Impaired vascular function in patients with chronic kidney disease who have elevated symmetric dimethylarginine but not asymmetric dimethylarginine

Abstract

Chronic kidney disease (CKD) is associated with heightened cardiovascular disease (CVD) risk, partly due to impaired peripheral vascular function. Symmetric dimethylarginine (SDMA) and asymmetric dimethylarginine (ADMA) are emerging biomarkers implicated in nitric oxide (NO) regulation and vascular health. While ADMA is a well-established inhibitor of NO synthesis, recent evidence suggests that SDMA may also play a critical role in vascular health, especially in CKD prior to end-stage. Thus, in 23 stage 3 and 4 patients with CKD (66±9 years) and 32 age-matched controls (64±8 years), we compared serum SDMA and ADMA levels and examined their associations with vascular function, including flow-mediated dilation (FMD), peak blood velocity to reactive hyperemia, and carotid-femoral pulse wave velocity (cfPWV). SDMA was significantly elevated in patients with CKD (163±37 vs. 100±15 ng/mL, p<0.0001), while ADMA did not differ significantly between groups (111±22 vs. 103±12 ng/mL, p=0.083). Patients with CKD had lower FMD (3.66±2.45 vs. 4.47±2.45%, p=0.048) and peak blood velocity (47.43±16.67 vs. 60.18±16.88 cm/s, p=0.009), but higher cfPWV (8.82±1.53 vs. 7.69±1.35 m/s, p=0.004) than controls. Pooled analysis revealed that SDMA correlated inversely with eGFR (r = −0.86, p<0.0001), FMD (r_s= −0.28, p=0.039), and peak blood velocity (r_s= −0.40, p=0.001) but not cfPWV (r=0.14, p=0.338). ADMA correlated inversely with peak blood velocity (r_s= −0.28, p=0.042) but not eGFR (r= −0.25, p=0.063), FMD (r_s= −0.06, p=0.664), or cfPWV (r=0.21, p=0.146). Collectively, these findings suggest that SDMA, relative to ADMA, may be a stronger marker of vascular dysfunction in stage 3–4 CKD. However, the predictive value of SDMA for vascular function was modest, which may limit its overall potential as a biomarker for vascular function in CKD.

New and Noteworthy

The associations between symmetric dimethylarginine (SDMA) and asymmetric dimethylarginine (ADMA) and measures of vascular function were investigated in patients with stage 3–4 chronic kidney disease (CKD). We found that SDMA exhibited stronger relationships with vascular function than ADMA. However, the strength of associations was modest, potentially limiting their role as standalone predictors of vascular dysfunction. Nonetheless, these data support emerging evidence of a differential impact of SDMA and ADMA in patients with CKD.

Graphical Abstract

Introduction:

Chronic kidney disease (CKD) is a significant global health concern, affecting approximately 10% of the world’s population (1–3) and is associated with high morbidity and mortality rates (2). Notably, with the aging population, the condition is expected to increase in prevalence and become the 5^th leading cause of years of life lost by 2040 (4). CKD is defined by a progressive decline in kidney function, measured by estimated glomerular filtration rate (eGFR), and is classified into five stages. Stages 1 and 2 are considered mild and often reflect normal age-related changes. In contrast, stages 3 and 4 indicate more substantial impairment, where clinical management focuses on slowing disease progression and preventing the onset of stage 5, or end-stage renal disease. Importantly, most individuals with CKD do not die from kidney failure but rather die prematurely due to cardiovascular disease (CVD), with a linear increase in mortality risk observed once CKD has progressed to stage 3 (eGFR < 60 ml/min/1.73 m²) (5, 6). Therefore, understanding the underlying contributing factors and mechanisms of CVD development in CKD is crucial.

It is well established that impairments in vascular function contribute to cardiovascular complications in CKD (6–8). Indeed, endothelial dysfunction, increased arterial stiffness, and a greater propensity for atherosclerosis substantially increase CVD risk in patients with CKD (6, 9, 10). One mechanism mediating vascular impairments in CKD is reduced bioavailability of the potent vasodilator nitric oxide (NO) (11). Attenuated NO bioavailability in CKD stems partly from the kidneys’ inability to eliminate waste and toxins, which can lead to the accumulation of endogenous uremic toxins that obstruct NO formation (10, 12–15). Among these, asymmetric dimethylarginine (ADMA) and symmetric dimethylarginine (SDMA) have garnered significant attention as strong non-traditional CVD risk factors (12, 14, 16–18).

