SGLT2 Inhibitors in Older Adults With Cardiovascular Disease: A Systematic Review and Meta‐Analysis

ABSTRACT

Background

Sodium‐glucose cotransporter‐2 (SGLT2) inhibitors, developed for type 2 diabetes mellitus (T2DM), have demonstrated cardiorenal benefits in conditions including cardiovascular (CV) disease. However, few meta‐analyses have synthesized outcomes in older adults with CV disease.

Methods

A systematic review and meta‐analysis of randomized controlled trials published from January 2015 to January 2025 was conducted using MEDLINE (PubMed), Embase (Ovid), and CENTRAL. We included studies that reported the risk of CV outcomes for subgroups of older adults (≥ 65 years) with CV disease. The primary outcome was a composite of hospitalization for heart failure (HHF), urgent heart failure (HF) visits, and cardiovascular death (CVD). Secondary outcomes included all‐cause mortality, CVD, and HHF individually. Subgroup analyses were conducted in patients with HF, T2DM, age strata (65–74 vs. ≥ 75), SGLT2 inhibitor agent, and adverse events.

Results

Analyzing nine studies, SGLT2 inhibitors were associated with reducing the risk of composite outcome (HR: 0.75, 95% CI: 0.67–0.83, I ² = 51%), all‐cause mortality (HR: 0.80, 95% CI: 0.66–0.97, I ² = 68%), CVD (HR: 0.78, 95% CI: 0.65–0.94, I ² = 61%), and HHF (HR: 0.73, 95% CI: 0.65–0.83, I ² = 0%). Benefits were consistent in subgroups of HF only, T2DM only, and those aged ≥ 75 years. No significant differences were observed by SGLT2 inhibitor type (p = 0.090). SGLT2 inhibitors increased the risk of genital infections (RR: 3.18, 95% CI: 2.35–4.30, I ² = 0%) but decreased that of other serious adverse events (RR: 0.92, 95% CI: 0.86–0.97, I ² = 64%).

Conclusions

In adults aged ≥ 65 years with CV disease, SGLT2 inhibitors significantly reduce the composite risk of HHF, urgent HF visits, and CVD and secondary outcomes of all‐cause mortality, CVD, and HHF, supporting their use in this population with careful monitoring of age‐related risks.

Abbreviations

AKI

acute kidney injury

CI

confidence interval

CKD

chronic kidney disease

CV

cardiovascular

CVD

cardiovascular death

HF

heart failure

HFpEF

heart failure with preserved ejection fraction

HFrEF

heart failure with reduced ejection fraction

HHF

hospitalization for heart failure

HR

hazard ratio

PRISMA

preferred reporting items for systematic review and meta‐analysis

RCT

randomized controlled trial

RR

risk ratio

SGLT2

sodium‐glucose cotransporter‐2

T2DM

Type 2 diabetes mellitus

Summary.

• Key points

  • ○
    In older adults (≥ 65 years) with cardiovascular disease, SGLT2 inhibitors significantly reduced the risk of a composite outcome comprising hospitalization for heart failure, urgent heart failure visits, and cardiovascular death (HR: 0.75).
  • ○
    Secondary outcomes indicated that SGLT2 inhibitors also significantly lowered the risk of all‐cause mortality (HR: 0.80), cardiovascular death (HR: 0.78), and hospitalization for heart failure (HR: 0.73).
  • ○
    SGLT2 inhibitors were associated with a significantly higher risk of genital infections but a lower risk of other serious adverse events in this population.
• Why does this paper matter?

  • ○
    This systematic review and meta‐analysis addresses an evidence gap by focusing on older patients with cardiovascular disease, an often underrepresented high‐risk population in clinical trials. The findings provide robust evidence that SGLT2 inhibitors confer substantial cardiovascular and survival benefits in this group, supporting their use to improve outcomes in this high‐risk population.

