Abstract
Objective
Systemic lupus erythematosus (SLE) is characterized by widespread organ inflammation. Metformin, commonly used for diabetes mellitus type 2, has been explored for its anti‐inflammatory potential in SLE. This study investigates the association of metformin use on renal and cardiovascular outcomes in patients with SLE.
Methods
This is a retrospective study. We used the multicenter research network (TriNetX) database from 88 health care organizations globally. Patients with SLE aged 18 and above, admitted between January 1, 2014, and April 21, 2024, were included. Propensity score matching compared patients with SLE on metformin with those not on metformin, considering demographics, laboratory results, comorbidities, and baseline medication use. The study assessed outcomes, including lupus nephritis (LN), chronic kidney disease (CKD), and major adverse cardiovascular events (MACEs) at one and five years after SLE diagnosis.
Results
We identified 9,178 patients with SLE on metformin and 78,983 patients with SLE not on metformin. After propensity score matching, patients with SLE on metformin had higher levels of hemoglobin A1C, whereas patients not on metformin had higher levels of urea nitrogen. When comparing both groups, the risk of developing LN (risk ratio [RR] = 1.70 [1.17–2.41]; P = 0.004), CKD (RR = 1.27 [1.07–1.52]; P = 0.007), and MACEs (RR = 1.21 [1.00–1.46]; P = 0.04) was significantly higher among patients not on metformin at one year after SLE diagnosis. After five years, the risk of LN and CKD was also higher in patients with SLE not on metformin. MACE risk was no longer significant after five years of diagnosis between both groups.
Conclusion
Patients with SLE not on metformin have a higher risk of developing LN, CKD, and MACEs compared with patients treated with metformin. Metformin's anti‐inflammatory potential offers promise as a complementary therapy for SLE. Nonetheless, further research and clinical trials are needed to clarify its mechanisms, optimal dosage, and long‐term effects.
INTRODUCTION
Systemic lupus erythematosus (SLE) is a systemic autoimmune disease that can involve many organs, and most of the clinical manifestations share a vascular component triggered by endothelial dysfunction. 1 In these patients, the most important causes of long‐term mortality are renal failure, infections, and cardiovascular disease (CVD), including ischemic heart disease and cerebrovascular and peripheral artery disease. In a published metanalysis, data from a prior metanalysis indicated a three‐fold increase in all‐cause mortality in patients with SLE compared with the general population and an increase in the cause‐specific mortality rate, with renal disease having the highest mortality risk (standardized mortality ratios [SMRs]: 7.90, 95% confidence interval [95% CI] 5.50–11.00]. 2
Lupus nephritis (LN) is a severe complication of SLE, affecting about 40% of adults with SLE. More than 10% of patients with LN progress to end‐stage renal disease (ESRD) in the first 10 years of diagnosis, even with adequate treatment. The risk of progression from LN to ESRD has not decreased for nearly two decades, and LN remains a significant cause of death in patients with SLE. 3 Kidney damage, in turn, exerts a negative impact on the cardiovascular (CV) system, worsening risk factors for CV diseases such as hypertension (HTN), stroke, and coronary syndrome, among others. Several factors contribute to this phenomenon, including chronic inflammation, which stimulates atherogenesis; use of inflammatory suppressive therapies (eg, glucocorticoids), which can have CV side effects; and high prevalence of CV risk factors among patients with SLE (eg, smoking, HTN, and obesity). 4 Therefore, SLE treatment should target systemic inflammation using anti‐inflammatory drugs, thereby suppressing endothelial dysfunction.
Metformin is a well‐known hypoglycemic agent with increasing evidence of exerting antitumor, antiaging, cardioprotective, anti‐inflammatory, and immunomodulatory effects through AMP‐activated protein kinase (AMPK)–dependent and independent pathways. Metformin has also been shown to decrease immune cell activation and proliferation, proinflammatory cytokine production, and oxidative stress. 5 A recent in vivo study testing metformin as a treatment for SLE showed that administration of metformin reduced kidney damage, improved kidney function, and improved mice survival in experimental LN models via suppression of the systemic and intrarenal inflammatory response in the kidney, primarily through necroptosis and inflammasome activation through AMPK‐mediated inhibition of STAT3. 3
Despite current evidence demonstrating the cardioprotective properties of metformin and intensive research linking SLE to CV risk, especially LN and chronic kidney disease (CKD), the role of endothelial dysfunction and the underlying mechanisms remain incompletely understood without a specific targeted pharmacological treatment. Therefore, updated data regarding these complex mechanisms and new treatments targeting autoimmune‐driven inflammatory responses are urgently needed. This study aims to describe the potential effects of metformin use on renal and CV outcomes in patients with SLE.
