Lithium In The Drinking Water

 

Lithium In The Drinking Water

 

The title is an inside joke for any physicians trained in my generation. It was a standard line to indicate how common it is to prescribe a certain medication.  The first time I heard it I was an intern in the emergency department (ED) at St. Paul-Ramsey Medical Center.  It was (and is) a very busy ED and one of 3 Level I trauma centers in the Minneapolis St. Paul area with a population of about 3.6M people.  In those days there were no urgent care centers so the ED was informally split into a trauma and high acuity side and a low acuity side.  The interns would rotate from one side to the other every other day.  I had just assessed a couple of sisters on the low acuity side and diagnosed otitis media (ear infection) and was writing scripts for amoxicillin while I waiting for my attending to confirm the diagnosis.  He came out, agreed with the diagnosis and treatment plan and said: “We should put amoxicillin in the drinking water.”

And so it went.  Since that day I have heard the same thing about H-2 blockers, proton pump inhibitors, and statins.  All medications that are commonly prescribed for common problems.  Nobody has ever said that about lithium. In the conversations I have had about lithium over the past 30 years – people generally slow down, look concerned, and say something like: “That is a heavy-duty med isn’t it doc?” 

Lithium apparently got that reputation after it when it started to be widely used by psychiatrists for the treatment of bipolar disorder.  It was approved by the FDA in 1970 but was used as early as 1894 that for both bipolar disorder and melancholic depression (1).  It was also used in popular beverages and sought in the form of mineral water.  From 1929 to 1948 it was in 7-UP Lithiated Soda – a brand that eventually became 7-UP.  Lithium citrate was the active form and there is no reliable information on the concentration it originally contained.  One source suggests 5 mg/L (8) but it is not clear if this is as Li or a compound.  In psychiatry, that would be a trivial dose as either lithium carbonate or lithium citrate.  If it was really 5 milliequivalents (mEq) of lithium that would be roughly equivalent to 300 mg of lithium carbonate (Li₂CO₃) or 550 mg of lithium citrate (Li₃C₆H₅O₇).  Practically all the lithium prescribed by psychiatrists in the US is lithium carbonate in a range of 600-1800 mg/day.      

Lithium is considered a disease modifying drug in psychiatry for long term stabilization of bipolar disorder.  It is probably underutilized in the United States for both antidepressant augmentation and treatment of depression.  It may be underprescribed in general because it requires monitoring, has a narrow therapeutic index, can cause renal and thyroid complications, and has the potential for significant drug-drug interactions with a variety of medications that are commonly prescribed.  Investigations of its mechanism of action has led to some speculation that it may prevent neurodegeneration and be effective against psychiatric disorders even in very low doses.  These studies look at lithium exposure in the water supply and in animal models of neurodegeneration.  A recent paper suggests that lithium deficiency may cause Alzheimer’s dementia.

Before I get to a discussion of that paper, I thought I would review the ecology of lithium in the environment that is primarily focused on water chemistry.  I am referencing two major studies of lithium in the drinking water in the United States.  The first (3) looks at groundwater measurements at 18,027 states and uses that data to model lithium in the groundwater across the US.  They map that data and the maps are shown along with the original sampling sites at the maps at the top of this post.  As noted in the table, about 15% of these sites have a concentration of lithium that is greater than >30 μg/L.  That is significant because the Health Based Screening Level (HBSL) is 10 μg/L.  HBSLs are non-enforceable good faith benchmarks based on the latest drinking water and toxicity data.  Some of the sites measured in this study were exceeded by 1500 fold.  Lithium is the 32nd most abundant element in the Earth and distribution in nature is variable.  

The second study was more specific for drinking water because it looked at samples directly from drinking water treatment plants (DWTP) (4).  Even though DWTPs have no specific processes for removal of lithium, the levels are significantly lower in range than the groundwater survey.  The surface water had a median level below the HBSL and groundwater level was higher.   The authors noted that 56% of the groundwater and 13% of the surface water sources of DWTPs exceeded the HBSL. 

In terms of pharmaceutical doses of lithium – the lowest dose I have ever prescribed was 150 mg as lithium carbonate.  Lithium carbonate is 18.79% lithium; therefore, each tablet or capsule contains about 28.185 mg of Li or 28,185 μg.  Looking at the range of concentration in the Lombard study (3) it would take ingesting 1.88 to 28M liters of those waters to be equivalent to a single 150 mg capsule per day.  In the case of the median groundwater and surface water from the Sharma (4) study it would take 176 to 216 liters to take in the equivalent amount. That study also suggests that drinking water sourced for treatment for human use is less likely to have extraordinary levels but does have levels that are currently flagged as a potential health risk.  Most people on lithium maintenance for bipolar disorder have much higher exposure to lithium than is likely from any drinking water source.  There are some commercially available lithium mineral waters that advertise a lithium level of 490 μg as Lithium Bicarbonate (LiHCO₃). That is equivalent to 50.3 μg Li (per liter) putting it in the range of both studies.

What does all of this say about the ecology and water chemistry of lithium?  Cleary there is a lot of variability.  Most water sources are not problematic but some with very high levels may be.  Drinking water surveillance appears to be a good approach to reducing exposure to high levels and many municipalities test for uncommon elements and organic compounds. Any attempts to correlate lithium in the water with medical or psychiatric outcomes needs to account for this variability and it is significant.  In the study that used machine learning to predict lithium levels with meteorological and geological variables – the results were modest.  I agree with the opinion that since the long-term effects of Li as a micronutrient are unknown and there is some toxicological concern as evidenced by the HBSL it should be studied (4).

That brings me to the recent study that has been heavily covered in the news (9).  I have received several questions about it and the most common questions are: “Does lithium prevent Alzheimer’s Disease?” and “If it does should I take it.”  I will preface my comments by saying this is a very well-done study.  It is also an intense study that is typical of what is published in both Nature and Science.  There are a mix of experiments using state of the art technology and they are all presented as crowded and very small graphics in the paper.  There is also a supplemental document (in this case 13 pages) of additional graphs and figures). 