1. Introduction

Sodium‐glucose cotransporter‐2 (SGLT2) inhibitors are a class of medications initially developed for the treatment of type 2 diabetes mellitus (T2DM) [1]. By inhibiting SGLT2 within the proximal renal tubules, glucose reabsorption is reduced, promoting glycosuria and thereby lowering blood glucose levels [2]. Based upon clinical trial findings, their use has expanded beyond T2DM alone, and current indications include patients with T2DM and cardiovascular (CV) disease, including those with heart failure (HF) [3, 4]. Current clinical guidelines recommend the use of SGLT2 inhibitors in patients with HF, irrespective of left ventricular ejection fraction [3, 4]. SGLT2 inhibitors have demonstrated significant cardioprotective and renoprotective benefits, reducing major adverse CV events and improving clinical outcomes in diverse patient populations [5, 6].

Despite these advancements, there has been hesitancy in broadly adopting SGLT2 inhibition in the treatment of older adults, especially those with comorbidities or frailty, due to concerns regarding adverse event profiles [7], such as urinary tract infection, acute kidney injury (AKI), diabetic ketoacidosis, dehydration, and orthostatic hypotension [8]. Additionally, their use may contribute to polypharmacy due to the complexities of managing multiple medications in this population [9]. Current guidelines do not explicitly address the use of SGLT2 inhibitors in the older patient population, and despite randomized controlled trials (RCTs) suggesting efficacy in older patient subgroups, these sample sizes are limited. Although there is robust analysis of efficacy and safety for older patients with T2DM and chronic kidney disease (CKD), limited meta‐analyses have been conducted regarding older adults with CV disease [10, 11, 12, 13, 14]. Our study addresses this important knowledge gap by conducting a meta‐analysis of SGLT2 inhibitors specifically in older adults with CV disease.

2. Methods

This systematic review and meta‐analysis was performed and is presented in accordance with the Cochrane and Preferred Reporting Items for Systematic Review and Meta‐Analysis (PRISMA) guidelines [15]. A complete PRISMA 2020 checklist is provided as Table S1, with each item cross‐referenced to its location in this manuscript. The protocol for this systematic review was prospectively registered in PROSPERO (ID 1117792), and the full protocol is publicly accessible through that registry. No amendments have been made to the registered protocol since its initial submission.

2.1. Eligibility Criteria

We analyzed RCTs published in English from January 1, 2015 to July 31, 2025 that compared any SGLT2 inhibitors with placebo in adults aged 65 years or older who had documented CV disease. To qualify, a study had to report at least one hazard ratio (HR) for an outcome of interest. We excluded non‐randomized designs, trials whose populations were not limited exclusively to patients with CV disease, duplicate publications, and studies available only as abstracts. (Figure 1).

FIGURE 1.

PRISMA flow diagram of the study selection.

2.2. Search Strategy and Study Selection

A comprehensive search of MEDLINE (PubMed), Embase (Ovid), and CENTRAL (Cochrane Library) was conducted on August 1, 2025, covering records from January 1, 2015, through July 31, 2025. The full, database‐tailored search terms appear in Table S2. Two investigators (K.M. and R.T.) screened titles and abstracts of retrieved citations independently to identify eligible trials. Full texts of retained citations were then independently screened by the same two investigators. Discrepancies were resolved by consensus at each step. Data from the final list of retained citations after full text review were extracted by the same two authors, who double‐checked each other's work for inconsistencies.

2.3. Risk of Bias Assessment

The methodological quality of the included studies was assessed using the Cochrane Risk of Bias 2.0 tool, which evaluates five key domains: bias due to randomization, bias due to deviations from the intended intervention, bias due to missing data, bias in the measurement of outcomes, and bias in the selection of reported results. Each domain was rated as “low risk,” “some concerns,” or “high risk” based on pre‐specified criteria, in accordance with the Cochrane Risk of Bias 2.0 (RoB 2) tool. The Risk of Bias assessment was conducted on the entire RCT, not limited to the subgroup analysis included in our meta‐analysis.

