Abstract
Background/Aim: Although dipeptidyl peptidase-4 inhibitors (DPP4is) are widely used in the treatment of diabetes, a known risk factor for renal function decline, the potential association between DPP4i use and acute kidney injury (AKI) remains controversial. This study aimed to evaluate the association between DPP4i and AKI while considering the concomitant use of inducers of AKI in patients with diabetes.
Patients and Methods: Data from the US Food and Drug Administration Adverse Event Reporting System were analyzed, focusing on seven DPP4is in daily use, namely alogliptin, anagliptin, linagliptin, saxagliptin, sitagliptin, teneligliptin, and vildagliptin. Disproportionality analysis was performed to determine the association between DPP4i use and AKI using reporting odds ratios and information components, stratified by age groups, in the presence or absence of 10 types of AKI inducer used concomitantly.
Results: Of 215,051 reports, a positive association between DPP4i and AKI was found for linagliptin, sitagliptin, and vildagliptin in the overall analysis. Age-stratified analysis revealed associations between the DPP4is linagliptin, saxagliptin, sitagliptin and vildagliptin and AKI in the middle-aged group of patients, whereas only sitagliptin and vildagliptin were associated with AKI in the elderly group. Exclusion of concomitant renin-angiotensin-aldosterone system inhibitors, diuretics, or non-steroidal anti-inflammatory drugs altered signal detection for certain DPP4is, with these changes varying by age group.
Conclusion: This study suggests that some DPP4is, including linagliptin, sitagliptin and vildagliptin, are associated with AKI, even without concomitant use of an AKI inducer. Given the widespread use of DPP4is and the severity of AKI, clinicians should be sufficiently informed about their potential relationship.
Introduction
Acute kidney injury (AKI) is a heterogeneous group of conditions characterized by a sudden decrease in glomerular filtration rate accompanied by an increase in serum creatinine level and oliguria (1). Its major complications include volume overload, electrolyte abnormalities, uremic complications, and drug toxicity. Although various medications can induce different forms of kidney injury, drug-induced damage to the tubulointerstitial compartment is a common cause of AKI. This damage can result from immune-mediated acute injury or hemodynamic effects, both leading to an increased serum creatinine level (2). The management of AKI includes specific treatment for the underlying cause and supportive care to prevent and manage complications. The incidence of AKI has been increasing over the years, potentially due to multiple factors, including an aging population, a rise in predisposing comorbidities, increased use of nephrotoxic drugs, and more frequent invasive procedures (3). Diabetes is a known comorbidity that increases the risk of AKI (4, 5).
Dipeptidyl peptidase-4 inhibitors (DPP4is), one of the types of drug most frequently used in the treatment of type 2 diabetes, are extremely well tolerated by elderly individuals (6). Interestingly, recent studies have reported several cases of AKI associated with the use of certain DPP4is (7-9). However, the association between DPP4i use and AKI currently remains controversial (10-14). Therefore, the present study evaluated the association between DPP4i and AKI in patients with diabetes using data from the US Food and Drug Administration Adverse Event Reporting System (FAERS) while considering the effects of concomitant use of AKI inducers.
Patients and Methods
Data source. We analyzed data from the FAERS, covering the period from the first quarter of 2013 through the end of 2019. Four out of the seven datasets comprising the FAERS database, namely patient demographic and administrative information (DEMO), drug/biological information (DRUG), adverse events (REAC), and indications for use/diagnosis (INDI), were used for analysis. In cases of redundant case identification numbers in the FAERS data, the most recent one was selected for analysis. Data containing no information regarding sex and age were excluded. Patients’ ages were recalculated based on the age code in the DEMO dataset, and individuals younger than 39 years were excluded. In this study, patients were stratified into two groups for analysis: middle-aged (40 to 64 years) and elderly (≥65 years) groups. A flowchart depicting the data extraction process used in this study is presented in Figure 1.
Flowchart of patient data extraction for this study of the association between acute kidney injury (AKI) and dipeptidyl peptidase-4 inhibitor (DPP4i) use in patients with diabetes mellitus (DM).
