Abstract
Background:
Bipolar disorder (BD) is associated with a higher prevalence of chronic kidney disease (CKD) and increased mortality, yet individuals with both BD and CKD are less likely to receive kidney replacement therapy (KRT). Sodium-glucose cotransporter-2 inhibitors (SGLT2is) provide renal protection in diabetes, but their long-term effects in this high-risk population remain unknown.
Objectives:
To evaluate whether SGLT2i use reduces the risk of KRT and all-cause mortality in adults with BD and mild-to-moderate CKD (stage ⩽ 3).
Design:
Observational cohort study with propensity score matching (PSM).
Method:
In this study using TriNetX (2009–2024) real-world database, 89,369 adults with BD and mild-to-moderate CKD were classified as SGLT2i users (n = 12,736) or nonusers (n = 76,633). Primary outcomes were progression to KRT and all-cause mortality over 5 years. Adjusted hazard ratios (aHRs) were estimated using Cox proportional hazards models, and 1:1 PSM reduced confounding. Subgroup analyses examined sex, race/ethnicity, diabetes status, CKD stage, and mood stabilizer use.
Results:
SGLT2i use was associated with lower risk of KRT (aHR 0.47 (95% CI, 0.42–0.53)) and all-cause mortality (aHR 0.69 (95% CI, 0.65–0.73); both p < 0.001). Protective effects were consistent across subgroups, including individuals receiving lithium, lamotrigine, valproate, or antipsychotics. After PSM (10,967 matched pairs), 5-year KRT-free survival was 94.61% versus 90.99%, and overall survival was 78.67% versus 67.53% (both p < 0.001). Given the observational design, the possibility of residual or unmeasured confounding canot be excluded.
Conclusion:
SGLT2i therapy in individuals with BD and mild-to-moderate CKD is associated with substantially lower risks of KRT and all-cause mortality. Prospective trials are needed to confirm these findings.
Keywords: bipolar disorder, chronic kidney disease, kidney replacement therapy, mortality, SGLT2 inhibitors
Plain language summary
SGLT2-inhibitors may lower risk of kidney replacement therapy and mortality in bipolar disorder with chronic kidney disease
Why this study matters
People with bipolar disorder face higher risks of kidney disease and death compared to the general population, yet they are less likely to receive life-saving treatments like dialysis or kidney transplants when they need them. While SGLT2 inhibitors (a type of medication) are known to protect the kidneys in people with diabetes, no one has studied whether they work for people with bipolar disorder and kidney disease.
What we set out to do
We conducted a long-term observational study using medical records from 89,369 adults with bipolar disorder who had mild to moderate kidney disease. We followed their health from 2009 to 2024 to see whether taking SGLT2 inhibitors reduced their risk of needing dialysis or a kidney transplant, and whether it lowered their risk of death.
Key findings
People with bipolar disorder who took SGLT2 inhibitors had better outcomes compared to those who did not:
• 53% lower risk of needing dialysis or a kidney transplant
• 31% lower risk of death from any cause over five years
These benefits were seen across different patient groups, including those taking lithium and other mood stabilizers. After accounting for other health factors, patients on SGLT2 inhibitors had much better five-year survival rates: 94.61% stayed off dialysis (compared to 90.99%) and 78.67% were still alive (compared to 67.53%).
What this means for patients and clinicians
This is the first large study showing that SGLT2 inhibitors may significantly protect kidney function and extend life in people with bipolar disorder—a group that has been left out of major research trials. For psychiatrists treating patients with lithium or other mood stabilizers, these findings suggest an important opportunity to prevent severe kidney problems.
Introduction
Bipolar disorder (BD) is a chronic mental health condition affecting approximately 40 million individuals worldwide. 1 Individuals with BD have almost twice the risk of developing type 2 diabetes mellitus (DM) compared with age- and sex-matched controls, along with higher rates of impaired fasting glucose, obesity, and increased abdominal obesity; these conditions are independent risk factors for cardiovascular disease.2,3 Data from the UK Biobank demonstrate increased phenotypic cardiometabolic abnormalities in individuals with BD, with the condition being associated with higher polygenic risk for coronary artery disease, DM, and abnormal lipid profiles. 4 Studies consistently highlight significantly elevated cardiovascular disease risk in individuals with BD compared to the general population. 5 Together, these factors reflect a complex interplay involving the disorder itself (e.g., systemic inflammation and hypothalamic–pituitary–adrenal axis dysregulation), lifestyle factors (such as smoking, sedentary behavior, and poor diet), and adverse effects of psychotropic medications (particularly antipsychotic-associated weight gain and metabolic syndrome).6 –8 This constellation of risks predisposes individuals with BD to a higher risk of chronic kidney disease (CKD). 9 Atypical antipsychotics commonly prescribed for BD 10 increase cardiovascular risk,11,12 which may in turn further elevate the likelihood of developing CKD. 13 Notably, although patients with CKD and BD may experience a faster decline in kidney function, they are less likely to receive kidney replacement therapy (KRT), such as dialysis or kidney transplantation. 14 Pharmacologic interventions that reduce the risk of renal decompensation and the need for KRT could improve both patient acceptance and outcomes in this population.
Sodium-glucose cotransporters (SGLTs) mediate glucose reabsorption in the proximal tubule, with SGLT2—primarily expressed in the kidneys—accounting for approximately 90% of this process.15,16 SGLT2 inhibitors (SGLT2is) selectively block SGLT2, providing an insulin-independent mechanism to lower blood glucose through glycosuria and natriuresis, thereby reducing hypoglycemia risk. 17 Six SGLT2is—canagliflozin, dapagliflozin, empagliflozin, ertugliflozin, bexagliflozin, and sotagliflozin (a dual SGLT1/SGLT2i)—are FDA-approved for DM or heart failure (canagliflozin, dapagliflozin, empagliflozin, and sotagliflozin).18 –20 Canagliflozin additionally treats diabetic nephropathy, while dapagliflozin and empagliflozin have broader approvals for CKD, independent of diabetes status. Beyond glycemic control, these agents reduce blood pressure and may slow CKD progression even in individuals without DM.21 –24 However, data in individuals with BD are limited, despite their elevated risk of kidney dysfunction due to higher cardiovascular burden and prevalence of CKD. 25 Many individuals with BD take lithium, which further increases CKD risk, particularly in the context of hypertension or DM, and recent evidence suggests that co-administration of SGLT2is may lower lithium levels, 26 potentially affecting mood stability if not closely monitored. A retrospective cohort study from our group at Mayo Clinic 27 provides initial evidence that SGLT2i initiation is associated with improved estimated glomerular filtration rate trajectories in individuals with mood disorder on long-term lithium, suggesting a potential kidney-protective benefit in this high-risk population. However, there are currently no data on their efficacy in reducing the need for KRT or improving mortality outcomes in BD and CKD.
Given the established cardioprotective and renoprotective effects of SGLT2is beyond glycemic control, individuals with BD and CKD may derive particular benefit from this drug class, underscoring the clinical importance of this research. These observations highlight the need for systematic investigation of SGLT2is in BD. To address this, we conducted a large cohort study using data from the TriNetX real-world database, which encompasses over 135 million patients. Our primary objective was to assess whether SGLT2i use, compared with nonuse, in adults with BD and mild to moderate CKD (stage ⩽ 3) is associated with a lower risk of receiving KRT and reduced mortality. We hypothesized that patients with BD and CKD stage ⩽3 receiving SGLT2is would have lower rates of KRT and mortality.