METHODS
This is a retrospective cohort study. We used TriNetX, a global federated administrative database with real‐time data of electronic medical records. TriNetX allowed the analysis of approximately 106,499,019 patients from 88 major health care organizations across 4 countries (United States of America, Brazil, Georgia, and Taiwan). As a federated network, TriNetX received a waiver from Western Institutional Review Board because only aggregated counts, statistical summaries of deidentified information, and no protected health information were obtained. Details of the network have been described elsewhere. 6 , 7 All analyses were done in the TriNetX “Analytics” network using the browser‐based real‐time analytics features. We included patients with SLE older than 18 years old with a diagnosis of SLE and admitted to inpatient settings between January 1, 2014, and April 21, 2024, using the International Classification of Disease‐10 and 9 (ICD‐10 and ICD‐9) codes. We compared patients with SLE on metformin with patients with SLE not on metformin, using a 1:1 propensity score and accounting for demographic variables (age, gender, race), laboratory results (C‐reactive protein [CRP], erythrocyte sedimentation rate [ESR], creatinine, urea nitrogen, and hemoglobin A1C [HbA1C] levels), associated comorbidities (diabetes mellitus type 2 [DMT2], pre–diabetes mellitus, polycystic ovarian syndrome, essential HTN, hyperlipidemia [HLD], smoking), and use of medications (prednisone, statins, hydroxychloroquine [HCQ], immunosuppressants, N‐acetylcysteine [NAC], mammalian target of rapamycin [mTOR] inhibitors, calcineurin inhibitors, aspirin [ASA], aldosterone receptor blockers [ARBs], angiotensin‐converting enzyme inhibitors [ACEi], and antidiabetics) at baseline for at the last 10 years before the diagnosis of SLE. The outcomes of interest include the development of biopsy‐proven LN; CKD stage 1, 2, or 3; and major adverse cardiovascular events (MACEs), which include acute myocardial infarction and stroke. Patients with ESRD stage 4 or 5 or on dialysis were completely excluded of our analysis because these patients can have higher levels of markers of inflammation, urea nitrogen, and creatinine at baseline, which might imply greater risk of MACEs and death overall.
The exposure for outcomes was at one and five years after the index event (SLE diagnosis) during the time window. Risk ratios (RRs) with 95% CIs were calculated for each outcome. The greedy nearest‐neighbor algorithm with a caliper of 0.1 pooled SDs was used for matching. Continuous variables are represented as mean ± SD and were compared between the groups using independent‐samples Student's t‐tests. Categorical variables are reported as count (percentage) and were compared between the groups using the chi‐square test. RRs were calculated for each cohort of patients. A two‐sided P value less than 0.05 was considered statistically significant. All statistical analyses were performed on the TriNetX network.
RESULTS
We identified 9,178 patients with SLE on metformin and 78,983 patients with SLE not on metformin. At baseline, the mean ± SD age at diagnosis was 55.9 ± 14.1 years for patients with SLE on metformin and 51.0 ± 17.6 years for patients with SLE not on metformin. In both groups, most of the patients were female (84%) and White (around 50%). At baseline, patients with SLE on metformin had a greater prevalence of DMT2 (69.6%), HTN (69.9%), HLD (45.6%), prediabetes (9.9%), and smoking (7.0%). Also, these patients were more likely to be on other medications, including SLE‐controlling medications (prednisone in 49.9%, HCQ in 40.2%, mycophenolate mofetil in 8.3%, and calcineurin inhibitors, especially cyclosporin, in 4.2%), statins (most commonly atorvastatin in 37.0%), ARBs (most commonly losartan in 20.1%), ACEi (most commonly lisinopril in 30.6%), ASA (43.2%), and other antidiabetic medications (most commonly insulin in 46.9% of patients). Finally, patients with SLE on metformin were more likely to be on other antioxidant medications, including NAC in 2.95% and other immunomodulators such as mTOR inhibitors (sirolimus in 0.12%) (Table 1).