The experimental sections of the paper can be broken down into a naturalistic look at a panel of metals concentrations in the brain and blood of subjects with normal cognition versus Alzheimer’s Disease (AD) or mild cognitive impairment, the effect of lithium deficiency in normal and mouse models of AD (3xTg and J20Ag), the effect of lithium on glycogen synthase kinase 3β  (GSK3β), the effect of lithium replacement, the impact of lithium on brain ageing in wild type mice, and determining an optimal form of lithium for replacement. 

The human brain samples depending on the experiment were from groups with no cognitive impairment (NCI)(n=22 - 133), Alzheimer’s Disease (AD)(n=5 - 105), and mild cognitive impairment (MCI)(n=7 - 66).  Brain samples were fractionated to check for metallic ion gradients and 27 ions were investigated.  Only Li showed lower levels in the cortex in both MCI and AD and it was also concentrated in the Aβ plaques.  The authors conclude that this shows a significant problem with Li homeostasis in both MCI and AD.  Some of the sampled ions like lead and arsenic are known neurotoxins.

Results of the experiments in mice are depicted in the diagram below.  3xTg mice are transgenic mice with three mutations (APPswe, PS1M146V, and TauP301L) associated with AD.  As noted, they accumulate amyloid- β protein (Aβ) and tau protein.  The J20Ag mice are transgenic mice that results in overexpression of amyloid precursor protein (APP).  Both mouse lines are considered models of AD.  In all cases lithium deficiency leads to accumulation of amyloid-β protein and other processes (where measured) consistent with AD like neurodegeneration at the behavioral, ultrastructural, and biochemical level. 

The authors demonstrate that lithium supplementation in mice prevents these changes in wild type mice.  They illustrate how blocking GSK3β prevents AD like changes.  They also demonstrate how a lithium compound (lithium orotate) with low solubility prevents lithium form being sequestered in Aβ plaques.  All experiments considered they provide a compelling backdrop for considering lithium as a therapy for MCI and AD.  Are there any other considerations?

First, there have been clinical trials of lithium in a number of neurodegenerative diseases including AD. After some initial isolated enthusiasm for Li in the 1970s and 1980s for Parkinson's, some syndromes associated with Parkinson's (on-off, anergia), and Huntington’s – most of the reported research started in 2009.   Since then, there has been research on AD (10-13), ALS (14-19), MSA (20), MCI (21-23), Niemann Pick Disease (24), Machado Joseph Disease (26-27), and Spinocerebellar Ataxia, Type 2 (28).  In most of these papers the authors cite putative mechanisms of action of lithium based on preclinical trials and some positive pilot studies – but the overwhelming results were negative.  In all cases, the trials were approached from the perspective of using lithium in pharmaceutical ranges except for the trial that states it used microdosing on the range of 300 μg.  Tolerability varied widely among research subjects due to varying diagnoses, but even from study to study using modest doses.

My experience treating people with lithium in all age ranges leads me to a few conclusions that may apply here.  First, I have treated patients with lithium who have been on it for decades and developed AD. I recall one patient in particular who had marked cortical atrophy on brain imaging despite all of those years on lithium and no episodes of toxicity.  That obviously does not rule out lithium as a neuron sparing therapy but it does suggest that it will not work for everybody.  Second, part of the population will not be able to tolerate it or will not want to take it. I am fairly certain that any psychiatrist experienced in prescribing it will have no problems minimizing or preventing side effects.  In most 50-70 year olds that would be lithium carbonate in the range of 150-600 daily.  In the experiments cited above, microdoses of lithium orotate (4.3 μEq/L) for 12-14 months was the dose that prevented brain aging in mice (microgliosis and astrogliosis) and reduced pro-inflammatory cytokines.  That oral dose is roughly 1/1,000 the lowest prescribed pharmaceutical dose.  Third, I am not aware of any cases of lithium toxicity from people drinking groundwater or surface water with high Li concentrations.  The commonest reasons for lithium toxicity in people taking pharmaceutical doses include water and salt imbalances like dehydration, drug-drug interactions, and improper dosing and/or monitoring.   

The question on many minds is “Should I start taking lithium to prevent Alzheimer’s Disease?”  The answer lies in the way the authors frame the discussion section of their paper.  Despite a lot of positive findings they state that “Disruption of Li homeostasis may contribute to the long prodromal period of neuropathological changes that occur prior to the onset of clinical AD.”    And – “Li deficiency is therefore a potential common mechanism for the multisystem degeneration of the brain that leads to the onset of AD”.  It is going to take replication and more work before these findings are widely accepted.  There are still a lot of unknowns about GSK3β signaling.  There have been mistakes made extrapolating from the preclinical studies in mice in the past.  Some of those mistakes were attributed to differences in mouse and human genetics and less heterogeneity in mice.   

That said, I know that will not prevent people from attempts at biohacking and taking supplemental lithium. If you are in that category, you should keep a few things in mind.  First, you have to know about drug dosing and the difference between pharmacological doses and supplemental doses. Lithium at pharmacological doses has a low therapeutic index (toxic dose range to therapeutic dose range) and it can cause kidney, thyroid, and parathyroid problems.  Second, there are many lithium orotate supplements currently available in a wide range of doses (5, 10, 20, 130 mg). They are advertised for "memory, state of mind, and behavioral health". None of these are clinical or FDA approved indications.  Any use of these products has to be considered experimental at this time and I recommend waiting for further data.

In summary, this is an excellent study that synthesizes clinical and preclinical data across a wide array of parameters that I have just touched on here.  If I was a young researcher just starting out, this would be the kind of research team I would be interested in joining.  It was an exciting paper to read.  At the same time it is a good test of how research may or may not be reproducible.  A common misconception about a lack of reproducibility is that it means the researchers did something wrong.  It seems obvious that it can happen just based on the sheer complexity. 

