Sodium-glucose cotransporter 2 inhibitors vs. glucagon-like peptide-1 receptor agonists and dipeptidyl peptidase-4 inhibitors for incident and recurrent urolithiasis: a European Association of Urology Endourology systematic review

Introduction

Urolithiasis is a common, recurrent, and increasingly prevalent condition closely linked to systemic metabolic health. The European Association of Urology (EAU) guidelines report that urinary stone prevalence varies widely across populations, can exceed 10% in high-income countries, and has increased by more than 37% in some areas over recent decades. They also identify metabolic syndrome among diseases associated with stone formation and recommend general preventive measures such as generous fluid intake sufficient to achieve a urine volume greater than 2.5 L/day (1). Similarly, the American Urological Association (AUA) guidelines discuss obesity, hypertension, and diabetes mellitus as epidemiological/ metabolic risk factors associated with increased stone risk. However, current pharmacologic prevention strategies remain focused on stone-specific interventions, including thiazides, potassium citrate, allopurinol, cystine-binding thiol drugs, and acetohydroxamic acid, rather than antidiabetic drug classes (2). This highlights a potential disconnect between the recognized metabolic basis of stone disease and the limited integration of modern metabolic therapies into stone-prevention guidance.

In recent years, the treatment of type 2 diabetes (T2Ds) has evolved substantially with the widespread use of sodium-glucose cotransporter 2 inhibitors (SGLT2i), glucagon-like peptide-1 receptor agonists (GLP-1 RA), and dipeptidyl peptidase-4 inhibitors (DPP-4i). SGLT2i reduce renal glucose reabsorption in the proximal tubule, promoting glucosuria and osmotic diuresis; GLP-1 RA enhance glucose-dependent insulin secretion, suppress glucagon secretion, slow gastric emptying, and promote weight loss; and DPP-4i increase endogenous incretin activity by preventing degradation of GLP-1 and glucose-dependent insulinotropic polypeptide (3-5). Given their effects on body weight, insulin resistance, renal solute handling, urine volume, and urinary chemistry, these agents may plausibly influence urinary stone risk.

Among these drug classes, SGLT2i have attracted particular interest in relation to urolithiasis (6).

Observational studies have suggested a lower risk of urolithiasis among patients initiating SGLT2i compared with those receiving GLP-1 RA or DPP-4i, although the magnitude and consistency of this association vary across study designs, populations, and comparator groups (7,8). Proposed mechanisms include osmotic diuresis, increased urinary citrate excretion, increased urine pH, and favorable effects on inflammation and metabolic parameters (7). By contrast, the evidence regarding GLP-1 RA and DPP-4i remains less clearly defined, despite their potential indirect effects through weight loss, glycemic control, and improvement of metabolic syndrome (9).

However, important gaps remain. Although current urolithiasis guidelines acknowledge diabetes and other metabolic conditions as relevant risk factors, they do not provide specific recommendations regarding SGLT2i, GLP-1 RA, or DPP-4i for prevention of incident or recurrent stones. Moreover, urolithiasis is rarely a primary endpoint in diabetes trials, and available data are often observational, heterogeneous, and limited with respect to stone recurrence, stone composition, and urinary metabolic parameters. Therefore, this systematic review aims to evaluate the existing evidence on the association between modern antidiabetic drug classes, particularly SGLT2i, GLP-1 RA, and DPP-4i, and the risk of incident or recurrent urolithiasis. We present this article in accordance with the PRISMA reporting checklist (available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0541/rc) (10,11).

Methods

Protocol and registration

The review evaluated the comparative association between SGLT2i and urolithiasis outcomes, with GLP-1 RA and DPP-4i as the primary comparator and other antidiabetic drug classes considered as secondary active comparators.

Review question and PICO framework

The aim of the review was to systematically evaluate comparative clinical-event evidence on the association between SGLT2i use and incident and recurrent urolithiasis outcomes among adults receiving antidiabetic pharmacotherapy, using GLP-1 RA, DPP-4i, or other antidiabetic therapies as active comparators.

