Nephrotoxic medication burden and drug-related problems in patients with chronic kidney disease using SGLT2 inhibitors Authors: Zeynep Ozturk Yalcin, Arda Erdut, Guldehan Haberal, Mustafa Arici, Aygin Bayraktar-Ekincioglu Published: 2026 ABSTRACT AbstractBackgroundSodium-glucose cotransporter 2 inhibitors (SGLT2i) are increasingly prescribed because of their renal and cardiovascular benefits. However, polypharmacy is common among patients with chronic kidney disease (CKD), raising concerns about cumulative nephrotoxic medication exposure. This study aimed to evaluate drug-related problems (DRPs), nephrotoxic medication burden, kidney failure risk, and patient-reported outcomes in patients receiving SGLT2i in a nephrology outpatient clinic.MethodsThis observational study was conducted in a nephrology outpatient clinic of a university hospital between September 2022–February 2023. Patients ≥18 years who were newly prescribed an SGLT2 inhibitor were included. A clinical pharmacist performed comprehensive medication reviews, identified and categorized DRPs using the Pharmaceutical Care Network Europe (PCNE) classification system, and assessed nephrotoxic exposure by a structured nephrotoxicity score based on drug-specific risk categories. Kidney failure risk was estimated using the Kidney Failure Risk Equation, and quality of life was assessed using the KDQOL-36 questionnaire.ResultsSeventy-two patients were included, and 69 DRPs were identified through clinical pharmacist-led medication review. The most common DRPs were inappropriate duration of therapy (20.3%), medication without indication (15.9%), and untreated symptoms (13.0%). Pharmacist’s recommendations were highly accepted (85.5%). The median number of nephrotoxic drugs per patient was 2 (IQR:1–3), while the median nephrotoxicity score was 1.5 (IQR: 1–2). No clinically overt nephrotoxic events were observed during follow-up, and nephrotoxicity burden was not significantly associated with estimated kidney failure risk.ConclusionThis study provides real-world descriptive data on nephrotoxic medication exposure, drug-related problems, and clinical pharmacist-led medication review among CKD patients receiving SGLT2i in a nephrology outpatient setting. Further prospective controlled studies are needed to clarify the clinical impact of pharmacist-led medication review on patient. Introduction Chronic kidney disease (CKD) is a major public health problem and about 10% of the adult population has at least one form of kidney injury [1,2]. It is estimated that by 2030, 643 million people will be living with diabetes and CKD, where the prevalence will rise up due to increased incidence of diabetes [3]. In recent years, sodium-glucose cotransporter 2 inhibitors (SGLT2i) have emerged not only as effective agents for glycemic control but also as agents conferring substantial cardiovascular and renal protection. Large-scale clinical trials [4–6] have demonstrated that SGLT2i including canagliflozin, empagliflozin, dapagliflozin, and ertugliflozin can significantly slow the progression of CKD in both diabetic and non-diabetic patients, highlighting their therapeutic potential beyond glucose lowering in the nephrology care settings [5–7]. Several studies have demonstrated the effectiveness of pharmacist-led initiation and monitoring of SGLT2i treatment in patients with diabetes [8,9]. However, the use of SGLT2i in nephrology settings remains relatively underexplored, particularly in the context of pharmacist-led interventions [9,10]. The management of CKD is further complicated by polypharmacy, which is highly prevalent, affecting between 62% and 86% of patients [11]. Polypharmacy has been consistently identified as an independent predictor of drug-related problems (DRPs), including inappropriate prescribing, drug–drug interactions, and medication-related toxicities [12,13]. Moreover, the use of several concomitant medications increases the likelihood of drug–drug interactions and drug-related toxicities, which may ultimately compromise treatment outcomes [14]. However, pharmacist-led medication review and patient assessment have been associated with improved identification and management of DRPs and potential drug-drug interactions in previous studies [13]. Studies have shown that the