Methods
This observational study utilized de-identified real-world data from TriNetX, LLC, a global federated health research network providing access to electronic health record (EHR) data from large healthcare organizations. 28 Data were collected in October 2025 from the TriNetX Research Network, which includes EHRs from approximately 135 million patients aged 18 years and older across 112 healthcare organizations. The dataset spans 2009–2024 and includes information from outpatient clinics, specialty clinics, inpatient units, and emergency departments. Available data include laboratory results, medication prescriptions, procedures, diagnoses, and genomic information. TriNetX maps International Classification of Diseases (ICD)-9 codes to ICD-10 codes using General Equivalence Mappings. 29 All analyses were performed on aggregated counts and statistical summaries of de-identified data within the TriNetX environment. No Protected Health Information (PHI) or personal data were accessed or downloaded by the investigators. The de-identification process was conducted by a qualified expert, ensuring compliance with federal privacy standards. 30 As this study involved a secondary analysis of de-identified data, it was deemed exempt from Institutional Review Board review in accordance with the National Human Research Protections Advisory Committee guidelines, and patient consent was not required. The study adheres to the principles of the Declaration of Helsinki and complies with the Strengthening the Reporting of Observational Studies in Epidemiology guidelines (Supplemental Material). 31
Study population
Eligible participants were adults aged 18 years or older with a diagnosis of BD (ICD-10-CM F30–F31)32,33 and CKD stage ⩽ 3 (hereafter referred to as CKD). The exposure group included individuals who newly initiated SGLT2i therapy, while the comparator group consisted of those with BD and CKD without SGLT2i exposure. Individuals with a documented history of dialysis or renal transplantation before the index date were excluded.
A total of 89,369 eligible individuals were identified at their index visit, defined as the first documented diagnosis of BD (ICD-10-CM F30–F31) with CKD (stage ⩽ 3; ICD-10-CM N18.1–N18.3) in the EHR during the study period. Among them, 12,736 (14.25%) were prescribed SGLT2is (canagliflozin, dapagliflozin, empagliflozin, ertugliflozin, bexagliflozin, or sotagliflozin) and classified as SGLT2i users, while the remaining 76,633 were classified as SGLT2i non-users. These individuals were followed longitudinally for 5 years.
The cohort was geographically diverse, with 24.6% from the Northeast, 24.8% from the Midwest, 35.7% from the South, 13.9% from the West, and 1.0% from outside the USA. Diagnostic, procedural, and medication data were identified using ICD-10-CM and Current Procedural Terminology (CPT) codes. Additional details are provided in the Supplemental Tables 1 and 2.
Variables
Dependent variables (outcomes)
The primary outcome was progression to KRT, and the secondary outcome was all-cause mortality (medical, psychiatric, or combined), with both assessed over a 5-year follow-up period. These outcomes were selected due to their clinical relevance in individuals with BD and CKD, capturing disease progression and overall risk. As a negative control, the incidence of acute appendicitis (ICD-10 code K35) was assessed to evaluate potential residual confounding. Similar strategy has been used in prior studies, as no association is expected between SGLT2i use and the incidence of acute appendicitis or other neutral outcomes.34,35
Independent variable
The primary independent variable was SGLT2i use, classified as user versus non-user. Individuals newly prescribed SGLT2i during the follow-up period were included in the SGLT2i user group from the date of their first prescription, which served as their index date. Follow-up began at SGLT2i initiation and continued until outcome occurrence, loss to follow-up, or study end. This time-dependent approach ensured exposure classification accurately reflected each patient's medication status throughout follow-up. None of the patients in the non-SGLT2i group ever received an SGLT2i.
Covariates
Covariates included demographic characteristics (age at index date, sex, race, and ethnicity), psychiatric and somatic comorbidities, and relevant pharmacotherapy, as documented in the EHR.
The target trial emulation framework for observational studies has been proposed to improve the quality of observational studies. 36 We used a similar methodology to account for baseline imbalances and performed propensity score matching (PSM) for the covariates that were significant in the univariate analyses and examined outcomes (KRT and all-cause mortality) at 5 years.
Statistical analyses
All analyses were conducted in October 2025 using the secure TriNetX platform. Descriptive analyses summarized baseline characteristics. Categorical data were reported as numbers and percentages, and continuous data were reported as means with standard deviations (SDs). Standardized mean differences (SMDs) with corresponding p-values were estimated to compare demographic characteristics, psychiatric and somatic comorbidities, and pharmacotherapy between SGLT2i users and non-users.
Cox proportional hazards models were applied over a 5-year follow-up to assess time-to-event outcomes, including the occurrence of KRT or all-cause mortality. Patients were censored at the time of the event, loss to follow-up, or the end of follow-up, whichever occurred first. Models were adjusted for sociodemographic characteristics, psychiatric and somatic comorbidities, and pharmacotherapy to estimate adjusted hazard ratios (aHRs) with 95% confidence intervals (CIs). Covariates were coded as follows: age (continuous), sex (male = 1, female/other = 0), race (White = 1, non-White = 0), ethnicity (Hispanic/Latino = 1, non-Hispanic/Latino = 0), and psychiatric/somatic comorbidities and pharmacotherapy (present = 1, absent = 0). Reference categories were female/other for sex, non-White for race, non-Hispanic/Latino for ethnicity, and absence of comorbidities or pharmacotherapy.
Sensitivity analyses evaluated the robustness of findings across clinically relevant subgroups, stratified by DM status (present vs absent), sex, race, ethnicity, and psychopharmacologic treatments (antipsychotics, valproate, lamotrigine, and lithium). Additional analyses examined less severe CKD cases (CKD ⩽ 2) and restricted the cohort to patients prescribed only the three FDA-approved SGLT2is for CKD or diabetic nephropathy (canagliflozin, dapagliflozin, and empagliflozin). Multiple-comparison adjustment was applied only to exploratory subgroup analyses, with false discovery rate correction performed separately for KRT and mortality outcomes. A neutral outcome (acute appendicitis) was also assessed to detect potential bias unrelated to SGLT2i use. To address concerns about possible exposure misclassification, an additional sensitivity analysis was conducted using a more stringent definition of SGLT2i exposure, requiring at least two prescriptions during the study period to ensure exposure reflected more sustained treatment. 37 Among all the individuals on SGLT2is, we conducted a sensitivity analysis comparing outcomes across three mutually exclusive psychotropic medication groups: (1) lithium + SGLT2i, (2) mood-stabilizing anticonvulsants (valproate, carbamazepine, or lamotrigine) + SGLT2i, and (3) antipsychotics + SGLT2i. Multivariable Cox proportional hazards regression models were used to estimate aHR and 95% CI for KRT and all-cause mortality with the non-lithium + SGLT2i group as reference. Models were adjusted for age, sex, comorbidities, and pharmacotherapy, including adjustment for the remaining psychotropic class not under comparison (for example when comparing lithium with mood-stabilizing anticonvulsants, models additionally adjusted for antipsychotic use).
Survival analyses
Kaplan–Meier survival analyses compared event-free probabilities for KRT and all-cause mortality over the 5-year follow-up period between SGLT2i users and nonusers. To account for baseline imbalances, PSM was performed using the same covariates as in the Cox models. Greedy nearest-neighbor matching with a 0.1 SD caliper was used to achieve 1:1 matching without replacement, minimizing confounding, ensuring balanced comparison groups, and improving the interpretability of the survival analyses. 38 Detailed descriptions of matching procedures, outcomes, and sample sizes are provided in the Supplemental Table 3.
