03. Can GLP-1 Agonists Elevate Lithium Levels?

Text version

One of the more critical mistakes we as clinicians can make is to fail to monitor lithium levels appropriately. This is due to lithium’s relatively narrow therapeutic index. The clinically effective serum range from 0.6 to 1.2 is only slightly below the level for toxicity of 1.5.

Even after a stable dose has been achieved, regularly scheduled checks every 6 to 12 months may miss changes in kidney function, salt or water balance that can affect lithium levels. Often, these changes are induced by the addition or discontinuation of concurrent medications including diuretics, NSAIDs and ACE inhibitors among many more.

As new medications enter common practice, they can have unexpected interactions. A new paper published in the Journal of Clinical Psychopharmacology reports three cases of elevated lithium levels after the addition of semaglutide with two showing signs of lithium toxicity.

DIPS: Assessing Drug Interaction Probability

Since a complete description of each case exceeds the scope of a Quick Take, the best way to summarize is with the Drug Interaction Probability Scale (DIPS) [1]. The DIPS is a 10-item checklist to assess the likelihood of a drug-drug interaction based on various factors.

These factors include:

• The presence of previous credible reports
• The time course
• Persistence during de-challenge and rechallenge
• Alternative causes
• Other objective evidence
• Dose-response changes

The scale ranges from doubtful for scores less than 2 to highly probable for scores over 8.

Case 1: Dehydration From Reduced Oral Intake?

The most compelling elements in the first case is that the patient, a 62-year-old man with bipolar disorder, had been psychiatrically stable on lithium for four decades. This included the last six years with a constant dose of 600 mg ER each night with a baseline lithium level of 1.0. His baseline creatinine was 1.41.

After starting semaglutide, he was eating less with a reduction in fluid intake by half. Two weeks later, he had the onset of unclear thinking, difficulty focusing, unsteadiness, worsening tremor and drowsiness.

Upon arrival at the hospital, his serum lithium was 1.8. His creatinine declined slightly to 1.29. Relevant comorbid factors included stable chronic kidney disease stage 3a and type 2 diabetes along with other medications but no recent changes. The semaglutide was stopped and lithium was restarted after discharge.

The first case received a DIPS score of 4 indicating an interaction between semaglutide and lithium was possible.

Case 2: Recurrent Elevation on Rechallenge

The second case described a 22-year-old woman with bipolar disorder who had also taken lithium for an extended time—since she was 9 years old—with a relatively constant dose of 900 mg ER. She had no previous episodes of toxicity.

She had taken semaglutide on and off over two years with two instances of elevated lithium while taking semaglutide. The first occurred a month prior to presentation without signs of toxicity. She had stopped the semaglutide but restarted it again with a larger dose eight days before presentation.

Five days before presentation, she had the onset of dizziness, abdominal pain, tremors, visual disturbances, nausea, vomiting and decreased appetite. On presentation to the hospital, her lithium level was 2.1 with a creatinine of 1.1, slightly above her baseline of 1.0.

Comorbid factors included a BMI of 62 and hypothyroidism along with other medications, again, without recent changes. Semaglutide was stopped and the lithium restarted after discharge.

The DIPS score for this case was 3 indicating a possible interaction.

Case 3: Increased Levels Without Overt Toxicity

The third and last case yielded the highest DIPS score of 5 indicating a probable interaction. This case involved a 22-year-old man with unspecified bipolar disorder.

As the dose of semaglutide was started and titrated upward, the dose of lithium was tapered downward from 2100 mg ER daily with a level of 1.1 to a dose of 1200 mg daily with a level of 0.8. The authors argue the serum level of lithium should have been lower after a proportionally larger reduction in dose indicating an interaction with semaglutide.

Unlike the other cases, the patient did not show signs of toxicity—nausea, vomiting or diarrhea. Creatinine did not deviate significantly from 1.1. Over two months, his weight decreased from 133 kg to 122 kg. Of note is he was also taking olanzapine among other medications without recent changes.

Potential Mechanisms of Interaction

In scoring the DIPS, 2 points were discounted for each case as the interaction was not consistent with the known properties of either semaglutide or lithium. The authors reviewed the properties of semaglutide to determine if this interaction could have been anticipated.