patients with CKD are exposed to various potentially nephrotoxic medications; a community-based analysis of inappropriate prescription of nephrotoxic drugs in Sweden found that 18% of CKD patients received nephrotoxic drugs, mainly non-steroidal anti-inflammatory drugs (NSAIDs) [15,16]. Stottlemyer et al. (2023) conducted a modified Delphi study to establish expert consensus on the nephrotoxic potential of 195 medications used in non-intensive care settings. The panel reached consensus for the majority of medications: most were classified as having low to moderate nephrotoxic potential, while ten drugs were rated as probably or probably/definitely nephrotoxic, including amikacin, amphotericin B, cisplatin, colistimethate, colistin, cyclosporine, foscarnet, gentamicin, lithium, tobramycin and vancomycin [17]. Clinical pharmacists play a critical role in optimizing pharmacotherapy through comprehensive medication review, identification of DRPs, and mitigation of medication-related risks. Pharmacist-led interventions have been shown to reduce the frequency and severity of DRPs and prevent clinically significant drug–drug interactions [13]. Moreover, previous studies have demonstrated the effectiveness of pharmacist-led initiation and monitoring of SGLT2i therapy in patients with diabetes [8,9]. However, evidence regarding such interventions in nephrology-specific settings remains limited. While the efficacy of SGLT2i is well-documented, limited data exist regarding the specific burden of concomitant medication use in nephrology setting and its relationship with DRPs and patient-reported outcomes [15,18–21]. In particular, patient access to SGLT2i and their utilization were previously limited due to regional reimbursement restrictions. In this context, pharmacist-led interventions may play a key role in facilitating appropriate use and improving access to these agents [22]. Therefore, this study aimed to comprehensively evaluate patients receiving SGLT2i in a nephrology outpatient setting by identifying drug-related problems, assessing the nephrotoxic potential of concomitant medications, estimating kidney failure risk, and examining patient-reported outcomes during the early phase of treatment. Methods This was an observational study conducted at a tertiary-care university hospital nephrology outpatient clinic between September 2022 and February 2023. The primary objective of the study was to describe drug-related problems and nephrotoxic medication use among CKD patients receiving SGLT2i in a nephrology outpatient setting. Secondary objectives were considered to explore associations between nephrotoxicity risk, kidney failure risk estimates, and KDQOL-36 domains. Prior to the initiation of the study, no structured clinical pharmacy services were routinely integrated into the clinic workflow. This study was approved by the Hacettepe University Health Sciences Research Ethics Committee (approval number: GO 22/951). Written informed consent was obtained from all participants prior to study participation. All procedures performed in this study involving human participants were conducted in accordance with the ethical standards of the institutional research committee and the Declaration of Helsinki. Study population Adult patients (≥18 years) who were newly prescribed SGLT2i during routine clinical care were consecutively screened for eligibility. Patients were included if they initiated SGLT2i therapy within the study period and attended at least one follow-up visit. Exclusion criteria comprised history of kidney transplantation, pregnancy, refusal to participate, or discontinuation of SGLT2i therapy within two weeks of initiation due to administrative reasons (lack of access to medication), to ensure sufficient treatment exposure for early evaluation. Clinical pharmacist intervention and medication review Following initiation of SGLT2i therapy, eligible patients were referred to a clinical pharmacist for counseling and medication review. The pharmacist performed a comprehensive evaluation of all concomitant medications, focusing on the identification of DRPs, including inappropriate dosing, potential adverse drug reactions, and clinically relevant drug–drug interactions. DRPs were classified according to the Pharmaceutical Care Network