Drugs of interest. The drugs of interest comprised seven DPP4is prescribed for daily use: alogliptin, anagliptin, linagliptin, saxagliptin, sitagliptin, teneligliptin, and vildagliptin. Additionally, drugs reported to be associated with AKI, hereafter referred to as AKI inducers, were identified based on a previous study (15). These included: angiotensin-converting enzyme inhibitors (ACEIs: alacepril, benazepril, captopril, cilazapril, delapril, enalapril, fosinopril, imidapril, lisinopril, perindopril, quinapril, ramipril, temocapril, and trandolapril), angiotensin receptor blockers (ARBs: candesartan, irbesartan, losartan, olmesartan, telmisartan, and valsartan), sodium-glucose cotransporter-2 inhibitors (SGLT-2is: canagliflozin, dapagliflozin, empagliflozin, ipragliflozin, lusegliflozin, and tofogliflozin), loop diuretics (azosemide, bumetanide, furosemide, and torasemide), thiazide diuretics (benzylhydrochlorothiazide, hydrochlorothiazide, indapamide, and trichlormethiazide), nonsteroidal anti-inflammatory drugs (NSAIDs: ampiroxicam, celecoxib, diclofenac, etodolac, flurbiprofen, ibuprofen, indomethacin, loxoprofen, mefenamic acid, meloxicam, naproxen, and piroxicam), anti-herpes simplex virus drugs (acyclovir, famciclovir, and valacyclovir), calcineurin inhibitors (cyclosporine and tacrolimus), cisplatin, and vancomycin. All drugs including the combined formulations were used for analysis.
Definitions of diabetes mellitus and AKI. Indications in the INDI dataset and adverse events in the REAC dataset were coded using the Preferred Terms from the Medical Dictionary for Regulatory Activities (MedDRA) terminology. Diabetes mellitus was identified using the Standardized MedDRA Query (SMQ) Hyperglycaemia/new onset diabetes mellitus (SMQ 20000041). AKI as an adverse event was defined using the Preferred Term ‘Acute kidney injury’ (10069339).
Data analysis. Disproportionality analyses with different algorithms were used to evaluate the association between DPP4i use and AKI using reporting odds ratios (RORs), a frequentist measure (16), and information component (IC) values, an index derived from the Bayesian Confidence Propagation Neural Network (17), to detect signals. Signal scores were calculated using a case/non-case method (18, 19).
All statistical analyses and data visualization were performed using JMP Pro version 17.1 (SAS Institute, Cary, NC, USA). An association between DPP4i and AKI was considered detected when the lower limit of the 95% confidence interval (CI) for the ROR exceeded 1 and the lower limit of the 95% CI for the IC was above 0. Conversely, an inverse association (i.e., a protective effect) was considered when the upper limit of the 95% CI for the ROR was below 1 and the upper limit of the 95% CI for the IC was below 0.
Results
Patient background. After omitting redundant numbers and entries with missing information on sex or age, a total of 215,051 entries related to diabetes mellitus remained from an initial total of 8,224,296 (Figure 1). The characteristics of the patient cohort are summarized in Table I. A total of 7,190 reports of AKI were identified.
Characteristics of cohort of the patient with diabetes in this study.
Disproportionality analysis of the whole dataset for the association between DPP4i use and AKI. Disproportionality analysis of the whole dataset revealed that DPP4i use was associated with AKI (ROR=1.42, 95% CI=1.34-1.51; IC=0.40, 95% CI=0.32-0.49). Individually, associations were observed for linagliptin (ROR=1.30, 95% CI=1.14-1.48; IC=0.35, 95% CI=0.16-0.54), sitagliptin (ROR=1.46, 95% CI=1.36-1.57; IC=0.47, 95% CI=0.37-0.58), and vildagliptin (ROR=1.98, 95% CI=1.71-2.28; IC=0.91, 95% CI=0.70-1.12) (Table II). On the other hand, an inverse association was observed for alogliptin (ROR=0.45, 95% CI=0.31-0.67; IC=−1.08, 95% CI=−1.65-−0.51).
Association between dipeptidyl peptidase-4 inhibitor (DPP4i) and acute kidney injury (AKI) in patients with diabetes by age.