Table 1.
Baseline patient characteristics of study groups, before and after propensity score matching*
| SLE + metformin group (N = 9,178) | SLE nonmetformin group (N = 78,983) | P value | SLE + metformin group (N = 7,242) | SLE nonmetformin group (N = 7,242) | P value | |
|---|---|---|---|---|---|---|
| Characteristics | Before matching | After matching | ||||
| Demographics | ||||||
| Age at index, mean ± SD | 55.9 ± 14.1 | 50.1 ± 17.6 | <0.001 | 55.3 ± 14.4 | 56.3 ± 15.6 | <0.00 |
| Female, % | 84.1 | 84.7 | 0.16 | 84.0 | 83.8 | 0.85 |
| Male, % | 12.4 | 11.7 | 0.05 | 12.4 | 12.2 | 0.78 |
| Race, % | ||||||
| White | 50.2 | 53.3 | <0.001 | 50.6 | 51.3 | 0.36 |
| Black or African American | 28.4 | 22.5 | <0.001 | 27.5 | 28.3 | 0.27 |
| Asian | 5.9 | 9.2 | <0.001 | 6.3 | 5.3 | 0.013 |
| Diagnosis, % | ||||||
| Diabetes mellitus type 2 | 69.6 | 8.2 | <0.001 | 61.6 | 62.3 | 0.35 |
| Essential hypertension | 69.9 | 34.0 | <0.001 | 65.3 | 66.6 | 0.09 |
| Hyperlipidemia | 45.6 | 16.8 | <0.001 | 40.4 | 40.3 | 0.89 |
| Smoking | 7.0 | 3.3 | <0.001 | 6.1 | 5.9 | 0.67 |
| Prediabetes | 9.9 | 2.1 | <0.001 | 8.1 | 9.0 | 0.05 |
| Polycystic ovarian syndrome | 4.3 | 0.8 | <0.001 | 3.7 | 3.8 | 0.82 |
| Impaired fasting glucose | 3.9 | 1.3 | <0.001 | 3.6 | 4.1 | 0.10 |
| Impaired glucose tolerance | 2.4 | 0.6 | <0.001 | 1.9 | 2.1 | 0.40 |
| Medications, % | ||||||
| Prednisone | 49.9 | 31.8 | <0.001 | 46.6 | 48.5 | 0.02 |
| Hydroxychloroquine | 40.2 | 32.0 | <0.001 | 39.2 | 39.9 | 0.33 |
| Mycophenolate mofetil | 8.3 | 6.3 | <0.001 | 8.0 | 8.5 | 0.20 |
| Mycophenolic acid | 1.9 | 1.7 | 0.185 | 1.9 | 1.9 | 0.95 |
| Calcineurin inhibitors | ||||||
| Cyclosporin | 4.2 | 2.4 | <0.001 | 3.8 | 4.2 | 0.17 |
| Tacrolimus | 3.3 | 1.7 | <0.001 | 3.1 | 3.3 | 0.48 |
| Voclosporin | 0.1 | 0.0 | 0.004 | 0.1 | 0.2 | 0.67 |
| Antioxidants | ||||||
| N‐acetylcysteine | 2.95 | 2.00 | <0.001 | 2.84 | 2.98 | 0.60 |
| mTOR inhibitors | ||||||
| Sirolimus | 0.12 | 0.07 | 0.16 | 0.12 | 0.17 | 0.41 |
| Statins | ||||||
| Atorvastatin | 37.0 | 10.6 | <0.001 | 31.0 | 30.7 | 0.73 |
| Simvastatin | 12.0 | 3.7 | <0.001 | 12.4 | 12.7 | 0.57 |
| Pravastatin | 8.7 | 2.5 | <0.001 | 10.0 | 10.1 | 0.76 |
| Rosuvastatin | 10.7 | 2.9 | <0.001 | 8.5 | 8.4 | 0.83 |
| Insulin | 46.9 | 6.7 | <0.001 | 38.1 | 37.4 | 0.38 |
| Sulfonylureas | ||||||
| Glipizide | 9.5 | 0.4 | <0.001 | 5.5 | 3.9 | <0.001 |