SDMA and ADMA are both naturally occurring circulating amino acids that can interfere with NO production (14). Historically, the majority of research has focused on ADMA as it is a potent endogenous NO synthase inhibitor shown to significantly predict CVD risk and mortality in patients with CKD (11, 14, 17, 19). Likewise, much attention has been given to end-stage CKD patients (i.e., stage 5), with fewer studies on the earlier stages of CKD (17). Indeed, this work has been mainly performed in end-stage renal disease patients on dialysis, and transplant populations (16). Additionally, these findings, although convincing, are largely correlational and based on clinical endpoints, with fewer studies directly examining vascular function. To our knowledge, only two studies have investigated the relationship between ADMA and vascular function in pre-dialysis CKD patients, reporting that eGFR correlates negatively with ADMA and positively with flow-mediated dilation (FMD), a marker of NO-mediated dilation (10). Moreover, ADMA has been reported to be inversely associated with FMD and positively associated with vascular stiffness in patients with CKD (10, 20). Thus, the limited available data suggest that ADMA and its associations with vascular dysfunction appear to vary with disease severity, becoming more pronounced and apparent in the advanced stages of CKD (10, 20–23).

Unlike ADMA, SDMA is not a direct inhibitor of NO but can indirectly reduce NO bioavailability by limiting L-Arginine availability, increasing superoxide production, and contributing to the uncoupling of endothelial NO synthase (24–26). Although less studied, emerging evidence suggests that SDMA may also be an important clinical marker of CVD risk (16, 26, 27). Likewise, few studies have focused on the role of SDMA in mediating vascular dysfunction; however, a large multicenter prospective study suggested that SDMA may be a more potent risk factor than ADMA (28). Specifically, Emrich et al. (2018) found that SDMA, compared to ADMA and other metabolites, was the only marker that strongly predicted disease progression and incident cardiovascular events after adjusting for potential confounders (e.g., age, sex, eGFR, inflammation, comorbidities, and medications) in non-dialysis patients with CKD. Thus, emerging evidence suggests that SDMA may be an equally strong, or even stronger, predictor of CVD risk in CKD compared to ADMA (16, 26–29). While the mechanisms are not well understood, one study found an association between elevated SDMA and impaired FMD in patients with stage 2–5 CKD and renal transplant recipients (29). However, ADMA was not measured and evaluated in parallel, limiting direct comparisons. In addition to its potential vascular relevance, SDMA is also considered a more reliable marker of renal function than ADMA (30). Notably, the available studies have examined either ADMA or SDMA in isolation, not together, have focused predominantly on clinical endpoints or a limited range of vascular outcomes, and have largely been conducted in dialysis populations. This highlights a critical gap in the literature and underscores the need for further research to examine ADMA and SDMA together with vascular outcomes in the earlier stages of CKD, where preventive interventions may have the greatest impact.

With this background in mind, we aimed to compare serum SDMA and ADMA concentrations in stages 3 and 4 patients with CKD with age-matched controls and to assess their association with vascular function. We hypothesized that elevated SDMA and ADMA would be associated with impaired vascular function. Furthermore, based on emerging research suggesting that SDMA is a stronger predictor of CVD risk (28), we also explored the hypothesis that SDMA would be more strongly related to measures of vascular function than ADMA. To the best of our knowledge, this is the first study to comprehensively explore the roles of these two molecules with direct measures of vascular function in patients with stage 3 and 4 CKD.

Methods

Participants

Twenty-three patients (12 males, 11 females) with a clinician diagnosis of stage 3 and 4 CKD (Age: 66±9 years; eGFR: 37±11 ml/min/1.73 m², mean ± standard deviation) were recruited through the Renal Specialists of North Texas in Arlington, and thirty-two age-matched controls (13 males, 19 females) without a diagnosis of CKD (Age: 64±8 years; eGFR: 84±13 ml/min/1.73 m²) were recruited from the Arlington area. All participants provided verbal and written informed consent and completed a comprehensive medical history form and medication list. All experimental procedures conformed to the Declaration of Helsinki and were approved by the Institutional Review Board at the University of Texas at Arlington. All CKD participants had a diagnosis of at least one comorbidity, with the most common being hypertension (n=22), type 2 diabetes (n=14), and high cholesterol (n=13). The majority of patients with CKD were also prescribed one or more medications for these conditions, including calcium channel blockers (n=14), angiotensin-converting enzyme inhibitors (n=9), beta-blockers (n=8), angiotensin 2 inhibitors (n=7), insulin (n=7), metformin (n=5), glucagon-like peptide-1 agonist (n=5), thiazolidinediones (n=4), statins (n=17), and cholesterol absorption inhibitor (n=5). Some of the control participants had a diagnosis of hypertension (n=12), pre-diabetes or type 2 diabetes (n=6), and high cholesterol (n=14). Prescribed medications in the control group included calcium channel blockers (n=2), angiotensin-converting enzyme inhibitors (n=4), beta-blockers (n=4), angiotensin 2 inhibitors (n=4), metformin (n=4), glucagon-like peptide-1 agonist (n=1), statins (n=10), and cholesterol absorption inhibitor (n=1).