Europe (PCNE) Classification for Drug-Related Problems version 9.1 [23]. In addition, the nephrotoxic potential of concomitant medications was systematically assessed. When DRPs were identified, the clinical pharmacist provided evidence-based recommendations to the treating physician. Final clinical decisions, including acceptance and implementation of recommendations, were made solely by physicians, as clinical pharmacists in the study setting do not have prescribing authority. When necessary, therapeutic decisions were discussed collaboratively between the pharmacist and physician. Data collection Baseline data were collected using standardized data collection forms and electronic medical records. The following variables were recorded: demographic characteristics, past medical history, medications, polypharmacy (use of ≥5 drugs) status, relevant laboratory parameters. Estimated glomerular filtration rate (eGFR) was automatically calculated using the CKD-EPI 2021 equation and recorded in the patients’ medical records as mL/min/1.73 m2. Kidney failure risk was estimated using the validated Kidney Failure Risk Equation (KFRE) (https://kidneyfailurerisk.com/), providing 2- and 5-year risk projections based on routinely available clinical parameters. Clinical and laboratory parameters, including eGFR, Albumin:Creatinine Ratio (ACR), metabolic variables, and kidney failure risk estimates, were evaluated longitudinally between baseline and follow-up visit. Assessment of nephrotoxicity The nephrotoxic potential of each concomitant medication was evaluated according to the risk classification proposed by Stottlemyer et al. [17]. Medications were categorized on a scale from 0 (no known nephrotoxic potential) to 3 (definite nephrotoxicity). The nephrotoxicity was calculated as the cumulative sum of drug-specific nephrotoxicity risk scores across concomitant medications per patient. The scoring system was used to describe potential nephrotoxic exposure and was not intended to imply direct medication-related harm or inappropriate prescribing. Assessment of patient-reported outcomes The patient-reported outcomes were assessed by the Kidney Disease Quality of Life (KDQOL)-36 questionnaire, which consists of 36 items and 5 sub-dimensions. The dimensions entitled as; physical status (6 items), mental status (6 items), symptoms and problems (12 items), impact of kidney disease (8 items), and the burden of kidney disease (4 items), and the KDQOL-36 questionnaire was administered once during the study at the next follow-up visit. Responses within each domain are transformed to a 0–100 scale using the standard KDQOL-36 scoring algorithm, with higher scores indicating a better quality of life (QoL). The questionnaire has been validated in multiple languages and cultural contexts, demonstrating good internal consistency, test–retest reliability, and construct validity in CKD populations. In the present study, the KDQOL-36 was administered in its validated native language version, and domain-specific as well as overall scores were calculated according to the standardized scoring protocol [21,24]. Statistical analysis All data were analyzed by using IBM SPSS Statistics v23.0 (IBM Corp., Armonk, NY). The Shapiro–Wilk test was used to assess the normality of continuous variables. Continuous variables were expressed as mean ± standard deviation (SD) or median (interquartile range-IQR), as appropriate. Comparisons between groups were performed using the independent-samples t-test or one-way ANOVA for normally distributed variables, and the Mann–Whitney U test or Kruskal-Wallis test for non-normally distributed variables. Categorical variables were compared using the chi-square test or Fisher’s exact test. Multivariable regression analyses were planned, adjusting for clinically relevant covariates, in order to explore factors associated with the outcomes (e.g. nephrotoxicity burden or kidney failure risk). A p-value <0.05 was considered statistically significant. Sample size and bias considerations Eligible patients were recruited using a consecutive sampling approach to minimize selection bias. Given the exploratory nature of the study, a formal a priori sample size calculation was not performed. Potential sources of bias, including selection bias and