Future use of a 12-hour standardized lithium serum concentration notation: the first step towards more interpretable scientific literature and the return of the therapeutic importance of lithium.
If I asked you to meet me at a restaurant, but never mentioned a time for us to meet, it would be a pretty useless invitation. A place without a time, (or a time without a place), is incomplete information. Space and time are inherently linked.1
So, consider this extract from Nolen et al2:
“For maintenance treatment of BD several reviews recommend differing minimum effective lithium serum levels ranging from 0.40 mmol/L and 0.50 mmol/L or 0.60 mmol/L…to as high as 0.80 mmol/L.”
The author has given the equivalent of a place (a lithium serum concentration range) without a time. How can we accurately interpret these values without knowing when they were taken, how long it has been since the last dose of lithium, and what the dosing schedule is?
You might also be wondering now, how common is this problem in the psychiatric literature?
Unfortunately, the answer is that nearly every psychiatric research paper that mentions a lithium serum concentration neglects to mention a time. For example, from Camus et al3:
“In affective disorders, the generally accepted steady-state therapeutic range of plasma Li+ concentrations is 0.6–1.2 mmol/l.”
From Vestergaard & Licht4:
“The therapeutic index of lithium is narrow. With serum lithium levels below 0.4 mmol/l, it is questionable whether lithium exerts any measurable therapeutic effects…With levels above 0.8 mmol/l, most patients experience unwanted side effects, and with levels above 1.2-1.5 mmol/l, signs of intoxication may appear.”
From Mauer et al5:
“Although standard lithium concentrations of 0.6–0.8 have the most benefit for enhancing neuronal viability, even ‘low’ levels of 0.2–0.4 have that benefit.”
These authors are most likely referencing the lithium serum concentration of a blood sample taken from a patient 12-hours (+/- 30 mins) after the previous dose.1 The procedure was first proposed by the Danish psychiatrist Amdi Amdisen in a 1975 paper, which became standard practice soon after.6
Amdisen chose 12 hours because, according to the experimental data he collected, it was the earliest time a sample could be taken to safely avoid the absorption and distribution phase of lithium. Through analysis of lithium elimination half-life data of 226 participants, Amdisen showed that the later a blood sample was taken, the less qualified a control value became as a guide for therapeutic range adjustment and as an indicator of toxicity. As he explains in a paper published in 19807:
“After an interval of 12 hr the serum values in most patients will be below the mean of the 24 hr concentration course of a day on lithium. The great interindividual differences in the steepness of the concentration fall mean that the longer the time interval, the greater the difference of the mean lithium level in patients with the same control value.”
In a 1977 paper8, Amdisen gave a definition for this standardized serum concentration:
“The standardised 12 hour serum lithium concentration (12h-stSLi) is defined as the Li + concentration in the serum from a blood sample drawn in the morning exactly 12 hours (± 30 min) after the last dose of lithium in a patient who has been taking his daily lithium in 2 or more divided doses and who has been taking all his tablets at the scheduled hours during the past 48 hours.”
Within his definition, Amdisen introduced a notation to signify a standardized 12 hour serum concentration, which he used extensively throughout the paper (eg, “a 12h-stSLi around 1.20mmol/L”).
While the practice of taking lithium serum samples from patients at 12 hours was adopted by the psychiatric community, the notation in the literature was not. Amdisen, himself, noted this. In a 1990 paper9, he observed that “Most case reports do not specify the time interval between the last lithium dose and the time of venesection.”
He went on to explain why this was a significant problem:
“Without an accurate specification of the control value (12-hour standard) the term ‘serum lithium level’ is almost impossible to interpret…individual patients on the same lithium dosage but with different half-lives would have different mean levels and show different concentration profiles. It should also be acknowledged that the 12-hour serum lithium level is not identical to the mean serum lithium level and, furthermore, the magnitude of that deviation varies with the patient’s half-life. A shorter half-life gives a larger deviation between the 24-hour and the 12-hour serum lithium level.”
It is astounding that the practice of omitting the notation for the standardized 12-hour lithium concentration not only continues today, but has become so widespread. It is remarkably poor scientific practice and has detrimental consequences for the future of research involving lithium serum concentrations.
Many analyses concerning experimental lithium pharmacokinetics seem to have been constrained to the data collection of only 12-hour lithium serum values. Examples include: optimal serum concentration,2 brain to serum concentration ratio,10 optimal dosing regimen,11 and renal effects.12 Collection of lithium serum concentrations outside of 12 hours do not seem to be considered by current investigators as useful. We believe this is a mistake and restricting pharmacokinetic investigation to the 12-hour time point only has resulted in a rudimentary understanding of lithium pharmacokinetics.