| Glimepiride | 5.9 | 0.3 | <0.001 | 3.5 | 2.5 | 0.001 |
| DPP‐4 inhibitors | ||||||
| Sitagliptin | 8.3 | 0.3 | <0.001 | 4.4 | 3.2 | <0.001 |
| Linagliptin | 1.9 | 0.1 | <0.001 | 1.3 | 1.0 | 0.07 |
| GLP‐1 agonists | ||||||
| Dulaglutide | 4.4 | 0.1 | <0.001 | 2.0 | 1.4 | 0.005 |
| Exenetide | 1.6 | 0.1 | <0.001 | 0.9 | 0.7 | 0.15 |
| TZDs | ||||||
| Pioglitazone | 2.9 | 0.1 | <0.001 | 1.8 | 1.3 | 0.006 |
| Rosiglitazone | 0.1 | 0.0 | <0.001 | 0.2 | 0.1 | 0.82 |
| SGLT‐2 inhibitors | ||||||
| Empagliflozin | 4.1 | 0.2 | <0.001 | 2.0 | 1.4 | 0.007 |
| Dapaglifozin | 2.0 | 0.1 | <0.001 | 1.0 | 0.8 | 0.08 |
| ARBs | ||||||
| Losartan | 20.1 | 7.0 | <0.001 | 17.3 | 17.5 | 0.69 |
| Valsartan | 7.4 | 2.8 | <0.001 | 6.5 | 6.6 | 0.89 |
| Olmesartan | 2.7 | 0.9 | <0.001 | 2.1 | 2.3 | 0.53 |
| ACEi | ||||||
| Lisinopril | 30.6 | 10.4 | <0.001 | 25.8 | 26.0 | 0.74 |
| Enalapril | 2.4 | 1.1 | <0.001 | 2.1 | 2.1 | 0.90 |
| Aspirin | 43.2 | 20.4 | <0.001 | 39.1 | 39.3 | 0.82 |
| Laboratory results | ||||||
| HbA1C | 7.0 ± 1.9 | 5.8 ± 1.4 | <0.001 | 6.8 ± 1.8 | 6.4 ± 1.8 | <0.001 |
| Creatinine (mg/dL), mean ± SD | 0.9 ± 2.2 | 0.8 ± 0.8 | 0.001 | 0.9 ± 2.5 | 0.9 ± 0.6 | 0.54 |
| Urea nitrogen (mg/dL), mean ± SD | 15.5 ± 7.7 | 14.7 ±8.5 | <0.001 | 15.3 ± 7.6 | 16.4 ± 9.7 | <0.001 |
| CRP (mg/dL), mean ± SD | 19.6 ± 40.5 | 17.4 ± 38.8 | 0.001 | 19.1 ± 38.9 | 20.9 ± 41.7 | 0.07 |
| ESR (mm/h), mean ± SD | 34.0 ± 27.2 | 30.3 ± 27.7 | <0.001 | 33.7 ± 27.2 | 33.5 ± 27.7 | 0.78 |
ACEi, angiotensin‐converting enzyme inhibitor(s); ARB, aldosterone receptor blocker; CRP, C‐reactive protein; DPP‐4, dipeptidyl peptidase‐4; ESR, erythrocyte sedimentation rate; GLP‐1, glucagon‐like peptide‐1 agonists; HbA1C, hemoglobin A1C; SGLT‐2, sodium‐glucose co‐transporter‐2; SLE, systemic lupus erythematosus; TZD, thiazolidinedione.
After propensity score matching, patients with SLE who were on metformin had higher levels of HbA1C (6.8 ± 1.8 vs 6.4 ± 1.8, P < 0.001). On the other side, patients with SLE not on metformin had higher levels of urea nitrogen (16.4 ± 9.7 mg/dL vs 15.3 ± 7.6 mg/dL, P < 0.001) when compared with patients with SLE on metformin. Levels of creatinine, CRP, and ESR were higher among patients with SLE not on metformin at baseline before matching; however, these differences became not statistically significant after balancing both cohorts (Table 1).
