Highlight
• A SMART-C meta-analysis of 8 large randomized trials (58,816 participants) demonstrates that SGLT2 inhibitors reduce kidney disease progression, acute kidney injury (AKI), all-cause hospitalization and death in people both with and without diabetes.
• Relative benefits were broadly consistent across diabetes status and by urine albumin-to-creatinine ratio (UACR) strata; larger absolute kidney benefit occurs when baseline UACR is ≥200 mg/g due to higher baseline event rates.
• Net benefits remained when focusing on non–heart-failure populations and participants with eGFR <60 mL/min/1.73 m2, supporting broad applicability in clinical practice.
Background
Chronic kidney disease (CKD) is a major global public health problem, driving morbidity, all-cause hospitalization, cardiovascular disease and premature mortality. Albuminuria (measured by urine albumin-to-creatinine ratio, UACR) stratifies risk and guides treatment decisions: higher UACR confers substantially greater risk of kidney progression and cardiovascular events. Sodium-glucose cotransporter 2 (SGLT2) inhibitors were developed as glucose-lowering agents but have shown kidney and cardiovascular protection in several randomized trials. Clinical practice guidelines have increasingly recommended SGLT2 inhibitors for CKD, but the strength and applicability of recommendations vary with diabetes status and degree of albuminuria, leaving clinicians uncertain about absolute benefits across subgroups.
Study design and methods (SMART-C meta-analysis)
The analysis by Staplin et al. (SMART-C consortium) pooled individual trial results from eight randomized clinical trials that compared an SGLT2 inhibitor with placebo in people with kidney disease and reported longitudinal kidney outcomes and baseline albuminuria. Trials included major kidney-directed SGLT2 inhibitor studies with label indications for kidney disease. The pooled population comprised 58,816 participants (mean age 64 years, 35% female), of whom 48,946 had diabetes and 9,870 did not.
Principal methods: inverse-variance weighted meta-analysis to estimate pooled relative effects (hazard ratios) stratified by diabetes status and baseline UACR (<200 mg/g vs ≥200 mg/g). Absolute effects were derived by applying pooled relative risks to observed placebo event rates to estimate events per 1000 patient-years and absolute risk differences for each subgroup. Heterogeneity by baseline UACR was assessed within diabetes strata.
Key findings
Overall population and subgroups
Across the pooled population, SGLT2 inhibitor allocation produced consistent relative risk reductions for several clinically important outcomes. Results are reported separately for participants with diabetes and without diabetes when available; absolute rates are presented as events per 1000 patient-years (placebo vs SGLT2 inhibitor), with pooled hazard ratios (HRs) and 95% confidence intervals (CIs).
Kidney disease progression
With diabetes: 48 vs 33 events per 1000 patient-years (placebo vs SGLT2 inhibitor); HR 0.65 (95% CI, 0.60–0.70).
Without diabetes: 46 vs 32 events per 1000 patient-years; HR 0.74 (95% CI, 0.63–0.85).
Interpretation: Relative risk reductions were substantial and clinically meaningful in both groups. Although point estimates suggest slightly greater relative risk reduction in those with diabetes, the difference was small, and confidence intervals overlap. Importantly, participants with higher baseline UACR (≥200 mg/g) had higher absolute event rates; applying similar relative reductions produced larger absolute reductions in events per 1000 patient-years in the higher-UACR subgroup.
Acute kidney injury (AKI)
With diabetes: 18 vs 14 events per 1000 patient-years; HR 0.77 (95% CI, 0.69–0.87).
Without diabetes: 18 vs 13 events per 1000 patient-years; HR 0.72 (95% CI, 0.56–0.92).
Interpretation: SGLT2 inhibitors were associated with lower rates of investigator-defined AKI across diabetes status—an important safety finding that counters earlier concerns about hemodynamic effects leading to kidney harm in some patients.
All-cause hospitalization
With diabetes: 231 vs 202 events per 1000 patient-years; HR 0.90 (95% CI, 0.87–0.92).
Without diabetes: 237 vs 203 events per 1000 patient-years; HR 0.89 (95% CI, 0.83–0.95).
Interpretation: Hospitalization reductions were observed irrespective of diabetes, and the absolute number of hospitalizations prevented was notable, particularly in lower-UACR groups where baseline kidney event rates are lower but hospitalization rates remain substantial.
All-cause mortality
With diabetes: 47 vs 42 events per 1000 patient-years; HR 0.86 (95% CI, 0.80–0.91).
Without diabetes: 48 vs 42 events per 1000 patient-years; HR 0.91 (95% CI, 0.78–1.05).
Interpretation: Mortality reductions were clear and statistically robust in participants with diabetes; in those without diabetes the point estimate favored benefit but confidence intervals crossed unity, reflecting fewer events and reduced power in that subgroup. The overall pattern supports a mortality benefit across the studied populations.
Effect modification by albuminuria (UACR)
Relative treatment effects (HRs) were broadly consistent when stratified by baseline UACR (<200 mg/g vs ≥200 mg/g) within diabetes strata. Because baseline event rates are higher when UACR ≥200 mg/g, the same relative risk reduction translates into larger absolute risk reductions in that subgroup—i.e., more kidney events prevented per 1000 patient-years when albuminuria is high. For non-kidney outcomes such as hospitalization and death, meaningful absolute benefits were also observed even in participants with UACR <200 mg/g.