Study design

Participants arrived at the laboratory in the morning after being instructed to abstain from food, caffeine, medication or supplements, and nicotine for >12 hours and alcohol and exercise for >24 hours. Participants then transitioned to a cushioned medical exam table and laid quietly, resting in a dimly lit temperate controlled room (21⁰C) while measures of resting blood pressure (BP), heart rate (HR), and vascular function were performed following standard procedures.

Experimental measurements and protocol

Cardiovascular assessments:

Participants were instrumented with standard lead II electrocardiogram (model Q710, Quinton, Bothell, WA) and pneumobelt (Pneumotrace II 1132, UFI, Morro Bay, CA) to continuously assess heart rate (HR) and monitor major respiratory excursions, respectively continuously. The latter is performed to avoid the impact of major respiratory excursions (i.e., deep sigh, yawn) during data collection. Participants rested for 20 minutes while a high-quality image of the brachial artery was obtained. Peripheral macro (endothelial dilation; %) and micro- (reactive hyperemia; peak blood velocity) vascular function was non-invasively assessed using the brachial artery FMD technique according to current guidelines (31) and as previously reported (32). Importantly, this vascular assessment is thought to indicate, at least in part, NO-mediated vasodilation capacity and has a good predictive value for future cardiovascular outcomes (33, 34). Briefly, using duplex Doppler ultrasonography (GE LOGIQ P5 or GE Logiq P9, Milwaukee, WI) with an 11-MHz linear array transducer, brachial artery diameter and blood velocity were assessed at rest and following a 5-min period of forearm ischemia induced by rapidly inflating a cuff (Hokanson, Bellevue, WA) placed approximately 2 cm distal to the antecubital fossa to supra systolic pressure (220 mmHg). Stable continuous data recordings were obtained over a 5-minute baseline period and from 30 seconds before to 3 minutes after cuff release.

After FMD assessment, central arterial stiffness was non-invasively assessed using carotid-femoral pulse wave velocity (cfPWV) according to the manufacturer’s instruction (SphygmoCor, AtCor Medical device, and XCEL 1.3 software, Sydney, Australia) and as previously reported (32). For this assessment, a cuff is placed around the upper thigh, and handheld tonometry is positioned at the carotid artery at the strongest pulsation point found by palpation. Distances between the carotid artery to the sternal notch, the sternal notch to the top of the thigh cuff, and between a strong pulsation point of the femoral artery (also identified through palpation) to the top of the thigh cuff are obtained and input into the software. Pulse waves were visually inspected before recording at least two measures (> 1 min apart) of stable and robust pulse waves that also passed the quality control of the software. Due to technical difficulty, cfPWV was not obtained in 2 controls and 2 patients with CKD. The same software system assesses the augmentation index corrected for heart rate (AIx75), central systolic, diastolic, and mean BP using an analysis of arterial waveforms from an automated brachial cuff. cfPWV and AIx75 were recorded in duplicates and averaged for each participant. Standard resting brachial BP was obtained every minute during a 10-minute quiet rest using an automated sphygmomanometer (Welch Allyn, Skaneateles Falls, NY).

Blood markers

A venous blood sample was obtained following vascular assessments using serum and plasma collection tubes and handled according to standard protocols by a commercial laboratory (Quest Diagnostics and LabCorp). Blood samples were assessed for eGFR (calculated from serum creatinine), metabolic markers (e.g., blood glucose), as well as ADMA and SDMA. Metabolic markers were assessed using an enzymatic assay (Labcorp, Burlington, NC, USA), while ADMA and SDMA were measured using Liquid Chromatography/Tandem Mass Spectrometry (LC/MS/MS; Quest Diagnostics, Secaucus, NJ, USA).

