Type 2 diabetes mellitus (T2DM) is a rapidly growing global epidemic, with a continuously increasing number of affected individuals worldwide (1). One of its most serious complications is diabetic kidney disease (DKD), a leading cause of chronic kidney disease (CKD) and end-stage kidney disease (ESKD) (2, 3). The global prevalence of DKD has risen significantly over recent decades, with an estimated annual incidence of 8% among patients with T2DM, and a 2 – 4% yearly decline in kidney function, culminating in an ESKD incidence of up to 0.8% annually (3).

Despite major advancements in glycemic control and cardiovascular risk management, the burden of DKD remains high and continues to grow (4, 5). DKD not only accelerates the progression to ESKD, but also significantly increases cardiovascular morbidity and mortality (6, 7). Optimal glycemic control plays a central role in preventing microvascular complications such as DKD (8–10), yet achieving and maintaining adequate glucose control remains challenging, particularly in patients with advanced kidney disease who face therapeutic limitations (11, 12).

For many years, standard DKD management has relied heavily on inhibition of the renin–angiotensin–aldosterone system (RAAS), particularly in patients with moderate-to-severe albuminuria (13, 14). In recent years, however, sodium-glucose co-transporter 2 inhibitors (SGLT2i) have emerged as disease-modifying agents with proven benefits in slowing DKD progression and reducing cardiovascular events when added to RAAS inhibitors (15–20). Similarly, glucagon-like peptide-1 receptor agonists (GLP1RAs) have demonstrated cardiovascular and potentially renal benefits in patients with T2DM and DKD (21–23).

In 2020, the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines recommended SGLT2i for patients with T2DM and CKD (eGFR >30 ml/min/1.73 m²) as first-line agents for cardiorenal protection, and GLP1RAs as second-line therapy in patients not achieving glycemic targets with metformin and/or SGLT2i, or when those agents are contraindicated (24). These recommendations have since been endorsed by other leading professional societies (25–31).

Nevertheless, real-world evidence suggests underutilization of these agents remains widespread. In the United States, from 2017 to 2020, only 5.8% of adults with T2DM were treated with SGLT2i, and 4.4% with GLP1RAs, with little difference between patients with or without DKD (32). Similarly, in Canada, less than 8% of high-risk diabetic patients with DKD or cardiovascular disease received SGLT2i or GLP1RAs during the same period (33, 34).

To date, there is a lack of published data assessing the real-world implementation of the 2020 KDIGO guidelines, particularly in nephrology settings. Timely initiation of guideline-directed therapy in patients with DKD can significantly reduce the risk of kidney disease progression, ESKD, and cardiovascular events. Identifying gaps between evidence-based recommendations and actual prescribing practices is critical to improving patient outcomes.

The primary aim of this study was to evaluate the implementation of the 2020 KDIGO recommendations among adult patients with T2DM and CKD treated in a tertiary academic medical center nephrology clinic. Additionally, we sought to identify patient characteristics associated with the underuse of SGLT2i and GLP1RA therapies.

Diabetic kidney disease is a common and serious complication of type 2 diabetes mellitus, associated with an increased risk of ESKD, cardiovascular morbidity, and premature mortality (2). For decades, the management of DKD has focused on optimizing glycemic control, controlling blood pressure, and inhibiting the RAAS system. In recent years, the therapeutic landscape has evolved with the emergence of SGLT2i and GLP1RAs, which have demonstrated cardiorenal protective effects in patients with diabetes and CKD. Reflecting this evidence, the 2020 KDIGO guidelines position these agents as first- and second-line therapies for patients with T2DM and DKD (24).

The current study provides real-world evidence from a nephrology clinic setting, evaluating the implementation of the 2020 KDIGO guidelines approximately one year after their publication. Despite broad eligibility, we observed low rates of SGLT2i and GLP1RA use among patients with T2DM and CKD, both at baseline and at the end of follow-up. These findings are consistent with previous reports documenting a persistent gap between guideline-based recommendations and real-world prescribing practices (32, 33).

Most patients in our cohort had long-standing diabetes and significant comorbidities, including cardiovascular disease and hypertension, characteristics that place them at high risk for adverse outcomes and make them prime candidates for SGLT2i and GLP1RA therapy. Nevertheless, only 12.9% were treated with SGLT2i and 10.4% with GLP1RAs at the first clinic visit, despite eligible kidney function in the majority of patients. SGLT2i use increased modestly during the study period (to 18.9%, p=0.005), but the uptake of GLP-1 RAs remained unchanged.

Since, at the time of the study, guideline recommendations limited initiation of SGLT2i to patients with eGFR >29 ml/min/1.73m², we performed a subgroup analysis restricted to this cohort. Consistent with the main results, treatment rates remained very low, underscoring the substantial implementation gap. Following the subsequent KDIGO update extending eligibility to patients with eGFR ≥20 ml/min/1.73m², we also examined this subgroup. Similar to patients with higher eGFR, treatment rates in the 20 – 29 ml/min/1.73m² group were extremely low and showed no meaningful change throughout follow-up, further emphasizing the limited real-world adoption of guideline-directed therapy even among eligible patients.

Importantly, when considering only patients with eGFR>29 ml/min/1.73m², UACR data were available in 53% of cases, considerably higher than the 35% observed in the entire cohort. In this group, albuminuria levels were relatively lower compared to patients with eGFR 20 – 29 ml/min/1.73m², who exhibited much higher UACR values, consistent with more advanced kidney damage and greater cardiorenal risk. The observation that treated patients had significantly lower UACR compared with untreated patients suggests that appropriate initiation of guideline-directed therapy may be associated with reduced albuminuria burden, although causality cannot be established in this retrospective study.

