Toward a unified understanding of lithium action in basic biology and its significance for applied biology.
      ). There is indeed strong evidence that these channels show distinct distribution across the different regions of the human brain (
      • Leterrier C.
      • Brachet A.
      • Fache M.P.
      • Dargent B.
      Voltage-gated sodium channel organization in neurons: Protein interactions and trafficking pathways.
      ,
      • Wang J.
      • Ou S.W.
      • Wang Y.J.
      Distribution and function of voltage-gated sodium channels in the nervous system.
      ,
      • Whitaker W.R.
      • Clare J.J.
      • Powell A.J.
      • Chen Y.H.
      • Faull R.L.
      • Emson P.C.
      Distribution of voltage-gated sodium channel alpha-subunit and beta-subunit mRNAs in human hippocampal formation, cortex, and cerebellum.
      ,
      • Hladky S.B.
      • Barrand M.A.
      Fluid and ion transfer across the blood-brain and blood-cerebrospinal fluid barriers; a comparative account of mechanisms and roles.
      ). The heterogeneous distribution of Li across the brain could in turn explain why its neurotrophic effects would manifest differently across the brain and across patients with BD. Indeed, only 30% of patients show an optimal outcome (
      • Severus E.
      • Bauer M.
      • Geddes J.
      Efficacy and effectiveness of lithium in the long-term treatment of bipolar disorders: An update 2018.
      ), and variability in Li response and tolerability is poorly understood.
      It is at this point that nuclear magnetic resonance (NMR) provides a unique solution to detect and quantify in a noninvasive manner the local concentration of Li ([Li]). Indeed, Lithium-7 (7Li), the most abundant stable Li isotope, possesses a good NMR sensitivity (∼29% of 1H), and its detection using in vivo 7Li NMR spectroscopy has been performed successfully in animal models as well as in humans (
      • Renshaw P.F.
      • Wicklund S.
      In vivo measurement of lithium in humans by nuclear magnetic resonance spectroscopy.
      ,
      • Smith F.E.
      • Cousins D.A.
      • Thelwall P.E.
      • Ferrier I.N.
      • Blamire A.M.
      Quantitative lithium magnetic resonance spectroscopy in the normal human brain on a 3 T clinical scanner.
      ,
      • Machado-Vieira R.
      • Otaduy M.C.
      • Zanetti M.V.
      • de Sousa R.T.
      • Dias V.V.
      • Leite C.C.
      • et al.
      A selective association between central and peripheral lithium levels in remitters in bipolar depression: A 3T-7Li magnetic resonance spectroscopy study.
      ). More recently, 7Li NMR spectroscopic imaging (
      • Komoroski R.A.
      • Newton J.E.
      • Sprigg J.R.
      • Cardwell D.
      • Mohanakrishnan P.
      • Karson C.N.
      In vivo 7Li nuclear magnetic resonance study of lithium pharmacokinetics and chemical shift imaging in psychiatric patients.
      ,
      • Ramaprasad S.
      • Ripp E.
      • Pi J.
      • Lyon M.
      Pharmacokinetics of lithium in rat brain regions by spectroscopic imaging.
      ,
      • Lee J.H.
      • Adler C.
      • Norris M.
      • Chu W.J.
      • Strakowski S.M.
      • Komoroski R.A.
      4-T 7Li 3D MR spectroscopy imaging in the brains of bipolar disorder subjects.
      ,
      • Gyulai L.
      • Wicklund S.W.
      • Greenstein R.
      • Bauer M.S.
      • Ciccione P.
      • Whybrow P.C.
      • et al.
      Measurement of tissue lithium concentration by lithium magnetic resonance spectroscopy in patients with bipolar disorder.
      ) and magnetic resonance imaging (MRI) (
      • Ramaprasad S.
      • Newton J.E.O.
      • Cardwell D.
      • Fowler A.H.
      • Komoroski R.A.
      In vivo 7Li NMR imaging and localized spectroscopy of rat brain.
      ,
      • Komoroski R.A.
      • Pearce J.M.
      • Newton J.E.
      The distribution of lithium in rat brain and muscle in vivo by 7Li NMR imaging.
      ,
      • Smith F.E.
      • Thelwall P.E.
      • Necus J.
      • Flowers C.J.
      • Blamire A.M.
      • Cousins D.A.
      3D 7Li magnetic resonance imaging of brain lithium distribution in bipolar disorder.
      ) have been applied for the noninvasive determination of brain Li distribution in patients with BD. However, the spatial resolution and precision of those images remain limited by the low [Li] at therapeutic range (0.3–1.0 mmol/L in the plasma). Even so, those 7Li-NMR studies have made important observations about Li compartmentation and its cerebral distribution (
      • Machado-Vieira R.
      • Otaduy M.C.
      • Zanetti M.V.
      • de Sousa R.T.
      • Dias V.V.
      • Leite C.C.
      • et al.
      A selective association between central and peripheral lithium levels in remitters in bipolar depression: A 3T-7Li magnetic resonance spectroscopy study.
      ,
      • Ramaprasad S.
      • Ripp E.
      • Pi J.
      • Lyon M.
      Pharmacokinetics of lithium in rat brain regions by spectroscopic imaging.
      ,
      • Gyulai L.
      • Wicklund S.W.
      • Greenstein R.
      • Bauer M.S.
      • Ciccione P.
      • Whybrow P.C.
      • et al.
      Measurement of tissue lithium concentration by lithium magnetic resonance spectroscopy in patients with bipolar disorder.
      ,
      • Komoroski R.A.
      • Pearce J.M.
      • Newton J.E.
      The distribution of lithium in rat brain and muscle in vivo by 7Li NMR imaging.
      ,
      • Stout J.
      • Hanak A.S.
      • Chevillard L.
      • Djemaï B.
      • Bellivier F.
      • Boumezbeur F.
      • et al.
      Investigation of lithium distribution in the rat brain ex vivo using lithium-7 magnetic resonance spectroscopy and imaging at 17.2 T.
      ). Yet so far, those studies have failed to identify regions of the brain where Li would accumulate significantly. We determined that such limits could be addressed by acquiring 7Li MRI at 7T to improve the spatial resolution of our Li images and consequently help in determining the heterogeneity of Li distribution in the brain of euthymic patients with BD. In addition, we aimed at quantifying the total brain [Li]. For this purpose, we chose to use an ultra-short echo-time steady-state free-precession (SSFP) sequence combined with a twisted projection imaging k-space sampling scheme (
      • Boada F.E.
      • Gillen J.S.
      • Shen G.X.
      • Chang S.Y.
      • Thulborn K.R.
      Fast three dimensional sodium imaging.
      ) with a reasonably short repetition time (TR) as was proposed by Boada et al. (

      Qian Y, Boada FE; University of Pittsburgh (2010): Method for producing a magnetic resonance image of an object having a short T2 relaxation time. U.S. patent 7,750,632. Patent granted July 6, 2010.

      ) and a quantification pipeline that we validated in a previous ex vivo preclinical study on rats pretreated with Li (
      • Stout J.
      • Hanak A.S.
      • Chevillard L.
      • Djemaï B.
      • Bellivier F.
      • Boumezbeur F.
      • et al.
      Investigation of lithium distribution in the rat brain ex vivo using lithium-7 magnetic resonance spectroscopy and imaging at 17.2 T.
      ).