Results
In a cohort of 89,369 individuals with BD and CKD, 12,736 (14.3%) were prescribed SGLT2is. Overall, lithium use was low, with only 13.2% of patients receiving it, whereas the majority were treated with other psychotropics: 75.4% received any anticonvulsant, 73.3% an antipsychotic, 79.1% an antidepressant, and 81.2% a benzodiazepine (Table 1). Compared to nonusers, SGLT2i users were slightly older (mean age 61.5 vs 60.6 years) and more likely to be male (43.6% vs 39.9%), Hispanic (5.85% vs 4.13%), or Black (22.6% vs 18.1%). Psychiatric comorbidities were more prevalent among SGLT2i users, including anxiety disorders (78.6% vs 70.4%), substance use disorders (57.6% vs 51.6%), sleep disorders (17.8% vs 12.7%), and eating disorders (3.19% vs 2.60%). Somatic conditions were markedly higher in the SGLT2i group, with DM (87.1% vs 50.2%), heart failure (57.1% vs 32.4%), and obesity (71.7% vs 49.2%) showing the largest differences (all SMD ⩾ 0.474). SGLT2i users were more frequently prescribed angiotensin-converting enzyme inhibitors (62.3% vs 41.7%), angiotensin II receptor blockers (58.0% vs 29.1%), and hypoglycemic agents (46.1% vs 10.2%). Most group differences were statistically significant (p < 0.001), indicating clinically relevant distinctions in demographic, psychiatric, and medical profiles between users and nonusers.
Table 1.
Characteristics of bipolar disorder patients with CKD ⩽ 3 by SGLT2 inhibitor use. a
| Variables | Full cohort 89,369 (%) |
SGLT2i users 12,736 (%) |
SGLT2i nonusers 76,633 (%) |
SMD | p-Value |
|---|---|---|---|---|---|
| Sociodemographic | |||||
| Age at index in years, mean (SD) | 60.7 ± 13.5 | 61.5 ± 11.4 | 60.6 ± 13.8 | 0.075 | <0.001 |
| Gender | |||||
| Male | 35,373 (39.58) | 5551 (43.61) | 29,822 (39.92) | 0.095 | <0.001 |
| Female | 53,961 (60.38) | 7181 (56.39) | 46,780 (61.04) | 0.095 | <0.001 |
| Unknown | 35 (0.04) | 4 (0.03) | 31 (0.04) | 0.016 | 0.063 |
| Ethnicity | |||||
| Not Hispanic or Latino | 68,784 (76.97) | 9850 (77.38) | 58,934 (76.90) | 0.010 | 0.280 |
| Hispanic or Latino | 3909 (4.37) | 745 (5.85) | 3164 (4.13) | 0.079 | <0.001 |
| Unknown Ethnicity | 16,396 (18.35) | 2141 (16.82) | 14,255 (18.60) | 0.056 | <0.001 |
| Race | |||||
| White | 64,572 (72.25) | 8562 (67.23) | 56,010 (73.09) | 0.128 | <0.001 |
| Black or African America | 16,741 (18.73) | 2872 (22.55) | 13,869 (18.10) | 0.111 | <0.001 |
| American Indian or Alaska Native | 549 (0.61) | 102 (0.80) | 447 (0.58) | 0.026 | 0.002 |
| Asian | 1474 (1.65) | 281 (2.21) | 1193 (1.56) | 0.048 | <0.001 |
| Native Hawaiian or Pacific Islander | 493 (0.55) | 90 (0.71) | 403 (0.53) | 0.023 | 0.011 |
| Other | 1904 (2.13) | 331 (2.60) | 1573 (2.05) | 0.036 | <0.001 |
| Unknown Race | 3636 (4.07) | 498 (3.91) | 3138 (4.09) | 0.019 | 0.329 |
| Psychiatric comorbidity | |||||
| Anxiety disorders | 63,953 (71.56) | 10,007 (78.57) | 53,946 (70.40) | 0.188 | <0.001 |
| Psychotic disorders | 24,036 (26.90) | 3369 (26.45) | 20,667 (26.97) | 0.012 | 0.224 |
| Substance use disorders | 46,866 (52.44) | 7340 (57.63) | 39,526 (51.58) | 0.122 | <0.001 |
| Sleep disorders | 11,976 (13.40) | 2272 (17.84) | 9704 (12.66) | 0.144 | <0.001 |
| Eating disorders | 2400 (2.69) | 406 (3.19) | 1994 (2.60) | 0.035 | <0.001 |
| Somatic comorbidity | |||||
| Hypertension | 79,013 (88.41) | 12,354 (97.00) | 66,659 (86.98) | 0.376 | <0.001 |
| Ischemic heart diseases | 40,473 (45.29) | 8078 (63.43) | 32,395 (42.27) | 0.434 | <0.001 |
| Heart failure | 32,125 (35.95) | 7267 (57.06) | 24,858 (32.44) | 0.511 | <0.001 |
| Cerebrovascular diseases | 26,862 (30.06) | 4613 (36.22) | 22,249 (29.03) | 0.154 | <0.001 |
| COPD | 29,609 (33.13) | 5286 (41.50) | 24,323 (3174) | 0.204 | <0.001 |
| Overweight and obesity | 46,813 (52.38) | 9134 (71.71) | 37,679(49.17) | 0.474 | <0.001 |
| Diabetes mellitus | 49,554 (55.45) | 11,089 (87.07) | 38,465 (50.19) | 0.866 | <0.001 |
| Thyroid disorders | 39,016 (43.66) | 5748 (45.13) | 33,268 (43.41) | 0.035 | <0.001 |
| Pharmacotherapy | |||||
| Lithium | 11,775 (13.18) | 1401 (11.00) | 10,374 (13.54) | 0.077 | <0.001 |
| Anticonvulsants | 67,351 (75.36) | 10,427 (81.87) | 56,924 (74.28) | 0.184 | <0.001 |
| Valproate | 20,984 (23.48) | 2931 (23.01) | 18,053 (23.56) | 0.013 | 0.180 |
| Carbamazepine | 4,355 (4.87) | 615 (4.83) | 3740 (4.88) | 0.002 | 0.802 |
| Lamotrigine | 22,193 (24.83) | 3278 (25.74) | 18,915 (24.68) | 0.024 | 0.011 |
| Antipsychotics | 65,503 (73.29) | 9689 (76.08) | 55,814 (72.83) | 0.074 | <0.001 |
| Antidepressants | 70,719 (79.13) | 10,944 (85.93) | 59,775 (78.00) | 0.207 | <0.001 |
| SSRI/SNRI | 69,320 (77.57) | 10,732 (85.27) | 58,588 (76.45) | 0.198 | <0.001 |
| TCAs | 16,026 (17.93) | 2905 (22.81) | 13,121 (17.12) | 0.143 | <0.001 |
| Benzodiazepines | 72,533 (81.16) | 11,090 (87.08) | 61,443 (80.18) | 0.187 | <0.001 |
| ACE inhibitors | 39,890 (44.64) | 7940 (62.34) | 31,950 (41.69) | 0.423 | <0.001 |
| Angiotensin II inhibitor/ARB | 29,676 (33.21) | 7380 (57.95) | 22,296 (29.09) | 0.608 | <0.001 |
| Hypoglycemic agents | 13,658 (15.28) | 5865 (46.05) | 7793 (10.17) | 0.871 | <0.001 |
| NSAIDs | 34,980 (39.14) | 5,891 (46.25) | 29,089 (37.96) | 0.169 | <0.001 |
SGLT2 inhibitor use includes canagliflozin, dapagliflozin, empagliflozin, ertugliflozin, bexagliflozin, or sotagliflozin.
ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; CKD, chronic kidney disease; COPD: Chronic obstructive pulmonary disease; NSAID, nonsteroidal anti-inflammatory drug; SGLT2i, sodium–glucose cotransporter-2 inhibitor; SMD, standardized mean difference; SNRI, serotonin–norepinephrine reuptake inhibitor; SSRI, selective serotonin reuptake inhibitor; TCA, tricyclic antidepressant.
After adjusting for covariates in the Cox proportional hazards models, SGLT2i use was associated with a significantly lower risk of incident KRT compared with non-SGLT2i use (aHR 0.47, 95% CI 0.42–0.53; p < 0.001) and a reduced risk of mortality (aHR 0.69, 95% CI 0.65–0.73; p < 0.001; Table 2 and Supplemental Figure 1(A)). Sensitivity analysis using a more stringent definition of SGLT2i exposure (requiring at least two prescriptions during the study period) demonstrated consistent and more robust findings. Overall, 78% (n = 9,989) of patients met this threshold. After adjusting for covariates, sustained SGLT2i use was associated with a significantly lower risk of KRT (aHR 0.41, 95% CI 0.36–0.47; p < 0.001) and mortality (aHR 0.63, 95% CI 0.59–0.67; p < 0.001) among patients with BD and CKD. There was no significant difference in the risk of incident acute appendicitis between the groups (aHR 1.30, 95% CI 0.87–1.94; p = 0.20). Subgroup analyses showed consistent protective effects across demographic and clinical strata, including individuals with (aHR 0.40) and without diabetes (aHR 0.57), males (aHR 0.43) and females (aHR 0.51), and racial/ethnic subgroups: White (aHR 0.46), Black (aHR 0.48), Hispanic (aHR 0.43), and non-Hispanic (aHR 0.42), all p < 0.001 (Table 2, Supplemental Figure 1(B)). Benefits were also observed across all mood stabilizers—lithium (aHR 0.40), lamotrigine (aHR 0.47), antipsychotics (aHR 0.47), and valproate (aHR 0.54)—as well as in patients with CKD severity ⩽2 (aHR 0.43) and those prescribed FDA-approved SGLT2is for CKD or diabetic nephropathy (aHR 0.46), all p ⩽ 0.001. These results indicate robust benefits of SGLT2i use across clinical and demographic subgroups, with similar patterns observed for reduced mortality (Supplemental Figure 1(C)).
Table 2.
Adjusted hazard ratios a for clinical outcomes with SGLT2i use b in bipolar disorder with CKD stage ⩽ 3.
| Outcomes | aHR (95% CI) | p-Value | FDR-adjusted p-value c |
|---|---|---|---|
| Dialysis/renal transplant (Kidney replacement therapy) | 0.47 (0.42, 0.53) | <0.001 | |
| Mortality | 0.69 (0.65, 0.73) | <0.001 | |
| Sustained SGLT2i use (at least two prescriptions during the study period) | |||
| Dialysis/renal transplant | 0.41 (0.36–0.4) | <0.001 | |
| Mortality | 0.63 (0.59–0.67) | <0.001 | |
| Subgroup analysis a of kidney replacement therapy in bipolar disorder with CKD ⩽ 3 by SGLT2i use b | |||
| Subgroup | aHR a (95% CI) | p-Value | |
| DM | |||
| Present | 0.40 (0.34, 0.46) | <0.001 | 0.001 |
| Absent | 0.57 (0.38, 0.85) | 0.006 | 0.006 |
| Sex | |||
| Male | 0.43 (0.36, 0.51) | <0.001 | 0.001 |
| Female | 0.51 (0.44, 0.60) | <0.001 | 0.001 |
| Race | |||
| White | 0.46 (0.40, 0.53) | <0.001 | 0.001 |
| Black | 0.48 (0.39, 0.59) | <0.001 | 0.001 |
| Ethnicity | |||
| Hispanic | 0.43 (0.30, 0.63) | <0.001 | 0.001 |
| Non-Hispanic | 0.42 (0.37, 0.48) | <0.001 | 0.001 |
| Medication | |||
| Antipsychotics | 0.47 (0.42, 0.54) | <0.001 | 0.001 |
| Lithium | 0.40 (0.23, 0.69) | 0.001 | 0.001 |
| Valproate | 0.54 (0.43, 0.64) | <0.001 | 0.001 |
| Lamotrigine | 0.47 (0.37, 0.59) | <0.001 | 0.001 |
| Others | |||
| Severity (CKD ⩽ 2) | 0.43 (0.36, 0.52) | <0.001 | 0.001 |
| FDA approved d SGLT2i for CKD or DN | 0.46 (0.42, 0.52) | <0.001 | 0.001 |
| Subgroup analysis a of mortality in bipolar disorder with CKD ⩽ 3 by SGLT2i use b | |||
| Subgroup | aHR a (95% CI) | p-Value | |
| DM | |||
| Present | 0.71 (0.67, 0.76) | <0.001 | 0.001 |
| Absent | 0.62 (0.51, 0.75 | <0.001 | 0.001 |
| Sex | |||
| Male | 0.68 (0.62, 0.74 | <0.001 | 0.001 |
| Female | 0.71 (0.65, 0.78) | <0.001 | 0.001 |
| Race | |||
| White | 0.71 (0.66, 0.77) | <0.001 | 0.001 |
| Black | 0.62 (0.54, 0.72) | <0.001 | 0.001 |
| Ethnicity | |||
| Hispanic | 0.61 (0.45, 0.83) | <0.001 | 0.001 |
| Non-Hispanic | 0.58 (0.54, 0.62) | <0.001 | 0.001 |
| Medication | |||
| Antipsychotics | 0.73 (0.68, 0.79) | <0.001 | 0.001 |
| Lithium | 0.59 (0.47, 0.74) | <0.001 | 0.001 |
| Valproate | 0.75 (0.66, 0.85) | <0.001 | 0.001 |
| Lamotrigine | 0.65 (0.56, 0.74) | <0.001 | 0.001 |
| Others | |||
| Severity (CKD ⩽ 2) | 0.73 (0.65, 0.81) | <0.001 | 0.001 |
| FDA approved d SGLT2i for CKD or DN | 0.68 (0.64, 0.73) | <0.001 | 0.001 |
Cox regression models were adjusted for demographic factors, psychiatric comorbidities, and treatment variables.
SGLT2 inhibitor use includes canagliflozin, dapagliflozin, empagliflozin, ertugliflozin, bexagliflozin, or sotagliflozin
FDR-adjusted p-values were calculated using the Benjamini-Hochberg method to control for multiple comparisons.
FDA-approved SGLT2 inhibitor for CKD or DN includes canagliflozin, dapagliflozin, empagliflozin.
aHR, adjusted hazard ratio; CI, confidence interval; CKD, chronic kidney disease; DM, diabetes mellitus; DN, diabetic nephropathy; FDR, false discovery rate; SGLT2i, sodium–glucose cotransporter-2 inhibitor.
In the sensitivity analysis restricted to SGLT2i users, we compared outcomes based on concurrent psychotropic medication use. Compared to lithium + SGLT2i users, individuals receiving mood-stabilizing anticonvulsants + SGLT2i or antipsychotics+SGLT2i had significantly higher risks of both KRT and mortality. For KRT, aHR was 0.41 (95% CI 0.26–0.66; p < 0.001) for lithium + SGLT2i versus antipsychotics + SGLT2i and 0.42 (95% CI 0.26–0.67; p = 0.006) versus mood-stabilizing anticonvulsants + SGLT2i. For mortality, the aHR was 0.75 (95% CI 0.61–0.92; p = 0.005) for lithium + SGLT2i versus antipsychotics + SGLT2i and 0.79 (95% CI 0.64–0.98; p = 0.028) versus mood-stabilizing anticonvulsants + SGLT2i.