Analysis of data subsets stratified by age. An age-stratified analysis is presented in Table II. Subset analysis of all DPP4is showed an association with AKI in both groups. However, when analyzed individually, more DPP4is tended to be associated with AKI in the middle-aged group than in the elderly group (Table II). In the middle-aged group, associations were observed for linagliptin (ROR=1.43, 95% CI=1.16-1.77; IC=0.48, 95% CI=0.17-0.79), saxagliptin (ROR=1.48, 95% CI=1.14-1.93; IC=0.53, 95% CI=0.15-0.91), sitagliptin (ROR=1.32, 95% CI=1.17-1.49; IC=0.35, 95% CI=0.18-0.35), and vildagliptin (ROR=1.60, 95% CI=1.20-2.13; IC=0.63, 95% CI=0.21-1.04). Conversely, in the elderly group, only sitagliptin (ROR=1.52, 95% CI=1.38-1.66; IC=0.51, 95% CI=0.37-0.64) and vildagliptin (ROR=2.03, 95% CI=1.72-2.39; IC=0.93, 95% CI=0.69-1.17) were associated with AKI (Table II). In addition, inverse association was observed for alogliptin (ROR=0.24, 95% CI=0.13-0.45; IC=−1.90, 95% CI=−2.77-−1.02) in the only elderly group.
Analysis of data subsets stratified by age without concomitant use of AKI inducers. In this sub-analysis, which excluded reports with concomitant AKI inducers, we focused on the four DPP4is that had shown positive associations in Table II. The results of the evaluation of the association between DPP4i use and AKI after excluding 10 types of AKI inducer are presented in Table III, Table IV, Table V, and Table VI.
Association between dipeptidyl peptidase-4 inhibitor (DPP4i) and acute kidney injury (AKI) without concomitant AKI inducer use in patients with diabetes (whole dataset).
Association between dipeptidyl peptidase-4 inhibitor (DPP4i) and acute kidney injury (AKI) without concomitant AKI inducer use in patients with diabetes aged 40-64 years.
Association between m dipeptidyl peptidase-4 inhibitor (DPP4i) and acute kidney injury (AKI) without concomitant AKI inducer use in patients with diabetes aged over 65 years.
Summary of signal detection for dipeptidyl peptidase-4 inhibitor (DPP4i) and acute kidney injury (AKI).
Analysis of the whole dataset revealed that linagliptin, sitagliptin, and vildagliptin were associated with AKI even after excluding the 10 types of AKI inducers (Table III and Table VI). However, an association between saxagliptin and AKI was observed only when concomitant SGLT-2i use was excluded (Table III and Table VI). In the middle-aged group, an association between saxagliptin and AKI was found after excluding use of ACEi, SGLT-2i, NSAID, anti-herpes simplex virus drug, CNi, cisplatin, and vancomycin (Table IV and Table VI). However, for linagliptin, the association with AKI was not observed when reports involving concomitant ACEI use were excluded. Similarly, for sitagliptin, no association was detected when ARB users were excluded, nor for vildagliptin when loop diuretic users were excluded. In the elderly group, sitagliptin and vildagliptin were associated with AKI even after excluding concomitant use of the 10 types of AKI inducer (Table V and Table VI). However, no association was observed between saxagliptin and AKI even after excluding SGLT-2i users. Linagliptin was associated with AKI only after excluding reports involving concomitant ACEI and NSAID use (Table V and Table VI).
Discussion
This study evaluated the association between DPP4i and AKI in patients with diabetes stratified by age and considering the concomitant use of AKI inducer. Notably, our analysis of the whole dataset revealed that linagliptin, sitagliptin, and vildagliptin were associated with AKI. After stratifying patients into age groups, we found that linagliptin, saxagliptin, sitagliptin, and vildagliptin were associated with AKI in the middle-aged group, whereas only sitagliptin and vildagliptin were associated with AKI in the elderly group. Furthermore, after excluding concomitant use of 10 types of AKI-inducing drugs (including renin-angiotensin-aldosterone system inhibitors, diuretics, and NSAIDs), our analysis revealed age-dependent differences in the association between certain DPP4i and AKI compared to the analysis without such exclusions.
One proposed mechanism for AKI development with DPP4i involves a decline in the estimated glomerular filtration rate due to hemodynamic effects, similar to renin-angiotensin-aldosterone system inhibitors (2). Furthermore, an in silico and in vivo study reported that some DPP4is can potentially inhibit ACE at concentrations similar to those required for DPP4 inhibition (20). In particular, linagliptin and sitagliptin inhibit ACE (20), whereas anagliptin does not (21) Moreover, sitagliptin and vildagliptin have been reported to reduce the estimated glomerular filtration rate (22, 23), which may explain why both drugs were associated with AKI even after stratifying patients by age and considering AKI inducer use.