      Methods and Materials

      Participants

      Twenty-one euthymic adult outpatients with BD (10 women and 11 men, mean age ± SD = 41.2 ± 12.3 years) were recruited from the Center of Expertise for Bipolar Disorders of the Saint-Louis-Lariboisière-Fernand Widal University Hospitals (Paris, France). The criteria for inclusion were age18–65 years, a diagnosis of BD type I or type II (by DSM-IV-R criteria), scores on the Montgomery–Åsberg Depression Rating Scale (
      • Montgomery S.A.
      • Åsberg M.
      A new depression scale designed to be sensitive to change.
      ) and Young Mania Rating Scale (
      • Young R.C.
      • Biggs J.T.
      • Ziegler V.E.
      • Meyer D.A.
      A rating scale for mania: Reliability, validity and sensitivity.
      ) ≤ 7, and a current daily treatment of Li carbonate extended-release tablets ingested daily for at least 2 years before inclusion. Exclusion criteria were contraindications to NMR examination and an ongoing pregnancy. Trained practitioners established the diagnosis using the Diagnostic Interview for Genetic Studies (
      • Nurnberger J.I.
      • Blehar M.C.
      • Kaufmann C.A.
      • York-Cooler C.
      • Simpson S.G.
      • Reich T.
      • et al.
      Diagnostic interview for genetic studies: Rationale, unique features, and training.
      ). Co-prescriptions (including antipsychotics, anticonvulsants, and antidepressants) and history of substance misuse were recorded. All subjects had taken their Li treatment between 8 and 10 pm in the evening before the MRI examination, which was performed systematically between 9:00 and 11:00 am. Blood samples were drawn just before the scanning session to determine serum Li levels using inductively coupled plasma–mass spectrometry (Elan DRCe; Perkin Elmer, Courtaboeuf, France) by standard addition calibration, as previously reported (
      • Hanak A.S.
      • Chevillard L.
      • El Balkhi S.
      • Risede P.
      • Peoc’h K.
      • Megarbane B.
      Study of blood and brain lithium pharmacokinetics in the rat according to three different modalities of poisoning.
      ). For each participant, response to Li was evaluated using the Retrospective Criteria of Long-Term Treatment Response in Research Subjects with Bipolar Disorder (Alda scale) (
      • Manchia M.
      • Adli M.
      • Akula N.
      • Ardau R.
      • Aubry J.M.
      • Alda M.
      • et al.
      Assessment of response to lithium maintenance treatment in bipolar disorder: A Consortium on Lithium Genetics (ConLiGen) report.
      ). Clinical details are summarized in Table 1. The institutional review board and ethics committee (Comité de Protection des Personnes Ile-de-France VI) approved the study protocol in October 2016. Every patient received a complete description of the study and gave written informed consent.
      Table 1Demographics and Clinical Characteristics of the Cohort of Patients With Bipolar Disorder (N = 21)
      CharacteristicsValue
      Categorical Variables, n
      All values represent the number of participants with the characteristic.
       Male/female11/10
       Other current medication
      Antiepileptic10
      Antipsychotic8
      Antidepressant2
      Nonpsychotropic drug4
       History of substance misuse
      Tobacco14
      Alcohol2
      Cannabis4
      Others0
      Numerical Variables, Mean ± SD
       Age at MRI, years41.2 ± 12.3
       Lithium treatment duration, years5.5 ± 3.5
       Lithium dosage, mg/day976 ± 241
       Serum lithium concentration, mmol/L0.80 ± 0.19
       Erythrocyte lithium concentration, mmol/L0.39 ± 0.19
       Alda scores
      A = Lithium response; B = weight against lithium factors; Total = A − B.
      Total6.1 ± 2.5
      A7.8 ± 2.2
      B1.8 ± 1.4
       MADRS3.1 ± 2.6
       YMRS1.3 ± 1.9
      MADRS, Montgomery–Åsberg Depression Rating Scale; MRI, magnetic resonance imaging; YMRS, Young Mania Rating Scale.
      a All values represent the number of participants with the characteristic.
      b A = Lithium response; B = weight against lithium factors; Total = A − B.

      Magnetic Resonance Data Acquisition

      7Li-NMR acquisitions were performed on a 7T Magnetom scanner (Siemens Healthineers, Erlangen, Germany) with a dual-resonance 1H/7Li RF birdcage coil (Rapid Biomedical, Rimpar, Germany). The protocol consisted of the acquisition of a T1-weighted image (magnetization prepared rapid gradient echo sequence, 2-mm isotropic resolution) for anatomical reference; the acquisition of two B0 field-maps, one for B0 shimming and another for the correction of B0 inhomogeneities in postprocessing; and the calibration of the reference voltage for the 7Li channel using nonlocalized spectra (echo time [TE]/TR = 0.3/3000 ms), before the acquisition of the 3D 7Li image (SSFP sequence, TE/TR = 0.3/200 ms, α = 20°, 1769 projections, 50% linear fraction, 352 points per spoke, 10.6-ms readout duration, 15-mm isotropic nominal resolution, 17-mm effective resolution, 24-minute acquisition time). Global T1 and T2 relaxation times for 7Li were estimated from nonlocalized spectra acquired from a subset of our cohort of patients with BD (5 out of 21) using the progressive saturation technique (
      • Freeman R.
      • Hill H.D.W.
      Fourier transform study of NMR spin–lattice relaxation by “progressive saturation.
      ) (TR ranging from 0.4 to 20 seconds) and by varying the TE (ranging from 30 to 120 ms).

      Magnetic Resonance Data Reconstruction

      Non-Cartesian reconstruction of the 7Li MRI was realized using a homemade Python gridding algorithm (
      • Jackson J.I.
      • Meyer C.H.
      • Nishimura D.G.
      • Macovski A.
      Selection of a convolution function for Fourier inversion using gridding (computerised tomography application).
      ) with additional correction for B0 inhomogeneities using the second experimental B0 field-map (
      • Noll D.C.
      • Pauly J.M.
      • Meyer C.H.
      • Nishimura D.G.
      • Macovskj A.
      Deblurring for non-2D Fourier transform magnetic resonance imaging.
      ,
      • Man L.C.
      • Pauly J.M.
      • Macovski A.
      Multifrequency interpolation for fast off-resonance correction.
      ,
      • Nylund A.
      Off-resonance correction for magnetic resonance imaging with spiral trajectories [dissertation].
      ). After [Li] maps were aligned with their 1H anatomical reference, all clinical Li distribution images were interpolated and co-registered into the MNI152 space provided by the Statistical Parameter Mapping software (
      Wellcome Centre for Human Neuroimaging
      Statistical parametric mapping.
      ).