Following PSM, 10,967 individuals with BD and CKD prescribed SGLT2i were matched to 10,967 nonusers. The matched cohorts were well balanced in terms of demographic characteristics, psychiatric and somatic comorbidities, and concurrent medication use (Supplemental Table 3). Kaplan–Meier survival analysis demonstrated significantly higher survival probabilities among SGLT2i users compared to nonusers for both progression to KRT and all-cause mortality (Tables 3 and 4). At 5 years, survival without KRT was 94.61% among SGLT2i users compared to 90.99% among nonusers (χ2 = 44.365, df = 1, p < 0.001). Similarly, survival probability for mortality was 78.67% in the SGLT2i group versus 67.53% in the nonuser group (χ2 = 166.325, df = 1, p < 0.001). These findings suggest a robust association between SGLT2i use and improved long-term renal and overall survival outcomes in individuals with BD and CKD.
Table 3.
Kaplan–Meier survival analysis of kidney replacement therapy in bipolar disorder with CKD stage ⩽ 3 by SGLT2i use. a .
| Cohort | Patients with outcome (n = 10,967) b |
Survival probability at the end of time window | χ2 | df | p-Value |
|---|---|---|---|---|---|
| 1 year | |||||
| SGLT2i users | 98 | 98.83% | 26.346 | 1 | <0.001 |
| SGLT2i nonusers | 183 | 97.85% | |||
| 2 Years | |||||
| SGLT2i users | 166 | 97.50% | 30.356 | 1 | <0.001 |
| SGLT2i nonusers | 297 | 95.95% | |||
| 3 years | |||||
| SGLT2i users | 195 | 96.53% | 38.906 | 1 | <0.001 |
| SGLT2i nonusers | 376 | 94.17% | |||
| 4 years | |||||
| SGLT2i users | 213 | 95.42% | 41.820 | 1 | <0.001 |
| SGLT2i nonusers | 431 | 92.45% | |||
| 5 years | <0.001 | ||||
| SGLT2i users | 220 | 94.61% | 44.365 | 1 | |
| SGLT2i nonusers | 465 | 90.99% | |||
SGLT2 inhibitor use includes canagliflozin, dapagliflozin, empagliflozin, ertugliflozin, bexagliflozin, or sotagliflozin.
Propensity score matching (n = 10,967) was used to adjust for demographic factors, psychiatric comorbidities, and treatment variables. Patients with kidney replacement therapy recorded prior to the index event (n = 307 in the SGLT2i user cohort and n = 642 in the nonuser cohort) were excluded to minimize bias from pre-existing outcomes and potential EHR data irregularities, such as delayed reporting or inaccurate event dates.
CKD, chronic kidney disease; df, degrees of freedom; SGLT2i, sodium–glucose cotransporter-2 inhibitor.
Table 4.
Kaplan–Meier survival analysis of mortality in bipolar disorder with CKD stage ⩽ 3 by SGLT2i use. a .
| Cohort | Patients with outcome (n = 10,967) b | Survival probability at the end of time window | χ2 | df | p-Value |
|---|---|---|---|---|---|
| 1 year | |||||
| SGLT2i users | 627 | 93.05% | 90.783 | 1 | <0.001 |
| SGLT2i nonusers | 1027 | 89.26% | |||
| 2 years | |||||
| SGLT2i users | 869 | 88.78% | 118.694 | 1 | <0.001 |
| SGLT2i nonusers | 1467 | 83.12% | |||
| 3 years | |||||
| SGLT2i users | 1001 | 84.94% | 142.471 | 1 | <0.001 |
| SGLT2i nonusers | 1791 | 77.28% | |||
| 4 years | |||||
| SGLT2i users | 1066 | 81.52% | 155.732 | 1 | <0.001 |
| SGLT2i nonusers | 2006 | 72.19% | |||
| 5 years | <0.001 | ||||
| SGLT2i users | 1096 | 78.67% | 166.325 | 1 | |
| SGLT2i nonusers | 2157 | 67.53% | |||
SGLT2 inhibitor use includes canagliflozin, dapagliflozin, empagliflozin, ertugliflozin, bexagliflozin, or sotagliflozin.
Propensity score matching (n = 10,967) was used to adjust for demographic factors, psychiatric comorbidities, and treatment variables. Patients with mortality recorded on or before the index date (n = 49 in the SGLT2i user cohort and n = 67 in the nonuser cohort) were excluded to minimize bias from potential EHR data irregularities, such as delayed reporting or inaccurate death dates).
CKD, chronic kidney disease; df, degrees of freedom; SGLT2i, sodium–glucose cotransporter-2 inhibitor.
Discussion
To our knowledge, this is the first study demonstrating that SGLT2i use is associated with a substantially lower risk of incident KRT and reduced mortality in patients with BD and CKD. Specifically, SGLT2i users had a 53% lower risk of requiring KRT and a 31% lower risk of death compared with non-users. These associations were highly statistically significant, suggesting that SGLT2is may provide meaningful renal and survival benefits in this high-risk population. The findings are consistent with prior evidence of the renal- and cardioprotective effects of SGLT2is17,39 –41 and underscore their potential clinical utility in patients with BD and CKD, who are likely to undergo kidney transplantation.
Patients with BD are at heightened risk of CKD, particularly in the presence of underlying risk factors such as DM, hypertension, and lithium therapy.9,13,25,42,43 As CKD advances, some patients may ultimately require KRT. However, concerns regarding mood instability often result in lower rates of kidney transplantation and higher mortality in this population. 14 Identifying interventions that reduce the need for KRT and improve survival has important public health implications. In this study, the benefits of SGLT2i use were consistent across clinical and demographic subgroups—including patients with and without DM, males and females, Black and White, Hispanic and non-Hispanic, and those on various mood stabilizers—highlighting the robustness of the findings. Notably, SGLT2i use did not affect the incidence of acute appendicitis, supporting that residual confounding is unlikely.
The reduced risk of requiring KRT may reflect the renal-protective effects of SGLT2is, which preserves kidney function in patients both with and without DM.44,45 These benefits are thought to arise from reductions in glomerular hyperfiltration, intraluminal pressure, and oxidative stress. 17 SGLT2is also decrease the risk of cardiovascular disease, thereby reducing kidney burden, lowering the likelihood of KRT, and supporting improved long-term survival. 17 In addition, data from animal models suggest potential neuroprotective effects via reductions in inflammation and oxidative stress, 46 which may enhance treatment outcomes and survival in individuals with BD. Preclinical data further indicate that SGLT2is enhance neurotrophic signaling, reduce inflammation, and may exert antidepressant effects, 47 potentially through mild ketosis, a state associated with improved mood and reduced depressive symptoms.48 –50 However, this study cannot directly assess the impact of SGLT2is on depression or mood stability, and future research should explore their effects on mood symptoms in individuals with BD.
