You are here
Home > Blog > Aging in Place > Lithium Effects On Renal Function In Geriatric Patients

Lithium Effects On Renal Function In Geriatric Patients

Lithium Effects On Renal Function In Geriatric Patients


The practice of augmenting antidepressant therapy with lithium, known as Lithium Augmentation (LA), is considered a primary treatment option for individuals suffering from Treatment-Resistant Depression (TRD). However, in the case of geriatric patients, this approach is notably underutilized due to concerns surrounding potential kidney toxicity. This study endeavors to investigate changes in the estimated glomerular filtration rate (eGFR) and occurrences of acute kidney injuries (AKI) associated with LA, comparing geriatric and non-geriatric patients.


In the framework of a prospective multicenter cohort study, the study monitored changes in eGFR in a cohort of 201 patients diagnosed with unipolar depression, with 29 of them being geriatric (aged 65 years or older) and 172 non-geriatric patients (aged less than 65 years). This assessment was conducted both at the baseline and over a span of 2 to 6 weeks during the course of LA treatment. To analyze these eGFR fluctuations, linear mixed models were employed, and the incidence of AKIs was evaluated according to the guidelines established by Kidney Disease: Improving Global Outcomes (KDIGO).


The findings of this study unveiled a significant decline in eGFR for both the geriatric and non-geriatric patient groups during the LA treatment regimen. Notably, geriatric patients exhibited a lower eGFR when compared to their younger counterparts. The study also highlighted the impact of lithium serum levels, which can be interpreted as the “effect of LA” on eGFR decline. It was determined that the influence of age group and lithium serum levels on eGFR decline was not interdependent, signifying that the effect of LA on eGFR decline did not significantly differ between the two age groups.


Furthermore, the study identified two instances of acute kidney injuries in the geriatric patient group when serum lithium levels exceeded the therapeutic threshold of greater than 0.8 mmol/L. These findings underscore the importance of closely monitoring lithium levels, particularly in geriatric patients, to mitigate the risk of acute kidney injuries.


This research sheds light on the impact of Lithium Augmentation on eGFR in both geriatric and non-geriatric populations. It emphasizes the need for vigilant monitoring of kidney function, especially in older patients, to ensure the safe and effective use of LA in the treatment of TRD.



Lithium augmentation (LA) is considered the foremost treatment approach for managing treatment-resistant depression (TRD) in both geriatric and non-geriatric patients. Previous research has demonstrated the effectiveness of LA, especially in geriatric individuals with TRD. However, it is essential to recognize that lithium, the key component in this treatment, is predominantly excreted through the kidneys. Consequently, its potential for toxicity is closely tied to kidney function. Given its narrow therapeutic range, regular monitoring of lithium serum levels is imperative to prevent severe lithium intoxication, which can have life-threatening consequences, including acute kidney injuries (AKI).


Geriatric patients, in particular, raise safety concerns regarding lithium use due to age-related declines in glomerular filtration rate (GFR) and a higher incidence of somatic comorbidities that often result in polypharmacy. To mitigate the risk of lithium intoxication, it is crucial to be mindful of potential drug interactions, conduct frequent laboratory assessments, provide patients with adequate instructions, and educate them about possible side effects.


Numerous studies have explored the renal toxicity of lithium, with a predominant focus on patients with bipolar disorders. While some meta-analyses have suggested a minimal impact on kidney function, recent research by Schoretsanitis et al. has indicated an increased risk of kidney impairment with lithium treatment, particularly in older patients and with prolonged treatment duration. 


Despite the acknowledged benefits of LA in geriatric patients, there has been limited investigation into the kidney toxicity of lithium in this age group. To date, only a small number of randomized controlled trials and observational studies have explored the effects of lithium on kidney function in geriatric patients. These studies have reported varied results, including a decline in estimated glomerular filtration rate (eGFR), improvements after discontinuation of lithium, or no significant kidney impairment. Notably, few retrospective studies have focused on acute effects of lithium on eGFR, and these have primarily examined cases of acute lithium intoxication.