Regarding the outcomes, within one year after the index diagnosis, patients with SLE not on metformin had a higher risk of developing biopsy‐proven LN (RR = 1.70 [1.17–2.41]; P = 0.004); CKD stage 1, 2, or 3 (RR = 1.27 [1.07–1.52]; P = 0.007); and MACEs (RR = 1.21 [1.00–1.46]; P = 0.04) when compared with patients with SLE who were taking metformin (Tables 2 and 3). Similarly, the risks were higher at five years for the outcome of LN (RR = 1.82 [1.35–2.45]; P < 0.001) and CKD stage 1, 2, or 3 (RR = 1.17 [1.04–1.31]; P = 0.006) among patients with SLE not on metformin. MACE risk was no longer statistically significant after five years of diagnosis in both groups of patients with SLE.
Table 2.
Cardiovascular and renal outcomes of patients with SLE grouped by metformin usage after one year of diagnosis*
| Outcomes a | SLE + metformin group (N = 9,178) | SLE nonmetformin group (N = 78,983) | RR (95% CI), P value | |
|---|---|---|---|---|
| Lupus nephritis | n after matching b | 7,118 | 7,108 | 1.70 (1.17–2.47), P = 0.004 |
| Patients with the outcome | 44 | 75 | ||
| CKD stage 1, 2, or 3 | n after matching b | 6,518 | 6,202 | 1.27 (1.07–1.52), P = 0.007 |
| Patients with the outcome | 212 | 258 | ||
| MACEs | n after matching b | 6,240 | 5,959 | 1.21 (1.00–1.46), P = 0.04 |
| Patients with the outcome | 198 | 229 |
CKD, chronic kidney disease; MACE, major adverse cardiovascular event; RR, risk ratio; SLE, systemic lupus erythematous; 95% CI, 95% confidence interval.
Propensity matching balanced cohorts according to demographic variables, laboratory results, associated comorbidities, and use of medications at baseline.
Patients who had the outcome before the time window were excluded from each cohort.
Table 3.
Cardiovascular and renal outcomes of patients with SLE grouped by metformin usage after five years of diagnosis*
| Outcomes a | SLE + metformin group (N = 9,17)8 | SLE nonmetformin group (N = 78,983) | RR (95% CI), P value | |
|---|---|---|---|---|
| Lupus nephritis | n after matching b | 7,118 | 7,108 | 1.82 (1.35–4.45), P < 0.001 |
| Patients with the outcome | 66 | 120 | ||
| CKD stage 1, 2, or 3 | n after matching | 6,518 | 6,202 | 1.17 (1.04–1.31), P = 0.006 |
| Patients with the outcome | 518 | 577 | ||
| MACE | n after matching | 6,240 | 5,959 | 1.00 (0.88–1.14), P < 0.001 |
| Patients with the outcome | 462 | 444 |
Abbreviations: CKD, chronic kidney disease; MACE, major adverse cardiovascular events; RR, risk ratio; SLE, systemic lupus erythematosus; 95% CI, 95% confidence interval.
Propensity matching balanced cohorts according to demographic variables, laboratory results, associated comorbidities, and use of medications at baseline.
Patients who had the outcome before the time window were excluded from each cohort.
DISCUSSION
This retrospective real‐world database study showed that patients with SLE who did not receive metformin had a greater risk of adverse events, including biopsy‐proven LN; CKD stage 1, 2, or 3; and MACEs. Our results also showed that the same cohort of patients who were not on metformin had higher levels of urea nitrogen, which is an indirect marker of worsening kidney function in these patients. Dysregulation of cytokines and adipokines is known to be a common feature of both SLE and other cardiometabolic conditions, suggesting a complex relationship between autoimmunity, obesity, inflammation, and atherosclerosis and a possible role of metformin as a new immunoregulatory agent in the treatment of SLE. 8
The effectiveness of metformin not only as a glucose‐lowering agent but also through multidirectional cardioprotective, metabolic, and anti‐inflammatory effects have been reported. Recent data show how metformin reduces the incidence of CVD and all‐cause mortality in patients with DMT2 and coronary artery disease. 9 , 10 The positive properties of the drug in the CV system are associated with a beneficial effect on the endothelium, protection against oxidative stress, and reduction of smooth muscle cell proliferation. 11 Similarly, metformin can effectively limit ischemia‐reperfusion injury and reduce infarct area in animal models of myocardial infarction, which is also seen in nondiabetic animals. 12
Similarly, several studies have evaluated the effect of metformin on the incidence of stroke. Yu et al studied the incidence of stroke and found that metformin use was associated with a lower risk of stroke in metformin users compared to other patients with DMT2 not treated with metformin. 13 In addition, a meta‐analysis conducted in 2017 reported a significant reduction in the incidence of stroke in patients taking metformin. 14 These prior findings are consistent with the results of our study, in which patients with SLE who did not receive metformin had a higher risk of MACEs, including myocardial infarction and stroke, one year after the index admission and diagnosis of SLE. This risk appears to equalize after five years of SLE diagnosis with patients taking metformin, most likely due to a higher incidence of MACE cases in both groups as the chronic inflammatory process continues.