 

George Dawson, MD, DFAPA

 

Supplementary 1:

I am really interested in the mechanism of action of Li and all the links to GSK3β signaling.  I ran out of space in the above post and hope to elaborate on the mechanism soon.

Graphics Credit:

The lead graphic for this post is from reference 3 per the open access CC-BY-NC-ND 4.0 license. The remaining graphic and table were made by me from data in the given references.

 

References:

1:  Shorter E. The history of lithium therapy. Bipolar Disord. 2009 Jun;11 Suppl 2(Suppl 2):4-9. doi: 10.1111/j.1399-5618.2009.00706.x. PMID: 19538681; PMCID: PMC3712976.

2:  Coppen A. 50 years of lithium treatment of mood disorders. Bipolar Disord. 1999 Sep;1(1):3-4. doi: 10.1034/j.1399-5618.1999.10102.x. PMID: 11256652.

3:  Lombard MA, Brown EE, Saftner DM, Arienzo MM, Fuller-Thomson E, Brown CJ, Ayotte JD. Estimating Lithium Concentrations in Groundwater Used as Drinking Water for the Conterminous United States. Environ Sci Technol. 2024 Jan 16;58(2):1255-1264. doi: 10.1021/acs.est.3c03315. Epub 2024 Jan 2. PMID: 38164924; PMCID: PMC10795177.

4:  Sharma N, Westerhoff P, Zeng C. Lithium occurrence in drinking water sources of the United States. Chemosphere. 2022 Oct;305:135458. doi: 10.1016/j.chemosphere.2022.135458. Epub 2022 Jun 23. PMID: 35752313; PMCID: PMC9724211.

5:  Norman JE, Toccalino PL, Morman SA, 2018. Health-Based Screening Levels for evaluating water-quality data (2d ed.). U.S. Geological Survey web page, accessible at https://water.usgs.gov/water-resources/hbsl/, doi:10.5066/F71C1TWP

6:  Seidel U, Jans K, Hommen N, Ipharraguerre IR, Lüersen K, Birringer M, Rimbach G. Lithium Content of 160 Beverages and Its Impact on Lithium Status in Drosophila melanogaster. Foods. 2020 Jun 17;9(6):795. doi: 10.3390/foods9060795. PMID: 32560287; PMCID: PMC7353479.

7:  El-Mallakh RS, Roberts RJ. Lithiated lemon-lime sodas. Am J Psychiatry. 2007 Nov;164(11):1662. doi: 10.1176/appi.ajp.2007.07081255. PMID: 17974929.

8:  Neves MO, Marques J, Eggenkamp HGM. Lithium in Portuguese Bottled Natural Mineral Waters-Potential for Health Benefits? Int J Environ Res Public Health. 2020 Nov 12;17(22):8369. doi: 10.3390/ijerph17228369. PMID: 33198207; PMCID: PMC7696288.

One of the most popular soft drinks in the world was launched in 1929; the “Lithiated Lemon Soda” that was supplemented with 5 mg Li (as Li citrate/L) until 1948 [16], when it was banned by the government. It was believed to cure alcohol-induced hangover symptoms, make people more energetic and give lust for life and on the top of that shinier hair and brighter eyes [17]. In fact, it is still on the market but since 1936 its name changed to 7UP. In 1949, John Cade discovered that higher Li concentrations were toxic. Nowadays, according Seidel et al. [16] 7UP only contains 1.4 µg Li/L.

9:  Aron L, Ngian ZK, Qiu C, Choi J, Liang M, Drake DM, Hamplova SE, Lacey EK, Roche P, Yuan M, Hazaveh SS, Lee EA, Bennett DA, Yankner BA. Lithium deficiency and the onset of Alzheimer's disease. Nature. 2025 Aug 6. doi: 10.1038/s41586-025-09335-x. Open Access.

The paper suggests that Li may be a critical micronutrient in terms of brain function.

AD: 

10:  Nunes MA, Viel TA, Buck HS. Microdose lithium treatment stabilized cognitive impairment in patients with Alzheimer's disease. Curr Alzheimer Res. 2013 Jan;10(1):104-7. doi: 10.2174/1567205011310010014. PMID: 22746245.

As lithium is highly toxic in regular doses, our group evaluated the effect of a microdose of 300 μg, administered once daily on AD patients for 15 months

11:  Hampel H, Ewers M, Bürger K, Annas P, Mörtberg A, Bogstedt A, Frölich L, Schröder J, Schönknecht P, Riepe MW, Kraft I, Gasser T, Leyhe T, Möller HJ, Kurz A, Basun H. Lithium trial in Alzheimer's disease: a randomized, single-blind, placebo-controlled, multicenter 10-week study. J Clin Psychiatry. 2009 Jun;70(6):922-31. PMID: 19573486.

“The current results do not support the notion that lithium treatment may lead to reduced hyperphosphorylation of tau protein after a short 10-week treatment in the Alzheimer's disease target population.”

12:  Macdonald A, Briggs K, Poppe M, Higgins A, Velayudhan L, Lovestone S. A feasibility and tolerability study of lithium in Alzheimer's disease. Int J Geriatr Psychiatry. 2008 Jul;23(7):704-11. doi: 10.1002/gps.1964. PMID: 18181229.

“Lithium treatment in elderly people with AD has relatively few side effects and those that were apparently due to treatment were mild and reversible. Nonetheless discontinuation rates are high. The use of lithium as a potential disease modification therapy in AD should be explored further but is not without problems.”

13:  Leyhe T, Eschweiler GW, Stransky E, Gasser T, Annas P, Basun H, Laske C. Increase of BDNF serum concentration in lithium treated patients with early Alzheimer's disease. J Alzheimers Dis. 2009;16(3):649-56. doi: 10.3233/JAD-2009-1004. PMID: 19276559.