      Lithium Concentration Quantification

      To estimate [Li] from the intensity of our 7Li images, we adopted a modified phantom replacement approach that we validated in a previous ex vivo 7Li MRI study of rats pretreated with Li (
      • Stout J.
      • Hanak A.S.
      • Chevillard L.
      • Djemaï B.
      • Bellivier F.
      • Boumezbeur F.
      • et al.
      Investigation of lithium distribution in the rat brain ex vivo using lithium-7 magnetic resonance spectroscopy and imaging at 17.2 T.
      ). The same 7Li MRI protocol was used to acquire at room temperature an image from a 7-L cylindrical phantom of Li chloride (50 mmol/L, pH = 7.0). Copper sulfate (1.2 g/L) was added to the solution to lower the T1 and T2 relaxation times of 7Li in this reference phantom. In addition, in vitro T1 and T2 relaxation times for 7Li were estimated from nonlocalized spectra acquired from this phantom using the same progressive saturation technique (
      • Freeman R.
      • Hill H.D.W.
      Fourier transform study of NMR spin–lattice relaxation by “progressive saturation.
      ) (TR ranging from 0.4 to 20 seconds) and by varying the TE (ranging from 30 to 2000 ms). This in vitro 7Li MR image was used as a 3-dimensional external reference of concentration. By considering the experimental in vivo and in vitro T1 and T2 relaxation times of 7Li and the proper signal equation (
      • Bernstein M.A.
      • King K.F.
      • Zhou X.J.
      Handbook of MRI Pulse Sequences.
      ), individual in vivo 7Li intensity images were converted into [Li] maps:
      SSSFPFID[Li].tanα2.(1(E1cosα).r)


      with: r=1E22(1E1cosα)2E22(E1cosα)2 and E1,2=exp(T1,2/TR)

      Data Analysis

      First, the average brain [Li] ([Li]B) was estimated from the individual [Li] map (Figure 1). Then, average [Li] value were evaluated over 7 broad anatomical regions of interest (broad ROIs [bROIs]): the frontal, parietal, temporal, and occipital lobes, the brainstem, the midbrain region, and the cerebellum. These large anatomical areas are well suited for a primary description of the brain Li distribution, particularly at the individual scale. The masks were defined with a combination of probabilistic atlases provided by the FMRIB Software Library (
      University of Oxford
      Templates and atlases included with FSL.
      ), in particular the Montreal Neurological Institute and Harvard-Oxford atlases (
      • Makris N.
      • Goldstein J.M.
      • Kennedy D.
      • Hodge S.M.
      • Tsuang M.T.
      • Seidman L.J.
      • et al.
      Decreased volume of left and total anterior insular lobule in schizophrenia.
      ,
      • Frazier J.A.
      • Chiu S.
      • Breeze J.L.
      • Makris N.
      • Caviness V.S.
      • Biederman J.
      • et al.
      Structural brain magnetic resonance imaging of limbic and thalamic volumes in pediatric bipolar disorder.
      ). For a better visualization of the brain Li distribution across our cohort of patients, each [Li] map was normalized (i.e., the mean [Li] was set to 1.0), co-registered to the Montreal Neurological Institute brain atlas, and averaged (Figure 2). Those regional and whole-brain [Li] value were then compared with the individual plasma concentrations ([Li]P) via a linear regression (Figure 3) to compute average brain-to-plasma ratios and the corresponding Pearson correlation coefficients (R) (Table 2).
      Figure thumbnail gr1
      Figure 1Gallery of lithium concentration ([Li]) maps acquired at 7T from 10 patients with bipolar disorder. Individual 7Li images were interpolated at the resolution of their anatomical reference for transformation into the MNI152 template space.
      Figure thumbnail gr2
      Figure 2Normalized brain lithium distribution averaged across all our patients with bipolar disorder. Coronal, sagittal, and axial views. [Li], lithium concentration.
      Figure thumbnail gr3
      Figure 3Average brain [Li]B vs. plasma [Li]P lithium concentrations. Overall, a good correlation between both [Li] levels was observed across our cohort, as illustrated by the linear regression, with [Li]B being ∼40% of [Li]P (R = .74).
      Table 2Mean Regional Lithium Concentrations ([Li]), Normalized [Li] Levels, Brain-to-Plasma Ratios ([Li]B/[Li]P), Slopes of the Linear Regressions, and Pearson’s Correlation Factors
      Brain Region of Interest[Li], mmol/LNormalized [Li] mmol/L[Li]B/[Li]PSlopeR
      Brainstem0.46 ± 0.171.38 ± 0.360.65 ± 0.230.57.52
      Cerebellum0.36 ± 0.081.10 ± 0.070.51 ± 0.120.44.70
      Frontal Lobe0.28 ± 0.050.87 ± 0.070.40 ± 0.060.34.72
      Midbrain0.39 ± 0.101.16 ± 0.110.54 ± 0.140.48.69
      Occipital Lobe0.33 ± 0.071.00 ± 0.110.47 ± 0.100.40.56
      Parietal Lobe0.32 ± 0.060.98 ± 0.060.45 ± 0.080.39.71
      Temporal Lobe0.34 ± 0.081.03 ± 0.080.48 ± 0.110.42.74
      Whole Brain0.33 ± 0.061.000.46 ± 0.090.40.74
      Values are mean ± SD, N = 21.
      As a second analysis step, we used a refined version of the Harvard-Oxford atlas covering 48 cortical and 21 subcortical areas (parcellation ROIs [pROIs]) to investigate whether Li tends to concentrate preferentially in those more specific structures. Individual [Li] maps were centered and scaled with global individual means and SDs computed over a mask of intracerebral voxels. The average of normalized images over the refined pROIs were computed and regressed onto a design matrix made of an intercept with the patient age and sex. The intercept captures the deviation of Li within each pROI from zeros, i.e., from the global individual means. Then, we computed the p values associated with the Student t test statistic corrected from multiple comparisons over the 69 pROIs using the Benjamini-Hochberg (false discovery rate) procedure (
      • Benjamini Y.
      • Hochberg Y.
      Controlling the false discovery rate: A practical and powerful approach to multiple testing.
      ). A standard threshold of .05 was considered on Q values (adjusted p values) for statistical significance. For visualization purpose, Figure 4 shows the statistical map at the voxel level, thresholded at p < 10−3 and uncorrected for multiple comparisons.
      Figure thumbnail gr4
      Figure 4Voxel-based statistical map (Student’s t test score) of increased lithium concentration over the individual mean, thresholded at p < 10−3. Top: coronal, sagittal, and axial glass views highlight a large cluster in the left hippocampus and a small cluster in the right pallidum. Bottom: The hippocampus, as defined by the Harvard-Oxford atlas (blue contour), overlaps with our thresholded statistical map. L, left; R, right.

      Results

      In Vivo T1 and T2 Relaxation Times of Lithium-7

      Although 7Li is a 3/2 spin with a priori biexponential longitudinal and transversal relaxation decays in complex media such as the brain (
      • Komoroski R.A.
      • Lindquist D.M.
      • Pearce J.M.
      Lithium compartmentation in brain by 7Li MRS: Effect of total lithium concentration.
      ), apparent monoexponential longitudinal and transverse relaxation times were estimated in vivo from our first 5 patients with BD at T1,mono = 3950 ms and T2,mono = 63 ms. For comparison, the T1 and T2 relaxation times measured for our external reference were estimated at 4.1 and 1.7 seconds, respectively.

      Cerebral Lithium Distribution

      As illustrated in Figure 1, the adopted 7Li MRI protocol allowed us to image the whole-brain Li distribution and estimate its concentration with a satisfactory sensitivity (normalized signal-to-noise ratio of 7.3 105 mol−1.min−½). As a consequence, all 21 7Li images were kept for our group analysis. For all patients, brain Li distribution was highly heterogeneous. Figure 2 shows the average normalized Li distribution across our cohort. Of note, we observed higher Li levels in the white matter and in subcortical areas. Table 2 summarizes the average (as well as the normalized) [Li] in our 7 bROIs and the whole brain. Those values confirm that higher [Li] values are observed in the brainstem, the midbrain (constituted with a large portion of white matter), and the cerebellum, while lower [Li] values could be found in the frontal and occipital lobes.