Sensitivity analyses and special populations
The authors report that net absolute benefits persisted when analyses were restricted to participants without heart failure and when baseline estimated glomerular filtration rate (eGFR) was <60 mL/min/1.73 m2. These findings support applicability across a range of CKD severities and in patients without clinical heart failure.
Safety signals
The meta-analysis showed no excess in AKI and demonstrated reductions instead. Prior randomized trials have noted increased risks of genital mycotic infections and, in some studies, small increases in diabetic ketoacidosis among insulin-deficient patients; these adverse events remain important but were not the focus of the pooled kidney-centric analysis. Clinicians should continue to counsel patients about these known, largely manageable adverse events and monitor appropriately.
Expert commentary and interpretation
Clinical implications: The SMART-C pooled analysis strengthens the evidence that SGLT2 inhibitors provide kidney, cardiovascular and survival benefits across people with and without diabetes, and across a spectrum of albuminuria levels. The central translational point is that relative risk reductions are broadly homogeneous, but absolute benefit varies with baseline risk—principally albuminuria and eGFR. Therefore, treatment decisions based on expected absolute benefit favor prioritizing patients with higher UACR for rapid implementation, but important benefits for hospitalization and mortality remain even at lower UACR.
How this informs guidelines: KDIGO and other societies have moved toward recommending SGLT2 inhibitors for patients with CKD and albuminuria, particularly when diabetes is present. SMART-C provides additional support for use beyond diabetes, mirroring the inclusion of non-diabetic CKD populations in DAPA-CKD and EMPA-KIDNEY and endorsing a broader clinical application.
Mechanistic considerations: Benefits likely reflect multiple mechanisms—hemodynamic effects that reduce intraglomerular pressure, reductions in tubular workload and inflammation, improved cardiac-renal interplay, reduction in heart failure exacerbations, and favorable metabolic and natriuretic effects. These pleiotropic effects plausibly explain consistent benefits across diabetes status.
Limitations and generalizability: Although pooled, the analysis is constrained by the inclusion criteria of the underlying trials, which enrolled patients selected for clinical trials and often underrepresented certain demographic groups (e.g., women and some racial/ethnic minorities). Power for some subgroup analyses (notably non-diabetic mortality) was limited. Heterogeneity in event definitions, follow-up duration and background therapies across trials may influence pooled estimates, although inverse-variance methods and sensitivity checks were used to mitigate biases.
Practical recommendations for clinicians
• Consider SGLT2 inhibitor therapy in patients with CKD who meet label indications irrespective of diabetes status, prioritizing those with higher UACR due to greater absolute kidney risk and benefit.
• Counsel about expected benefits (reduced kidney progression, fewer hospitalizations, lower mortality) and known risks (genital infections, rare ketoacidosis), and implement routine monitoring of volume status, renal function after initiation, and metabolic status as clinically indicated.
• Recognize that SGLT2 inhibitors reduce AKI risk in randomized settings; however, individual monitoring during intercurrent illness or volume depletion remains prudent.
Conclusion
In a large pooled analysis of 58,816 participants from eight randomized trials, SGLT2 inhibitors reduced kidney disease progression, AKI, all-cause hospitalization and mortality across participants with and without diabetes. Relative effects were consistent across levels of albuminuria; higher baseline UACR (≥200 mg/g) yielded larger absolute kidney benefits because baseline event rates were higher. These results reinforce current guideline directions toward broader use of SGLT2 inhibitors in CKD and provide clinicians with quantitative estimates of absolute and relative benefit to guide shared decision-making.
Funding and clinicaltrials.gov
The SMART-C meta-analysis and constituent trials received funding reported in their primary publications; funding sources vary by trial and include government, academic and industry sponsors. Key trial identifiers include: CREDENCE (NCT02065791), DAPA-CKD (NCT03036150), EMPA-KIDNEY (NCT03594110). Clinicians should consult original trial reports for detailed sponsor disclosures.
Selected references
1. Staplin N, Roddick AJ, Neuen BL, et al.; SGLT2 Inhibitor Meta-Analysis Cardio-Renal Trialists’ Consortium (SMART-C). Effects of Sodium Glucose Cotransporter 2 Inhibitors by Diabetes Status and Level of Albuminuria: A Meta-Analysis. JAMA. 2025 Nov 7. doi:10.1001/jama.2025.20835.
2. Perkovic V, Jardine MJ, Neal B, et al.; CREDENCE Trial Investigators. Canagliflozin and Renal Outcomes in Type 2 Diabetes and Nephropathy. N Engl J Med. 2019;380(24):2295–2306. doi:10.1056/NEJMoa1811744.
3. Heerspink HJL, Stefánsson BV, Correa-Rotter R, et al.; DAPA-CKD Trial Committees and Investigators. Dapagliflozin in Patients with Chronic Kidney Disease. N Engl J Med. 2020;383(15):1436–1446. doi:10.1056/NEJMoa2024816.
4. Herrington WG, Staplin N, Wanner C, et al.; EMPA-KIDNEY Collaborative Group. Empagliflozin in Patients with Chronic Kidney Disease. N Engl J Med. 2022;388(7):690–702. doi:10.1056/NEJMoa2205233.
5. Kidney Disease: Improving Global Outcomes (KDIGO) Diabetes Work Group. KDIGO 2022 Clinical Practice Guideline for Diabetes Management in Chronic Kidney Disease. Kidney Int. 2022;102(4S):S1–S127.