“We assessed the influence of a lithium treatment on BDNF serum concentration in a subset of a greater sample recruited for a randomized, single-blinded, placebo-controlled, parallel-group multicenter 10-week study, investigating the efficacy of lithium treatment in AD patients. In AD patients treated with lithium, a significant increase of BDNF serum levels, and additionally a significant decrease of ADAS-Cog sum scores in comparison to placebo-treated patients, were found.”

ALS:

14:  Boll MC, Alcaraz-Zubeldia M, Rios C, González-Esquivel D, Montes S. A phase 2, double-blind, placebo-controlled trial of a valproate/lithium combination in ALS patients. Neurologia (Engl Ed). 2025 Jan-Feb;40(1):32-40. doi: 10.1016/j.nrleng.2022.07.003. Epub 2022 Aug 29. PMID: 36049647.

15:  UKMND-LiCALS Study Group; Morrison KE, Dhariwal S, Hornabrook R, et al. Lithium in patients with amyotrophic lateral sclerosis (LiCALS): a phase 3 multicentre, randomised, double-blind, placebo-controlled trial. Lancet Neurol. 2013 Apr;12(4):339-45. doi: 10.1016/S1474-4422(13)70037-1. Epub 2013 Feb 27. Erratum in: Lancet Neurol. 2013 Sep;12(9):846. PMID: 23453347; PMCID: PMC3610091.

16:  Verstraete E, Veldink JH, Huisman MH, Draak T, Uijtendaal EV, van der Kooi AJ, Schelhaas HJ, de Visser M, van der Tweel I, van den Berg LH. Lithium lacks effect on survival in amyotrophic lateral sclerosis: a phase IIb randomised sequential trial. J Neurol Neurosurg Psychiatry. 2012 May;83(5):557-64. doi: 10.1136/jnnp-2011-302021. Epub 2012 Feb 29. PMID: 22378918.

17:  Miller RG, Moore DH, Forshew DA, Katz JS, Barohn RJ, Valan M, Bromberg MB, Goslin KL, Graves MC, McCluskey LF, McVey AL, Mozaffar T, Florence JM, Pestronk A, Ross M, Simpson EP, Appel SH; WALS Study Group. Phase II screening trial of lithium carbonate in amyotrophic lateral sclerosis: examining a more efficient trial design. Neurology. 2011 Sep 6;77(10):973-9. doi: 10.1212/WNL.0b013e31822dc7a5. Epub 2011 Aug 3. PMID: 21813790; PMCID: PMC3171956.

18:  Chiò A, Borghero G, Calvo A, Capasso M, Caponnetto C, Corbo M, Giannini F, Logroscino G, Mandrioli J, Marcello N, Mazzini L, Moglia C, Monsurrò MR, Mora G, Patti F, Perini M, Pietrini V, Pisano F, Pupillo E, Sabatelli M, Salvi F, Silani V, Simone IL, Sorarù G, Tola MR, Volanti P, Beghi E; LITALS Study Group. Lithium carbonate in amyotrophic lateral sclerosis: lack of efficacy in a dose-finding trial. Neurology. 2010 Aug 17;75(7):619-25. doi: 10.1212/WNL.0b013e3181ed9e7c. Epub 2010 Aug 11. PMID: 20702794.

“Lithium was not well-tolerated in this cohort of patients with ALS, even at subtherapeutic doses. The 2 doses were equivalent in terms of survival/severe disability and functional data. The relatively high frequency of AEs/SAEs and the reduced tolerability of lithium raised serious doubts about its safety in ALS.”

19:  Aggarwal SP, Zinman L, Simpson E, McKinley J, Jackson KE, Pinto H, Kaufman P, Conwit RA, Schoenfeld D, Shefner J, Cudkowicz M; Northeast and Canadian Amyotrophic Lateral Sclerosis consortia. Safety and efficacy of lithium in combination with riluzole for treatment of amyotrophic lateral sclerosis: a randomised, double-blind, placebo-controlled trial. Lancet Neurol. 2010 May;9(5):481-8. doi: 10.1016/S1474-4422(10)70068-5. Epub 2010 Apr 1. PMID: 20363190; PMCID: PMC3071495.

Multiple System Atrophy (MSA)

20:  Saccà F, Marsili A, Quarantelli M, Brescia Morra V, Brunetti A, Carbone R, Pane C, Puorro G, Russo CV, Salvatore E, Tucci T, De Michele G, Filla A. A randomized clinical trial of lithium in multiple system atrophy. J Neurol. 2013 Feb;260(2):458-61. doi: 10.1007/s00415-012-6655-7. Epub 2012 Aug 30. PMID: 22932748.

MCI: 

21:  Damiano RF, Loureiro JC, Pais MV, Pereira RF, Corradi MM, Di Santi T, Bezerra GAM, Radanovic M, Talib LL, Forlenza OV. Revisiting global cognitive and functional state 13 years after a clinical trial of lithium for mild cognitive impairment. Braz J Psychiatry. 2023 Mar 11;45(1):46-49. doi: 10.47626/1516-4446-2022-2767. PMID: 36049127; PMCID: PMC9976922.

22:  Forlenza OV, Radanovic M, Talib LL, Gattaz WF. Clinical and biological effects of long-term lithium treatment in older adults with amnestic mild cognitive impairment: randomised clinical trial. Br J Psychiatry. 2019 Nov;215(5):668-674. doi: 10.1192/bjp.2019.76. PMID: 30947755.

23:  Forlenza OV, Diniz BS, Radanovic M, Santos FS, Talib LL, Gattaz WF. Disease-modifying properties of long-term lithium treatment for amnestic mild cognitive impairment: randomised controlled trial. Br J Psychiatry. 2011 May;198(5):351-6. doi: 10.1192/bjp.bp.110.080044. PMID: 21525519.

Lithium treatment was associated with a significant decrease in CSF concentrations of P-tau (P = 0.03) and better performance on the cognitive subscale of the Alzheimer's Disease Assessment.