      Brain-to-Plasma Lithium Comparison

      As illustrated by the linear regression shown in Figure 3, we observed a good correlation between individual [Li]P and average [Li]B values. Assuming a direct proportionality, we found that [Li]B values were ∼40% of [Li]P values (R = .74). As summarized in Table 2, the brain-to-plasma ratios were systematically superior to the slopes of the linear regression, ranging from 0.40 to 0.65, while the corresponding slopes ranged from 0.34 to 0.57. At the regional level, the strongest correlations were observed in the temporal and frontal lobes (R = .74 and .72, respectively), while the weakest correlations were seen in the brainstem and occipital lobes (R = .52 and .56, respectively).

      Areas Where Li Is Consistently More Concentrated

      Across our cohort and pROIs, [Li] was found to be significantly higher than the average [Li]B in the left hippocampus (coefficient = 1.85, SE = 0.47, 95% confidence interval [CI] = 0.87–2.83, t = 4, corrected p = .01) and in the right pallidum (coefficient = 1.98, SE = 0.58, 95% CI = 0.76–3.2, t = 3.4, corrected p = .02). At a voxel level, the statistical map thresholded at p < 10−3 clearly fits specifically the left hippocampus shape defined in the atlas (Figure 4).

      Discussion

      In this study, we managed to map the cerebral distribution of Li in a cohort of 21 euthymic patients with BD using 7Li MRI at 7T. While [Li]P and [Li]B were strongly correlated, our results highlighted the heterogeneity of Li distribution in the brain as well as the interindividual variability of this distribution between patients. Remarkably, our results suggest that the left hippocampus displayed high Li content consistently across our cohort. Finally, our data further validate the use of advanced MRI approaches to map Li—performed here in 24 minutes, a clinically feasible time.
      The heterogeneity of Li distribution corroborates similar results obtained at lower spatial resolutions (
      • Lee J.H.
      • Adler C.
      • Norris M.
      • Chu W.J.
      • Strakowski S.M.
      • Komoroski R.A.
      4-T 7Li 3D MR spectroscopy imaging in the brains of bipolar disorder subjects.
      ,
      • Smith F.E.
      • Thelwall P.E.
      • Necus J.
      • Flowers C.J.
      • Blamire A.M.
      • Cousins D.A.
      3D 7Li magnetic resonance imaging of brain lithium distribution in bipolar disorder.
      ). It supports the hypothesis of an active transport of Li inside the brain, possibly through sodium and magnesium transporters (
      • Luo H.
      • Gauthier M.
      • Tan X.
      • Landry C.
      • Cisternino S.
      • Declèves X.
      Sodium transporters are involved in lithium influx in brain endothelial cells.
      ,
      • Jakobsson E.
      • Argüello-Miranda O.
      • Chiu S.W.
      • Fazal Z.
      • Kruczek J.
      • Pritchet L.
      • et al.
      Toward a unified understanding of lithium action in basic biology and its significance for applied biology.
      ). In particular, the distribution of sodium channels across the central nervous system is complex, owing to the molecular diversity of each subtype of sodium transporter and the differential regulation of their expression. Consequently, distinct localization patterns of those sodium channel subtypes have been observed for the different types of neurons within the brain parenchyma and the various epithelial cells at the level of the blood-brain or blood–cerebrospinal fluid barriers, leading to widely variable levels of expression and density across the brain (
      • Leterrier C.
      • Brachet A.
      • Fache M.P.
      • Dargent B.
      Voltage-gated sodium channel organization in neurons: Protein interactions and trafficking pathways.
      ,
      • Wang J.
      • Ou S.W.
      • Wang Y.J.
      Distribution and function of voltage-gated sodium channels in the nervous system.
      ,
      • Whitaker W.R.
      • Clare J.J.
      • Powell A.J.
      • Chen Y.H.
      • Faull R.L.
      • Emson P.C.
      Distribution of voltage-gated sodium channel alpha-subunit and beta-subunit mRNAs in human hippocampal formation, cortex, and cerebellum.
      ,
      • Hladky S.B.
      • Barrand M.A.
      Fluid and ion transfer across the blood-brain and blood-cerebrospinal fluid barriers; a comparative account of mechanisms and roles.
      ). This could explain the heterogeneity of Li distribution in our study.
      Thanks to the additional net magnetization available at 7T, we acquired our 7Li images with an effective spatial resolution of 17 mm isotropic and a sensitivity threshold of ∼0.05 mmol/L, thus reaching the highest spatial resolution for clinical Li imaging obtained so far. By exploiting the experimental T1 and T2 relaxation times of 7Li in a subset of patients with BD, [Li] maps were established. Those global monoexponential T1,mono and T2,mono values were well within the range of previous reports (
      • Smith F.E.
      • Cousins D.A.
      • Thelwall P.E.
      • Ferrier I.N.
      • Blamire A.M.
      Quantitative lithium magnetic resonance spectroscopy in the normal human brain on a 3 T clinical scanner.
      ,
      • Komoroski R.A.
      • Newton J.E.
      • Sprigg J.R.
      • Cardwell D.
      • Mohanakrishnan P.
      • Karson C.N.
      In vivo 7Li nuclear magnetic resonance study of lithium pharmacokinetics and chemical shift imaging in psychiatric patients.
      ,
      • Komoroski R.A.
      • Lindquist D.M.
      • Pearce J.M.
      Lithium compartmentation in brain by 7Li MRS: Effect of total lithium concentration.
      ). It is of interest to note here that the use of an SSFP sequence with an ultra-short TE minimizes the T2∗ weighting and reduces sensibly the T2 weighting of the 7Li image than a balanced SSFP sequence (
      • Smith F.E.
      • Thelwall P.E.
      • Necus J.
      • Flowers C.J.
      • Blamire A.M.
      • Cousins D.A.
      3D 7Li magnetic resonance imaging of brain lithium distribution in bipolar disorder.
      ). Yet owing to our short TR, our quantification approach is vulnerable to local variations in T1. Therefore, local differences in Li compartmentation (e.g., between intra- and extracellular compartments leading to varying apparent T1 relaxation times) could contribute to the observed heterogeneity.