2.4. Outcomes and Subgroup Analyses

The primary outcome of this meta‐analysis was the composite outcome of hospitalization for heart failure (HHF), urgent HF visits, and cardiovascular death (CVD). Secondary outcomes included all‐cause mortality, CVD, and HHF, each analyzed individually. Subgroup analyses were conducted to evaluate the treatment effects in specific populations for the primary outcome. These included studies that only enrolled patients with HF, studies that only enrolled patients with diagnosed T2DM, and studies that stratified the age ≥ 65 years old group into age 65–74 years old and age ≥ 75 years old. Additionally, subgroup analyses were performed on all included studies stratified by the type of SGLT2 inhibitor used. We also evaluated adverse events, namely serious adverse events, genital infections, AKI, diabetic ketoacidosis, volume depletion, hypotension, lower limb amputation, fractures, and falls. Adverse events documented in at least two trials were included in the pooled analysis.

2.5. Statistical Analysis

We extracted both HRs and risk ratios (RRs), prioritizing HRs when both were available for the same endpoint, and extracted 95% confidence Intervals (CIs) from included studies. For studies that reported HRs stratified into age 65–74 years old and age ≥ 75 years old, we combined the HRs using a fixed‐effects model to calculate one HR for age ≥ 65 years old. For studies that only reported a subgroup HR for age ≥ 75 years old, we included the unadjusted HR in our final analysis. We assessed the heterogeneity between studies in each outcome using Q‐statistics, τ, and I ². Q‐statistics showed whether the observed heterogeneity was statistically significant, while I ² was interpreted as follows: 0%–40% might not be important; 30%–60% may represent moderate heterogeneity; 50%–90% may represent substantial heterogeneity; 75%–100% considerable heterogeneity [16]. Conversely, when I ² was ≥ 50%, indicating substantial heterogeneity, a random‐effects model was employed to account for variability across studies. In contrast, when I ² was < 50%, suggesting minimal heterogeneity, a fixed‐effects model was used. Publication bias was assessed via visualization of the funnel plot. All statistical work, inclusive analysis, and graphical illustrations were conducted using RStudio Version 2024.12.0467.

2.6. Ethical Considerations

All studies included in this meta‐analysis were assessed for quality and adherence to ethical guidelines. Ethical approval for this analysis was not required, as only publicly available data were used.

3. Results

3.1. Study Selection and Characteristics

Our initial search of 5268 studies (Figure 1) yielded nine studies that met the inclusion criteria, enrolling a total of 24,889 older adults (aged ≥ 65 years) with CV disease [17, 18, 19, 20, 21, 22, 23, 24, 25]. Table S3 summarizes the baseline characteristics of these trials. Hernandez 2024 [17] and Butler 2024 [18] analyzed the same trial (EMPACT‐MI) but evaluated different outcomes, so both were included in our final analysis. The trials evaluated SGLT2 inhibitors (empagliflozin, ertugliflozin, sotagliflozin, and dapagliflozin) versus placebo in CV disease populations with varying baseline comorbidities, including HF, myocardial infarction, and T2DM. The proportion of patients aged ≥ 75 years ranged from 9.3% to 42.4%, with follow‐up periods of 9–42 months. Table S4 details how the composite CV outcome (HHF, urgent HF visit, and CVD) was defined in each trial.

3.2. Risk of Bias Assessment

All studies were judged low risk for bias arising from the randomization process and for bias due to missing outcome data. One study showed some concerns for bias due to deviations from the intended intervention. Five studies showed some concerns for bias in outcome measurement. Three studies showed some concerns for bias arising from selection of the reported result. Consequently, the overall judgment was low risk for most studies, while Martinez [24], Monterio [25], and Pitt [20] were classified as some concerns (Figure 2).

FIGURE 2.

Risk of bias assessment using RoB 2 tool.

3.3. Primary Outcome

Seven studies [19, 20, 21, 22, 23, 24, 25] evaluated the effect of SGLT2 inhibitors on the composite outcome comprising HHF, urgent HF visits, and CVD (Figure 3). Funnel plot analysis did not reveal significant asymmetry, suggesting no major publication bias (Figure 4A‐2). The pooled analysis demonstrated a statistically significant reduction in the risk of the composite outcome of HHF, urgent HF visits, and CVD (HR: 0.75, 95% CI: 0.67–0.83, p = 0.058, I ² = 51%) (Figure 3A‐1).