Niemann Pick Disease:

24:  Han S, Zhang H, Yi M, Liu X, Maegawa GHB, Zou Y, Wang Q, Wu D, Ye Z. Potential Disease-Modifying Effects of Lithium Carbonate in Niemann-Pick Disease, Type C1. Front Pharmacol. 2021 Jun 9;12:667361. doi: 10.3389/fphar.2021.667361. PMID: 34177581; PMCID: PMC8220070.

MS:

25:  Rinker JR 2nd, Meador WR, King P. Randomized feasibility trial to assess tolerance and clinical effects of lithium in progressive multiple sclerosis. Heliyon. 2020 Jul 28;6(7):e04528. doi: 10.1016/j.heliyon.2020.e04528. PMID: 32760832; PMCID: PMC7393418.

Machado Joseph Disease:

26:  Saute JA, Rieder CR, Castilhos RM, Monte TL, Schumacher-Schuh AF, Donis KC, D'Ávila R, Souza GN, Russo AD, Furtado GV, Gheno TC, Souza DO, Saraiva-Pereira ML, Portela LV, Camey S, Torman VB, Jardim LB. Planning future clinical trials in Machado Joseph disease: Lessons from a phase 2 trial. J Neurol Sci. 2015 Nov 15;358(1-2):72-6. doi: 10.1016/j.jns.2015.08.019. Epub 2015 Aug 14. PMID: 26297649.

27:  Saute JA, de Castilhos RM, Monte TL, Schumacher-Schuh AF, Donis KC, D'Ávila R, Souza GN, Russo AD, Furtado GV, Gheno TC, de Souza DO, Portela LV, Saraiva-Pereira ML, Camey SA, Torman VB, de Mello Rieder CR, Jardim LB. A randomized, phase 2 clinical trial of lithium carbonate in Machado-Joseph disease. Mov Disord. 2014 Apr;29(4):568-73. doi: 10.1002/mds.25803. Epub 2014 Jan 7. PMID: 24399647.

Spinocerebellar Ataxia Type 2:

28:  Saccà F, Puorro G, Brunetti A, Capasso G, Cervo A, Cocozza S, de Leva M, Marsili A, Pane C, Quarantelli M, Russo CV, Trepiccione F, De Michele G, Filla A, Morra VB. A randomized controlled pilot trial of lithium in spinocerebellar ataxia type 2. J Neurol. 2015 Jan;262(1):149-53. doi: 10.1007/s00415-014-7551-0. Epub 2014 Oct 28. PMID: 25346067.

Huntington’s Disease:

29:  Aminoff MJ, Marshall J. Treatment of Huntington's chorea with lithium carbonate. A double-blind trial. Lancet. 1974 Jan 26;1(7848):107-9. doi: 10.1016/s0140-6736(74)92339-3. PMID: 4130308.

A New Book From An Expert Psychotherapist

 

I first became aware of Mardi Horowitz’s work when I was researching adjustment disorders many years ago.  As an acute care psychiatrist that is one of the disorders that ends up on your unit that you must separate from severe mental illnesses and significant risks.  I wanted to do more than just make the diagnosis.  I also wanted to assist these folks with psychotherapy that might prove useful, even if I ended up discharging them the same day.  Dr. Horowitz has written extensively about that and many other topics.  I decided to buy his recent book Clinician Technique in Personalized Psychotherapy.  In the introduction he mentions watching decades of watching psychotherapy videotapes and trying to figure out what helped people change. 

In the forward by Roberta Isberg, MD – she mentions that therapists might see something in the book that they have been doing in practice for years.  That happened to me when I read the Chapter Confronting Dilemmas by Assertion of the Therapeutic Alliance.  In fact, I had mentioned this intervention just a few hours earlier in psychotherapy seminar that I coteach. In that seminar I discussed how making the therapeutic alliance explicit could be useful in resolving impasses.  Dr. Horowitz’s chapter uses a dyadic diagram of the therapeutic alliance (p. 103) that is good in that it delineates the roles of both the patient and the therapist and what the expected exchanges might be.  For example, the patient is disclosing and focused on problems while the therapist is intervening, supporting, and emphasizing adaptive changes by the patient. 

Dr.  Horowitz also presents a table of Common Dilemmas for a Psychotherapist.  He defines dilemmas as binaries where both poles are unlikely to be helpful.  A common example is encouraging further elaboration of a problem that the patient may find very problematic in terms of external relationships, the relationship with the therapist, or longstanding internalized patterns of thinking and behavior.  In the table he presents ten common dilemmas, the therapist’s intervention, and how it might be interpreted.  In the case of these dilemmas, he suggests clarifying the situation and trying to reach a middle ground:  “ The middle ground between the binaries of the dilemma may be reached if the therapists state the properties of the periodically experienced therapeutic alliance.” (p. 100).   

I thought I would present a frequent acute care dilemma as a vignette, but before doing that borrow another definition from Dr. Horowitz.  That is the idea that the vignettes are fictionalized composites of multiple therapeutic encounters.  In the case below it is hundreds of encounters:

Patient: “Are you the one holding me here?  Are you the one I have to talk to to get out of here?  I want to be released as soon as possible.”

MD: “I am the person who will make that decision….”

Patient: “Well what’s the hold up?  You can’t just keep me here.  There is no reason why I should be sitting in this hospital.”  

MD: “I will do what I can but I have to be able to make an independent assessment in order to do that….”

Patient: “Look – I don’t care about that.  You have no right to hold me here.  I want to go home right now and you are in my way.”

MD: “OK – this is the first time I am seeing you. None of the people who brought you to the hospital or admitted you to my unit have been in touch with me.  I have nothing to do with who is admitted to my unit and in fact have been told that I am supposed to discharge people as soon as possible.  But I can’t do that unless I am fairly certain that they will be safe….”

Patient: “So you’re just covering your ass doc?  Really?  You are just worried about getting sued?”

