Journal Article Annotations
2023, 3rd Quarter


Annotations by Sahil Munjal, MD
October, 2023

  1. Absolute and Relative Risks of Kidney Outcomes Associated With Lithium vs Valproate Use in Sweden.

PUBLICATION #1 — Psychonephrology

Absolute and Relative Risks of Kidney Outcomes Associated With Lithium vs Valproate Use in Sweden.
Alessandro Bosi, Catherine M Clase, Laura Ceriani, Arvid Sjölander, Edouard L Fu, Björn Runesson, Zheng Chang, Mikael Landén, Rino Bellocco, Carl-Gustaf Elinder, Juan Jesus Carrero.


The finding:
This cohort study compared patients newly starting lithium with those newly starting valproate and found no difference in patients’ risk of chronic kidney disease (CKD) progression, acute kidney injury (AKI), albuminuria, or yearly GFR decrease. However, people with higher lithium levels (>1.0 mmol/L) were at elevated risk of CKD progression and AKI.

Strength and weaknesses:
This is a large study (>10,000 individuals) with an intention-to-treat design, inclusion of albuminuria as a novel study outcome, and use of rolling 1-year assessments to evaluate long-term toxic effects. The generalizability of the study may be limited as the study population was limited to one city in Sweden with limited diversity. There may have been confounding as valproate has broader indications for treatment initiation. The accuracy of trough lithium levels was hard to determine. Approximately 15% of patients in each group had at least 1 dispensation of the other drug. The median duration of follow-up was short (4.5 years) relative to the natural history of lithium nephropathy.

Lithium has been implicated over multiple studies with a clinically significant decrease in renal function. Even though initiation of lithium in this study was not associated with a significant decrease in renal function compared to valproate, higher lithium levels (levels >1.0mmol/L) did increase renal risk. We reviewed a similar study here by Rej et al in 2021 which showed a 14% increased risk of renal decline in older patients(>65years) on lithium, compared to valproate, suggesting age to be a risk factor in addition to duration and higher levels. The median age in this study was 45 years. In both studies, absolute risk remains low which helps frame a risk/benefit analysis given the effectiveness, neuroprotective, and anti-suicidal properties of lithium in bipolar patients.

Type of study (EBM guide):
Cohort study