confounding, were addressed through predefined inclusion criteria, standardized data collection, and multivariable statistical adjustment where appropriate. Results A total of 98 patients were initially screened, of whom 92 met the inclusion criteria. Patients who discontinued treatment within two weeks of initiation of drug therapy (n = 6) were excluded. Moreover, patients who did not attend the next follow-up appointment (n = 20) were also excluded due to incomplete information. Consequently, 72 patients were included in the final analysis. Baseline demographics and clinical characteristics of excluded patients were comparable to those included in the final cohort with respect to age (p = 0.403), sex (p = 0.313), comorbidities (p > 0.05), CKD stage (p > 0.05) and polypharmacy status (p = 0.432) (Figure 1). Study flow diagram. Among the 72 included patients, 36 (50%) were female, with a median (IQR) age of 65 (57–72) years. Thirteen patients (18.1%) were current smokers. The majority of patients had type 2 diabetes (91.7%) and hypertension (70.8%), reflecting a typical CKD population with a high comorbidity burden. Polypharmacy was common, with 69.5% of patients receiving five or more medications. The median (range) duration for the next clinic visit was 4 (3–6) months. Baseline characteristics of patients are presented in Table 1. Table 1.Patients’ baseline characteristics (n = 72). n (%) SGLT2 inhibitor used Dapagliflozin26 (36.1)Empagliflozin24 (33.3)Empagliflozin + Metformin22 (30.6) Chronic diseases Type 2 Diabetes66 (91.7)Hypertension51 (70.8)Chronic Kidney Disease50 (69.4)Coronary Artery Disease20 (27.8)Hypothyroidism13 (18.1)Chronic Obstructive Pulmonary Disease9 (12.5)Asthma6 (8.3)Congestive Heart Failure6 (8.3)Gout5 (6.9)Dyslipidemia5 (6.9)Other*13 (18.1) eGFR Stages (mL/min/1.73m2) G1 (≥90)12 (16.7)G2 (60–89)18 (25.0)G3a (45–59)21 (29.2)G3b (30–44)20 (27.8)G4 (15–29)1 (1.4)G5 (<15)0 (0.0) ACR (mg/g) categories A1 (<30)11 (15.3)A2 (30–300)21 (29.2)A3 (>300)40 (55.6)*Addison’s disease (n = 1), Alzheimer’s disease (n = 1), Atrial fibrillation (n = 2), Benign prostatic hyperplasia (n = 2), Cancer (n = 3), Celiac disease (n = 1), Deep vein thrombosis (n = 1), Sleep apnea (n = 1), Osteoporosis (n = 1).eGFR: Estimated glomerular filtration rate (CKD-EPI 2021), ACR: Urine albumin creatinine ratio. Patients’ baseline characteristics (n = 72). Addison’s disease (n = 1), Alzheimer’s disease (n = 1), Atrial fibrillation (n = 2), Benign prostatic hyperplasia (n = 2), Cancer (n = 3), Celiac disease (n = 1), Deep vein thrombosis (n = 1), Sleep apnea (n = 1), Osteoporosis (n = 1). eGFR: Estimated glomerular filtration rate (CKD-EPI 2021), ACR: Urine albumin creatinine ratio. During the study period, a total of 69 DRPs were identified in 72 patients, corresponding to nearly one DRP per patient, revealing that treatment safety (n = 25) and other problems (n = 29) constituted the majority of identified DRPs. The other problems primarily comprised inappropriate duration of drug therapy (n = 15), medication use without indication (n = 11), and need for additional drug therapy (n = 4). These were followed by treatment effectiveness (n = 13) and patient non-adherence (n = 2) (Table 2). Physicians were informed about the identified DRPs by a clinical pharmacist and appropriate interventions were undertaken following the discussion between pharmacist and physician. Table 2.Drug-related problems and pharmacist’s recommendations.1. PCNE Domain & Drug-Related Problem Categoryn (%) Treatment effectiveness Untreated symptoms9 (13.0) Subtherapeutic dosing4 (5.8) Treatment safety Adverse drug events4 (5.8) Overdosing6 (8.7) Drug–drug interactions7 (10.1) Inappropriate dose selection according to renal function8 (11.6) Other Inappropriate duration of drug therapy14 (20.3) Medication use without indication11 (15.9) Need for additional drug therapy4 (5.8) Patient-related Patient non-adherence2 (2.9) Total DRPs 69 (100.0) 2. Pharmacist Recommendation Outcomes n (%) Accepted59 (85.5) Not accepted10 (14.5) Total Recommendations 69 (100.0) 3. Medications Commonly Implicated in DRPs n (%) SGLT2 inhibitors18 (26.1)Urate-lowering agents9 (13.0)Phosphate binders8 (11.6)Ferrous sulfate7 (10.1)Erythropoiesis-stimulating agents6 (8.7)RAAS blockers5 (7.2)Antimicrobials3 (4.3)Analgesics3 (4.3)Vitamin D3 (4.3)Diuretics2 (2.9)Proton pump inhibitors2 (2.9)Lipid-lowering agents2 (2.9) Total Medications 69 (100.0) Drug-related problems and pharmacist’s recommendations. The acceptance rate of pharmacist recommendations was high (85.5%, n = 59), suggesting the feasibility and potential applicability of pharmacist-led medication review in this setting. The reasons for non-acceptance (n = 10) were primarily related to physicians’ preference for continued clinical monitoring rather than immediate therapeutic modification. Most non-accepted recommendations were not associated with urgent safety concerns or clinically significant deterioration during the follow-up period. The medications most commonly implicated in DRPs were SGLT2i (n = 18), urate-lowering agents (n = 9), phosphate binders (n = 8), and ferrous sulfate (n = 7) (Table 2). Among 18 DRPs associated with SGLT2i, the majority (n = 11) were related to mild and manageable adverse effects; including nausea (n = 3), increased urinary frequency (n = 2), shortness of breath (n = 2), fatigue (n = 2), pruritus (n = 1), and transient creatinine elevation (n = 1). Importantly, none of these events required treatment discontinuation, supporting the tolerability of SGLT2i in this patient population. Changes between baseline and follow-up visit in kidney function, glycemic parameters, and lipid parameters observed during the study are summarized in Table 3. While eGFR values remained stable, markers of kidney damage and metabolic parameters showed significant improvements. Table 3.Baseline and follow-up clinical parameters. Median (minimum-maximum) BaselineFollow-up visit p Creatinine (mg/dL) 1.21 (0.51–2.43)1.24 (0.51–3.16) 0.001 eGFR (mL/min/1.73 m²) 55.57 (29.2-117)55.01 (18.5-110) 0.016 (2.474) ACR (mg/g) 247.73 (2.11–14481.55)119.39 (0.84–9446.06) <0.001 Glucose (mg/dL) 127.5 (73–366)118 (68–241) 0.001 HbA1c (%) 6.9 (5.4–13.2)6.8 (5-11.8) 0.029 HDL-C (mg/dL) 46.50 (28–105)50.00 (23–105) 0.001 LDL-C (mg/dL) 125.5 (64–353)116 (49–332) 0.002 (3.166) Total Cholesterol (mg/dL) 193.5 (109–564)176.50 (91–504) 0.001 (3.447) 2-year Risk of Kidney Failure (%) 0.22 (0-16.61)0.16 (0-22.15)0.227 5-year Risk of Kidney Failure (%) 0.55 (0-55.82)0.55 (0-67.58)0.165 Baseline and follow-up clinical parameters. Assessment of nephrotoxic medication exposure showed that only 4 patients (5.6%) were not receiving any nephrotoxic drugs, whereas more than one-third of patients were receiving one (37.5%) or two (33.3%) nephrotoxic drugs. The median number of nephrotoxic drugs per patient was 2 (IQR: 1–3) with a median nephrotoxicity score of 1.5 (IQR: 1–2). Although most patients (68.1%) were categorized as having low nephrotoxicity risk, a small subset (2.8%) exhibited moderate-to-high risk profiles (Table 4). Table 4.Medication and nephrotoxicity profile. Number of drugs used per patients 11 (1.4)27 (9.7)37 (9.7)47 (9.7)≥5 (polypharmacy)50 (69.5) Number of nephrotoxic drugs 04 (5.6)127 (37.5)224 (33.3)316 (22.2)41 (1.4) Nephrotoxicity Score 0 (no nephrotoxicity)4 (5.56)0.5 (very low)6 (8.33)1 (low)49 (68.06)1.5 (low-moderate)11 (15.28)2 (moderate)02.5 (moderate-high)2 (2.78) Medication and nephrotoxicity profile. The analysis of 127 concomitant drugs used by the patients (excluding SGLT2i) revealed that nearly all drugs (98.4%) were classified as having very low to moderate nephrotoxic potential, while only 1.6% were categorized as moderate-to-high risk (Table 5). Table 5.The nephrotoxicity risk of concomitant drugs (n = 127).Risk of nephrotoxicityn (%)Very low (0.5 points)Allopurinol, digoxin, lansoprazole, pantoprazole, metformin, torsemide57 (44.9)Low (1 point)Irbesartan, candesartan, losartan, olmesartan, telmisartan, valsartan, lisinopril, perindopril, ramipril56 (44.1)Low-moderate (1.5 points)Furosemide, colchicine12 (9.4) Moderate (2 points) 0 (0)Moderate-high (2.5 points)Diclofenac sodium, cyclosporine2 (1.6) Total 127 (100)Note: Reported concomitant drug counts represent patient-level medication exposures identified during medication review rather than unique medication types or prescription counts. The nephrotoxicity risk of concomitant drugs (n = 127). Note: Reported concomitant drug counts represent patient-level medication exposures identified during medication