PICO—(P) adults aged 18 years or older receiving pharmacologic therapy for T2Ds, obesity, chronic kidney disease, heart failure, cardiovascular risk reduction, or another relevant clinical indication; (I) SGLT2i, including empagliflozin, dapagliflozin, canagliflozin, ertugliflozin, sotagliflozin, ipragliflozin, luseogliflozin, tofogliflozin, or remogliflozin; (C) active antidiabetic comparators, primarily GLP-1 RA and DPP-4i. GLP-1 RA included semaglutide, liraglutide, dulaglutide, exenatide, lixisenatide, and related incretin-based therapies when reported within GLP-1-based exposure frameworks. DPP-4i and other antidiabetic drug classes were eligible when used as active comparators in comparative clinical-event studies; (O) incident urolithiasis, recurrent urolithiasis, symptomatic stone events, emergency department visits, hospitalizations, urologic procedures, and severe or procedure-requiring stone events.

Eligibility criteria

Inclusion criteria

Studies were eligible for inclusion if they met the following criteria:

• Included adults aged 18 years or older;
• Evaluated SGLT2i as the exposure of interest;
• Included GLP-1 RA, DPP-4i, placebo, usual care, no exposure, or another antidiabetic drug class as a comparator;
• Reported incident urolithiasis, recurrent urolithiasis, symptomatic stone events, or stone-related healthcare outcomes;
• Used a randomized, non-randomized interventional, cohort, case-control, target trial emulation, or other comparative observational design.

Exclusion criteria

Studies were excluded if they met any of the following criteria:

• Included pediatric or pregnant populations;
• Did not evaluate SGLT2i or an eligible comparator;
• Did not report urolithiasis, stone recurrence, or stone-related healthcare outcomes;
• Reported only urinary biochemical or supersaturation outcomes without clinical stone events;
• Were animal, in vitro, or purely mechanistic studies;
• Were case reports, small uncontrolled case series, editorials, narrative reviews, conference abstracts;
• Were pharmacovigilance or disproportionality studies unable to estimate incidence or comparative risk;

Studies limited to urinary chemistry, pharmacovigilance, or mechanistic outcomes were not included in the primary clinical-event synthesis, but were considered for background or discussion when relevant.

Information sources and search strategy

A comprehensive literature search was performed across MEDLINE, Embase, Cochrane CENTRAL, and Scopus from database inception through April 25^th, 2026.

The search strategy combined free-text terms and MeSH terms where applicable. Search terms included: “SGLT2 inhibitor”, “sodium-glucose cotransporter-2 inhibitor”, “gliflozin”, “empagliflozin”, “dapagliflozin”, “canagliflozin”, “GLP-1 receptor agonist”, “glucagon-like peptide-1 receptor agonist”, “semaglutide”, “liraglutide”, “dulaglutide”, “DPP-4 inhibitor”, “dipeptidyl peptidase-4 inhibitor”, “urolithiasis”, “urolithiasis”, “kidney stone”, “renal stone”, “ureteral stone”, and “urinary stone”. Boolean operators (AND/OR) were used to combine search terms. No publication date restrictions were applied. Reference lists of included studies and relevant review articles were also screened to identify additional eligible studies.

Study selection

Search results were imported into Rayyan for screening. Duplicates were removed automatically and then checked manually. Two reviewers (A.P. and F.E.) independently screened titles and abstracts. Potentially eligible articles underwent full-text review by the same two reviewers. Disagreements were resolved by discussion, with adjudication by a third senior reviewer (B.K.S.) when necessary.

Studies for which only abstracts were available and full-text methods or outcome definitions could not be assessed were not included in the primary synthesis.

Data extraction

Data were collected in a standardized Google Sheet. Data extraction was performed independently by two reviewers (A.P. and F.E.), with disagreements resolved by discussion and, when necessary, adjudication by a senior reviewer (B.K.S.). Extracted variables included author, year, country, database or data source, study design, population, sample size, eligibility criteria, baseline stone history, exposure definition, comparator definition, follow-up duration, outcome definition, analytic method, adjustment strategy, effect estimates, confidence intervals (CIs), absolute risk measures, subgroup analyses, sensitivity analyses, funding source, and key limitations.

For each study, effect measures were extracted as reported by the original studies, including hazard ratios (HRs), rate or risk ratios (RRs), and odds ratios (ORs), together with their 95% CIs; absolute measures, including rate differences and numbers needed to treat, were recorded when reported. No effect estimates were recalculated or converted to a common metric. When studies reported multiple outcomes or comparator groups, data were extracted separately at the outcome level. If a variable was not directly reported (NDR), it was marked as NDR.

Risk of bias assessment

Risk of bias was assessed independently by A.P. and F.E., with disagreements resolved by discussion or senior adjudication by B.K.S. Non-randomized studies were assessed using ROBINS-I, focusing on confounding, participant selection, exposure classification, deviations from intended interventions, missing data, outcome measurement, and selective reporting (12).