Interestingly, the immunomodulatory properties of metformin have been proven not only in patients with diabetes mellitus but also in other medical conditions. Metformin affects several intracellular signaling pathways, including AMPK. Because AMPK and subsequent intracellular signaling promote T and B cell activation and differentiation, as well as inflammatory responses, metformin may have positive immunomodulatory and anti‐inflammatory effects in chronic inflammatory conditions, including SLE. 15 In 2020, Jang et al investigated whether metformin could improve the immunoregulatory function and potential therapeutic properties of mesenchymal stem cells (MSCs) in an animal lupus model. 16 The researchers found that metformin optimized the immunoregulatory properties of MSCs, inhibited the expansion of CD4‐CD8 T cells, and reduced the ratio of Th17/Treg cells, leading to significant improvements in disease activity, including inflammatory phenotype, glomerulonephritis, proteinuria, and anti–double‐stranded DNA IgG antibody production in mice. 16
Similarly, LN is known to be a significant complication and cause of death in patients with SLE, and it is mainly driven by the effects of the AMPK/STAT3 pathway, which lead to kidney dysfunction via releasing of proinflammatory cytokines such as interleukin‐1β (IL‐1β), IL‐18, and high‐mobility group box 1 and through pyroptosis. 5 Multiple in vivo and in vitro studies have investigated the therapeutic potential of metformin in the treatment of kidney damage caused by LN‐induced inflammation. Chen et al found that the administration of metformin in lupus‐prone mice improved renal function, as measured by reductions in urea nitrogen, urinary proteins, IgG, and complement C3 deposition in glomeruli, and overall alleviated kidney injury in LN through suppressing renal necroptosis and inflammation via the AMPK/STAT3 pathway. 3 Similar findings were observed in our study, in which patients with SLE treated with metformin had lower levels of kidney dysfunction biomarkers (especially urea nitrogen) and overall less risk of LN and CKD (stage 1, 2, and 3) compared with patients with SLE not on metformin. A recent clinical trial from Renji Hospital also found that metformin administration reduced subsequent disease flare in patients with SLE, consistent with findings in other animal studies. 17 , 18
Furthermore, other possible mechanisms implied in the anti‐inflammatory effects of metformin have also been associated with the alterations in AMPK–mTOR–STAT3 signaling. The inhibition of this pathway may have therapeutic value in patients with SLE by suppressing B cell differentiation into plasma cells and germinal centers 19 and restoring the balanced redox state in the mitochondria of immune cells. 20 , 21
Nevertheless, other medications with antioxidants properties have also been described as possible target therapies in patients with SLE. Oaks et al described how treatment with sirolimus in lupus‐prone mice selectively blocked mTORC1 activation and antiphospholipid antibody (aPL) production. 22 Previous studies have described how aPLs are a major source of cardiovascular and renal disease in patients with SLE with or without nephritis, 23 suggesting that such mechanisms may represent a treatment target in patients with SLE. Likewise, Lai et al described the protective effects of sirolimus, a well‐known mTOR inhibitor, in patients with SLE. 24 In this single‐arm, open‐label, phase 1/2 trial, sirolimus improved disease activity in patients with SLE after 12 months of treatment. Mechanisms described were associated with the inhibition of antigen‐induced T cell proliferation and correction of proinflammatory T cell lineage specification in patients with SLE. 24 The same author also described in another study the use of NAC in patients with SLE and how this drug can safely improve lupus disease activity by blocking mTOR in T lymphocytes and decreasing T cell dysfunction in patients with SLE. 25 Other mechanisms associated with NAC are the suppression of autoantibody formation and a modest survival benefit in patients with SLE. 26 As seen in our study, the use of sirolimus and NAC was higher in patients with SLE on metformin. Despite the potential enhancing beneficial effects of these medications in patients with SLE, the protective effects of metformin remained significant for the outcomes of the study after propensity score matching.