review rather than unique medication types or prescription counts. Patient-reported outcomes were assessed using the KDQOL-36 questionnaire. Patients demonstrated the highest scores in the ‘symptoms and problems’ domain (mean ± SD: 88.51 ± 13.50), indicating relatively well-controlled symptom burden. In contrast, the lowest scores were observed in the ‘physical functioning’ (42.53 ± 10.33) and ‘mental health’ (44.01 ± 8.83) domains, highlighting substantial deficiency in these domains. Intermediate scores were observed for ‘impact of kidney disease’ (74.83 ± 20.17) and ‘burden of kidney disease’ (54.34 ± 34.64). The number of prescribed nephrotoxic drugs and their corresponding nephrotoxicity score were considered to reflect the potential drug-related nephrotoxicity burden in patients. Exploratory analyses demonstrated a weak positive correlation between nephrotoxicity scores and the ‘symptoms/problems’ domain of the KDQOL-36 (p = 0.046). Specifically, the number of nephrotoxic drugs was weakly associated with both 2-year (Spearman’s r = 0.06, p = 0.614) and 5-year KFRE risk (r = 0.10, p = 0.421). Similarly, nephrotoxicity score showed weak positive correlations with 2-year (r = 0.18, p = 0.14) and 5-year KFRE risk (r = 0.21, p = 0.08). These findings should therefore be interpreted cautiously given the small effect sizes and exploratory nature of the analyses. In multivariable linear regression analyses including 72 patients, adjusting for age, sex, diabetes status, and baseline eGFR, nephrotoxicity burden was not associated with estimated kidney failure risk at either 2 (p = 0.224) or 5 years (p = 0.227) (Supplementary Table S1). Exploratory analyses integrated into the multivariable models demonstrated weak negative correlations between nephrotoxicity score and QoL domains; however, these associations were small and inconsistent across domains, suggesting limited clinical relevance. Therefore, these findings should be considered descriptive and interpreted cautiously, as multiple QoL domains were evaluated simultaneously and the observed association demonstrated only a small effect size. Discussion In this observational study conducted in a nephrology outpatient setting, we demonstrated that patients initiated on SGLT2i prone to a substantial burden of polypharmacy, DRPs and exposure to potentially nephrotoxic medications. Importantly, clinical pharmacist-led medication review identified nearly one DRP per patient and was associated with a high physician acceptance rate, highlighting the feasibility of integrating pharmacist services into nephrology care in a real-world outpatient setting. One of the most notable findings of this study is the high prevalence of DRPs (n = 69 in 72 patients), which underscores the complexity of pharmacotherapy in CKD populations. Consistent with previous literature, polypharmacy was highly prevalent (69.5%) and emerged as a key contributor to drug-related issues [25,26]. The predominance of DRPs related to inappropriate treatment duration and medication use without clear indication suggests that not only drug selection but also longitudinal treatment management may require further optimization in routine practice. These findings reinforce the need for systematic medication review, particularly in high-risk populations such as patients with CKD. Our results also highlight the critical role of clinical pharmacists in identifying and resolving DRPs. The high acceptance rate of pharmacist recommendations (85.5%) is consistent with prior studies and reflects both the clinical validity of the recommendations and the willingness of physicians to collaborate in optimizing pharmacotherapy. Notably, even in a healthcare setting where pharmacists do not have prescribing authority, their contribution to patient care was substantial. These findings may provide supportive real-world observations regarding the potential role of structured pharmacist-led medication review models in nephrology clinics. Another key finding is the widespread exposure to potentially nephrotoxic medications. Nearly all patients were receiving at least one nephrotoxic drug, and the median nephrotoxicity score was clinically relevant. While most medications were classified as having low to moderate nephrotoxic potential, the cumulative