Data analysis

Brachial artery diameter and mean blood velocity were assessed using specialized offline software for wall tracking and edge detection (LabVIEW, National Instruments, Austin, TX). Macrovascular function was defined as FMD%, calculated by the formula: [(three-beat average peak diameter − baseline diameter) / baseline diameter] × 100. Microvascular function was defined as the three-beat average peak blood velocity following cuff release. The shear rate was determined by the equation: [(8 × mean blood velocity/diameter]. Using the trapezoidal method, the shear stimulus for brachial artery dilation was estimated by calculating the area under the curve (AUC) for the hyperemic shear rate up to peak brachial artery dilatation. For cfPWV, the average of two measures was used if the difference between measures was <0.5 m/s. A third assessment was performed when the difference was >0.5, and the median was used per current recommendations (35). The software calculates cfPWV as the average transit time based on the entered distance measures and recorded pulses. For central BP and AIx75, an average of two consistent measurements was used. An average of five measures were taken during a 10-minute resting period for brachial artery BP.

Statistical analysis

The normality of data was assessed using the Shapiro-Wilk test and QQ plots. Data are presented as mean ± standard deviation. Comparisons between CKD and the control group were analyzed using Student’s t-test or Mann-Whitney U test based on the normality of data. A one-tailed analysis was chosen for the group comparison of vascular function outcomes given the strong evidence of impairments in CKD (10, 11), while all other analyses were two-tailed. FMD was corrected for shear stress AUC to peak diameter minus baseline shear using analysis of covariance to consider the impact of shear stress. Associations were analyzed using the Pearson or Spearman rank correlation coefficients as appropriate, and the strength of relationships was estimated as negligible (r=0.00–0.10), weak (r=0.10–0.39), moderate (r=0.40–0.69), strong (r=0.70–0.89), and very strong (r=0.90–1.00) (36). A priori significance was set at p < 0.05. GraphPad Prism Software version 10 (La Jolla, California, USA) and IBM SPSS Statistics version 30 (Armonk, NY, USA) were used for all statistical analyses.

Results

Participant characteristics

Table 1 summarizes the characteristics of the study participants. The CKD group had higher body weight, BMI, and waist-to-hip ratio than controls. Resting brachial systolic and mean BP were also elevated in patients with CKD, though diastolic BP and HR did not differ between groups. Among metabolic markers, fasting blood glucose was elevated in CKD, while triglycerides and total cholesterol were not different between groups.

Table 1.

Participant characteristics

	Control (n=32)	CKD (n=23)		P value
Anthropometrics
Height (cm)	166.6±8.4	167.3±10.0		0.788
Body weight (kg)	78.1±17.3	89.9±15.8		0.012
Body mass index (kg/m2)	28.1±5.6	32.0±4.5		0.007
Waist-hip ratio	0.90±0.1	0.97±0.1		0.020
Cardiovascular
Systolic BP (mmHg)	129±15	146±23		0.001
Diastolic BP (mmHg)	77±9	81±10		0.113
Mean BP (mmHg)	94±9	103±13		0.004
Heart rate (bpm)	60±9	64±9		0.104
Metabolic
Glucose (mg/dL)	102±33	132±50		0.009
Triglycerides (mg/dL)	105±39	121±50		0.191
Total cholesterol (mg/dL)	175±37	173±45		0.853

Values are mean ± standard deviation. Abbreviations: n, number of participants; cm, centimeters; kg, kilograms; kg/m2, kilograms divided by height (in meters) squared; BP, blood pressure; mmHg, millimeters of mercury; mg/dL, milligrams per deciliter. Group comparisons were made using an unpaired Student’s t-test.

Serum SDMA and ADMA in patients with CKD and controls are reported in Figure 1. SDMA was elevated in patients with CKD compared to controls (163±37 vs. 100±15 ng/mL, p<0.0001) and showed a strong inverse relationship with eGFR in CKD (r= −0.77, p<0.0001) and controls (r= −0.55, p= 0.001). ADMA, however, did not differ between patients with CKD and controls (111±22 vs. 103±12 ng/mL, p=0.083) and was not associated with eGFR in CKD (r= −0.31, p=0.150) or controls (r=0.12, p=0.508). Pooled data across both groups (n=55) demonstrated a strong inverse correlation between SDMA and eGFR (r= −0.86, p<0.0001), while ADMA had only a weak, but not significant, relationship with eGFR (r= −0.25, p= 0.063).