As of now, there is a notable gap in the research, with no study comparing changes in eGFR between geriatric and non-geriatric patients undergoing LA. The primary objective of this study is to investigate alterations in eGFR and the incidence of AKIs in geriatric patients in comparison to their non-geriatric counterparts when employing LA as part of the acute therapy for TRD. This prospective multi-center cohort study seeks to address this critical gap in understanding and provide insights into the safety and efficacy of lithium augmentation in the context of geriatric patients.



This study constitutes a prospective observational multicenter cohort investigation carried out within the Berlin Research Network on Depression. It focuses on patients meeting specific inclusion criteria, namely those diagnosed with unipolar depression (as defined by ICD-10 codes F32.1-3 and F33.1-3), aged 18 or older, showing insufficient response to antidepressant treatment over a minimum period of four weeks, and exhibiting a clinical need for Lithium Augmentation (LA). The patients’ eligibility also required a Hamilton Depression Rating Scale (HDRS-17) score equal to or greater than 12, and their informed consent.


For patients receiving co-medication that could potentially be nephrotoxic or influence lithium serum levels, the study mandated that their medication regimen remain stable throughout the treatment period, meaning no new medication or dosage adjustments after the commencement of LA.


Exclusion criteria encompassed contraindications for LA, such as severe kidney insufficiency or chronic kidney disease as per Kidney Disease: Improving Global Outcomes (KDIGO) 2012 definitions. Other exclusions were cases of depressive syndromes stemming from other somatic or psychiatric diagnoses, dementia, substance abuse disorders with less than six months of abstinence, and antisocial personality disorders. Recruitment of patients occurred between February 2008 and February 2020 across 12 psychiatric departments within the Berlin Research Network on Depression.


For the purpose of this study, geriatric patients were defined as those aged 65 or older. Confirmation of the diagnosis of unipolar depression was carried out using the Mini-International Neuropsychiatric Interview. A significant majority of patients were inpatients, and the study received ethical approval. The patients’ somatic comorbidity severity was assessed using the updated Charlson comorbidity index, with hypertension and diabetes mellitus recorded as potential risk factors for kidney function decline based on the International Statistical Classification of Diseases and Related Health Problems (ICD-10) codes from patient records.


During the study, patients were monitored for a maximum of six weeks under LA, a timeframe typically corresponding to their initial hospitalization. All patients underwent a personalized titration of lithium carbonate over the first two weeks, with dosages adjusted based on individual lithium serum levels to attain therapeutic concentrations. The target range was set at 0.5–0.8 mmol/L for non-geriatric patients and 0.4–0.8 mmol/L for geriatric patients. 


Sufficient lithium levels were defined as at least 0.4 mmol/L for geriatric patients and at least 0.5 mmol/L for non-geriatric patients, maintained for a minimum of two weeks. Toxic lithium serum levels were set at 1.5 mmol/L for non-geriatric patients and 1.2 mmol/L for geriatric patients, in accordance with international guidelines. Lithium serum levels were monitored on a weekly basis and recorded as trough levels, immediately before the next dose.


The study also involved laboratory assessments obtained from patients’ medical records, including weekly serum creatinine data, measured through standard laboratory methods. The interval between two measurements was typically one week, ranging from 5 to 9 days based on weekly study visits. Patients had to possess at least one pre-LA and one during-LA laboratory value, with the final creatinine value within a maximum six-week observation period setting the study endpoint. The Chronic Kidney Disease (CKD) Epidemiology Collaboration (CKD-EPI) 2021 equation was utilized to calculate the estimated glomerular filtration rate (eGFR) and classify CKD based on the KDIGO 2012 criteria.


Moreover, the study assessed outcomes related to kidney function, including the occurrence of acute kidney injuries (AKIs). These injuries were categorized by the extent of serum creatinine increase over baseline during LA, specifically as a ≥1.5 fold (AKI1), ≥2 fold (AKI2), or ≥3 fold (AKI3) increase, following the KDIGO 2012 criteria.


Statistical Analysis

To investigate the changes in estimated glomerular filtration rate (eGFR) in response to Lithium Augmentation (LA), a random-intercept linear mixed-effects model was employed for analyzing repeated measurements. In this model, eGFR served as the dependent variable, while the age group was used as the independent variable, distinguishing between “geriatric patients” (age ≥ 65 years) and “non-geriatric patients” (age < 65 years). The aim was to examine the interaction between age group and lithium serum levels concerning their impact on eGFR, assessing whether eGFR changes during treatment varied across age groups. It’s important to note that the lithium serum level was considered an indicator of the “effect of LA” because all study participants initiated lithium treatment at the study’s outset, and baseline eGFR was measured before lithium intake began.