FIGURE 3.

Forest plot for primary and secondary outcomes. (A‐1) Forest plot (random‐effects model) for composite of hospitalization for heart failure (HHF), urgent heart failure (HF) visits, and cardiovascular death (CVD). (B‐1) Forest plot (random‐effects model) for all‐cause mortality. (C‐1) Forest plot (random‐effects model) for cardiovascular death (CVD). (D‐1) Forest plot (random‐effects model) for heart failure hospitalization (HHF).

FIGURE 4.

Funnel plot for primary and secondary outcomes. (A‐2) Funnel plot for composite of hospitalization for heart failure (HHF), urgent heart failure (HF) visits, and cardiovascular death (CVD). (B‐2) Funnel plot for all‐cause mortality. (C‐2) Funnel plot for cardiovascular death (CVD). (D‐2) Funnel plot for heart failure hospitalization (HHF).

3.4. Secondary Outcomes

3.4.1. All‐Cause Mortality

Three trials [21, 24, 25] evaluated the effect of SGLT2 inhibitors on all‐cause mortality (Figure 3). The funnel plot did not suggest marked asymmetry, although formal quantitative testing was limited by the small number of studies (Figure 4B‐2). The pooled analysis demonstrated a statistically significant reduction in the risk of all‐cause mortality (HR: 0.80, 95% CI: 0.66–0.97, p = 0.044, I ² = 68%) (Figure 3B‐1).

3.4.2. CVD

Four studies [19, 21, 24, 25] evaluated the effect of SGLT2 inhibitors on CVD (Figure 3). Funnel plot inspection showed no overt publication bias, though the limited number of trials precluded definitive conclusions (Figure 4C‐2). The pooled analysis indicated a statistically significant reduction in the risk of CVD (HR: 0.78, 95% CI: 0.65–0.94, p = 0.053, I ² = 61%) (Figure 3C‐1).

3.4.3. HHF

Four studies [17, 19, 21, 25] evaluated the effect of SGLT2 inhibitors on HHF (Figure 3). Funnel plot inspection showed no overt publication bias, though the limited number of trials precludes definitive conclusions (Figure 4D‐2). The pooled analysis indicated a statistically significant reduction in the risk of HHF (HR: 0.73, 95% CI: 0.65–0.83, p = 0.67, I ² = 0%) (Figure 3D‐1).

3.5. Adverse Events

Across the included RCTs, adverse event reporting was as follows: serious adverse events were reported in five studies [19, 21, 23, 24, 25], genital infections in three [19, 23, 25], and AKI in two [23, 25]. Baseline characteristics of the included studies are summarized in Table S3. HRs for adverse events were unavailable; therefore, RRs with 95% CIs were used for the pooled analysis.

3.5.1. Serious Adverse Events

The effect of SGLT2 inhibitors on serious adverse events in patients aged ≥ 65 years was evaluated in five studies [19, 21, 23, 24, 25] (Figure 5A‐1). Definitions of serious adverse events are summarized in Table S5. The pooled analysis showed a statistically significant reduction in the risk of serious adverse events (RR: 0.92, 95% CI: 0.86–0.97, p = 0.025, I ² = 64%). The subgroup analysis showed a statistically significant reduction in the risk of serious adverse events with SGLT2 inhibitors in both patients aged 65–74 (RR: 0.92, 95% CI: 0.87–0.98, p = 0.18, I ² = 36%) and patients aged ≥ 75 years (RR: 0.91, 95% CI: 0.84–0.97, p = 0.16, I ² = 39%) (p interaction = 0.68) (Figure 5A‐2).

FIGURE 5.

Forest plot for adverse events. (A‐1) Forest plot (random‐effects model) for adverse serious events in age ≥ 65 years old. (A‐2) Forest plot (random‐effects model) for adverse serious events in age 65–74 years old and age ≥ 75 years old. (B‐1) Forest plot (fixed‐effects model) for genital infections in age ≥ 65 years old. (B‐2) Forest plot (fixed‐effects model) for genital infections in age 65–74 years old and age ≥ 75 years old. (C‐1) Forest plot (fixed‐effects model) for acute kidney injury in age ≥ 65 years old. (C‐2) Forest plot (fixed‐effects model) for acute kidney injury in age 65–74 years old and age ≥ 75 years old.