This near-exclusive reliance on the 12-hour measurement was not recommended by the pioneers in this area. In Chapter 22 of the Handbook of Lithium therapy, Amdisen noted that “Attention should be drawn to the fact that the specifications of I2h-stSLi have only aimed at isolating a serum concentration which is both sufficiently comparable between patients... and satisfactorily reproducible from time to time within the same patient. Concerning the therapeutic efficacy or the toxic side-effects of lithium, 12h-stSLi has no special significance as might be the case for the minimum, the maximum, or the mean concentration during the 24 hours of the day.”13
Amdisen went on to say “that a multitude of 'natural' influencing factors is continuously at work within the same subject, resulting in a limited, but substantial, inaccuracy in the reproducibility of 12h-stSLi from time to time…Therefore, a single 12h-stSLi should never be evaluated in isolation but only in the light of the trend in previous values and, in the event of a limited but apparently substantial deviation, the measurement should be immediately repeated at least twice.”13
One data point does not provide sufficient information for rigorous scientific analysis. Any conclusions drawn from comparison between patients are limited in nature and susceptible to high levels of noise. The limited experimental data available for periodic lithium serum samples taken over the course of a dose, show that pharmacokinetic curves can differ significantly between patients.14 One of us (AM) has previously made a case that lithium pharmacokinetic curves over a range of doses could be generated and simulated15 to show clinicians what the pharmacokinetic curve looks like and the times at which the concentrations peak. We are of the view that there is need to move beyond the 12-hour time point to include data from other time points.
Even 2 data points in the elimination phase of the lithium pharmacokinetic curve, while nowhere near optimal, provide crucial benefits over 1 data point. A gradient can be measured which gives the elimination half-life for each patient. Assuming elimination phase serum concentrations obey exponential decay, or concentrations follow a straight line on a semi-log plot, it does not matter when the 2 serum samples are taken for each patient (as long as there is sufficient time in between the 2 measurements to minimize error from natural fluctuations of serum concentration). Elimination half-lives are time independent, and assuming linear pharmacokinetics, dose independent. Therefore, comparison between patients has much greater validity.
For analytical and replicability purposes, it is clear that conducting such research in the future requires explicit and meticulous record of the time of measurement. However, we believe that the lack of 12-hour notation in the contemporary literature has led to a lack of curiosity among practitioners regarding the gaps in existing pharmacokinetic knowledge. As we have discussed, researchers such as Amdisen were acutely aware of the need to gather comprehensive pharmacokinetic data, but subsequent generations have taken the 12-hour measurement to be a matter of settled science.
The problem is also relevant to lithium toxicity case reports where patient serum samples are more likely to be taken at times other than 12 hours post dose. The patient may not be even able to communicate the time of the previously ingested lithium dose. If temporal information is not included in the case reports, the serum lithium concentration values can be easily misinterpreted by readers.
Many current guidelines (as well as some academic papers) allow for a broader window for lithium blood samples to be taken. For instance, a paper by Tondo et al16 states that “lithium remains unique in not being dosed adequately by the mg dose of drug given per day, but instead by achieving serum concentrations 10–14 h after the last taken dose.” Worryingly, the lithium patient information booklet (the “purple book”),17 given to every UK patient taking lithium, advises an open window of “at least 12 hours” to take a blood test. In the light of such recommendations, it may not always be accurate to assume that a lithium serum concentration value mentioned in a clinical report was obtained at the 12-hour mark.
Lithium prescriptions in most countries are slowly declining18 despite the clear evidence for the effectiveness of lithium therapy in treating bipolar disorder.19,20 It is postulated that some of the reasons for the decline are: the drug is seen as dangerous/difficult to manage,21,22 reduced marketing compared to newer medications,23 and conflicting evidence regarding the long term renal effects of lithium.24
Fundamentally, however, for psychiatric and public confidence in lithium therapy to improve, rigorous scientific practice is essential. The simple misuse of notation is more than enough to cause uncertainty and confusion in the reader, especially if they are unfamiliar with lithium pharmacokinetics.
In the 2020 article “Make lithium great again!”, Malhi et al states23:
“Lithium has never been under patent…so there has never been a pharmaceutical company to promote it per se. Instead, its therapeutic importance has relied solely on empirical evidence and the scientific literature.”
Future use of a 12-hour standardized lithium serum concentration notation by authors, whether Amdisen’s “12h-stSLi” or another, would be a first step towards more interpretable scientific literature and the return of the therapeutic importance of lithium.
Dr Mendelsohn is a current psychiatric patient residing in the UK. He holds a doctorate in physics and writes using a pseudonym at The Anxious Physicist. Dr Aftab is a clinical assistant professor of psychiatry at Case Western Reserve University and a member of the Psychiatric Times editorial board. He writes online at Psychiatry at the Margins.