effect of multiple such agents may still be clinically relevant, particularly in patients with already compromised renal function. This observation aligns with previous reports indicating that CKD patients are frequently exposed to nephrotoxic agents, often without systematic risk assessment [27,28]. It should be noted that nephrotoxicity burden may vary across healthcare settings. A higher burden might be expected in primary care or general internal medicine settings, where prescribing practices may differ from those in specialized nephrology clinics. In addition, nephrotoxic risk in CKD populations is highly dynamic and the nephrotoxic potential of medications is context-dependent which may vary according to CKD stage, baseline kidney function, intravascular volume status, comorbidities, and treatment indication. Several drugs classified as potentially nephrotoxic by the implemented nephrotoxicity score, such as angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), loop diuretics, and colchicine, they remain essential components of CKD management due to their well-established clinical benefits. Therefore, the cumulative nephrotoxicity scoring approach used in this study should not be interpreted as a direct measure of medication-related harm, but rather as a structured framework for describing potential nephrotoxic exposure within a complex clinical setting. Despite this high exposure, we did not observe a statistically significant association between nephrotoxicity burden and estimated kidney failure risk. Several factors may explain this finding. First, the relatively short follow-up period may have limited the ability to detect meaningful changes in long-term risk estimates. Second, the majority of medications were categorized as low to moderate risk, which may attenuate the overall impact on kidney outcomes. Finally, the use of risk prediction models such as KFRE, although validated, may not be sufficiently sensitive to short-term changes in drug-related risk. Although albuminuria and glycemic parameters demonstrated numerical improvement during follow-up, estimated 2-year and 5-year kidney failure risks did not significantly change over the study period. This finding may be explained by the relatively short follow-up duration and the chronic progressive nature of CKD, as KFRE is primarily driven by stable long-term parameters such as age, eGFR, and albuminuria. Therefore, short-term clinical improvements may not immediately translate into measurable changes in estimated kidney failure risk. Moreover, the absence of significant change in estimated kidney failure risk over the relatively short follow-up period was expected, given that KFRE is based on relatively stable clinical variables and was originally developed to estimate long-term kidney outcomes rather than short-term changes. Moreover, the numerically small decline observed in eGFR may fall within the expected therapeutic effects of SGLT2i and also biological and analytical variability of creatinine-based eGFR estimation, particularly over a relatively short follow-up period, and therefore may not represent a clinically meaningful change. Importantly, SGLT2i demonstrated a favorable safety profile in our cohort. Although mild adverse effects were observed, none required treatment discontinuation. It should also be noted that several SGLT2i related DRPs represented mild and expected pharmacological effects requiring monitoring or clinical evaluation rather than severe medication-related harm. However, inclusion of these problems may have contributed to the overall number of DRPs identified in this population. In parallel, we observed significant improvements in albuminuria and metabolic parameters, including glycemic and lipid control. These results are generally consistent with findings reported in previous large-scale clinical trials evaluating SGLT2i [4–6]. However, given the observational and uncontrolled nature of the study, these improvements cannot be attributed solely to SGLT2i initiation or pharmacist’s interventions. Potential confounding factors, concomitant treatment modifications, and follow-up bias may also have contributed to the observed changes. The assessment of patient-reported outcomes provides additional insight into the multidimensional