Although lithium is traditionally associated with adverse renal outcomes in a small subset of patients, 13 our analysis of individuals with BD and CKD receiving lithium + SGLT2is revealed a protective association compared with other antipsychotics + SGLT2is and mood-stabilizing anticonvulsants + SGLT2is. Furthermore, in our BD and CKD cohort, lithium use (13.2%) was substantially lower than anticonvulsant (75.4%) and antipsychotic (73.3%) use. Moreover, the proportion of patients receiving lithium was significantly lower among SGLT2i users compared to non-users (11.0% vs 13.5%). This observation may reflect selection bias favoring medically stable patients who continued lithium + SGLT2i therapy, the kidney-protective effects of SGLT2is mitigating lithium-associated renal decline, and enhanced renal monitoring among lithium-treated patients. Crucially, the finding that SGLT2is confer renal protection even in patients on lithium carries important clinical implications, as lithium is often under prescribed due to concerns about kidney disease and the potential need for future interventions such as KRT.13,51 Patients on long-term lithium therapy who are stable but subsequently develop CKD face a high risk of mood decompensation if lithium is discontinued. 52 Initiating SGLT2is in this context could reduce or prevent the need to discontinue lithium, preserve kidney function, and improve overall outcomes—not only through the direct renal benefits of SGLT2is but also by enabling continued mood stabilization with lithium. These findings underscore a potentially transformative approach for managing patients who are stable on long-term lithium therapy yet develop CKD, with significant implications for both renal and psychiatric outcomes over the long term.
Strengths and limitations
A major strength of our study was the large-scale real-world dataset from diverse healthcare organizations across the USA, thus enhancing generalizability. To our knowledge, this is the first study demonstrating the role of SGLT2is in reducing risks of KRT and mortality in patients with BD and CKD. The consistency of findings in PSM analyses further supported the robustness of our results. This real-world evidence approach is particularly valuable for studying outcomes in individuals with BD and CKD—a population often excluded from randomized controlled trials due to psychiatric comorbidity and polypharmacy—enabling examination of treatment effects in routine clinical practice with adequate sample size for rare outcomes such as KRT initiation. Additionally, we conducted multivariable analyses and several sensitivity analyses, demonstrating the robustness of findings across important subgroups and exposure definitions.
This study has several limitations that merit consideration. As an observational study, causality cannot be inferred. Although the study benefits from a large sample size and longitudinal follow-up, the database lacks detailed information on BD symptoms and symptom severity, treatment efficacy, adherence (including therapeutic drug levels), duration of SGLT2i use, and tolerability, limiting our ability to fully evaluate the influence of mood states and BD treatment on reductions in KRT and mortality. Data on insurance coverage and prescriber characteristics were unavailable, restricting assessment of systemic barriers to SGLT2i use and the potential influence of socioeconomic factors, such as income and education, on long-term outcomes. Mortality data were obtained from EHRs and may not capture all deaths or be affected by data irregularities; however, prior studies have validated this source for mortality analyses, making differential bias between SGLT2i and non-SGLT2i cohorts unlikely.
To preserve model stability and minimize overfitting, not all somatic comorbidities or concomitant treatments were included. We prioritized empirically supported demographic, psychiatric, somatic, and treatment covariates more directly related to clinical outcomes, though the omission of other covariates may introduce residual confounding. Despite these constraints, subgroup analyses stratified by sex, diabetes status, race, ethnicity, mood stabilizer type, and CKD severity demonstrated consistent direction and magnitude of benefit. Future research should incorporate more comprehensive comorbidity measures and detailed prescription data to enhance generalizability and validity. Collectively, these findings highlight the urgent need for randomized clinical trials in individuals with BD and CKD to evaluate the safety and long-term outcomes of SGLT2 inhibitors, establish optimal dosing ranges, and assess their impact on mood.
Conclusion
In individuals with bipolar disorder and mild-to-moderate chronic kidney disease, SGLT2 inhibitor therapy is associated with a lower risk of progression to kidney replacement therapy and reduced all-cause mortality over a 5-year follow-up period. These novel findings are consistent across subgroups defined by sex, diabetes status, race, ethnicity, mood stabilizer type, and CKD severity, suggesting a broad potential benefit. Although causality cannot be inferred from this retrospective analysis, the findings underscore the renal-protective and survival benefits of SGLT2is in individuals with BD and CKD. Overall, these findings support the need for prospective, randomized clinical trials of SGLT2is to confirm their safety, determine optimal dosing, evaluate long-term kidney and survival outcomes, and explore potential effects on mood in this high-risk population.
Supplemental Material
Supplemental material, sj-docx-1-tpp-10.1177_20451253261423437 for Sodium-glucose cotransporter-2 inhibitors lower risk of kidney replacement therapy and mortality in bipolar disorder with chronic kidney disease by Balwinder Singh, Maria L. Gonzalez Suarez, Ritika Baweja, Osama A. Abulseoud, Erika F. H. Saunders, Mark A. Frye and Raman Baweja in Therapeutic Advances in Psychopharmacology
Supplemental material, sj-docx-2-tpp-10.1177_20451253261423437 for Sodium-glucose cotransporter-2 inhibitors lower risk of kidney replacement therapy and mortality in bipolar disorder with chronic kidney disease by Balwinder Singh, Maria L. Gonzalez Suarez, Ritika Baweja, Osama A. Abulseoud, Erika F. H. Saunders, Mark A. Frye and Raman Baweja in Therapeutic Advances in Psychopharmacology
Acknowledgments
None.
Footnotes
ORCID iDs: Balwinder Singh 
https://orcid.org/0000-0001-7062-8192
Osama A. Abulseoud
https://orcid.org/0000-0002-0652-0862
Supplemental material: Supplemental material for this article is available online.
AI tools disclosure: During the preparation of this manuscript, the authors used Open AI ChatGPT 5 to edit and condense text. After using this tool, the authors reviewed and modified the content as pertinent and take full responsibility for the content of the manuscript.
Contributor Information
Balwinder Singh, Department of Psychiatry and Behavioral Health, Pennsylvania State College of Medicine, Hershey, PA, USA; Department of Psychiatry and Psychology, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA.
Maria L. Gonzalez Suarez, Division of Nephrology and Hypertension, Mayo Clinic, Rochester, MN, USA
Ritika Baweja, Department of Psychiatry and Behavioral Health, Pennsylvania State College of Medicine, Hershey, PA, USA.
Osama A. Abulseoud, Department of Psychiatry and Psychology, Mayo Clinic Arizona, Phoenix, AZ, USA
Erika F. H. Saunders, Department of Psychiatry and Behavioral Health, Pennsylvania State College of Medicine, Hershey, PA, USA
Mark A. Frye, Department of Psychiatry and Psychology, Mayo Clinic, Rochester, MN, USA
Raman Baweja, Department of Psychiatry and Behavioral Health, Pennsylvania State College of Medicine, Hershey, PA, USA.
Declarations
Ethics approval and consent to participate: Not required. This is a secondary analysis of de-identified data available from TriNetX. Study was deemed exempt from Institutional Review Board review, and patient consent was not required.
Consent for publication: Not applicable.
Author contributions: Balwinder Singh: Conceptualization: Investigation: Methodology: Project administration: Resources: Writing – original draft: Writing – review & editing.
Maria L. Gonzalez Suarez: Conceptualization: Writing – review & editing.
Ritika Baweja: Investigation: Writing – review & editing.
Osama A. Abulseoud: Investigation: Writing – review & editing.
Erika F. H. Saunders: Investigation: Writing – review & editing.
Mark A. Frye: Investigation: Writing – review & editing.
Raman Baweja: Conceptualization: Data curation: Formal analysis: Funding acquisition: Investigation: Methodology: Resources: Software: Writing – original draft: Writing – review & editing.
Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was supported by the National Center for Advancing Translational Sciences (NCATS), National Institutes of Health (NIH), through Grant UL1 TR002014 and Grant UL1 TR00045 to Penn State College of Medicine. This publication was supported by the KL2 TR00237 from the NCATS, a component of the NIH. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH. The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Balwinder Singh has received research grant support from Mayo Clinic, the National Network of Depression Centers, Breakthrough Discoveries for Thriving with Bipolar Disorder (BD2) and NIH. He is a KL2 Mentored Career Development Program scholar, supported by CTSA Grant Number KL2TR002379 from the NCATS. Dr. Singh has received honoraria (to Mayo Clinic) from Elsevier for editing a Clinical Overview on Treatment-Resistant Depression. Maria L. Gonzalez Suarez has served on the advisory board for Alexion and has participated as principal investigator/enrollment site for AstraZeneca. Mark A. Frye has been a consultant to or a member of the scientific advisory boards of Carnot Laboratories and American Physician Institute. He has been a principal or co-investigator on studies sponsored by Assurex Health, Baszucki Group, BD2, and Mayo Foundation. He also has financial interest/stock ownership/royalties with Chymia LLC.
Raman Baweja has received grant funding from Cardinal Health Foundation (Children's Hospital Association and Zero Suicide Initiative), research funding from Supernus and served on the advisory board for Ironshore. Other authors report no financial relationships with commercial interests.
Availability of data and materials: To gain access to the data in the TriNetX research network, a request can be made to TriNetX (https://live.trinetx.com), but costs may be incurred, a data sharing agreement would be necessary, and no patient identifiable information can be obtained.
References
- 1. Singh B, Swartz HA, Cuellar-Barboza AB, et al. Bipolar disorder. Lancet 2025; 406(10506): 963–978. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Liu YK, Ling S, Lui LMW, et al. Prevalence of type 2 diabetes mellitus, impaired fasting glucose, general obesity, and abdominal obesity in patients with bipolar disorder: a systematic review and meta-analysis. J Affect Disord 2022; 300: 449–461. [DOI] [PubMed] [Google Scholar]
- 3. Vancampfort D, Mitchell AJ, De Hert M, et al. Prevalence and predictors of type 2 diabetes mellitus in people with bipolar disorder: a systematic review and meta-analysis. J Clin Psychiatry 2015; 76(11): 1490–1499. [DOI] [PubMed] [Google Scholar]
- 4. Furtjes AE, Coleman JRI, Tyrrell J, et al. Associations and limited shared genetic aetiology between bipolar disorder and cardiometabolic traits in the UK Biobank. Psychol Med 2021; 52(16): 1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. SayuriYamagata A, Brietzke E, Rosenblat JD, et al. Medical comorbidity in bipolar disorder: The link with metabolic-inflammatory systems. J Affect Disord 2017; 211: 99–106. [DOI] [PubMed] [Google Scholar]
- 6. Rosenblat JD, McIntyre RS. Bipolar disorder and inflammation. Psychiatr Clin North Am 2016; 39(1): 125–137. [DOI] [PubMed] [Google Scholar]
- 7. Gardea-Resendez M, Pazdernik VK, Jezzini-Martinez S, et al. Unhealthy diet quality as a clinical phenotype in bipolar disorder: a potential treatment target for improved mood and metabolic outcomes. J Affect Disord 2026; 394(Pt B): 120632. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Kong L, Wang H, Yan N, et al. Effect of antipsychotics and mood stabilisers on metabolism in bipolar disorder: a network meta-analysis of randomised-controlled trials. EClinicalMedicine 2024; 71: 102581. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Parsaik AK, Chaudhary A, Sharma S, et al. Effects of long-term lithium therapy on kidney functioning in mood disorders: a population-based historical cohort study. Bipolar Disord 2025; 27(1): 28–35. [DOI] [PubMed] [Google Scholar]
- 10. Singh B, Ho AM, Coombes BJ, et al. Antipsychotic use in bipolar disorder: clinical and genomic correlates- a Mayo clinic bipolar disorder biobank study. Int J Bipolar Disord 2025; 14(1): 1. doi: 10.1186/s40345-025-00405-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Sinha A, Shariq A, Said K, et al. Medical comorbidities in bipolar disorder. Curr Psychiatry Rep 2018; 20(5): 36. [DOI] [PubMed] [Google Scholar]
- 12. Pahwa M, Kucuker MU, Ho MC, et al. Cardiometabolic and endocrine comorbidities in women with bipolar disorder: a systematic review. J Affect Disord 2022; 323: 841–859. [DOI] [PubMed] [Google Scholar]
- 13. Ercis M, Gonzalez Suarez ML, Singh B. Lithium and kidney disease. Mayo Clin Proc 2025; 100(1): 19–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Zhu N, Faucon AL, Kuja-Halkola R, et al. Prevalence of severe mental illness and its associations with health outcomes in patients with CKD: a swedish nationwide study. Am J Kidney Dis 2025; 85(5): 577–588 e571. [DOI] [PubMed] [Google Scholar]
- 15. Simes BC, MacGregor GG. Sodium-glucose cotransporter-2 (SGLT2) inhibitors: a clinician's guide. Diabetes Metab Syndr Obes 2019; 12: 2125–2136. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Romagnani P. SGLT2 inhibitors in CKD: are they really effective in all patients? Nephrol Dial Transplant 2025; 40(10): 1838–1842. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Wu JH, Foote C, Blomster J, et al. Effects of sodium-glucose cotransporter-2 inhibitors on cardiovascular events, death, and major safety outcomes in adults with type 2 diabetes: a systematic review and meta-analysis. Lancet Diabetes Endocrinol 2016; 4(5): 411–419. [DOI] [PubMed] [Google Scholar]
- 18. Verma S, Juni P, Mazer CD. Pump, pipes, and filter: do SGLT2 inhibitors cover it all? Lancet 2019; 393(10166): 3–5. [DOI] [PubMed] [Google Scholar]
- 19. Kalantar-Zadeh K, Jafar TH, Nitsch D, et al. Chronic kidney disease. Lancet 2021; 398(10302): 786–802. [DOI] [PubMed] [Google Scholar]
- 20. Tuttle KR, Rangaswami J. SGLT2 inhibitors as the bedrock of therapy for heart failure. Lancet 2022; 400(10354): 711–713. [DOI] [PubMed] [Google Scholar]
- 21. Heerspink HJL, Stefansson BV, Correa-Rotter R, et al. Dapagliflozin in patients with chronic kidney disease. N Engl J Med 2020; 383(15): 1436–1446. [DOI] [PubMed] [Google Scholar]
- 22. Perkovic V, Jardine MJ, Neal B, et al. Canagliflozin and renal outcomes in type 2 diabetes and nephropathy. N Engl J Med 2019; 380(24): 2295–2306. [DOI] [PubMed] [Google Scholar]
- 23. The EMPA-KIDNEY Collaborative Group. Empagliflozin in patients with chronic kidney disease. N Engl J Med 2023; 388(2): 117–127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Nuffield Department of Population Health Renal Studies Group and SGLT2 inhibitor Meta-Analysis Cardio-Renal Trialists' Consortium. Impact of diabetes on the effects of sodium glucose co-transporter-2 inhibitors on kidney outcomes: collaborative meta-analysis of large placebo-controlled trials. Lancet 2022; 400(10365): 1788–1801. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Pahwa M, Joseph B, Nunez NA, et al. Long-term lithium therapy and risk of chronic kidney disease in bipolar disorder: a historical cohort study. Bipolar Disord 2021; 23(7): 715–723. [DOI] [PubMed] [Google Scholar]