MD: “I am not worried about getting sued, but I do worry about not getting people the assessments and treatment they might need.  The way this is supposed to work is that you and I talk about what happened and try to determine if you have any problems that I can help you with. It is not me against you or you against me.  It is you and I working on an agreed upon set of problems.  Do you think we can do that?”

Patient: “I suppose…”

MD: “OK let’s give it a try.”

This is an example of a situation that many physicians find impossible to approach because their authority is questioned and the potential for escalation.  That escalation depends largely on the physician not taking the critical comments as a personal attack but rather as a process issue.  It requires the ability to remain neutral in addition to confronting the dilemma and establishing a middle ground to proceed on. This skill is critical in acute care psychiatry as well as in crisis outpatient situations.  And before I get too grandiose like all things in medicine there are no guarantees – only probabilities.  There are situations that will rapidly escalate out of control despite your best efforts – but in my experience they are rare.    

There are many other dilemmas facing therapists during assessments and in ongoing therapy.  More common examples arise from the situation where the patient is reluctant to disclose the details of certain events or has expectations of the therapist that are not consistent with the reality of the therapy situation.  The standard cinematic approach of reflecting the problem back to the patient (“well how do you feel about that?”) is generally not an optimal response and it is one that most people see as cliché these days.  Clarifying what is going on in the room and in the therapy is probably a better strategy.

I have written about the therapeutic alliance in several areas on this blog. Here is a post from 2012 and 2017.  I also posted diagrams of the therapeutic alliance in those posts and include my most recent modification below.  In the diagram I am using MD as the therapist since almost all the therapists I interact with are psychiatrists or psychiatric residents, but it also applies to non-physician therapists.

 

I use a tripartite diagram to highlight the problem space as being a specific focus of patient and therapist since it is the combined process of what is happening in the therapy rather than the isolated process of either person.  Some authors write about this as intersubjectivity (2) or the result of the interaction between two unique conscious states.  Most physicians are taught to observe and record findings from an objective third person position.  The exception is psychiatry where subjectivity has recognized value and the importance of the physician-patient relationship is emphasized.  

Intersubjectivity provides a more comprehensive look at what happens in the therapeutic alliance than seeing the interaction as orchestrated solely by an objective therapist.  The therapist and patient have complementary roles.  For example, empathy is a critical dimension of the therapeutic alliance and a critical skill for the therapist.  Empathy is also required on the part of the patient and its presence can be palpable to varying degrees.  Does the patient really understand what the therapist is trying to do?  Does the patient experience the therapist as a person who is trying to be helpful?  Intersubjectivity does not reduce the value of traditional concepts like transference and countertransference.  Both can exist in this intersubjective space.  It provides a more comprehensive framework for understanding.   

Intersubjectivity has developmental origins, is considered adaptive from an evolutionary perspective, and therefore most people have it to one degree or another. An exception might be autism where the absence of an intersubjective process has been considered as a deficit or a defense.  Along the same lines varying degrees of severe mental illness can impact it.   

In a therapy session, the process and content of the session are co-created rather than being dependent on the therapist.  As the therapy progresses the process may be more important than the content.  This is an obvious departure from criteria based diagnoses and highlights the social determinants of the problem. The underlying assumption of how the mind operates on an intersubjective basis is that the primary goal is to form object relations or real relationships and their internalized representations. That differs from some other assumptions of mind goals such as discharge for pleasure.  Like many technical terms used in therapy there is often confusion based on how they are used by different authors.  For example, when I have written about empathy on this blog I have used Sims very precise definition (par. 10). In the chapter I have referenced here, Stern suggests that intersubjectivity subsumes many dimensions including all the imprecise definitions of empathy, sympathy, and mind reading as ways to appreciate the subjective experience of another.  To further complicate matters, there are other descriptions of this phenomenon that are difficult to separate.  One is folk psychology which is defined as the intuitive way people understand and predict the behavior of others. Folk psychology (3) could be seen as the result of a long series of intersubjective encounters – the success of which will depend on both the quality of the interactions and the inherent properties of the subjects.

Before I get too far afield, I will add a brief comment about confusion over the objective and subjective in psychiatry. When physicians start out, the objective is highly valued.  What are the reproducible elements of diseases and treatments? Physicians leave medical school with a sense of medical science being like any other science until they start practicing and realize they are seeing hundreds of conditions that defy description and standard treatments.  In psychiatry there has been an historic move from an attempt at the highly objective approaches of the late 19^th century to the subjective wave of psychoanalytical dominance and back to the attempted objectivity of brain-based precision psychiatry.  That pendulum swing is more rhetoric than reality.  The reality is that in psychiatry we are privileged to work with the most complex organ in the body.  The brain has an obvious complex physical basis and an equally complex psychological basis.  Both must be understood as completely as possible.  That is difficult in that it takes a lot of time and effort – but that is the job.

On a practical note, what about the rest of the book and should you buy it?  I was pleasantly surprised to find what I have done for decades was recommended by an academic psychiatrist who is an expert in the field.  I am certain that most people who have been engaged in providing psychotherapy will find the same thing.  The overall advantage in this book is that it is an information dense text of 115 pages with additional pages for 84 references, an index, and a glossary.  It is set in what appears to be 10-point font and you can read it in one long sitting.  There is no elaboration on the history and technical details of schools of psychotherapy.  The chapters are matter-of-fact and straightforward. Every concept has a concise definition and definitions are added as needed as footnotes on the respective pages.  Since the author is a psychotherapy researcher there are some unique conceptualizations and jargon contained in the book.  There were well explained and not an impediment to understanding.  