Figure 1.

Serum levels of symmetric dimethylarginine (SDMA, A) but not asymmetric dimethylarginine (ADMA, B) were significantly elevated in patients with stage 3 and 4 chronic kidney disease (CKD) compared with age-matched controls. Data are presented as mean and individual values. Group comparisons were made using an unpaired Student’s t-test.

Macro- and microvascular function

Baseline brachial artery diameter (0.436 ± 0.083 cm vs. 0.362 ± 0.065 cm, p=0.001) and peak diameter following cuff release (0.452 ± 0.086 cm vs. 0.378 ± 0.061 cm, p=0.001) were higher in patients with CKD than in controls. FMD was significantly lower in CKD than in controls (Figure 2A). The shear stimulus for brachial artery dilation, reflected by the AUC to peak diameter minus baseline shear, was significantly lower in CKD than in controls (17,527 ± 8236 vs. 31,399 ± 20,418 arbitrary units (AU), p=0.0001]. FMD % after correcting for shear stimulus was not significantly different between groups (p=0.769). No group difference was observed for baseline mean blood velocity (p=0.327). Peak blood velocity following cuff release (reactive hyperemia) was significantly lower in CKD than in controls (Figure 2B).

Figure 2.

Flow-mediated dilation (FMD, A) and peak blood velocity after cuff release (B) were evaluated for macro- and microvascular function. FMD and peak blood velocity were significantly lower in patients with stage 3 and 4 chronic kidney disease (CKD) compared to age-matched controls. Data are presented as medians with individual values, and group comparisons were conducted using a Mann-Whitney U test.

Arterial stiffness and central blood pressure

cfPWV was significantly higher in CKD than in controls (Figure 3A), while no group differences were observed for AIx75 (Figure 3B). Central SBP (131 ± 17 mmHg vs. 121 ± 14 mmHg, p=0.018) and pulse pressure (48 ± 11 mmHg vs. 42 ± 9 mmHg, p=0.038) were also significantly higher in CKD than in controls. Central DBP (83 ± 12 mmHg vs. 78 ± 10 mmHg, p=0.091) and MAP (98 ± 13 mmHg vs. 92 ± 10 mmHg, p=0.054) were not statistically different between groups.

Figure 3.

Arterial stiffness, measured using the gold-standard carotid-femoral pulse wave velocity (cfPWV, A), was significantly higher in patients with stage 3 and 4 chronic kidney disease (CKD) compared to age-matched controls. In contrast, the heart rate-corrected augmentation index (AIx75, B), a less sensitive indicator of arterial stiffness, did not differ between the groups. Due to technical difficulty, cfPWV was not obtained in 2 controls and 2 patients with CKD. Data are presented as means with individual values, and group comparisons were conducted using an unpaired t-test.

Correlations

The main correlations are shown in Figure 4. Pooled data analysis revealed a weak but significant inverse correlation between SDMA and both FMD (r = −0.28, p = 0.039) and peak blood velocity following cuff release (r = −0.40, p = 0.003). Conversely, serum ADMA showed no relationship with FMD; however, a weak but significant inverse relationship with peak blood velocity was found (r = −0.28, p = 0.042). No significant correlations were found between SDMA or ADMA and any arterial stiffness indices (all: p>0.05). Likewise, no correlations were found between SDMA or ADMA and brachial or central blood pressure (all: p>0.05).

Figure 4.

The top section shows correlations between symmetric dimethylarginine (SDMA) and vascular function markers, while the bottom section presents correlations between asymmetric dimethylarginine (ADMA) and these same markers across both controls and patients with stage 3 and 4 chronic kidney disease (CKD). SDMA showed a significant inverse correlation with flow-mediated dilation (FMD, A) and peak blood velocity (B). No significant correlation was found between SDMA and carotid-femoral pulse wave velocity (cfPWV, C) or the heart rate-corrected augmentation index (AIx75, D). In contrast, ADMA displayed a significant inverse relationship only with peak blood velocity (F), with no significant associations observed for FMD (E), cfPWV (G), or AIx75 (H). Correlations were analyzed using Pearson and Spearman coefficients, as appropriate.