- 26. Powers J, Singh B, Stonerock LR, et al. Exploratory case series assessing the potential interaction between lithium and sodium-glucose cotransporter-2 inhibitors. J Clin Psychopharmacol 2025; 45(6): 662–665. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Ercis M, Tarikogullari I, Pazdernik VK, et al. Sodium-Glucose cotransporter 2 inhibitors for lithium-associated kidney dysfunction in mood disorders: a real-world historical cohort study. medRxiv 2025. [Google Scholar]
- 28. Baweja R, Waschbusch DA, Kong L, et al. Pediatric attention-deficit/hyperactivity disorder and disruptive mood dysregulation disorder: analyzing national treatment trends. J Pediatrics 2025; 279: 114471. [DOI] [PubMed] [Google Scholar]
- 29. Baweja R, Waschbusch DA, Rajalakshmi AK, et al. National trends in pediatric obsessive compulsive disorder management: insights from a retrospective cohort study. J Child Adolesc Psychopharmacol 2025; 35(8): 454–462. [DOI] [PubMed] [Google Scholar]
- 30. Institute of Medicine (US) Committee on Assessing the System for Protecting Human Research Participants; Federman DD, Hanna KE, Rodriguez LL, eds. Responsible research: a systems approach to protecting research participants. Washington (DC): National Academies Press (US), 2002. [PubMed] [Google Scholar]
- 31. Vandenbroucke JP, von Elm E, Altman DG, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. PLoS Med 2007; 4(10): e297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Singh B, Bakian AV, Newman M, et al. The association between thyrotropin and clinically relevant depression: a retrospective cross-sectional study. Thyroid 2025; 35(3): 245–254. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Kumar R, LeMahieu AM, Stan MN, et al. The association between thyroid stimulating hormone and depression: a historical cohort study. Mayo Clin Proc 2023; 98(7): 1009–1020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Su YC, Hsieh PC, Lai EC, et al. Comparison of sodium-glucose cotransporter 2 inhibitors versus glucagon-like peptide-1 receptor agonists and risks of osteoarthritis and arthroplasty in patients with type 2 diabetes mellitus: a propensity score-matched cohorts retrospective study. Diabetes Metab Res Rev 2025; 41(5): e70051. [DOI] [PubMed] [Google Scholar]
- 35. Pan HC, Chen JY, Chen HY, et al. Sodium-glucose cotransport protein 2 inhibitors in patients with type 2 diabetes and acute kidney disease. JAMA Netw Open 2024; 7(1): e2350050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Hubbard RA, Gatsonis CA, Hogan JW, et al. “Target Trial Emulation” for observational studies - potential and pitfalls. N Engl J Med 2024; 391(21): 1975–1977. [DOI] [PubMed] [Google Scholar]
- 37. Baweja R, Waschbusch DA, Padilla FM, et al. Attention-deficit/hyperactivity disorder treatment patterns and association with clinical outcomes in adolescents and young adults with co-occurring attention-deficit/hyperactivity disorder and substance use disorder: a retrospective analysis. J Am Acad Child Adolesc Psychiatry 2025: S0890–8567(25)02232-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Austin PC. A comparison of 12 algorithms for matching on the propensity score. Stat Med 2014; 33(6): 1057–1069. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Dong YH, Chang CH, Wu LC, et al. Continuation versus discontinuation of sodium-glucose cotransporter-2 inhibitors and cardiorenal outcomes among patients with type 2 diabetes and chronic kidney disease: a nationwide cohort study with a target trial emulation framework. Clin Transl Sci 2025; 18(8): e70319. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Natale P, Tunnicliffe DJ, Toyama T, et al. Sodium-glucose co-transporter protein 2 (SGLT2) inhibitors for people with chronic kidney disease and diabetes. Cochrane Database Syst Rev 2024; 5(5): CD015588. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Yankah RK, Anku EK, Eligar V. Sodium-glucose cotransporter-2 inhibitors and cardiovascular protection among patients with type 2 diabetes mellitus: a systematic review. J Diabetes Res 2024; 2024: 9985836. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Pahwa M, Singh B. Lithium therapy in patients with chronic kidney disease-A clinical dilemma? Eur Neuropsychopharmacol 2020; 34: 87. [DOI] [PubMed] [Google Scholar]
- 43. Singh B, Nunez NA, Schock L, et al. Lithium continuation for a patient with bipolar disorder and chronic kidney disease. Bipolar Disord 2020; 22(3): 316–318. [DOI] [PubMed] [Google Scholar]
- 44. Leaf DE, Bansal N. Another win for SGLT2 inhibitors in chronic kidney disease. N Engl J Med 2025; 392(8): 822–823. [DOI] [PubMed] [Google Scholar]
- 45. Fernandez-Fernandez B, Sarafidis P, Soler MJ, et al. EMPA-KIDNEY: expanding the range of kidney protection by SGLT2 inhibitors. Clin Kidney J 2023; 16(8): 1187–1198. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Muhammad RN, Albahairy MA, Abd El, Fattah MA, et al. Empagliflozin-activated AMPK elicits neuroprotective properties in reserpine-induced depression via regulating dynamics of hippocampal autophagy/inflammation and PKCzeta-mediated neurogenesis. Psychopharmacology 2024; 241(12): 2565–2584. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Sa-Nguanmoo P, Tanajak P, Kerdphoo S, et al. SGLT2-inhibitor and DPP-4 inhibitor improve brain function via attenuating mitochondrial dysfunction, insulin resistance, inflammation, and apoptosis in HFD-induced obese rats. Toxicol Appl Pharmacol 2017; 333: 43–50. [DOI] [PubMed] [Google Scholar]
- 48. Jensen NJ, Wodschow HZ, Nilsson M, et al. Effects of ketone bodies on brain metabolism and function in neurodegenerative diseases. Int J Mol Sci 2020; 21(22). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Brietzke E, Mansur RB, Subramaniapillai M, et al. Ketogenic diet as a metabolic therapy for mood disorders: evidence and developments. Neurosci Biobehav Rev 2018; 94: 11–16. [DOI] [PubMed] [Google Scholar]
- 50. Qiu H, Novikov A, Vallon V. Ketosis and diabetic ketoacidosis in response to SGLT2 inhibitors: basic mechanisms and therapeutic perspectives. Diabetes Metab Res Rev 2017; 33(5). [DOI] [PubMed] [Google Scholar]
- 51. Singh B, Yocum AK, Strawbridge R, et al. Patterns of pharmacotherapy for bipolar disorder: a GBC survey. Bipolar Disord 2024; 26(1): 22–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52. Kumar R, Joseph B, Pazdernik VM, et al. Real-world clinical practice among patients with bipolar disorder and chronic kidney disease on long-term lithium therapy. J Clin Psychopharmacol 2023; 43(1): 6–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental material, sj-docx-1-tpp-10.1177_20451253261423437 for Sodium-glucose cotransporter-2 inhibitors lower risk of kidney replacement therapy and mortality in bipolar disorder with chronic kidney disease by Balwinder Singh, Maria L. Gonzalez Suarez, Ritika Baweja, Osama A. Abulseoud, Erika F. H. Saunders, Mark A. Frye and Raman Baweja in Therapeutic Advances in Psychopharmacology
Supplemental material, sj-docx-2-tpp-10.1177_20451253261423437 for Sodium-glucose cotransporter-2 inhibitors lower risk of kidney replacement therapy and mortality in bipolar disorder with chronic kidney disease by Balwinder Singh, Maria L. Gonzalez Suarez, Ritika Baweja, Osama A. Abulseoud, Erika F. H. Saunders, Mark A. Frye and Raman Baweja in Therapeutic Advances in Psychopharmacology