      Our study corroborates previous studies concerning the relationship between [Li]P and [Li]B (
      • Machado-Vieira R.
      • Otaduy M.C.
      • Zanetti M.V.
      • de Sousa R.T.
      • Dias V.V.
      • Leite C.C.
      • et al.
      A selective association between central and peripheral lithium levels in remitters in bipolar depression: A 3T-7Li magnetic resonance spectroscopy study.
      ,
      • Lee J.H.
      • Adler C.
      • Norris M.
      • Chu W.J.
      • Strakowski S.M.
      • Komoroski R.A.
      4-T 7Li 3D MR spectroscopy imaging in the brains of bipolar disorder subjects.
      ,
      • Gyulai L.
      • Wicklund S.W.
      • Greenstein R.
      • Bauer M.S.
      • Ciccione P.
      • Whybrow P.C.
      • et al.
      Measurement of tissue lithium concentration by lithium magnetic resonance spectroscopy in patients with bipolar disorder.
      ), with average [Li]B values ∼40% of [Li]P in our study despite methodological differences. However, a stronger correlation was found in our study, probably due to the larger number of patients and the greater signal-to-noise ratio of our acquisitions.
      Another favorable technical factor was the fact that our spatial resolution and our ultra-short echo-time SSFP sequence limit the interference of Li contained in the cerebrospinal fluid compared with other in vivo 7Li MRI studies (
      • Smith F.E.
      • Thelwall P.E.
      • Necus J.
      • Flowers C.J.
      • Blamire A.M.
      • Cousins D.A.
      3D 7Li magnetic resonance imaging of brain lithium distribution in bipolar disorder.
      ). Still, this interference can be felt through the larger variabilities and lesser correlations observed for ROIs close to the ventricles (such as the midbrain or the brainstem).
      Using a parcellation-based t test analysis, we identified few regions displaying consistently high normalized [Li] across our cohort despite our modest cohort size and a quite stringent false discovery rate correction [compared with a correction procedure that would account for the correlation between imaging features such as the one proposed by Westfall and Young (
      • Westfall P.H.
      • Young S.S.
      • Paul Wright S.
      On adjusting p values for multiplicity.
      )]. Among those regions were the left middle temporal gyrus, the lateral occipital cortex, and the right pallidum. However, the most well-defined and most significant cluster was the one that corresponded closely with the left hippocampus. To the best of our knowledge, this is the first study reporting such results. This observation is of potential importance for several reasons. First, the hippocampus (along with the amygdala) plays a major role in emotion regulation and processing and in inhibition of stress responses (
      • Phillips M.L.
      • Swartz H.A.
      A critical appraisal of neuroimaging studies of bipolar disorder: Toward a new conceptualization of underlying neural circuitry and a road map for future research.
      ). Second, it has been demonstrated that individuals with BD exhibited GM atrophy throughout the entire cortex as well as in subcortical limbic structures (
      • Hibar D.P.
      • Westlye L.T.
      • van Erp T.G.
      • Rasmussen J.
      • Leonardo C.D.
      • Faskowitz J.
      • et al.
      Subcortical volumetric abnormalities in bipolar disorder.
      ,
      • Hibar D.P.
      • Westlye L.T.
      • Doan N.T.
      • Jahanshad N.
      • Cheung J.W.
      • Krämer B.
      • et al.
      Cortical abnormalities in bipolar disorder: An MRI analysis of 6503 individuals from the ENIGMA Bipolar Disorder Working Group.
      ). In particular, the ENIGMA (Enhancing NeuroImaging Genetics through Meta-Analysis) consortium has found that the hippocampus was the subcortical structure with the largest shrinkage in BD (
      • Hibar D.P.
      • Westlye L.T.
      • van Erp T.G.
      • Rasmussen J.
      • Leonardo C.D.
      • Faskowitz J.
      • et al.
      Subcortical volumetric abnormalities in bipolar disorder.
      ). The hippocampus thus plays a central role in current neural models of emotion dysregulation in BD (
      • Phillips M.L.
      • Swartz H.A.
      A critical appraisal of neuroimaging studies of bipolar disorder: Toward a new conceptualization of underlying neural circuitry and a road map for future research.
      ). Third, other neuroimaging studies have suggested the neuroprotective and neurotrophic effects of Li, which can slow down GM atrophy in BD or may even promote a normalization of GM volumes (
      • Phillips M.L.
      • Swartz H.A.
      A critical appraisal of neuroimaging studies of bipolar disorder: Toward a new conceptualization of underlying neural circuitry and a road map for future research.
      ,
      • Moore G.J.
      • Bebchuk J.M.
      • Wilds I.B.
      • Chen G.
      • Menji H.K.
      Lithium-induced increase in human brain grey matter.
      ), particularly in the hippocampus (
      • Hallahan B.
      • Newell J.
      • Soares J.C.
      • Brambilla P.
      • Strakowski S.M.
      • Fleck D.E.
      • et al.
      Structural magnetic resonance imaging in bipolar disorder: An international collaborative mega-analysis of individual adult patient data.
      ,
      • Hajek T.
      • Bauer M.
      • Simhandl C.
      • Rybakowski J.
      • O’Donovan C.
      • Pfennig A.
      Neuroprotective effect of lithium on hippocampal volumes in bipolar disorder independent of long-term treatment response.
      ,
      • López-Jaramillo C.
      • Vargas C.
      • Díaz-Zuluaga A.M.
      • Palacio J.D.
      • Castrillón G.
      • Vieta E.
      • et al.
      Increased hippocampal, thalamus and amygdala volume in long-term lithium-treated bipolar I disorder patients compared with unmedicated patients and healthy subjects.
      ,
      • Foland L.C.
      • Altshuler L.L.
      • Sugar C.A.
      • Lee A.D.
      • Leow A.D.
      • Thompson P.M.
      • et al.
      Increased volume of the amygdala and hippocampus in bipolar patients treated with lithium.
      ). Preclinical studies are also supportive of a neurotrophic effect of Li. In particular, Zanni et al. (
      • Zanni G.
      • Michno W.
      • di Martino E.
      • Tjärnlund-Wolf A.
      • Blomgren K.
      • Hanrieder J.
      • et al.
      Lithium accumulates in neurogenic brain regions as revealed by high resolution ion imaging.