impact of CKD and its treatment [21]. While symptom burden appeared relatively well controlled, patient-reported lower scores in physical and mental health domains, indicating persistent impairment in these areas. Interestingly, nephrotoxicity burden was not meaningfully associated with quality of life, suggesting that subclinical drug-related risks may not be directly perceived by patients. This finding emphasizes the importance of clinician-driven risk assessment alongside patient-reported measures. However, these QoL findings should be interpreted cautiously, as multiple KDQOL-36 domains were evaluated simultaneously and the observed association was limited to a small effect size within a single exploratory domain. Because KDQOL-36 assessments were performed only once at follow-up, the present study does not allow evaluation of longitudinal changes in QoL over time. In addition, QoL scores assessed at follow-up may already have been influenced by early effects of SGLT2i therapy, which may have affected the observed associations with nephrotoxicity burden. Our findings suggest that while SGLT2i are effective and well tolerated, their use occurs within a complex therapeutic landscape characterized by polypharmacy and potential nephrotoxicity. In this context, clinical pharmacist involvement may contribute to the identification of medication-related issues during routine nephrology care. This study has several important clinical implications. First, routine integration of pharmacist-led medication review may facilitate the identification and management of DRPs in nephrology settings. Second, systematic evaluation of nephrotoxic medication burden should be considered as part of comprehensive CKD management. Third, strategies to enhance appropriate use and monitoring of SGLT2i may support comprehensive medication management in CKD patients. Limitations Several limitations should be acknowledged. The single-center design may limit generalizability. The observational and uncontrolled nature of the study precludes causal inference, and the absence of a comparison group limits interpretation regarding the clinical effectiveness of pharmacist-led interventions. Therefore, the findings should be interpreted as descriptive real-world observations rather than evidence of causal benefit. The relatively short follow-up period may have limited the detection of long-term clinical outcomes, particularly kidney failure risk. Although excluded patients may have introduced survivorship bias regarding treatment discontinuation, this bias can be overcome by the fact that the baseline demographic and clinical characteristics of the excluded patients were comparable to those of the patients included in the final cohort. In addition, nephrotoxicity was assessed using a structured classification system, the cumulative scoring approach may not fully capture the dynamic and patient-specific nature of nephrotoxic risk in CKD populations, where risk may vary according to baseline kidney function, CKD stage, volume status, comorbidities, and therapeutic indication. DRPs identification and classification were performed by the same clinical pharmacist, and formal inter-rater reliability assessment was not conducted. Furthermore, subgroup or sensitivity analyses according to CKD stage or eGFR category were not performed because of the relatively limited sample size and exploratory nature of the study. Finally, potential residual confounding cannot be excluded despite adjustment in multivariable analyses. Conclusions In conclusion, patients with CKD receiving SGLT2i are exposed to a substantial burden of polypharmacy, DRPs, and potentially nephrotoxic medications. Clinical pharmacist-led medication review enabled the identification of DRPs and nephrotoxic medication exposure in this real-world nephrology outpatient setting, with high acceptance by physicians. While nephrotoxicity burden of concomitant drugs was not independently associated with short-term kidney failure risk, its high prevalence underscores the need for systematic assessment. Integrating clinical pharmacists into nephrology care may represent a feasible approach for supporting comprehensive medication review in routine nephrology outpatient practice. Supplementary Material Supplemental Material