Data synthesis

A narrative synthesis was performed for all included studies. Studies were grouped according to comparator class and outcome type: SGLT2i vs. GLP-1 RA, SGLT2i vs. DPP-4i, incident urolithiasis, recurrent urolithiasis, and severe or procedure-requiring urolithiasis.

A meta-analysis was not performed because of the small number of eligible studies and clinical and methodological heterogeneity across populations, comparator groups, outcome definitions, follow-up durations, and analytic approaches. Instead, effect estimates were summarized descriptively, prioritizing the most fully adjusted estimates reported by each study.

Heterogeneity was assessed qualitatively by comparing study design, data source, population characteristics, baseline urolithiasis history, comparator class, outcome definition, follow-up duration, and adjustment methods.

Certainty of evidence and reporting bias

The certainty of evidence for key outcomes was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework, considering risk of bias, inconsistency, indirectness, imprecision, and publication bias (13).

Assessment of small-study effects or publication bias was planned only if at least 10 studies were available for a given outcome. Because fewer than 10 studies were available, publication bias was described narratively rather than formally assessed.

Results

A total of 231 records were identified through database searching. After removal of 151 duplicate records, 80 records underwent title and abstract screening, of which 62 were excluded. A total of 18 full-text reports were assessed for eligibility. Of these, 12 reports were excluded because they did not meet the prespecified eligibility criteria, leaving 6 studies for inclusion in the final qualitative synthesis (7,14-18) (Figure 1).

Figure 1 PRISMA 2020 flow diagram. Flow diagram showing identification, screening, eligibility assessment, and inclusion of studies in the systematic review.

Characteristics of included studies

The main characteristics of the included studies are presented in Table 1. All six included studies were comparative observational clinical-event studies evaluating SGLT2i against active antidiabetic comparators. Two studies used a target trial emulation framework, two used an active-comparator new-user cohort design, and two used a retrospective active-comparator cohort design. The included studies used large administrative or registry-based data sources and were conducted over study periods ranging from approximately 68 to 144 months.

The study populations differed according to baseline stone history and comparator class. Kristensen et al. evaluated adults aged 40 years or older initiating SGLT2i or GLP-1 RA, with incident urolithiasis assessed in patients without prior urolithiasis and recurrent urolithiasis assessed in a prior-stone subgroup. In that study, 12,325 matched SGLT2i users and 12,325 matched GLP-1 RA users were included after propensity score matching, with a median follow-up of approximately 2 years (14).

Paik et al. used three US claims databases to compare new SGLT2i users with new GLP-1 RA or DPP-4i users among adults with T2Ds and no prior kidney or urinary tract stones. After propensity score matching, the study included 358,203 matched pairs for the SGLT2i vs. GLP-1 RA comparison and 331,028 matched pairs for the SGLT2i vs. DPP-4i comparison, with median follow-up under one year (7).

McCormick et al. specifically focused on adults with T2Ds and pre-existing urolithiasis. This study compared SGLT2i initiation with GLP-1 RA initiation, with DPP-4i used as an alternative comparator. The study included 20,146 patients with urolithiasis and T2Ds before weighting, and follow-up averaged 1.3 years among SGLT2i users and 0.93 years among GLP-1 RA users (15).

Shin et al. evaluated SGLT2i vs. DPP-4i among patients with T2Ds using the Korea National Health Insurance Service database. The study separately analyzed stone never-formers and stone ever-formers, allowing distinction between incident stone risk and events among patients with prior stone history (16).

Chung et al. also compared SGLT2i with DPP-4i, using both the Taiwan National Health Insurance Research Database (NHIRD) and TriNetX databases, and assessed both incident urolithiasis and urolithiasis requiring surgery (17).

Lukkunaprasit et al. evaluated Thai adults with T2Ds treated with SGLT2i, DPP-4i, sulfonylureas, or thiazolidinediones in a hospital-based cohort. This study included 17,821 patients, excluded those with urolithiasis before or within one month after treatment initiation, and defined incident urolithiasis using diagnostic codes confirmed by imaging reports (18).

SGLT2i vs. GLP-1 RA

Comparative outcomes for SGLT2i vs. GLP-1 RA are summarized in Table 2. Across studies, SGLT2i use was consistently associated with a lower risk of urolithiasis outcomes compared with GLP-1 RA use.