• A logical mechanism would involve a change in renal function but GLP-1 receptor agonists have beneficial effects on kidney function and increase natriuresis which would lower lithium levels.
• Also, semaglutide is known to delay gastric emptying but since lithium is absorbed in the small intestine, it is difficult to understand any connection.
• Finally, appetite suppression and decreased oral intake caused by semaglutide could lead to weight loss and dehydration resulting in an elevation in lithium levels. The authors also note that 20% of patients taking semaglutide report nausea with 6% reporting vomiting, another cause of dehydration and elevation of lithium levels. This effect would not be unique to semaglutide but could be caused by any nausea-inducing medication.

Clinical Takeaways and Monitoring

In conclusion, the authors present three cases that suggest a possible to probable pharmacokinetic interaction between semaglutide and lithium. In light of the beneficial effect of GLP-1 receptor agonist on metabolic profile, we can expect and hope that more of our patients have access to these medications.

My take-home point is to consider GLP-1 receptor agonists a class of medication that might elevate lithium levels although I will wait for more evidence before pre-emptively reducing the dose of lithium as suggested by the authors. I will, however, review the signs of lithium toxicity with my patients when they start semaglutide along with additional monitoring of lithium levels afterwards.

Additional Reference

1. Horn, J. R., Hansten, P. D., & Chan, L. N. (2007). Proposal for a new tool to evaluate drug interaction cases. The Annals of pharmacotherapy, 41(4), 674–680. https://doi.org/10.1345/aph.1H423

Abstract

Lithium Toxicity and Altered Clearance Following Initiation of Semaglutide in Patients With Bipolar Disorder: A Case Series and Literature Review

Al-Soleiti, Majd, M.D.; Leung, Jonathan G., PharmD, RPh, BCPP; Mubaydeen, Teeba, M.D.; Markota, Matej, M.D.; Abulseoud, Osama, M.D.; Singh, Balwinder, M.D., MS & Sola, Christopher L., D.O.

Background: 

Lithium is a mainstay treatment for bipolar disorder, but its narrow therapeutic index and susceptibility to pharmacokinetic interactions make appropriate monitoring crucial. Glucagon-like peptide-1 receptor agonists (GLP-1 RAs), such as semaglutide, are increasingly prescribed for type 2 diabetes and weight management. Scarce evidence exists on the potential interaction between semaglutide and lithium.

Methods: 

We present 3 cases involving patients on stable lithium regimens who were initiated on semaglutide, reviewing potential mechanisms underlying the interaction between them.

Findings: 

In 2 cases, lithium levels increased significantly, leading to toxicity despite stable renal function and no changes in concurrent medications. In the third case, preemptive reductions in lithium dosage mitigated toxicity, although lithium levels remained higher than anticipated. Mechanistic hypotheses that might contribute to semaglutide-associated elevated lithium levels include altered kidney function, dehydration from reduced oral intake, vomiting, or diarrhea, and delayed gastric emptying.

Conclusions: 

To our knowledge, this is one of the first documented case series describing a potential interaction between semaglutide and lithium in the medical literature. These cases underscore the importance of vigilant monitoring when combining lithium with semaglutide, and potentially other GLP-1 RAs. Baseline renal function, hydration status, and lithium levels should be assessed before initiating semaglutide, and lithium levels should be monitored more frequently during therapy. Clinicians prescribing semaglutide to patients on lithium should exercise caution, monitor for signs of toxicity, and provide appropriate patient education. Further research is needed to elucidate the mechanisms of this potential interaction and its clinical significance.

Reference

Majd, A. M.D.; Jonathan G, L. PharmD, RPh, BCPP; Teeba, M. M.D.; Matej, M. M.D.; Osama, A. M.D.; Balwinder, S. M.D., MS & Sola, Christopher L., D.O. (2025). Lithium Toxicity and Altered Clearance Following Initiation of Semaglutide in Patients With Bipolar Disorder: A Case Series and Literature Review. Journal of Clinical Psychopharmacology 45(6):p 613-618.