Dr Mendelsohn and Dr Aftab have no conflicts of interest to disclose.
References
1. Carroll S. How to think about relativity’s concept of space-time. Quanta Magazine. November 14, 2022. Accessed February 20, 2024. https://www.quantamagazine.org/how-to-think-about-relativitys-concept-of-space-time-20221114/
2. Nolen WA, Licht RW, Young AH, et al. What is the optimal serum level for lithium in the maintenance treatment of bipolar disorder? A systematic review and recommendations from the ISBD/IGSLI Task Force on treatment with lithium. Bipolar Disord. 2019;21(5):394-409.
3. Camus M, Henneré G, Peytavin G, et al. Comparison of lithium concentrations in red blood cells and plasma in samples collected for TDM, acute toxicity, or acute-on-chronic toxicity. Eur J Clin Pharmacol. 2003;59(8-9):583-587.
4. Vestergaard P, Licht RW. 50 Years with lithium treatment in affective disorders: present problems and priorities. World J Biol Psychiatry. 2001;2(1):18-26.
5. Mauer S, Vergne D, Ghaemi SN. Standard and trace-dose lithium: a systematic review of dementia prevention and other behavioral benefits. Aust N Z J Psychiatry. 2014;48(9):809-818.
6. Amdisen A. Monitoring of lithium treatment through determination of lithium concentration. Dan Med Bull. 1975;22(7):277-291.
7. Amdisen A. Serum concentration and clinical supervision in monitoring of lithium treatment. Ther Drug Monit. 1980;2(1):73.
8. Amdisen A. Serum level monitoring and clinical pharmacokinetics of lithium. Clin Pharmacokinet. 1977;2(2):73-92.
9. Amdisen A. Lithium neurotoxicity—the reliability of serum lithium measurements. Hum Psychopharmacol. 1990;5(3):281-285.
10. Komoroski RA. Biomedical applications of 7Li NMR. NMR Biomed. 2005;18(2):67-73.
11. Carter L, Zolezzi M, Lewczyk A. An updated review of the optimal lithium dosage regimen for renal protection. Can J Psychiatry. 2013;58(10):595-600.
12. Schoot TS, Molmans THJ, Grootens KP, Kerckhoffs APM. Systematic review and practical guideline for the prevention and management of the renal side effects of lithium therapy. Eur Neuropsychopharmacol. 2020;31:16-32.
13. Amidsen A. Monitoring lithium dose levels: clinical aspects of serum lithium estimations. In: Johnson FN, ed. Handbook of Lithium Therapy. Springer Netherlands; 1980:181.
14. Hunter R. Steady-state pharmacokinetics of lithium carbonate in healthy subjects. Br J Clin Pharmacol. 1988;25(3):375-380.
15. Mendelsohn A. Psychiatry needs more simulations: the case of serum lithium concentrations. Psychiatry at the Margins. July 3, 2023. Accessed February 5, 2024. https://www.psychiatrymargins.com/p/psychiatry-needs-more-simulations
16. Tondo L, Alda M, Bauer M, et al. Clinical use of lithium salts: guide for users and prescribers. Int J Bipolar Disord. 2019;7(1):16.
17. Lithium Therapy - Important information for patients. National Patient Safety Agency; National Health Service. December 2009.
18. Pérez de Mendiola X, Hidalgo-Mazzei D, Vieta E, González-Pinto A. Overview of lithium’s use: a nationwide survey. Int J Bipolar Disord. 2021;9(1):10.
19. Anmella G, Vieta E, Hidalgo-Mazzei D. Commentary on: “make lithium great again!” Bipolar Disord. 2021;23(1):90-91.
20. Severus E, Taylor MJ, Sauer C, et al. Lithium for prevention of mood episodes in bipolar disorders: systematic review and meta-analysis. Int J Bipolar Disord. 2014;2:15.
21. Malhi GS, Bell E, Jadidi M, et al. Countering the declining use of lithium therapy: a call to arms. Int J Bipolar Disord. 2023;11(1):30.
22. Rybakowski JK. Challenging the negative perception of lithium and optimizing its long-term administration. Front Mol Neurosci. 2018;11:349.
23. Malhi S, Bell, E, Boyce P, et al. Make lithium great again! Bipolar Disord. 2020;22(4):325-327.
24. Hidalgo-Mazzei D, Mantingh T, Pérez de Mendiola X, et al. Clinicians’ preferences and attitudes towards the use of lithium in the maintenance treatment of bipolar disorders around the world: a survey from the ISBD Lithium task force. Int J Bipolar Disord. 2023;11(1):20.