This study has several strengths, including the use of real‐world data and enhancing the external validity of its findings. The inclusion of a substantial number of patients further bolsters the statistical power and generalizability of the results, allowing for a more comprehensive understanding of the phenomena under investigation. Additionally, the multicenter and multigeographic design ensures that the study captures a broad spectrum of patient demographics, health care practices, and geographical variations, thereby enhancing its applicability across diverse populations and health care settings.
This analysis has several limitations. First, based on the TriNetX platform, it was not possible to confirm the precision of any electronic health records. There could be inaccuracies in the recorded ICD‐10 and ICD‐9 codes or their assignment. Likewise, important information such as SLE Disease Activity Index score, classes of LN among patients, and outcomes after kidney biopsy could not be obtained. Second, although we employed established outcome definitions and implemented propensity score matching to mitigate bias, it was not possible to eliminate misclassification bias and confounding because of inherent limitations in the health care database and characteristics of electronic health record studies. Finally, it is important to note that a study of this nature can only establish associations. Prospective cohort studies and other research designs will be necessary to pinpoint mechanisms and evaluate causality.
In summary, among patients with SLE, the use of metformin is associated with a decreased risk of developing LN, CKD, and MACEs. These findings underscore the potential protective effects of metformin in mitigating the progression of renal complications and cardiovascular events in individuals with SLE. Further studies and clinical trials may be warranted to validate these findings and explore the underlying mechanisms responsible for the observed protective effects of metformin in patients with SLE.
AUTHOR CONTRIBUTIONS
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr Gonzalez Moret had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design
Gonzalez Moret, Lo, Tan.
Acquisition of data
Gonzalez Moret.
Analysis and interpretation of data
Gonzalez Moret, Lo, Tan.
Supporting information
Disclosure form
Author disclosures are available at https://onlinelibrary.wiley.com/doi/10.1002/acr2.11698.
REFERENCES
- 1. Mancardi D, Arrigo E, Cozzi M, et al. Endothelial dysfunction and cardiovascular risk in lupus nephritis: new roles for old players? Eur J Clin Invest 2021;51(2):e13441. [DOI] [PubMed] [Google Scholar]
- 2. Yurkovich M, Vostretsova K, Chen W, et al. Overall and cause‐specific mortality in patients with systemic lupus erythematosus: a meta‐analysis of observational studies. Arthritis Care Res (Hoboken) 2014;66(4):608–616. [DOI] [PubMed] [Google Scholar]
- 3. Chen XC, Wu D, Wu HL, et al. Metformin improves renal injury of MRL/lpr lupus‐prone mice via the AMPK/STAT3 pathway. Lupus Sci Med 2022;9(1):e000611. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Nurmohamed MT, Heslinga M, Kitas GD. Cardiovascular comorbidity in rheumatic diseases. Nat Rev Rheumatol 2015;11(12):693–704. [DOI] [PubMed] [Google Scholar]
- 5. Peng X, Yang T, Liu G, et al. Piperine ameliorated lupus nephritis by targeting AMPK‐mediated activation of NLRP3 inflammasome. Int Immunopharmacol 2018;65:448–457. [DOI] [PubMed] [Google Scholar]
- 6. Stapff M. Use of electronic health data in clinical development. Pharm Ind 2017;79:204–210. [Google Scholar]
- 7. Stacey J, Mehta MD. Using EHR data extraction to streamline the clinical trial process. Clin Researcher 2017;April:2–7. Accessed February 27, 2024. https://trinetx.com/clinical-researcher-using-ehr-data-extraction-to-streamline-the-clinical-trial-process/ [Google Scholar]
- 8. Mok CC. Metabolic syndrome and systemic lupus erythematosus: the connection. Expert Rev Clin Immunol 2019;15(7):765–775. [DOI] [PubMed] [Google Scholar]
