Bipolar disorder (BD) is a chronic, recurrent mental disease that represents a significant problem to the affected individual, his family and society. According to the World Health Organization (WHO), it is one of the ten main causes of disability 1. Furthermore, suicidal behavior is one of its most common outcomes. Around 25%-50% patients with BD attempt suicide, among which 15% are successful 2.

Amongst the pharmacological therapeutical options, the Food and Drug Administration (FDA) have approved lithium carbonate for the treatment of manic episodes since 1970 3, and it is still indicated as the first-line drug for all phases and subtypes of BD. It is also recommended as an adjuvant to the treatment of antidepressant-resistant unipolar depression 1, 4, 5. Moreover, lithium’s chemical properties play a key role in suicide prevention, the reason why it is considered an anti-suicide drug for people with mood disorders, regardless of its action in controlling acute manic or depressive episodes 2, 4, 5, 6.

Despite countless pieces of evidence on lithium’s effectiveness in controlling BD, there are still many questions to be answered concerning its safety and tolerability, especially when it comes to its adverse effects on the kidneys. Thus, its capability of causing arginine vasopressin resistance (AVP-R), formerly known as nephrogenic diabetes insipidus (NDI), is well known, but the rates vary considerably from 20% to 87% amongst studies and whose emergence can occur from two to four months since the beginning of lithium treatment 3, 7, 8, 9, 10, 11. Conversely, the ability to cause chronic kidney disease (CKD) is much more intriguing, because the literature has many studies showing both positive and negative causality 3, 7, 8, 9, 10, 11.

The mechanism through which lithium may cause chronic kidney disease has not yet been fully elucidated, and chronic use of the medication has been associated with various types of renal injury, the main one being chronic tubulointerstitial nephropathy 12, 13.

Lithium enters the principal cells of the collecting duct via sodium epithelial channels in the luminal membrane. It is believed that, during chronic use, the medication increases the resistance of principal cells to the antidiuretic hormone (ADH), thereby reducing water permeability through aquaporin-2 water channels (AQP-2) 14. Among the proposed mechanisms to explain this phenomenon is lithium's stimulation of cyclooxygenase-2 super expression, which would lead to increased urinary excretion of prostaglandin E2 (PGE2) by medullary interstitial cells. These prostaglandins, in turn, induce degradation of AQP-2 water channels and consequently reduce the renal capacity to modulate urine concentration 15, 16.

It is also proposed that lithium may reduce AQP2 gene transcription independently of PGE2 action, and cause remodeling of the collecting duct with a reduction in the relative number of principal cells compared to intercalated cells, further reducing renal function 17.

Indeed, over the years, it can be noticed that the 1980s and 1990s initial studies were more indicative of a lithium-induced nephrotoxicity9, 10. More recent studies, however, showed lesser rates of significant kidney impairment, probably due to greater methodological refinement. But, even taking these differences into account, the weight of scientific evidence still suggests that this salt, abundantly found in various natural elements, has some potential of causing kidney disease, promoting chronic tubulointerstitial nephritis, particularly if the exposure period lasts more than 20 years. It is worth mentioning that, regarding the development of end-stage renal disease (ESRD), a greater exposure period of 27 years or more, according to some research, might be necessary 11, 18.

Nevertheless, beyond the time of exposure, other factors directly related to lithium, such as the average dosage used in the recurrence prevention period, the interaction with other substances potentially nephrotoxic, lithium intoxication in higher doses and the development of AVP-R also must be taken into consideration while evaluating the nephrotoxicity risk 18. Another relevant and unsolved point is that, even in the absence of lithium usage, mental disorders patients generally show a significant reduction in the urinary concentration, that is, the result may derive from the presence of other risk factors, such as smoking, dyslipidemia, diabetes mellitus, and hypertension, comorbidities that are frequently found in people with mental disorders 19.

Therefore, in sight of the diverging results obtained from lithium usage and kidney function studies, more research was deemed necessary. In this study, we investigated the Estimated Glomerular Filtration Rate (eGFR) of a group of patients with BD who were being treated with lithium, with or without the addition of other psychopharmaceuticals. We also evaluated the relationship between eGFR and the sociodemographic and clinical characteristics of these individuals.

A total of 121 patients were recruited, 26 were excluded due to not being treated with lithium. The characteristics of the selected sample are presented in Table 1. The median age was 49.0 (IQR: 37.0-57.5); 80% was female, 75 (78.9%) were self-declared black/mixed race. The majority of patients (85/89,5%) were diagnosed with BD type I. The median duration of the lithium treatment was 10 years (IQR: 6 - 16) and the median serum lithium level described on the medical records was 0,70 mmol/L (IQR: 0,6 - 0,9).