In the incident-stone analysis by Kristensen et al., SGLT2i initiators had a lower rate of incident urolithiasis than GLP-1 RA initiators (2.0 vs. 4.0 events per 1,000 person-years), corresponding to an HR of 0.51 (95% CI: 0.37–0.71) and a rate difference of −1.9 events per 1,000 person-years. In the recurrent urolithiasis subgroup, the event rate was also lower with SGLT2i (36 vs. 53 events per 1,000 person-years), with an HR of 0.68 (95% CI: 0.48–0.97) and a rate difference of −17 events per 1,000 person-years (14).

Paik et al. provided additional incident-stone evidence from three large US claims databases among adults with T2Ds and no prior kidney or urinary tract stones. After 1:1 propensity score matching, SGLT2i initiators had a lower urolithiasis event rate than GLP-1 RA initiators (14.9 vs. 21.3 events per 1,000 person-years), corresponding to an HR of 0.69 (95% CI: 0.67–0.72) and a rate difference of −6.4 events per 1,000 person-years (7).

McCormick et al. reported similar findings among patients with T2Ds and pre-existing urolithiasis. Recurrent urolithiasis occurred at a rate of 105.3 events per 1,000 person-years among SGLT2i initiators compared with 156.4 events per 1,000 person-years among GLP-1 RA initiators. This corresponded to a rate ratio of 0.67 (95% CI: 0.57–0.79), a rate difference of −51 events per 1,000 person-years, and a number needed to treat of 20 (15).

The absolute association was more pronounced among patients with urolithiasis in the preceding 12 months. In this high-risk subgroup, SGLT2i initiation was associated with 219 fewer recurrent urolithiasis events per 1,000 person-years compared with GLP-1 RA initiation, corresponding to a number needed to treat of 5. McCormick et al. also found lower rates of more clinically intensive outcomes, including recurrent urolithiasis requiring emergency department visit or hospitalization and recurrent urolithiasis requiring procedures (15).

Among patients with concomitant gout, the association remained directionally consistent. SGLT2i initiation was associated with a lower rate of recurrent urolithiasis compared with GLP-1 RA initiation, with a rate ratio of 0.67 (95% CI: 0.57–0.79), a rate difference of −53 events per 1,000 person-years, and a number needed to treat of 19. In a secondary target-trial analysis among patients without urolithiasis history at baseline, SGLT2i were also associated with lower incident urolithiasis compared with GLP-1 RA (7.4 vs. 12.8 events per 1,000 person-years; rate ratio 0.58, 95% CI: 0.52–0.64) (15).

SGLT2i vs. DPP-4i

Comparative outcomes for SGLT2i vs. DPP-4i are presented in Table 3. The findings were again consistent, with all included studies reporting lower urolithiasis or urolithiasis risk among SGLT2i users compared with DPP-4i users.

In the DPP-4i comparison from Kristensen et al., SGLT2i initiation was associated with a lower risk of incident urolithiasis, with event rates of 2.3 vs. 3.7 events per 1,000 person-years, an HR of 0.61 (95% CI: 0.41–0.88), and a rate difference of −1.5 events per 1,000 person-years (14).

Paik et al. provided additional large-scale US evidence among adults with T2Ds and no prior kidney or urinary tract stones. After 1:1 propensity score matching across three claims databases, SGLT2i initiators had a lower urolithiasis event rate than DPP-4i initiators (14.6 vs. 19.9 events per 1,000 person-years), corresponding to an HR of 0.74 (95% CI: 0.71–0.77) and a rate difference of −5.3 events per 1,000 person-years (7).

McCormick et al. also observed lower recurrent urolithiasis rates when DPP-4i were used as the active comparator. Among patients with T2Ds and pre-existing urolithiasis, recurrent urolithiasis occurred at a rate of 108.9 events per 1,000 person-years among SGLT2i initiators compared with 146.5 events per 1,000 person-years among DPP-4i initiators. The corresponding rate ratio was 0.73 (95% CI: 0.68–0.78), with a rate difference of −38 events per 1,000 person-years and a number needed to treat of 26. Among patients with concomitant gout, the association was stronger in absolute terms, with a rate difference of −62 events per 1,000 person-years and a number needed to treat of 16 (15).