3.5.2. Genital Infections

The effect of SGLT2 inhibitors on genital infections in patients aged ≥ 65 years was evaluated in three studies [19, 23, 25] (Figure 5B‐1). The pooled analysis indicated a significantly increased risk of genital infections with SGLT2 inhibitors compared to placebo (RR: 3.18, 95% CI: 2.35–4.30, p = 0.89, I ² = 0%). The subgroup analysis showed a statistically significant increased risk of genital infections with SGLT2 inhibitors in patients aged 65–74 (RR: 2.88, 95% CI: 2.06–4.03, p = 0.82, I ² = 0%) and patients aged ≥ 75 (RR: 4.57, 95% CI: 2.27–9.21, p = 0.86, I ² = 0%) (p interaction = 0.24) (Figure 5B‐2).

3.5.3. Acute Kidney Injury

The effect of SGLT2 inhibitors on AKI in patients aged ≥ 65 years was evaluated in two studies [23, 25] (Figure 5C‐1). There was no statistically significant increase in the risk with SGLT2 inhibitors (RR: 0.92, 95% CI: 0.74–1.15, p = 0.25, I ² = 25%). The subgroup analysis showed no difference in the risk of AKI with SGLT2 inhibitors in both patients aged ≥ 65 years (RR: 0.95, 95% CI: 0.71–1.28, p = 0.24, I ² = 28%) and patients aged ≥ 75 years (RR: 0.89, 95% CI: 0.64–1.25, p = 0.74, I ² = 0%). (Figure 5C‐2).

3.6. Subgroup Analyses

3.6.1. All HF Versus Less Than 50% of Participants With HF

The subgroup of studies that enrolled only participants with HF included five studies [20, 21, 22, 23, 24]. Table S3 summarizes the key inclusion criteria for HF across the studies. For this subgroup, the pooled analysis indicated a statistically significant reduction in the risk of the primary outcome (HR: 0.76, 95% CI: 0.70–0.82, I ² = 10%) (Figure S1A). The subgroup of studies that enrolled less than 50% of participants with HF included two studies [19, 25] and the pooled analysis showed no statistically significant reduction in the risk of the primary outcome (HR: 0.72, 95% CI: 0.46–1.11, I ² = 87%) (Figure S1A). The test for subgroup differences suggested no significant difference in treatment effect comparing studies specifically enrolling patients with HF and those that did not (p interaction = 0.80) (Figure S1A).

3.6.2. All T2DM Versus Less Than 50% of Participants With T2DM

The subgroup of studies that enrolled only participants with T2DM included three studies [19, 20, 25] and the pooled analysis indicated a statistically significant reduction in the risk of primary outcome (HR: 0.65, 95% CI: 0.44–0.95, I ² = 81%) (Figure S1B). The subgroup of studies that enrolled less than 50% of participants with T2DM included four studies [21, 22, 23, 24] and the pooled analysis showed a significant reduction in the risk of primary outcome (HR: 0.77, 95% CI: 0.71–0.83, I ² = 0%) (Figure S1B). The test for subgroup differences suggested no significant difference in treatment effect between studies specifically enrolling patients with T2DM and those that did not (p interaction = 0.39) (Figure S1B).

3.6.3. Age 65–74 Years Old Versus Age ≥ 75 Years Old

The subgroup of studies that stratified patients into age groups of 65–74 years and ≥ 75 years old included three studies [21, 24, 25]. The pooled analysis across all six HRs (two per study) indicated a statistically significant reduction in the risk of the primary outcome (HR: 0.73, 95% CI: 0.65–0.82, I ² = 26%) (Figure S1C). For age ≥ 75 years old and age 65–74 years old, the pooled analysis indicated a statistically significant reduction in the risk of the primary outcome (HR: 0.71, 95% CI: 0.57–0.88, I ² = 51%, and HR: 0.73, 95% CI: 0.63–0.85, I ² = 26%, respectively) (Figure S1C). The test for subgroup differences suggested no significant difference in treatment effect comparing age ≥ 75 years old and age 65–74 years old (p interaction = 0.83) (Figure S1C).