      ) have also reported higher concentration levels of Li in the hippocampus of mice using inductively coupled plasma atomic emission spectroscopy, which directly supports our finding. Identifying a consistently high normalized Li concentration specifically in a region could therefore be an intriguing finding and supports the neurogenic hypothesis of Li mode of action. Indeed, the hippocampus seems to be the main location of neurogenesis in the brain (
      • Moreno-Jiménez E.P.
      • Flor-García M.
      • Cafini F.
      • Pallas-Bazarra N.
      • Ávila J.
      • Llorens-Martín M.
      • et al.
      Adult hippocampal neurogenesis is abundant in neurologically healthy subjects and drops sharply in patients with Alzheimer’s disease.
      ).
      Interestingly, previous neuroimaging studies suggested that the physiopathological processes taking place in the hippocampus of patients with BD could manifest in an asymmetrical way. For instance, both Moorhead et al. (
      • Moorhead T.W.
      • McKirdy J.
      • Sussmann J.E.
      • Hall J.
      • Lawrie S.M.
      • Johnstone E.C.
      • McIntosh A.M.
      Progressive gray matter loss in patients with bipolar disorder.
      ) and Quigley et al. (
      • Quigley S.J.
      • Scanlon C.
      • Kilmartin L.
      • Barker G.J.
      • Cannon D.M.
      • McDonald C.
      • et al.
      Volume and shape analysis of subcortical brain structures and ventricles in euthymic bipolar I disorder.
      ) have reported smaller left hippocampus volumes among patients with BD compared with healthy control subjects. Also of interest is the observation, in a recent multimodal molecular imaging study (
      • Haarman B.
      • Burger H.
      • Doorduin J.
      • Renken R.J.
      • Mendes R.
      • Nolen W.A.
      • et al.
      Volume, metabolites and neuroinflammation of the hippocampus in bipolar disorder—A combined magnetic resonance imaging and positron emission tomography study.
      ), of decreased levels of N-acetyl-aspartate + N-acetyl-aspartyl-glutamate (NAA+NAAG) in the left hippocampus of patients with BD. This decrease was positively correlated with deregulated microglial activation [measured via the 11C-(R)-PK11195 binding potential] and depression symptoms in BD. Furthermore, our observation could be related to asymmetrical effects of Li on hippocampus volume. Indeed, several studies reported larger left hippocampus volume among patients with BD treated with Li compared with patients without Li (
      • Zung S.
      • Souza-Duran F.L.
      • Soeiro-de-Souza M.G.
      • Uchida R.
      • Busatto G.F.
      • Vallada H.
      • et al.
      The influence of lithium on hippocampal volume in elderly bipolar patients: A study using voxel-based morphometry.
      ,
      • Hajek T.
      • Cullis J.
      • Bauer M.
      • Young L.T.
      • Macqueen G.
      • Alda M.
      • et al.
      Hippocampal volumes in bipolar disorders: Opposing effects of illness burden and lithium treatment.
      ). Remarkably, Selek et al. (
      • Selek S.
      • Nicoletti M.
      • Zunta-Soares G.B.
      • Matsuo K.
      • Sanches M.
      • Soares J.C.
      • et al.
      A longitudinal study of fronto-limbic brain structures in patients with bipolar I disorder during lithium treatment.
      ) highlighted, in a longitudinal volumetric study of patients with BD treated for 4 weeks with Li, a decrease of the left hippocampus volume in Li nonresponders but not in Li responders, suggesting that a decreased left hippocampus volume might be an early marker of nonresponse to Li among patients with BD.
      Another interesting result of our study is that [Li] values were highest in the brainstem, midbrain, and cerebellum regions. However, the interindividual variance found for these regions was also very large, as most patients showed elevated concentration levels in those regions while for few others, [Li] in those regions did not exceed the whole-brain average values. Though it is too early to establish a relationship between those observations and the clinical response of Li, it seems relevant to ponder if [Li] in these areas could be linked to some of Li’s most problematic side-effects. In particular, chronic toxicity of Li can lead to physiological tremors and ataxia (
      • Sadosty A.T.
      • Groleau G.A.
      • Atcherson M.M.
      The use of lithium levels in the emergency department.
      ), while acute toxicity can lead to permanent loss of eye and speech coordination alongside cerebellar dysfunction (
      • Niethammer M.
      • Ford B.
      Permanent lithium-induced cerebellar toxicity: Three cases and review of literature.
      ). Future research is needed to confirm if an elevated [Li] in the brainstem and cerebellar regions could be an indicator that a patient is at risk of neurological Li toxicity.
      While those results are promising, especially our ability to identify for the first time, using 7Li MRI, the hippocampus as an area of interest to focus our attention regarding the investigation of Li therapeutic action, it is important to remain humble regarding the limited size of our cohort. As a consequence, we are unable at this stage to account for the potential bias linked to the many sources of interindividual variability, such as sex, age, and the administration of co-medication or eventual addictions, especially alcohol and tobacco.
      Finally, this 7Li MRI study represents important scientific and technical steps toward the identification of Li-targeted brain regions involved in Li clinical response in patients with BD. In particular, individual brain [Li] maps hold the promise to inform us about the relationship between the local [Li] and its neuroprotective action as it has been previously evaluated using structural, functional MRI, and 1H magnetic resonance spectroscopy data (
      • Machado-Vieira R.
      Lithium, stress, and resilience in bipolar disorder: Deciphering this key homeostatic synaptic plasticity regulator.
      ). It also opens avenues to explore to what extent local [Li] correlates with therapeutic response and/or tolerance, as well as the brain regions involved in the different levels of Li response. It would also be of great interest to exploit such data to predict in newly diagnosed patients with BD which ones will benefit from receiving Li.