Shin et al. provided the largest DPP-4i comparison and analyzed stone never-formers and stone ever-formers separately. In the overall matched cohort, SGLT2i initiation was associated with a lower risk of urolithiasis than DPP-4i initiation (0.65 vs. 1.12 events per 100 person-years; HR 0.54, 95% CI: 0.50–0.57). Among stone never-formers, the HR was 0.43 (95% CI: 0.39–0.48), whereas among stone ever-formers it was 0.64 (95% CI: 0.59–0.69). Although the relative risk reduction was greater among stone never-formers, the absolute incidence rate difference was larger among stone ever-formers (−2.26 vs. −0.32 events per 100 person-years), reflecting their higher baseline event risk. SGLT2i initiation was also associated with lower risk of stone-removing interventions overall and in both stone-history subgroups (16).

Chung et al. extended these findings to incident urolithiasis and surgery-requiring urolithiasis across two independent databases. In the Taiwan NHIRD cohort, SGLT2i users had lower incidence of urolithiasis than DPP-4i users (10.26 vs. 12.78 cases per 1,000 patient-years), with an adjusted HR of 0.82 (95% CI: 0.77–0.87). Surgery-requiring urolithiasis was also less frequent with SGLT2i (2.04 vs. 3.12 cases per 1,000 patient-years), with an adjusted HR of 0.72 (95% CI: 0.63–0.82). In the TriNetX cohort, the corresponding adjusted HRs were 0.84 (95% CI: 0.78–0.90) for incident urolithiasis and 0.71 (95% CI: 0.56–0.89) for surgery-requiring urolithiasis (17).

Lukkunaprasit et al. further supported the DPP-4i comparison in a Thai hospital-based cohort of adults with T2Ds. Incident urolithiasis, defined by diagnostic codes and confirmed by imaging reports, occurred at a lower rate among SGLT2i users than DPP-4i users (7.7 vs. 18.5 events per 1,000 person-years), corresponding to an HR of 0.45 (95% CI: 0.35–0.58) (18).

Risk assessment

Risk of bias assessment using ROBINS-I is summarized in Figures S1,S2. Overall, all six included studies were judged to have moderate risk of bias (7,14-18).

GRADE of certainty

The certainty of evidence was assessed using GRADE for prespecified outcome-comparison pairs. All evidence was derived from non-randomized comparative observational studies and therefore started at low certainty. Risk of bias was explicitly considered; although all included studies were judged to have moderate overall risk of bias using ROBINS-I, no additional downgrade was applied because no study was considered at serious or critical risk of bias and the identified limitations were not judged sufficient to lower certainty by a further level. Evidence was also not downgraded for inconsistency or indirectness as effect estimates were directionally consistent and aligned with the review PICO. No upgrading was applied because the large-effect criterion was not consistently met and residual confounding could not be excluded. Overall, certainty was judged as low for incident, recurrent, and severe or procedure-requiring urolithiasis outcomes (Table S1).

Discussion

This systematic review was designed to determine whether the emerging association between modern antidiabetic drug classes and stone risk is supported by comparative clinical-event evidence, with particular focus on SGLT2i relative to GLP-1 RA and DPP-4i. Across six large observational studies, SGLT2i use was consistently associated with fewer urolithiasis or urolithiasis events across multiple clinical settings, including incident stones among patients without prior urolithiasis, recurrent events among stone formers, and more severe outcomes requiring emergency department care, hospitalization, surgery, or stone-removing intervention (7,14-18). In comparisons with GLP-1 RA, SGLT2i were associated with lower incident urolithiasis in patients without prior stones in both Kristensen et al. and Paik et al. (HR 0.51, 95% CI: 0.37–0.71, and HR 0.69, 95% CI: 0.67–0.72, respectively), lower recurrent urolithiasis among prior stone formers (HR 0.68, 95% CI: 0.48–0.97), and lower recurrent urolithiasis in patients with established stone disease in the McCormick target-trial emulation (RR 0.67, 95% CI: 0.57–0.79) (7,14,15). The largest absolute difference was observed among patients with urolithiasis in the preceding 12 months, with event rates of 346.1 vs. 565.1 per 1,000 person-years, corresponding to 219 fewer events per 1,000 person-years and an NNT of 5 (15). In comparisons with DPP-4i, the direction of association remained consistent, including lower incident urolithiasis in Kristensen et al. and Paik et al., lower recurrent urolithiasis in McCormick et al. (RR 0.73, 95% CI: 0.68–0.78), lower urolithiasis risk in stone never-formers and ever-formers in Shin et al. (HR 0.43 and 0.64, respectively), and lower incident and surgery-requiring urolithiasis in both the Taiwan NHIRD and TriNetX cohorts in Chung et al. and lower imaging-confirmed incident urolithiasis in the Thai hospital-based cohort by Lukkunaprasit et al. (7,14-18). Importantly, these findings should be interpreted as a consistent comparative signal favoring SGLT2i, rather than evidence that GLP-1 RA or DPP-4i increase stone risk.