The model of therapy described is described as an integrative cognitive-psychodynamic approach that consider both conscious and unconscious elements.  He takes the secret handshake elements out of psychotherapy by clearly stating what he is doing and providing many clinical examples.  When therapists are starting out especially in psychodynamic therapy – the goal of therapy is often not very clear.  It can seem like therapy hinges on definitive interpretations of unconscious wishes and the residuals of past interactions.  Even when a therapist gets to the point where they feel more competent to make those interpretations, they may be skeptical of their accuracy and concerned that they be trying to convince the patient to accept an inaccurate interpretation.  Dr. Horowitz is very clear that interpretations are not necessary for change and reviews several cognitive and behavioral interventions that can be useful. I counted about 39 of these interventions in the obvious places, but there are probably more.  In some spots it assumes that the reader has working knowledge of basic behavioral interventions (breathing techniques, relaxation, etc) for application in the early stages.   

This method of therapy – supportive interventions used initially and intermittently in association with more interpretive therapy is often not explicit in therapy texts, but I am convinced that it is the norm for people who learn psychodynamic therapy and apply it outside the context of psychoanalysis.  There are clearly times when people being seen strictly for therapy or psychiatric treatment are in crisis and need supportive interventions for stabilization or to assist them toward an intersubjective state consistent with more exploration and interpretation.

 The book benefits therapists at both ends of the training and practice spectrum.  If you are starting out – it is a good overview of the topics and skills that you need to provide psychotherapy.  If you have been working in the field for years or decades, it leads to reflection on what you have been doing, whether there is potential for improvement, and how what you are doing fits into the general scheme of things.

Either way Dr. Horowitz does not disappoint.     

 

George Dawson, MD, DFAPA

 

References:

1:  Horowitz MJ.  Clinician Technique in Personalized Psychotherapy.  American Psychiatric Publishing, Inc, Arlington, VA, 2025.

2:  Stern D.  Intersubjectivity.  In: Person ES, Cooper AM, Gabbard GO. The American Psychiatric Publishing Textbook of Psychoanalysis.  American Psychiatric Publishing, Inc, Arlington, VA, 2005, 77-92.

3:  Hutto, Daniel and Ian Ravenscroft, "Folk Psychology as a Theory", The Stanford Encyclopedia of Philosophy (Fall 2021 Edition), Edward N. Zalta (ed.), https://plato.stanford.edu/archives/fall2021/entries/folkpsych-theory

 

An MGH Case For Acute Care Psychiatrists

 

I have been a New England Journal of Medicine (NEJM) subscriber since I left medical school.  It was a recommended practice in my first year Biochemistry class by the distinguished professors in that class. We had very close contact with them in my medical school for two reasons – daily seminars where we discussed research papers in the applicable topics and their graduate students talking the same course.  It was one of the more intellectually stimulating courses in medical school.   

Over the subsequent 43 years of subscribing, I have noticed a couple of trends.  The most significant one is that psychiatry has been increasingly represented on the pages especially in the past 10-15 years. It is more likely that authors and discussants in the weekly case presentations will be psychiatrists.  You can also get updates on relevant psychiatric papers sent by email. If you scan the table of contents each week it is likely that 2 or 3 papers will be relevant to psychiatric practice – more if you are a neuropsychiatrist or medical psychiatrist.

That brings me to Case 22-2025 from the July 25, 2025 issue.  I will describe the case as briefly as possible due to copyright considerations and the fact you can read all the details in your medical library copy or access.  I want to focus on the diagnostic process and what it implies for both psychiatric diagnosis and treatment.  I also want to focus on the fact that there are acute care psychiatrists in intensely medical settings and they are very knowledgeable and take care of very tough problems that nobody else does.  That can get lost on an almost daily basis as you see provocative headlines and social media posts seeking attention by distorting what psychiatrists do and what they are capable of.

The patient is a 19-year-old woman admitted with episodic right arm and leg shaking and unusual behavior.  The symptoms developed over the 10 days prior to admission with episodic shaking and numbness of the right arm, and slowed speech. A week before admission she collapsed in public and full body shaking was observed.  In the emergency department she was noted to be drooling, confused, and had bitten her tongue.  She gradually became more alert.  In the MGH ED her exam was normal and the only remarkable lab finding was an elevated lactate.  CT and MRI of the brain were normal.  An EEG was normal.  On day 1 she had sudden onset of intense fear and dread followed by whole body shaking lasting 1-1 ½ minutes.  With the last episode she had a decreased oxygen saturation to 50%.  She was started on lorazepam and levetiracetam.  On day 2 she was started on lamotrigine. She was also seen by a psychiatrist and was noted to have extension and stiffening of the right arm, flexion and stiffening of the left arm, turning the head to the right and whole-body stiffening. The episode lasted a minute and she described feeling like “brain and mind were disconnected”. She denied hallucinations, suicidal ideation, and aggressive ideation but did not think that she could return to college.  She became more agitated, tried to run out of her room, and thought the staff were trying to kill her.  She became agitated and required physical restraint and IM olanzapine.

Additional history was remarkable for a grandfather with schizophrenia and past treatment for anxiety and depression – most recently with psychotherapy and no medication.  Following a recent discharge from another hospital and a 5-day admission she was taking levetiracetam, lamotrigine, melatonin, and folic acid. She was rehospitalized after she developed symptoms on the way home from that hospitalization.

This is a severe and acute problem that every acute care psychiatrist should be able to analyze and treat.  The patient exhibits seizure like activity, catatonia, and psychosis in the form of disorganized behavior rather than any descriptions of hallucinations or delusions.  The concern about hospital staff trying to kill her could be paranoia – but unless there is corroboration that it was present for some time – it can also be due to the significant cognitive problems of poor memory and inattentiveness. 

In the subsequent discussion and unfolding events – Judith A. Restrepo, MD – a C-L psychiatrist at MGH presents a refined approach to the problem as outlined in the graphic at the top of this post.  After describing the observed characteristics of the three syndromes on the left she looks at groups of disorders that may account for the syndromes and how common they are.  Since the emergency screening has already been done, she can rule out any associated with obvious abnormalities of brain imaging studies or lab tests.  She goes through each major category and states why a diagnosis is likely of not.  For example, in the Rare Disorders Where Psychotic Sx Are Typical she mentions acute intermittent porphyria and Creutzfeldt-Jakob disease and how they are unlikely due to the illness pattern, lack of GI sx, and a normal EEG. 