3.6.4. SGLT2 Inhibitor Agent (Dapagliflozin vs. Empagliflozin Versus Sotagliflozin Versus Ertugliflozin)

When stratified by SGLT2 inhibitor agent, dapagliflozin [21, 24], empagliflozin [22, 23, 25] and sotagliflozin [20] indicated a statistically significant reduction in the risk of primary outcome (dapagliflozin HR: 0.77, 95% CI: 0.70–0.86, I ² = 0%; empagliflozin HR: 0.71, 95% CI: 0.60–0.84, I ² = 56%; sotagliflozin HR: 0.47, 95% CI: 0.28–0.79) (Figure S1D). Ertugliflozin, which included one study [19] showed no significant reduction in the primary outcome (HR: 0.89, 95% CI: 0.73–1.08) (Figure S1D). The test for subgroup differences suggested no significant difference in treatment effect by SGLT2 inhibitor type (p interaction = 0.09) (Figure S1D).

3.6.5. Heart Failure With Reduced Ejection Fraction (HFrEF) Versus Heart Failure With Preserved Ejection Fraction (HFpEF)

The subgroup of studies that enrolled only participants with HFrEF included three studies [20, 23, 24] and the pooled analysis indicated a statistically significant reduction in the risk of primary outcome (HR: 0.73, 95% CI: 0.65–0.82, I ² = 19%) (Figure S1E). The subgroup of studies that enrolled only participants with HFpEF included two studies [21, 22] and the pooled analysis showed a significant reduction in the risk of primary outcome (HR: 0.78, 95% CI: 0.71–0.86, I ² = 0%) (Figure S1E). The test for subgroup differences suggested no significant difference in treatment effect between studies specifically enrolling patients with HFrEF and those with HFpEF (p interaction = 0.39) (Figure S1E).

4. Discussion

To our knowledge, this is the first meta‐analysis restricted to RCTs of adults aged ≥ 65 years with established CV disease. In this meta‐analysis of nine studies comprising 24,889 older adults aged ≥ 65 years, SGLT2 inhibitors yielded a substantial reduction in the primary composite outcome of HHF, urgent HF visits, and CVD (HR: 0.75, 95% CI: 0.67–0.83). Significant benefits were also observed in each secondary endpoint, including all‐cause mortality (HR: 0.80, 95% CI: 0.66–0.97), CVD (HR: 0.78, 95% CI: 0.65–0.94), and HHF (HR: 0.73, 95% CI: 0.65–0.83). Although the included trials enrolled heterogeneous populations of patients with various comorbidities, less heterogeneous subgroup analyses showed that these benefits were consistent among subgroups of studies enrolling patients with HF, patients with T2DM, and older adults aged 65–74 years and ≥ 75 years. Furthermore, while the use of SGLT2 inhibitors was associated with an increased risk of genital infections, it decreased the risk of serious adverse events. Finally, there were no significant differences in treatment effect between HFrEF and HFpEF or among SGLT2 inhibitor types, and each agent (dapagliflozin, sotagliflozin, and empagliflozin) individually yielded a significant reduction in the primary outcome. These findings underscore the robust benefits of SGLT2 inhibitors in an older population with CV disease. Prior studies examining SGLT2 inhibitors in older adults with CV disease have yielded mixed results, and no definitive conclusion has yet been established [7, 10, 17, 18, 19, 25, 26]. Our study incorporates more recent RCTs, focuses on only older adults ≥ 65 years with CV disease, and performs further subgroup analyses by comorbidity (T2DM or HF), age (65–74 years and > 75 years), and the type of SGLT2 inhibitor agent.