The present findings are consistent with, but more narrowly focused than, the broader emerging literature on antidiabetic therapies and stone risk. Yeh et al. combined a real-world cohort study with a meta-analysis and concluded that SGLT2i may be associated with lower urolithiasis risk compared with GLP-1 RA and DPP-4i, while Kanbay et al. reported a lower pooled risk of urolithiasis with SGLT2i therapy compared with placebo or active control (8,19). However, these broader syntheses differ from the present review because they include heterogeneous study designs, control groups, and levels of evidence, whereas our analysis was restricted to comparative clinical-event studies and separated incident, recurrent, and procedure requiring outcomes. The literature on GLP-1 RA remains less definitive: available stone-specific evidence is largely limited to urinary chemistry data, with Feghali et al. reporting that GLP-1 based weight-loss therapy in obese stone formers was associated with lower urinary oxalate and sulfate and no significant worsening of urine supersaturation indices (9). This suggests that the lower event rates observed with SGLT2i should not be interpreted as evidence that GLP-1 RA are lithogenic. Radwan et al. similarly reported lower urolithiasis risk with SGLT2i compared with DPP-4i in older adults with T2Ds, though evidence for DPP-4i as a stone-modifying therapy remains limited (20). Overall, the current literature supports a consistent SGLT2i-favorable signal but also highlights the limited quantity and quality of prospective, stone-specific evidence across modern antidiabetic drug classes. Additional cross-sectional evidence from a large Japanese administrative database by Anan et al. showed that SGLT2i use was associated with lower odds of urolithiasis in patients with diabetes and, notably, in non-diabetic men (OR 0.42, 95% CI: 0.35–0.51), although the cross-sectional design precludes causal inference (21).

A possible explanation for the consistent clinical signal is that SGLT2i may affect stone risk through both urinary and metabolic pathways. Several studies that were not eligible for the primary clinical-event synthesis support this possibility. In the SWEETSTONE randomized crossover trial, empagliflozin improved urinary lithogenic risk profiles in nondiabetic calcium and uric acid stone formers, mainly through changes in calcium phosphate and uric acid relative supersaturation; however, this trial evaluated urinary surrogate outcomes rather than stone recurrence (22). Schaub et al. similarly reported that SGLT2i use was associated with higher urine volume and higher urinary citrate in a 24-hour urine cohort of stone formers, whereas GLP-1 RA showed fewer consistent changes in urine parameters related to stone risk (23). Mechanistic work by Anan et al. also suggested that SGLT2 inhibition may reduce calcium oxalate stone formation by limiting tubular injury, inflammation, osteopontin expression, and crystal adhesion, although these findings remain preclinical and hypothesis-generating (24). By contrast, available GLP-1-based data do not suggest a clear lithogenic effect as stated by Feghali et al. (9). Together, these studies suggest that the lower stone event rates observed with SGLT2i may reflect a specific favorable effect of this drug class rather than harm from GLP-1 RA or DPP-4i. More broadly, they support the idea that stone prevention may eventually need to consider systemic metabolic pathways in addition to traditional stone-specific treatments, although this remains to be proven in prospective clinical studies.

This review has several limitations. First, the included evidence base was small and consisted entirely of observational studies, so causal inference remains limited despite the use of active comparator designs, propensity score methods, and target trial emulation in several studies. Second, the available studies were based largely on administrative or registry data, meaning that urolithiasis and urolithiasis outcomes were generally identified using diagnostic or procedure codes rather than standardized imaging confirmation or stone analysis. Third, the included studies differed in baseline stone history, comparator class, follow-up duration, outcome definitions, and analytic approach, which precluded meta-analysis and required narrative synthesis. Finally, most evidence focused on patients with T2Ds and may not be generalizable to non-diabetic stone formers or to patients receiving SGLT2i primarily for heart failure, chronic kidney disease, or obesity alone. Importantly, the present review was designed to evaluate the relative association of SGLT2i with stone outcomes compared with other active antidiabetic therapies, rather than to determine the absolute effect of each drug class relative to an untreated or placebo control. Thus, the current evidence should be considered hypothesis-generating and insufficient to support selection of an antidiabetic therapy specifically for stone prevention. A quantitative synthesis may become more appropriate as additional prospective and clinically homogeneous studies become available.