      Acknowledgments and Disclosures

      The 7T MRI platform received support from the Equipement de Recherche et Plateformes Technologiques program of the Leducq Foundation. This work was funded by the French Ministries of Higher Education , Research, and Innovation and of Social Affairs and Health Inclusion, as well as the French National Research Agency (BIP-Li7 project 2014, Grant No. ANR-14-CE15-0003-01 [to FBe]).
      We thank Drs. Sarah Sportiche, Sunthavy Yem, Julia Maruani, and Gregory Gross for their help in recruiting the bipolar patients, as well our colleagues Chantal Ginisty, Severine Roger, Severine Desmidt, Yann Lecomte, Veronique Joly-Testault, Laurence Laurier, Gaëlle Médiouni, Bernadette Martins, and Damien Vanhoye for assisting in the preparation and conduction of the study and Dr. Alexandre Vignaud for his expertise and support.
      The authors report no biomedical financial interests or potential conflicts of interest.

      Supplementary Material

      References

        • Merikangas K.R.
        • Jin R.
        • He J.P.
        • Kessler R.C.
        • Lee S.
        • Sampson N.A.
        • et al.
        Prevalence and correlates of bipolar spectrum disorder in the world mental health survey initiative.
        Arch Gen Psychiatry. 2011; 68: 241-251
        • Grande I.
        • Berk M.
        • Birmaher B.
        • Vieta E.
        Bipolar disorder.
        Lancet. 2016; 387: 1561-1572
        • Shorter E.
        The history of lithium therapy.
        Bipolar Disord. 2009; 11: 4-9
        • Yatham L.N.
        • Kennedy S.H.
        • Parikh S.V.
        • Schaffer A.
        • Bond D.J.
        • Frey B.N.
        • et al.
        Canadian Network for Mood and Anxiety Treatments (CANMAT) and International Society for Bipolar Disorders (ISBD) 2018 guidelines for the management of patients with bipolar disorder.
        Bipolar Disord. 2018; 20: 97-170
        • Bauer M.S.
        • Mitchner L.
        What is a “mood stabilizer”? An evidence-based response.
        Am J Psychiatry. 2004; 161: 3-18
        • Baldessarini R.J.
        • Tondo L.
        • Davis P.
        • Pompili M.
        • Goodwin F.K.
        • Hennen J.
        Decreased risk of suicides and attempts during long-term lithium treatment: A meta-analytic review.
        Bipolar Disord. 2006; 8: 625-639
        • Rowe M.K.
        • Wiest C.
        • Chuang D.M.
        GSK-3 is a viable potential target for therapeutic intervention in bipolar disorder.
        Neurosci Biobehav Rev. 2007; 31: 920-931
        • Hibar D.P.
        • Westlye L.T.
        • van Erp T.G.
        • Rasmussen J.
        • Leonardo C.D.
        • Faskowitz J.
        • et al.
        Subcortical volumetric abnormalities in bipolar disorder.
        Mol Psychiatry. 2016; 21: 1710
        • Hallahan B.
        • Newell J.
        • Soares J.C.
        • Brambilla P.
        • Strakowski S.M.
        • Fleck D.E.
        • et al.
        Structural magnetic resonance imaging in bipolar disorder: An international collaborative mega-analysis of individual adult patient data.
        Biol Psychiatry. 2011; 69: 326-335
        • Hajek T.
        • Bauer M.
        • Simhandl C.
        • Rybakowski J.
        • O’Donovan C.
        • Pfennig A.
        Neuroprotective effect of lithium on hippocampal volumes in bipolar disorder independent of long-term treatment response.
        Psychol Med. 2014; 44: 507-517
        • López-Jaramillo C.
        • Vargas C.
        • Díaz-Zuluaga A.M.
        • Palacio J.D.
        • Castrillón G.
        • Vieta E.
        • et al.
        Increased hippocampal, thalamus and amygdala volume in long-term lithium-treated bipolar I disorder patients compared with unmedicated patients and healthy subjects.
        Bipolar Disord. 2017; 19: 41-49
        • Luo H.
        • Gauthier M.
        • Tan X.
        • Landry C.
        • Cisternino S.
        • Declèves X.
        Sodium transporters are involved in lithium influx in brain endothelial cells.
        Mol Pharm. 2018; 15: 2528-2538
        • Jakobsson E.
        • Argüello-Miranda O.
        • Chiu S.W.
        • Fazal Z.
        • Kruczek J.
        • Pritchet L.
        • et al.
        Toward a unified understanding of lithium action in basic biology and its significance for applied biology.
        J Membr Biol. 2017; 250: 587-604
        • Leterrier C.
        • Brachet A.
        • Fache M.P.
        • Dargent B.
        Voltage-gated sodium channel organization in neurons: Protein interactions and trafficking pathways.
        Neurosci Lett. 2010; 486: 92-100
        • Wang J.
        • Ou S.W.
        • Wang Y.J.
        Distribution and function of voltage-gated sodium channels in the nervous system.
        Channels (Austin). 2017; 11: 534-554
        • Whitaker W.R.
        • Clare J.J.
        • Powell A.J.
        • Chen Y.H.
        • Faull R.L.
        • Emson P.C.
        Distribution of voltage-gated sodium channel alpha-subunit and beta-subunit mRNAs in human hippocampal formation, cortex, and cerebellum.
        J Comp Neurol. 2000; 422: 123-139
        • Hladky S.B.
        • Barrand M.A.
        Fluid and ion transfer across the blood-brain and blood-cerebrospinal fluid barriers; a comparative account of mechanisms and roles.
        Fluids Barriers CNS. 2016; 13: 19
        • Severus E.
        • Bauer M.
        • Geddes J.
        Efficacy and effectiveness of lithium in the long-term treatment of bipolar disorders: An update 2018.
        Pharmacopsychiatry. 2018; 51: 173-176
        • Renshaw P.F.
        • Wicklund S.
        In vivo measurement of lithium in humans by nuclear magnetic resonance spectroscopy.
        Biol Psychiatry. 1988; 23: 465-475
        • Smith F.E.
        • Cousins D.A.
        • Thelwall P.E.
        • Ferrier I.N.
        • Blamire A.M.
        Quantitative lithium magnetic resonance spectroscopy in the normal human brain on a 3 T clinical scanner.
        Magn Reson Med. 2011; 66: 945-949
        • Machado-Vieira R.
        • Otaduy M.C.
        • Zanetti M.V.
        • de Sousa R.T.
        • Dias V.V.
        • Leite C.C.
        • et al.
        A selective association between central and peripheral lithium levels in remitters in bipolar depression: A 3T-7Li magnetic resonance spectroscopy study.
        Acta Psychiatr Scand. 2016; 133: 214-220
        • Komoroski R.A.
        • Newton J.E.
        • Sprigg J.R.
        • Cardwell D.
        • Mohanakrishnan P.
        • Karson C.N.
        In vivo 7Li nuclear magnetic resonance study of lithium pharmacokinetics and chemical shift imaging in psychiatric patients.
        Psychiatry Res. 1993; 50: 67-76
        • Ramaprasad S.
        • Ripp E.
        • Pi J.
        • Lyon M.
        Pharmacokinetics of lithium in rat brain regions by spectroscopic imaging.
        Magn Reson Imaging. 2005; 23: 859-863
        • Lee J.H.
        • Adler C.
        • Norris M.
        • Chu W.J.
        • Strakowski S.M.
        • Komoroski R.A.
        4-T 7Li 3D MR spectroscopy imaging in the brains of bipolar disorder subjects.
        Magn Reson Med. 2012; 68: 363-368
        • Gyulai L.
        • Wicklund S.W.
        • Greenstein R.
        • Bauer M.S.
        • Ciccione P.
        • Whybrow P.C.
        • et al.
        Measurement of tissue lithium concentration by lithium magnetic resonance spectroscopy in patients with bipolar disorder.
        Biol Psychiatry. 1991; 29: 1161-1170
        • Ramaprasad S.
        • Newton J.E.O.
        • Cardwell D.
        • Fowler A.H.
        • Komoroski R.A.
        In vivo 7Li NMR imaging and localized spectroscopy of rat brain.
        Magn Reson Med. 1992; 25: 308-318
        • Komoroski R.A.
        • Pearce J.M.
        • Newton J.E.
        The distribution of lithium in rat brain and muscle in vivo by 7Li NMR imaging.
        Magn Reson Med. 1997; 38: 275-278
        • Smith F.E.
        • Thelwall P.E.
        • Necus J.
        • Flowers C.J.
        • Blamire A.M.
        • Cousins D.A.
        3D 7Li magnetic resonance imaging of brain lithium distribution in bipolar disorder.
        Mol Psychiatry. 2018; 23: 2184
        • Stout J.
        • Hanak A.S.
        • Chevillard L.
        • Djemaï B.
        • Bellivier F.
        • Boumezbeur F.
        • et al.
        Investigation of lithium distribution in the rat brain ex vivo using lithium-7 magnetic resonance spectroscopy and imaging at 17.2 T.
        NMR Biomed. 2017; 30e3770
        • Boada F.E.
        • Gillen J.S.
        • Shen G.X.
        • Chang S.Y.
        • Thulborn K.R.
        Fast three dimensional sodium imaging.
        Magn Reson Med. 1997; 37: 706-715
      1. Qian Y, Boada FE; University of Pittsburgh (2010): Method for producing a magnetic resonance image of an object having a short T2 relaxation time. U.S. patent 7,750,632. Patent granted July 6, 2010.