There are multiple plausible explanations for the observed significant benefits of SGLT2 inhibitors in older adults aged over 65 years with CV disease, as revealed by our meta‐analysis. Older adults are at higher risk for hypertension and impaired renal function, which exacerbates CV disease progression [27, 28, 29]. SGLT2 inhibitors have been hypothesized to reduce the risk of HF by various mechanisms including natriuresis and reductions in plasma volume [5]. Potential benefits include reductions in preload and afterload, reversal of cardiac remodeling, attenuation of inflammatory markers, regression of myocardial mass, and modulation of renal sympathetic afferent tone [2, 5, 27].

These cardioprotective effects could possibly stem from SGLT2 inhibitors' capacity to dampen chronic inflammation: preclinical and clinical studies show they suppress NLR family pyrin domain containing 3‐inflammasome activation, lower circulating C‐reactive protein and Interleukin‐6, and mitigate oxidative stress [5]. These anti‐inflammatory actions may be especially beneficial because aging induces a chronic, low‐grade pro‐inflammatory state (“inflammaging”) [30]. Given older adults are at a higher risk for CV disease, they may benefit from SGLT2 inhibitors' cardioprotective and renoprotective effects.

Although SGLT2 inhibitors have demonstrated significant benefits, their use in older adults presents concerns in its safety with regard to adverse events [7], such as genital infections and diabetic ketoacidosis [8]. Our study demonstrated SGLT2 inhibitors increased the risk of genital infections but lowered the risk of serious adverse events in older adults. In a meta‐analysis comparing adverse events of SGLT2 inhibitors in older adults with T2DM, the risk of genital infections was similar between patients aged < 65 and those aged ≥ 65 (RR: 5.64 [95% CI: 2.75–11.55] vs. 6.55 [95% CI: 2.09–20.57]; p = 0.83) [26], underscoring that genital infections are common adverse events regardless of age. In prior meta‐analysis [10], SGLT2 inhibitors were well tolerated in frail and older patients with T2DM. The incidence of diabetic ketoacidosis was not significantly increased in people with frailty and older individuals [10]. This is important due to the high risks of volume depletion, reduced food intake, and risk of acute illness in this patient group, which might increase the risk of developing diabetic ketoacidosis with concurrent use of SGLT2 inhibitors. Additionally, a retrospective pharmacovigilance study using the Food and Drug Administration Adverse Event Reporting System specifically analyzed older adults (≥ 75 years old) receiving SGLT2 inhibitors, finding no overall increase in the reporting of adverse events such as genital infections and diabetic ketoacidosis, relative to younger recipients [31]. In combination with the survival benefits demonstrated in individual RCTs, as well as our study results, these findings reinforce the notion that, with careful monitoring including patient education and appropriate follow‐up visits, SGLT2 inhibitors can offer a favorable benefit–risk profile even in the older population.

Several approaches can be considered to ensure the appropriate use of SGLT2 inhibitors in older adults. In the Get With The Guidelines–Heart Failure registry, only 23.7% of eligible patients aged ≤ 75 years and 12.3% of those > 75 years were discharged on SGLT2 inhibitors, which signals therapeutic inertia potentially related to prescriber unfamiliarity, fragmented specialty oversight, and perceived cost barriers [32]. Introducing an electronic health record alert for outpatients with HFrEF subsequently increased initiation of guideline‐directed medical therapy, including SGLT2 inhibitors, within 30 days; this scalable strategy illustrates a practical route to accelerate adoption of high‐value therapies in older adults with HF [33]. In addition, a multidisciplinary approach is necessary for the appropriate use of SGLT2 inhibitors in older adults. Geriatrician‐led medication reviews enhance prescribing quality and cut rehospitalization [34]. RCTs confirm these prescribing gains [35], and a home‐based multidisciplinary program has shown feasibility [36]. Furthermore, geriatric co‐management on cardiology wards reduces functional decline and delirium [37]. SGLT2 inhibitors should not be deprescribed solely on the basis of polypharmacy, given their potential to improve multiple clinical outcomes. At the same time, because polypharmacy itself can be problematic, medication management should emphasize a geriatrician‐led, multidisciplinary, comprehensive approach to optimization.