In that same category, Dr. Restrepo discusses autoimmune encephalitis as a possibility and eventually lands on that diagnosis.  In the subsequent evaluation (anti-NMDA receptor antibodies, CSF studies, abdominal ultrasound and CT) the diagnosis of anti-NMDA receptor encephalitis secondary to a malignant mixed germ cell tumor of the left ovary was noted.

The case report is useful to read in full because of the complicated post diagnostic course and description of what is known about the treatment of this condition.  I am going to focus on a couple of additional diagnostic issues and the implications for psychiatrists.

Pattern matching remains a critical aspect of all medical diagnoses and that includes psychiatry.  It is still a popular trope that psychiatric diagnosis is DSM centric and nothing could be farther from the truth. The real value of psychiatry is the training and direct observation and assessment of real problems. Reading a checklist of symptoms is essentially worthless without knowing those patterns.  The obvious examples from this case are psychosis, seizures, and catatonia and their many variations.  The wording in the case report is often stacked to cause an association to those patterns.

An obvious example is whole body shaking followed by hypoxemia and an elevated lactate level should lead to an association to generalized seizures and probably similar patients seen in the past with that condition.  Similarly, the features of catatonia should be obvious without referring to a catatonia rating scale and lead to associations to past catatonia patients diagnosed and treated. Superimposed on these diagnostic patterns should be a general pattern of how to approach very ill patients – in this case patients who have either serious psychiatric disorders or psychiatric syndromes secondary to serious medical and neurological conditions. How should a stuporous or comatose patient be examined (2)?   In this specific patient could the arm movements be decorticate or decerebrate posturing?   Could they be a movement disorder?  That should include a triage pattern of how that patient needs to be stabilized until the diagnosis is determined.

As an example – what should happen if this patient is described to you as an admitting psychiatrist?  Should they be admitted to a typical inpatient psychiatric unit?  All that I would need to hear is hypoxemia following seizure like activity and my answer would be no.  They need to be in a unit that has telemetry and critical care nursing and psychiatric units do not.

There are also patterns on the rule-in side.  Are there any features of this illness that match typical patterns of schizophrenia, bipolar disorder, or depression?  Are there any features that match acute intoxication with commonly used substances?  Is the patient medically stable enough to be treated on a psychiatric unit? 

How do we prepare acute care psychiatrists who are based in medical neuroscience?  Thomas Insel the former NIMH director had the idea of a rotating clinical neuroscience fellowship where neurology, psychiatry, and neurosurgery residents would do a 2 year fellowship before moving on their respective residencies.  That is a hard sell when you come out of it needing to do another 3 – 5 years of residency. 

I propose getting ready in medical school.  Most MS4 courses are electives and there is probably room to further modify the MS3 year.  In addition to basic general medicine and surgery I recommend the following electives: neurosurgery, neurology, endocrinology, infectious disease, renal medicine, cardiology, emergency medicine, and allergy and immunology.  Just rotating through is not enough given what I have said about the pattern matching requirement.  As many acute care cases and unusual presentations of psychiatric disorders associated with brain and medical illnesses need to be seen as possible.  Only that will get residents ready to make diagnoses like the one in this case. It helps to have an attitude and interest in treating the most severe problems in psychiatry. 

And once you are out – keep reading the journals including the NEJM.

 

George Dawson, MD, DFAPA

  

Supplementary 1:  Thought I would add a couple of supplementary comments here rather than adding them to the main post.

Psychiatry is not DSM centric:  Practically every critic of the field and by default these days every social media venue and many journals have lengthy debates about the DSM, what it does or not do, and how it affects the future of psychiatry.  The above post is a shining example of what I did in over 35 years of practice.  At no point in that 35 years did I pull the old DSM off the shelf and think: “Gee I wonder what the DSM will tell me to diagnose?”  Just today I was in a seminar where the question was: “What diagnosis should we use for psychotherapy?” and reminded that the primary use of the DSM and why it was invented – initially for statistical and census purposes and now for billing and coding.  In other words, today – you do not get paid if you don’t have a diagnostic code but that code is technically an ICD-10 code and not a DSM code.

Should that role lead to protracted debates in journals and social media.  I guess I will take the lead role is saying emphatically no.  You can take all the debates about the precision and validity of psychiatric diagnoses and watch it explode in this case report. We see real psychiatry in action.  A psychiatry where patterns of illness are recognized and critical in making a diagnosis of a life-threatening condition.

Pathology in a psychiatric case:  I did not mention that this case report contains a pathology report including photos of the gross pathology and microscopic pathology of the left ovarian mass and the malignant germ cell tumor. This is reminiscent of late 19^th/early 20^th century psychiatry looking for neuropathological autopsy correlates of severe mental illness and the famous psychiatrists involved. It was a more intellectually stimulating approach and there were results but not for the major psychiatric disorders leading to asylum care at the time namely bipolar disorder, schizophrenia, and depression/melancholia. Now that the pathology is more specific should psychiatrists be taught the pathology and the pathophysiology of these disorders?  Should they be aware of all paraneoplastic syndromes that cause psychiatric symptomatology.  Of course they should. It is more important than a DSM and unless the DSM is serious about including real patterns and pathology it will be much less relevant in the future.  We have reached a limit when it comes to parsing words about psychiatric diagnoses and need to get back to reality.    

References:   

1:  Restrepo JA, Mojtahed A, Morelli LW, Venna N, Turashvili G. Case 22-2025: A 19-Year-Old Woman with Seizure like Activity and Odd Behaviors. N Engl J Med. 2025 Jul 31;393(5):488-496. doi: 10.1056/NEJMcpc2412531. PMID: 40742263.

2:  Plum F, Posner JB.  The Diagnosis of Stupor and Coma.  FA Davis Company, Philadelphia, 1980.