The use of linear mixed-effects models is advantageous for investigating the variability between patients (heterogeneity) and allows simultaneous adjustment for within-subject correlations. In this analysis, several covariates, including lithium serum level, sex, Charlson comorbidity index, hypertension, and co-medication affecting lithium serum levels, were included as fixed effects within the model. The results of this analysis were presented with regression coefficients, standard error estimates (SE), and 95% confidence intervals (CI) in Table 2. Additionally, an exploratory linear mixed model was utilized, incorporating the covariates time, geriatric versus non-geriatric status, and their interaction, as visually represented in Figure 1.


To ensure the normal distribution of the data, the Kolmogorov–Smirnov test was applied. Descriptive statistics were obtained using T-tests, Mann–Whitney U tests, Chi-square tests, and Fisher’s exact tests as needed. A significance level of p < 0.05 (two-sided) was set for all statistical analyses. The Statistical Analysis System (SAS) software (version 9.4) was employed for conducting the linear mixed models, while SPSS (version 27) was used for descriptive statistics.


In summary, this study utilized a robust statistical approach, the linear mixed-effects model, to investigate the effects of age group and lithium serum levels on eGFR changes in response to Lithium Augmentation. Various covariates were considered in the analysis, and statistical tests were rigorously applied to assess significance levels in the data.



A total of 201 patients were enrolled in the cohort, adhering to the predefined inclusion criteria outlined earlier. Of these, 29 patients were classified as geriatric, with an average age of 72.1 years (standard deviation [SD]: 5.4), while 172 were categorized as non-geriatric patients, with an average age of 46.5 years (SD: 10.7). Further details can be found in Table 1 for comprehensive descriptive statistics.


The study aimed to evaluate the impact of Lithium Augmentation (LA) on estimated glomerular filtration rate (eGFR) in geriatric versus non-geriatric patients, with the results summarized in Table 2. It was observed that both age groups experienced a significant decline in eGFR during the course of treatment, with geriatric patients displaying lower eGFR values (p < 0.01). Notably, the lithium serum level, indicative of the “effect of LA,” exerted a notable influence on eGFR decline (p < 0.01). Importantly, no statistically significant interaction was detected between lithium serum levels and age concerning eGFR decline (p = 0.51). This implies that the effect of LA on eGFR decline did not differ between the two age groups. On average, the eGFR decline from baseline to the last measured eGFR during LA was 5.89 ml/min/1.73 m^2 for geriatric patients and 4.76 ml/min/1.73 m^2 for non-geriatric patients. Notably, female patients and those with a higher somatic severity index experienced a more pronounced eGFR decline compared to male patients (p < 0.01) and patients with a lower somatic severity index (p = 0.02). However, the covariates “co-medication affecting lithium level” and “arterial hypertension” did not significantly impact eGFR (p > 0.05).


Crucially, two geriatric patients experienced acute kidney injuries (AKIs), while none of the non-geriatric patients encountered such events. Both AKIs occurred in the context of toxic lithium levels, indicative of clinically relevant lithium intoxication (detailed in Table S1). One patient, aged 83 years, developed an AKI stage 1, with creatinine levels rising from 0.82 mg/dl before LA to 1.28 mg/dl after the first week of treatment under a toxic lithium serum level of 1.52 mmol/L. Their eGFR declined from 70.93 ml/min/1.73 m^2 before LA to 41.57 ml/min/1.73 m^2. The initial lithium dosage was 12.2 mmol/day. The other patient, aged 82 years, experienced an AKI stage 3, with creatinine levels increasing from 0.90 mg/dl before LA to 2.93 mg/dl after 2 weeks under LA. Their eGFR declined from 63.83 ml/min/1.73 m^2 before LA to 15.48 ml/min/1.73 m^2 within 2 weeks. This occurred with a toxic lithium serum level of 2.0 mmol/L. The initial lithium dosage was 6.1 mmol/day, which was rapidly increased to 24.4 mmol/day after only 5 days of treatment. Both patients exhibited clinical symptoms of lithium intoxication (such as tremors, nausea, and dizziness) and were treated with intravenous fluid replacement, leading to full recovery. Importantly, acute dialysis was not required in either case (further details provided in Table S1).