Future research should prioritize high-quality prospective studies with prespecified clinical stone outcomes, longer follow-up, and standardized outcome ascertainment. This is particularly important given the rapid expansion of SGLT2i, GLP-1 RA, and related metabolic therapies in recent years. Few early stone-focused trials are beginning to address this gap specifically for SGLT2i, particularly the SWEETSTONE trial (22). Larger ongoing studies, including EMPASTONE, are now assessing whether empagliflozin and personalized dietary counseling can reduce kidney stone recurrence (25). In contrast, comparable prospective studies of GLP-1 RA, DPP-4i, or other antidiabetic drug classes remain limited, with existing GLP-1-based evidence largely being restricted to urinary chemistry data rather than long-term clinical stone outcomes (9).

Future studies should also clarify whether any stone-risk modification is specific to these metabolic therapies, ideally incorporating stone composition, 24-hour urine profiling, recurrence outcomes, and populations beyond T2Ds. Such studies should also account for established dietary and lifestyle factors, including fluid intake patterns and beverage consumption, which remain important modifiers of stone risk and may confound treatment effects (26). Furthermore, standardized risk stratification frameworks and validated predictive tools should be incorporated into study design and outcome reporting to facilitate comparisons across cohorts and improve individualized recurrence prediction (27).

Conclusions

In this systematic review of comparative clinical-event studies, SGLT2i use was consistently associated with a lower risk of incident, recurrent, and severe or procedure-requiring urolithiasis compared with GLP-1 RA or DPP-4i. The association was observed across different populations, comparator groups, and data sources, and appeared most clinically relevant among patients with prior or recently active urolithiasis, in whom the absolute risk reduction was greatest. However, the evidence remains based on observational studies with moderate overall risk of bias, and the findings should therefore be interpreted as associations rather than proof of a causal preventive effect. Future prospective studies and randomized trials with prespecified stone outcomes are needed to confirm these findings and determine whether these agents may have a role in stone risk reduction beyond their established metabolic, cardiovascular, and renal indications.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the PRISMA reporting checklist. Available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0541/rc