Some case studies have reported that linagliptin can cause AKI accompanied by hypotension and hyperkalemia when added to the treatment regimen of patients already receiving ACEis (7, 8). In the present study, although limited to the middle-aged group, no significant association was found for linagliptin or sitagliptin when excluding patients with concomitant ACEi/ARB use. Thus, DPP4i use might contribute to AKI, particularly in the presence of other predisposing conditions or risk factors.
Another case report suggested that sitagliptin caused acute tubulointerstitial nephritis (ATIN) (9), a significant and potential cause of AKI. The etiology of ATIN is diverse, with drugs having been implicated in over two-thirds (approximately 70%) of cases, followed by autoimmune diseases (5-20%) and infections (4-17%) (24). This disease is a hypersensitivity reaction directed against a renal antigen or mostly induced by extrarenal antigens, particularly drugs that bind to kidney structures (24). The results of this study suggest that some DPP4is may cause AKI via ATIN. Analysis according to age group showed that more DPP4is were associated with AKI in the middle-aged group than in the elderly group (Table VI), which is contrary to the general tendency of reduced renal function and increased susceptibility to AKI among elderly patients. However, similar results have been reported in studies examining the association between other drugs and AKI according to age groups using data from the FAERS (25, 26). This disparity in age among the patients who developed adverse events may be because drug-induced AKI is related to an immune hypersensitivity reaction, which is more common in relatively younger individuals (27). This finding might also be attributed to reporting bias given that serious adverse events occurring in relatively younger patients might be reported more frequently than those in the elderly patients.
In this study, alogliptin was the only drug that showed an inverse association with AKI. Such inverse associations between target drugs and adverse events are often present, which suggests potential drug-repositioning approaches may be possible (28, 29). Alogliptin has been reported to be renoprotective against renal injury induced by drugs such as doxorubicin and cisplatin in in vivo studies using rat models of drug-induced nephrotoxicity (30, 31). However, this effect has also been reported with other DPP4is, suggesting this might be a class effect mediated by anti-inflammatory effects via the glucagon-like peptide-1 pathway and reduction of oxidative stress (32), and is not necessarily unique to alogliptin. Therefore, this may be an early signal warranting further investigation to determine whether specific DPP4is are effective in preventing renal injury or if the observed inverse association for alogliptin is due to other factors.
Although our analysis of data from a spontaneous reporting system is a valuable method for signal detection, the potential limitations of such databases cannot be overlooked, as they may have affected the interpretation of the results. Firstly, several confounding factors can influence AKI, including comorbidities and concomitant medications (2). Although 10 different types of AKI-inducing factors were individually excluded in this study, other confounding factors, including other drugs and the concomitant use of multiple medications not considered in this study, may be involved in the development of AKI. Secondly, the FAERS database relies on spontaneous reports, potentially resulting in underreporting, false reporting, and incomplete information, which can introduce bias. Thirdly, the FAERS database is susceptible to missing data and inconsistencies in drug name reporting, including misspellings. This might have affected the accuracy of drug exposure assessment. For instance, some AKI inducers available in topical formulations might have been inadvertently included if the route of administration was not clearly specified. Finally, this study did not consider the dose and duration of use of the respective drugs of interest.
Conclusion
Our findings suggest that certain DPP4is, namely linagliptin, sitagliptin, and vildagliptin, were associated with AKI even without concomitant use of an AKI inducer. After stratifying patients by age, our analysis revealed that more DPP4is were associated with AKI in the middle-aged group than in the elderly group. Thus, clinicians should be aware of this potential adverse event.
Footnotes
Authors’ Contributions
K. Ohyama conceived and designed the study. T. Takahashi and S. Akiyama standardized drug nomenclature, and cleaned and analyzed the data. K. Ohyama drafted the manuscript. Y. Hori reviewed the manuscript. All Authors critically reviewed the manuscript and approved the final version.
Conflicts of Interest
The Authors declare that they have no conflicts of interest in relation to this study.
Artificial Intelligence (AI) Disclosure
No artificial intelligence (AI) tools, including large language models or machine-learning software, were used in the preparation, analysis, or presentation of this manuscript.
- Received December 24, 2025.
- Revision received February 9, 2026.
- Accepted February 10, 2026.
- Copyright © 2026 The Author(s). Published by the International Institute of Anticancer Research.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
References
Jump to section
Related Articles
Cited By...
- No citing articles found.