A total of 69 (72.5%) patients exhibits values of eGFR above 90 mL/min/1,73m²; 21 of them (22,1%), between 60 and 90 and 5 (5,3%) patients showed values under 60 (Table 1).

In the univariate analysis, age (p < 0.01), smoking (p = 0.01), use of BZD (p = 0.01), and sex (p = 0.07) met the criteria of p <0.1 (Table 2).

In the multivariate analysis, the final model was composed of serum lithium levels, duration of the lithium treatment, age, smoking, use BZD and sex. Higher serum lithium levels were associated to low eGFR levels (β = -18,06 [-34,70 to -1,42], p = 0,03); age was also associated with eGFR levels (β = -0,72 (-1,10 to -0,33), p = <0,01) (Figure 1).

This study evaluated the renal function of patients with BD by measuring the eGFR, calculated using the CKD/EPI formula, and the results showed that higher serum lithium levels and advanced age are associated with the significant decrease of the eGFR, even though the latter has already been extensively described in the literature for both users and non-users of lithium 22. Conversely, although factors like smoking, systemic hypertension and diabetes mellitus are classically associated with kidney dysfunction, this finding was not replicated in our sample, probably due to the small percentage of individuals with these comorbidities, or to their young age, that obviously reduced the likelihood of exposure 22, 23.

A multicentric study assessed the relationship between use of lithium and eGFR and found that higher levels of lithium were linked to decreased levels of eGFR in patients with BD aged 20-89 (mean 56) and treated with lithium for 8-48 (mean 18) years. Additionally, the authors described a positive correlation between daily lower doses of lithium and higher eGFR. In this case, the lithium dosage could have been lowered to compensate for decreasing eGFR with older age and keep serum lithium levels within the recommended therapeutic range 23. This result, as well as the one described in our study, reinforces the importance of considering the measurement of serum lithium levels during patient follow-up (maintenance phase) as an independent variable to be controlled in the prevention of damage to renal function in chronic users of lithium.

The association between higher serum lithium levels and reduced eGFR could be explained by impaired kidneys having a diminished capacity to metabolize lithium and, therefore, its serum levels increase; or higher levels of lithium could cause acute or chronic kidney damage, reducing eGFR 24. These two hypotheses do not exclude one another. In this sense, it is extremely important to elucidate if there is any component in lithium salts that can have a direct deleterious effect on the renal parenchyma, leading to some degree of nephrotoxicity, and know when lithium should be discontinued or not 25.

Contrary to what is described in the literature, we did not find an association between time of exposure to lithium and a decrease in eGFR 24. However, our study sample has a median duration of lithium treatment of 10 years, which is much shorter than that described in other studies that present values above 20 or more years of continuous therapy 23.

An Italian cohort from 1980 to 2012 evaluated 953 patients with BD and found maintenance therapy with lithium was associated with a decrease in eGFR to levels below 60mL/min/1.73m² in 50% of the sample, indicating some degree of impairment of renal function 26. The same study found female sex, age and exposure time to lithium as risk factors to decrease in eGFR. One strange point is this study is that over 4 years of follow-up, it was also found that the progression of renal failure after diagnosis of eGFR reduction to < 45mL/min/1.73 m2, did not differ between 3 subgroups observed later: 1 – which continued with lithium in the same dose as before; 2 – who discontinued its use at the time of diagnosis of the renal impairment; and 3 – who maintained the blood lithium at concentrations below the therapeutic range 11, 26. These results are difficult to explain, but they bring hope, as it may be reasonable to maintain lithium in patients with BD, even when renal function is already moderately compromised. But it is important to highlight that we don’t have clinical trials designed for answers these doubts. Furthermore, considering that the treatment with lithium is one of the most effective, greater efforts must be made to prevent and control the onset of kidney impairment in this population.

Therefore, although the precise details regarding the relationship between lithium usage and renal disease require further elucidation, existing literature consistently indicates a certain degree of reduction in renal function associated with long-term exposure to lithium. To address the existing uncertainties, it is imperative to conduct more methodologically rigorous longitudinal studies. Specifically, comparisons should be made between samples of bipolar disorder (BD) patients who do not use lithium and those who do, in order to shed light on this matter. These comprehensive investigations will contribute to a better understanding of the potential impact of lithium on kidney function and inform clinical decision-making regarding the management of BD.

 This cross-sectional data extracted from an ongoing cohort at a Specialized Center for the Treatment of BD patients revealed that serum lithium levels is associated, as an independent variable, with a reduction in eGFR. This result raises the suspicion that it may be a candidate to be an early marker of nephrotoxicity. Furthermore, the duration of treatment with lithium for a period of 10 years was not associated with a reduction in eGFR, a result that reinforces the theory that longer periods of exposure to lithium are necessary for there to be significant impairment of renal function.