Peer Review File: Available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0541/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0541/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs. 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. European Association of Urology. EAU Guidelines on Urolithiasis. EAU Guidelines 2024. Available online: https://uroweb.org/guidelines/urolithiasis
2. Pearle MS, Goldfarb DS, Assimos DG, et al. Medical management of kidney stones: AUA guideline. J Urol 2014;192:316-24. [Crossref] [PubMed]
3. Padda IS, Mahtani AU, Parmar M. Sodium-Glucose Transport 2 (SGLT2) Inhibitors. [Updated 2025 Sep 15]. In: StatPearls. Treasure Island (FL): StatPearls Publishing; 2026.
4. Collins L, Costello RA. Glucagon-Like Peptide-1 Receptor Agonists. [Updated 2024 Feb 29]. In: StatPearls. Treasure Island (FL): StatPearls Publishing; 2026.
5. Kasina SVSK, Baradhi KM. Dipeptidyl Peptidase IV (DPP IV) Inhibitors. [Updated 2023 May 22]. In: StatPearls. Treasure Island (FL): StatPearls Publishing; 2026.
6. Talyshinskii A, Nedbal C, Somani BK. Urological impact of flozins (SGLT2 inhibitors): an EAU Endourology review of risks, side effects and clinical considerations. Curr Opin Urol 2026;36:42-50. [Crossref] [PubMed]
7. Paik JM, Tesfaye H, Curhan GC, et al. Sodium-Glucose Cotransporter 2 Inhibitors and Nephrolithiasis Risk in Patients With Type 2 Diabetes. JAMA Intern Med 2024;184:265-74. [Crossref] [PubMed]
8. Yeh JA, Liu YC, Huang AH, et al. SGLT2 inhibitors and nephrolithiasis risk in patients with type 2 diabetes: A cohort study and meta-analysis. Diabetes Res Clin Pract 2025;222:112088. [Crossref] [PubMed]
9. Feghali K, Li X, Maalouf NM. Changes in 24-Hour Urine Chemistry in Patients with Nephrolithiasis during Weight Loss with Glucagon-Like Peptide 1-Based Therapies. Kidney360 2024;5:1706-12. [Crossref] [PubMed]
10. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71. [Crossref] [PubMed]
11. Cumpston M, Li T, Page MJ, et al. Updated guidance for trusted systematic reviews: a new edition of the Cochrane Handbook for Systematic Reviews of Interventions. Cochrane Database Syst Rev 2019;10:ED000142.
12. Sterne JA, Hernán MA, Reeves BC, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ 2016;355:i4919. [Crossref] [PubMed]
13. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ 2008;336:924-6. [Crossref] [PubMed]
14. Kristensen KB, Henriksen DP, Hallas J, et al. Sodium-glucose cotransporter 2 inhibitors and risk of nephrolithiasis. Diabetologia 2021;64:1563-71. [Crossref] [PubMed]
15. McCormick N, Yokose C, Lu N, et al. Comparative effectiveness of sodium-glucose cotransporter-2 inhibitors for recurrent nephrolithiasis among patients with pre-existing nephrolithiasis or gout: target trial emulation studies. BMJ 2024;387:e080035. [Crossref] [PubMed]
16. Shin A, Shin JY, Kang EH. Risk of Nephrolithiasis Associated With SGLT2 Inhibitors Versus DPP4 Inhibitors Among Patients With Type 2 Diabetes: A Target Trial Emulation Study. Diabetes Care 2025;48:193-201. [Crossref] [PubMed]
17. Chung MC, Lin CY, Chang CH, et al. Use of Sodium-Glucose Transport Protein 2 Inhibitors and the Incidence of Urolithiasis: A Multi-Database and Cross-Country Study in Patients With Type 2 Diabetes Mellitus. Clin Pharmacol Ther 2025;117:1775-83. [Crossref] [PubMed]
18. Lukkunaprasit T, Tansawet A, Siriyotha S, et al. Effects of sodium-glucose co-transporter-2 inhibitors on the risk of nephrolithiasis and urinary tract infections in Thai patients with type 2 diabetes: a hospital-based cohort study. Diabetol Metab Syndr 2025;17:400. [Crossref] [PubMed]
19. Kanbay M, Brinza C, Copur S, et al. SGLT2 inhibitors and nephrolithiasis risk: a meta-analysis. Nephrol Dial Transplant 2025;40:671-8. [Crossref] [PubMed]
20. Radwan RM, Huang W, Li Y, et al. Heterogeneous effects of sodium-glucose cotransporter-2 inhibitors compared to dipeptidyl peptidase-4 inhibitors on nephrolithiasis in older adults with type 2 diabetes. Pharmacotherapy 2025;45:426-34. [Crossref] [PubMed]
21. Anan G, Kikuchi D, Hirose T, et al. Impact of Sodium-Glucose Cotransporter-2 Inhibitors on Urolithiasis. Kidney Int Rep 2023;8:925-8. [Crossref] [PubMed]
22. Anderegg MA, Schietzel S, Bargagli M, et al. Empagliflozin in nondiabetic individuals with calcium and uric acid kidney stones: a randomized phase 2 trial. Nat Med 2025;31:286-93. [Crossref] [PubMed]
23. Schaub JA, Oerline MK, Crivelli JJ, et al. The Effect of Sodium-Glucose Cotransporter 2 Inhibitors and Glucagon-Like Peptide-1 Receptor Agonists on 24-Hour Urine Parameters: A Retrospective Cohort Study. Kidney360 2025;6:835-47. [Crossref] [PubMed]
24. Anan G, Hirose T, Kikuchi D, et al. Inhibition of sodium-glucose cotransporter 2 suppresses renal stone formation. Pharmacol Res 2022;186:106524. [Crossref] [PubMed]
25. Trial to Assess the Efficacy of EMPAgliflozin and Personalized Dietary Counseling for Kidney STONE Prevention (EMPASTONE). ClinicalTrials.gov 2026. Available online: https://clinicaltrials.gov/study/NCT06653738
26. Barghouthy Y, Corrales M, Doizi S, et al. Tea and coffee consumption and pathophysiology related to kidney stone formation: a systematic review. World J Urol 2021;39:2417-26. [Crossref] [PubMed]
27. Jones P, Pietropaolo A, Chew BH, et al. Atlas of Scoring Systems, Grading Tools, and Nomograms in Endourology: A Comprehensive Overview from the TOWER Endourological Society Research Group. J Endourol 2021;35:1863-82. [Crossref] [PubMed]