- 9. Soukas AA, Hao H, Wu L. Metformin as anti‐aging therapy: is it for everyone? Trends Endocrinol Metab 2019;30(10):745–755. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Paridari P, Jabermoradi S, Gholamzadeh R, et al. Can metformin use reduce the risk of stroke in diabetic patients? A systematic review and meta‐analysis. Diabetes Metab Syndr 2023;17(2):102721. [DOI] [PubMed] [Google Scholar]
- 11. Nesti L, Natali A. Metformin effects on the heart and the cardiovascular system: a review of experimental and clinical data. Nutr Metab Cardiovasc Dis 2017;27(8):657–669. [DOI] [PubMed] [Google Scholar]
- 12. Han Y, Xie H, Liu Y, et al. Effect of metformin on all‐cause and cardiovascular mortality in patients with coronary artery diseases: a systematic review and an updated meta‐analysis. Cardiovasc Diabetol 2019;18(1):96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Yu H, Yang RT, Wang SY, et al. [Metformin use and risk of ischemic stroke in patients with type 2 diabetes: a cohort study]. Beijing Da Xue Xue Bao Yi Xue Ban 2023;55(3):456–464. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Campbell JM, Bellman SM, Stephenson MD, et al. Metformin reduces all‐cause mortality and diseases of ageing independent of its effect on diabetes control: a systematic review and meta‐analysis. Ageing Res Rev 2017;40:31–44. [DOI] [PubMed] [Google Scholar]
- 15. Kim JW, Choe JY, Park SH. Metformin and its therapeutic applications in autoimmune inflammatory rheumatic disease. Korean J Intern Med 2022;37(1):13–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Jang SG, Lee J, Hong SM, et al. Metformin enhances the immunomodulatory potential of adipose‐derived mesenchymal stem cells through STAT1 in an animal model of lupus. Rheumatology (Oxford) 2020;59(6):1426–1438. [DOI] [PubMed] [Google Scholar]
- 17. Sun F, Geng S, Wang H, et al. Effects of metformin on disease flares in patients with systemic lupus erythematosus: post hoc analyses from two randomized trials. Lupus Sci Med 2020;7(1):e000429. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Yin Y, Choi SC, Xu Z, et al. Normalization of CD4+ T cell metabolism reverses lupus. Sci Transl Med 2015;7(274):274ra18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Lee SY, Moon, SJ , Kim EK, et al. Metformin suppresses systemic autoimmunity in Roquin san/san mice through inhibiting B cell differentiation into plasma cells via regulation of AMPK/mTOR/STAT3. J Immunol 2017;198(7):2661–2670. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Teng X, Brown J, Morel L. Redox homeostasis involvement in the pharmacological effects of metformin in systemic lupus erythematosus. Antioxid Redox Signal 2022;36(7–9):462–479. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Huang N, Winans T, Wyman B, et al. Rab4A‐directed endosome traffic shapes pro‐inflammatory mitochondrial metabolism in T cells via mitophagy, CD98 expression, and kynurenine‐sensitive mTOR activation. Nat Commun 2024;15(1):2598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Oaks Z, Winans T, Caza T, et al. Mitochondrial dysfunction in the liver and antiphospholipid antibody production precede disease onset and respond to rapamycin in lupus‐prone mice. Arthritis Rheumatol 2016;68(11):2728–2739. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. de Azevedo FVA, Maia DG, de Carvalho JF, et al. Renal involvement in antiphospholipid syndrome. Rheumatol Int 2018;38(10):1777–1789. [DOI] [PubMed] [Google Scholar]
- 24. Lai ZW, Kelly R, Winans T, et al. Sirolimus in patients with clinically active systemic lupus erythematosus resistant to, or intolerant of, conventional medications: a single‐arm, open‐label, phase 1/2 trial. Lancet 2018;391(10126):1186–1196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Lai ZW, Hanczko R, Bonilla E, et al. N‐acetylcysteine reduces disease activity by blocking mammalian target of rapamycin in T cells from systemic lupus erythematosus patients: a randomized, double‐blind, placebo‐controlled trial. Arthritis Rheum 2012;64(9):2937–2946. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Suwannaroj S, Lagoo A, Keisler D. Antioxidants suppress mortality in the female NZB × NZW F1 mouse model of systemic lupus erythematosus (SLE). Lupus 2001;10(4):258–265. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Disclosure form