        • Montgomery S.A.
        • Åsberg M.
        A new depression scale designed to be sensitive to change.
        Br J Psychiatry. 1979; 134: 382-389
        • Young R.C.
        • Biggs J.T.
        • Ziegler V.E.
        • Meyer D.A.
        A rating scale for mania: Reliability, validity and sensitivity.
        Br J Psychiatry. 1978; 133: 429-435
        • Nurnberger J.I.
        • Blehar M.C.
        • Kaufmann C.A.
        • York-Cooler C.
        • Simpson S.G.
        • Reich T.
        • et al.
        Diagnostic interview for genetic studies: Rationale, unique features, and training.
        Arch Gen Psychiatry. 1994; 51: 849-859
        • Hanak A.S.
        • Chevillard L.
        • El Balkhi S.
        • Risede P.
        • Peoc’h K.
        • Megarbane B.
        Study of blood and brain lithium pharmacokinetics in the rat according to three different modalities of poisoning.
        Toxicol Sci. 2014; 143: 185-195
        • Manchia M.
        • Adli M.
        • Akula N.
        • Ardau R.
        • Aubry J.M.
        • Alda M.
        • et al.
        Assessment of response to lithium maintenance treatment in bipolar disorder: A Consortium on Lithium Genetics (ConLiGen) report.
        PLoS One. 2013; 8e65636
        • Freeman R.
        • Hill H.D.W.
        Fourier transform study of NMR spin–lattice relaxation by “progressive saturation.
        J Chem Phys. 1971; 54: 3367-3377
        • Jackson J.I.
        • Meyer C.H.
        • Nishimura D.G.
        • Macovski A.
        Selection of a convolution function for Fourier inversion using gridding (computerised tomography application).
        IEEE Trans Med Imaging. 1991; 10: 473-478
        • Noll D.C.
        • Pauly J.M.
        • Meyer C.H.
        • Nishimura D.G.
        • Macovskj A.
        Deblurring for non-2D Fourier transform magnetic resonance imaging.
        Magn Reson Med. 1992; 25: 319-333
        • Man L.C.
        • Pauly J.M.
        • Macovski A.
        Multifrequency interpolation for fast off-resonance correction.
        Magn Reson Med. 1997; 37: 785-792
        • Nylund A.
        Off-resonance correction for magnetic resonance imaging with spiral trajectories [dissertation].
        (Available at:)
        • Wellcome Centre for Human Neuroimaging
        Statistical parametric mapping.
        (Available at:)
        • Bernstein M.A.
        • King K.F.
        • Zhou X.J.
        Handbook of MRI Pulse Sequences.
        Elsevier, 2004
        • University of Oxford
        Templates and atlases included with FSL.
        (Available at:)
        • Makris N.
        • Goldstein J.M.
        • Kennedy D.
        • Hodge S.M.
        • Tsuang M.T.
        • Seidman L.J.
        • et al.
        Decreased volume of left and total anterior insular lobule in schizophrenia.
        Schizophr Res. 2006; 83: 155-171
        • Frazier J.A.
        • Chiu S.
        • Breeze J.L.
        • Makris N.
        • Caviness V.S.
        • Biederman J.
        • et al.
        Structural brain magnetic resonance imaging of limbic and thalamic volumes in pediatric bipolar disorder.
        Am J Psychiatry. 2005; 162: 1256-1265
        • Benjamini Y.
        • Hochberg Y.
        Controlling the false discovery rate: A practical and powerful approach to multiple testing.
        J R Stat Soc Ser B. 1995; 57: 289-300
        • Komoroski R.A.
        • Lindquist D.M.
        • Pearce J.M.
        Lithium compartmentation in brain by 7Li MRS: Effect of total lithium concentration.
        NMR Biomed. 2013; 26: 1152-1157
        • Westfall P.H.
        • Young S.S.
        • Paul Wright S.
        On adjusting p values for multiplicity.
        Biometrics. 1993; 49: 941-945
        • Phillips M.L.
        • Swartz H.A.
        A critical appraisal of neuroimaging studies of bipolar disorder: Toward a new conceptualization of underlying neural circuitry and a road map for future research.
        Am J Psychiatry. 2014; 171: 829-843
        • Hibar D.P.
        • Westlye L.T.
        • Doan N.T.
        • Jahanshad N.
        • Cheung J.W.
        • Krämer B.
        • et al.
        Cortical abnormalities in bipolar disorder: An MRI analysis of 6503 individuals from the ENIGMA Bipolar Disorder Working Group.
        Mol Psychiatry. 2018; 23: 932-942
        • Moore G.J.
        • Bebchuk J.M.
        • Wilds I.B.
        • Chen G.
        • Menji H.K.
        Lithium-induced increase in human brain grey matter.
        Lancet. 2000; 356: 1241-1242
        • Foland L.C.
        • Altshuler L.L.
        • Sugar C.A.
        • Lee A.D.
        • Leow A.D.
        • Thompson P.M.
        • et al.
        Increased volume of the amygdala and hippocampus in bipolar patients treated with lithium.
        Neuroreport. 2008; 19: 221-224
        • Zanni G.
        • Michno W.
        • di Martino E.
        • Tjärnlund-Wolf A.
        • Blomgren K.
        • Hanrieder J.
        • et al.
        Lithium accumulates in neurogenic brain regions as revealed by high resolution ion imaging.
        Sci Rep. 2017; 7: 40726
        • Moreno-Jiménez E.P.
        • Flor-García M.
        • Cafini F.
        • Pallas-Bazarra N.
        • Ávila J.
        • Llorens-Martín M.
        • et al.
        Adult hippocampal neurogenesis is abundant in neurologically healthy subjects and drops sharply in patients with Alzheimer’s disease.
        Nat Med. 2019; 25: 554-560
        • Moorhead T.W.
        • McKirdy J.
        • Sussmann J.E.
        • Hall J.
        • Lawrie S.M.
        • Johnstone E.C.
        • McIntosh A.M.
        Progressive gray matter loss in patients with bipolar disorder.
        Biol Psychiatry. 2007; 62: 894-900
        • Quigley S.J.
        • Scanlon C.
        • Kilmartin L.
        • Barker G.J.
        • Cannon D.M.
        • McDonald C.
        • et al.
        Volume and shape analysis of subcortical brain structures and ventricles in euthymic bipolar I disorder.
        Psychiatry Res. 2015; 233: 324-330
        • Haarman B.
        • Burger H.
        • Doorduin J.
        • Renken R.J.
        • Mendes R.
        • Nolen W.A.
        • et al.
        Volume, metabolites and neuroinflammation of the hippocampus in bipolar disorder—A combined magnetic resonance imaging and positron emission tomography study.
        Brain Behav Immun. 2016; 56: 21-33
        • Zung S.
        • Souza-Duran F.L.
        • Soeiro-de-Souza M.G.
        • Uchida R.
        • Busatto G.F.
        • Vallada H.
        • et al.
        The influence of lithium on hippocampal volume in elderly bipolar patients: A study using voxel-based morphometry.
        Transl Psychiatry. 2016; 6: e846
        • Hajek T.
        • Cullis J.
        • Bauer M.
        • Young L.T.
        • Macqueen G.
        • Alda M.
        • et al.
        Hippocampal volumes in bipolar disorders: Opposing effects of illness burden and lithium treatment.
        Bipolar Disord. 2012; 14: 261-270
        • Selek S.
        • Nicoletti M.
        • Zunta-Soares G.B.
        • Matsuo K.
        • Sanches M.
        • Soares J.C.
        • et al.
        A longitudinal study of fronto-limbic brain structures in patients with bipolar I disorder during lithium treatment.
        J Affect Disord. 2013; 150: 629-633
        • Sadosty A.T.
        • Groleau G.A.
        • Atcherson M.M.
        The use of lithium levels in the emergency department.
        J Emerg Med. 1999; 17: 887-891
        • Niethammer M.
        • Ford B.
        Permanent lithium-induced cerebellar toxicity: Three cases and review of literature.
        Mov Disord. 2007; 22: 570-573
        • Machado-Vieira R.
        Lithium, stress, and resilience in bipolar disorder: Deciphering this key homeostatic synaptic plasticity regulator.
        J Affect Disord. 2018; 233: 92-99

      Linked Article

      • Mapping Lithium in the Brain: New 3-Dimensional Methodology Reveals Regional Distribution in Euthymic Patients With Bipolar Disorder
        Biological PsychiatryVol. 88Issue 5
        • Preview
          In this issue of Biological Psychiatry, Stout et al. (1) describe technological advances to achieve high-quality, 3-dimensional quantitative mapping of lithium levels throughout the brain using magnetic resonance spectroscopic imaging. In a group of 21 patients—10 women and 11 men who were euthymic and who were undergoing treatment with lithium chloride for at least 2 years—the authors report some intriguing results. Although lithium has a long and rich history as a mood stabilizer (2), the molecular and biological fundamentals that underlie its actions in bipolar disorder remain unclear.
        • Full-Text
        • PDF