Notably, even among patients for whom these therapies were explicitly recommended by the treating nephrologist, a substantial proportion did not initiate treatment by the last follow-up visit, highlighting potential barriers beyond the nephrology clinic, such as limited coordination with primary care or endocrinologists.

Our findings align with previous studies reporting underutilization of cardiorenal protective therapies in high-risk populations. For example, the multinational CAPTURE study (34) revealed low rates of SGLT2i and GLP1RA use, despite high cardiovascular disease prevalence, including in Israel, where these agents were prescribed to only 15% and 8.6% of patients, respectively. Similar trends were observed in national data from the United States and Canada (31–33).

In a study by Nargesi et al. (36), the use of SGLT2i and GLP1RA was evaluated among patients with diabetes and comorbid coronary artery disease, heart failure, or CKD, based on the treatment recommendations of the American College of Cardiology and the American Diabetes Association. Among the 1,104 patients included, 52.6% met the criteria for SGLT2i therapy, and 32.8% had an indication for treatment with a GLP1RA. Despite these indications, during the period between 2017 and 2018, only 4.5% received an SGLT2i and just 1.5% were treated with a GLP1RA.

In a later study from England (37) examined diabetes medication trends among ~150,000 primary care patients between 2017 and 2020, metformin remained the most commonly used drug (~62.5%), with minimal change over time. Although the use of SGLT2i increased from 6.7% to 13.8% in patients without cardiovascular disease, and from 4.3% to 20.1% in those with such disease, overall uptake remained modest. GLP1RA use rose only slightly, from 3% to 4.5%.

In a U.S. study by Limonte et al. (31), among 1,375 patients with T2DM (37% with CKD, average eGFR 88.4), only 5.8% received SGLT2i and 4.4% received GLP1RA. Among high-risk patients (CKD, heart failure, or CAD), usage was only 7.7% and 3.5%, respectively. Similarly, in a study by Hao et al. (38) of 7,168 diabetic patients in primary care (2018 – 2019), 77.7% were treated with metformin, 19.7% with SGLT2i, and 9.4% with GLP1RA. Among those with a cardiovascular indication (25% of the cohort), only 14.9% received SGLT2i and 4.6% received GLP1RA.

We identified in our study several factors independently associated with the use of recommended therapies. Younger age, higher eGFR, and lower serum creatinine were positively associated with treatment, while older age and impaired kidney function were linked to underuse. These findings echo previous literature indicating that physicians may hesitate to prescribe newer agents to older or frail patients, possibly due to concerns about adverse effects, polypharmacy, or perceived lack of benefit (31, 36).

Interestingly, we found a strong positive association between the presence of dyslipidemia and the use of metformin, SGLT2i or GLP1RAs. This may reflect a more proactive approach to cardiovascular risk management in patients already diagnosed with metabolic syndrome or established atherosclerotic disease.

Another important consideration is the persistence of insulin therapy without transitioning to guideline-recommended agents. In our cohort, nearly 18% of patients remained on insulin, with no adjustment toward SGLT2i or GLP1RA despite clear eligibility. Such therapeutic inertia is particularly concerning in insulin-dependent patients, as the continuation of insulin without modification is not easily justifiable. Insulin therapy carries well-documented risks, including weight gain, hypoglycemia, and local complications such as injection-induced lipohypertrophy, which increase both morbidity and healthcare costs (39).

Greater emphasis on proactive therapy modification, especially in collaboration with endocrinologists, diabetologists and primary care physicians, is warranted to reduce avoidable complications and to align real-world practice more closely with KDIGO guideline recommendations.

In addition to the underuse of diabetes-specific therapies, we also observed suboptimal use of long-established cardiovascular medications. Only 52% of patients received RAASi, 47% received statins and 27% were prescribed aspirin, despite decades of evidence supporting their role in reducing cardiovascular and renal events in high-risk populations. Suboptimal use of cardioprotective medications was also reported in other studies (34, 38). In the CAPTURE study (34), only 52% of patients received RAAS inhibitors, similar to the rate observed in our study, while 51% were treated with statins and just 39% with aspirin. These findings suggest that barriers to guideline implementation extend beyond newer agents and may reflect broader systemic issues.

Importantly, SGLT2i and GLP1RA therapies are reimbursed for most eligible patients, suggesting that cost or access limitations were unlikely to explain the low utilization observed. Instead, knowledge gaps, therapeutic inertia or lack of experience with newer agents, may play a role. Although our study did not explore physician-level attitudes or barriers, the discrepancy between nephrologist recommendations and actual implementation points to the need for better interdisciplinary coordination and follow-up.

To our knowledge, this is the first study to assess KDIGO 2020 guideline implementation in a nephrology clinic setting. Strengths include a relatively large sample size and detailed clinical data reflecting real-world care. However, the study has several limitations. As a retrospective analysis, we relied on patient-reported medication use, which may be subject to recall bias. We did not verify prescriptions or pharmacy dispensing records, nor did we assess medication adherence or reasons for treatment discontinuation. Although UACR was not available for the entire cohort, we performed subgroup analyses that partially mitigate this limitation, allowing more accurate assessment of albuminuria burden in the population eligible for treatments.

Taken together, our findings underscore a significant gap in the implementation of evidence-based therapies for patients with DKD. Bridging this gap requires targeted efforts to raise awareness among clinicians, improve interdisciplinary communication, and empower patients through education and shared decision-making. Ultimately, increasing the use of cardiorenal protective agents in eligible patients may reduce the burden of ESKD and cardiovascular complications in this vulnerable population.