HF affects 8%–9% of adults aged ≥ 65 years, and nearly 90% of those over 80 live with some form of CV disease [38, 39]. These figures underscore the pressing clinical relevance of our study, which closes a critical evidence gap for this large and vulnerable population.

Our study has several limitations. First, we relied on a subgroup of older adults aged ≥ 65 years that was identified after randomization in the original RCTs. Due to limited availability of subgroup‐specific baseline characteristics, we could not assess covariate balance between treatment arms. Second, although randomization and missing‐data domains were low risk, residual bias led to overall “some concerns” ratings for Martinez 2019 [24], Monteiro [25], and Pitt [20], slightly diminishing certainty in the pooled effect estimates. Third, limiting our database search to English‐language publications may have introduced language bias by excluding potentially relevant studies published in other languages. Fourth, our analysis focused on only three adverse events, which limits a more comprehensive assessment of safety in older adults. Fifth, important geriatric considerations such as frailty were not evaluated, and no RCTs provided HRs or other effect estimates on them, preventing subgroup analysis. Future RCTs should prospectively measure and report such outcomes to enable robust subgroup evaluation. Sixth, participants in RCTs are generally healthier and more self‐sufficient than the broader patient population, particularly among older adults, which may affect generalizability. Seventh, although we used a composite CV outcome as our primary outcome, there were slight differences in how this outcome was defined across the included studies, as detailed in Table S4. Eighth, data on overall hospitalizations were not available in the included trials, and future studies should determine whether these therapies reduce all‐cause hospitalizations. Ninth, the analysis of overall mortality, CVD outcomes, and subgroup results is limited by the small number of included studies and the high heterogeneity among them. Finally, other inherent limitations of meta‐analysis apply, including potential publication bias, variation in trial design, and heterogeneity in study populations.

In conclusion, SGLT2 inhibitors demonstrated significant CV benefits in older adults aged ≥ 65 years. These benefits were consistent across subgroups, regardless of HF status, diabetes status, or age strata. Given the growing aging population and the high burden of CV disease in older adults, these findings support the use of SGLT2 inhibitors in this population with appropriate monitoring.

Author Contributions

K.M.: conceptualization, data curation, writing – original draft preparation. R.T: formal analysis, methodology, visualization, writing – original draft preparation. L.F: writing – review and editing. Y.Y: writing – review and editing. A.R.V: writing – review and editing. Y.Y: writing – review and editing. T.N: writing – review and editing. S.M: conceptualization, methodology, writing – review and editing, supervision.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Forest plot for subgroup analyses. (A) Forest plot (random‐effects model) for primary composite outcome of hospitalization for heart failure (HHF), urgent heart failure (HF) visits, and cardiovascular death (CVD) in studies enrolling patients with HF and studies not limited to patients with HF. (B) Forest plot (random‐effects model) for primary composite outcome in studies enrolling patients with Type 2 Diabetes Mellitus (T2DM) and studies not limited to patients with T2DM. (C) Forest plot (random‐effects model) for primary composite outcome in studies that reported hazard ratios (HRs) for age 65–74 years old and age ≥ 75 years old. (D) Forest plot (random‐effects model) for primary composite outcome in all included studies stratified by Sodium‐glucose cotransporter‐2 (SGLT2) inhibitor agent. (E) Forest plot (fixed‐effect model) for primary composite outcome in studies enrolling patients with heart failure with reduced ejection fraction (HFrEF) or heart failure with preserved ejection fraction (HFpEF).

Table S1: Preferred Reporting Items for Systematic Review and Meta‐Analysis (PRISMA) 2020 Checklist.

Table S2: Search term for each database.

Table S3: Baseline characteristics of included studies.

Table S4: Definition of composite outcome.

Table S5: Serious adverse events.

Minami K., Terashima R., Freeman L., et al., “ SGLT2 Inhibitors in Older Adults With Cardiovascular Disease: A Systematic Review and Meta‐Analysis,” Journal of the American Geriatrics Society 73, no. 12 (2025): 3708–3718, 10.1111/jgs.70143.