Moreover, two additional instances of toxic lithium serum levels were observed in the geriatric group, although without clinical signs of lithium intoxication or AKI. Intriguingly, no cases of toxic lithium serum levels were noted in the non-geriatric group, underscoring the significance of this finding (p < 0.01).



This study represents the first investigation into changes in estimated glomerular filtration rate (eGFR), serving as an indicator of kidney function, in geriatric compared to non-geriatric patients undergoing Lithium Augmentation (LA) for Treatment-Resistant Depression (TRD). While previous research has explored lithium monotherapy, studies focusing on the acute effects of LA on kidney function in geriatric patients have been limited. In this study, both age groups exhibited a notable decline in eGFR during LA. Importantly, the interaction between age group and serum lithium levels had no significant impact on eGFR, signifying that the rate of eGFR decline was consistent across both groups.


Most earlier prospective studies have primarily concentrated on long-term lithium use over years or decades but often overlooked the acute effects of LA on kidney function. To address this, the study opted for a relatively short observation period of up to 6 weeks to capture the immediate effects of LA on the kidney. In both age groups, a considerable decline in eGFR (approximately 5 ml/min/1.73 m^2) was observed during this short-term treatment. To put this into perspective, adults typically experience an annual eGFR decline of approximately 1 ml/min/1.73 m^2 due to the natural loss of functioning nephrons. A decline exceeding 3 ml/min/1.73 m^2 per year is indicative of rapid kidney function deterioration, which was observed in both age groups during the acute LA therapy.


Existing studies have produced varying results, with some indicating a low risk of Chronic Kidney Disease (CKD) progression associated with lithium treatment, especially in the absence of acute intoxication episodes. In contrast, recent findings have suggested a higher risk of kidney impairment in patients undergoing long-term lithium treatment, with the risk increasing with age. Most studies focusing on long-term effects reported kidney function decline occurring over several years, with the average time to CKD progression after commencing lithium therapy ranging from 16.5 to 31 years. These disparities underscore the importance of continuous kidney function monitoring in clinical practice and future research. Since this study did not encompass a long-term follow-up, it remains uncertain whether the observed eGFR decline is temporary and if kidney function might recover over time.


Unsurprisingly, geriatric patients began with a lower eGFR before initiating lithium treatment, primarily due to the age-related reduction in functioning nephrons. However, while eGFR declined steadily in both groups, geriatric patients may experience a more pronounced decline in the long-term due to their lower baseline eGFR. Elevated serum lithium levels represent a critical risk factor for lithium-related adverse effects and are often a consequence of rapid kidney function decline. Maintaining serum lithium levels within the recommended therapeutic range is essential for a safe treatment. Toxic lithium serum levels can significantly impact kidney function and are associated with Acute Kidney Injury (AKI).


Notably, this study revealed a significant negative effect of female gender on eGFR during LA, which is in line with some previous research indicating that women may be more “sensitive” to lithium-related kidney damage. However, the literature on this topic remains controversial.


This study’s strengths include its longitudinal design, which provided eGFR data for each participant with at least two measurements during LA treatment. Additionally, comorbidities with potential effects on kidney function were assessed, including arterial hypertension and other severe chronic comorbidities. However, the absence of a control group comprising age-matched subjects not exposed to lithium is a key limitation. Moreover, the study focused on the acute phase of LA to capture AKIs, making it unable to predict the long-term effects on kidney function. Serum creatinine, although not ideal for detecting reduced GFR, was used as an indicator, and the study employed the CKD-EPI 2021 equation to estimate GFR.


In conclusion, this study suggests that geriatric patients may face a higher risk of AKI during LA due to their reduced ability to compensate for the treatment-related eGFR decline. The significant incidence of lithium intoxication among geriatric patients was promptly detected and treated, highlighting the importance of close monitoring through weekly laboratory tests and adverse effects assessments. With vigilant monitoring, clinicians may consider LA more frequently and at an earlier stage of TRD treatment for geriatric patients.


Oncology Related Tools


Latest Research


About Author

Similar Articles

Leave a Reply