Discussion

The primary novel finding of our study is that SDMA, but not ADMA, is significantly elevated in non-dialysis patients with CKD and more strongly associated with impaired macro- and microvascular function. This is particularly noteworthy because ADMA has historically been considered a key contributor to endothelial dysfunction in CKD due to its direct NO inhibitory effects. However, in our study, serum ADMA did not differ significantly from controls, was more variable, and showed weaker associations with macro- and micro-vascular function measures, as well as renal function. In contrast, SDMA was substantially elevated and more strongly correlated with macro- and micro-vascular function and renal function. However, although the association with SDMA and renal function was strong, the relationship with vascular function was modest, likely limiting its utility as a stand-alone predictor of vascular dysfunction in stage 3 and 4 CKD prior to end-stage renal disease and dialysis.

Our findings suggest a potential role of SDMA in mediating vascular dysfunction in CKD, which is consistent with emerging evidence (16, 28, 29). Unlike ADMA, which directly inhibits NO synthase, SDMA indirectly influences NO bioavailability through multiple mechanisms, including competition with L-arginine for cellular uptake, superoxide production, and NO synthase uncoupling (12, 14, 24–26). Thus, SDMA may contribute to endothelial dysfunction and vascular stiffening, particularly in CKD, where renal elimination of SDMA is impaired. Moreover, SDMA has been linked to higher superoxide production, which not only reduces NO bioavailability but also contributes to vascular oxidative stress and inflammation- hallmarks of CKD-related CVD (6, 13, 37, 38). The strong correlation between SDMA and eGFR in our cohort is also noteworthy and supports its potential utility as a marker of both renal and vascular health, as previously proposed (26, 30); however, additional research is warranted.

In contrast, serum ADMA was not significantly elevated in our CKD cohort and showed weaker associations with macro- and micro-vascular function and renal function. This somewhat conflicts with previous studies identifying ADMA as a significant predictor of cardiovascular events and CKD progression (16, 17, 19, 22, 39). A possible explanation is that ADMA is primarily eliminated through enzymatic degradation by dimethylarginine dimethylaminohydrolase (DDAH), with renal clearance contributing to a lesser extent (12, 14). If DDAH activity remains functional in earlier stages of CKD, ADMA levels may remain relatively stable until later stages of disease progression. This finding is consistent with studies reporting stronger associations between ADMA and cardiovascular outcomes in dialysis-dependent (stage 5) patients than in pre-dialysis CKD (16, 28). Additionally, comorbid conditions such as hypertension, diabetes, and dyslipidemia, common in CKD, may differentially influence ADMA metabolism, further contributing to the variability observed in our study and across previous studies (13, 14, 16, 40). Indeed, our study included controls with common age- and disease-related comorbidities, while previous studies reporting elevated ADMA and associations with vascular function (10, 20) have primarily compared patients with CKD to healthy controls.

Although limited, previous studies have reported associations between both SDMA and ADMA (29), with vascular function in pre-dialysis CKD (10, 20). However, these studies did not evaluate SDMA and ADMA in parallel. Thus, our study extends this literature by directly comparing ADMA and SDMA in relation to macro- and micro-vascular function and arterial stiffness measures in patients with stage 3–4 CKD. We found that SDMA, more so than ADMA, was associated with both reduced macro- and micro-vascular function. In contrast, neither SDMA nor ADMA showed significant associations with arterial stiffness or central blood pressure. Thus, by measuring both ADMA and SDMA along with multiple vascular outcomes in early-stage CKD, we were able to offer additional insights into their potential contributions to vascular and renal dysfunction in stages of CKD when preventive strategies may be most impactful.

Given that CVD remains the leading cause of death in patients with stage 3 and 4 CKD, there is an urgent need to identify the early contributors to vascular dysfunction before irreversible damage occurs. Our findings suggest that SDMA may be an earlier contributor to vascular impairment compared to ADMA in CKD stages prior to end-stage. SDMA’s strong inverse relationship with eGFR and its associations with endothelial dysfunction also support previous work emphasizing its potential utility for risk stratification and early intervention before patients progress to more severe disease, including dialysis dependence and CVD (16, 26). Notably, accumulating studies indicate that SDMA is not only a reliable indicator of renal function but may also outperform creatinine-based eGFR estimates in specific populations (26, 30). Furthermore, emerging evidence suggests that SDMA is more strongly associated with cardiovascular outcomes in patients with CKD who are not yet on dialysis (16, 26, 27). In contrast, ADMA’s predictive value appears more pronounced in critically ill patients and those with advanced renal failure (16, 28). Collectively, these data indicate that SDMA may potentially be useful for identifying patients at increased CVD risk before the development of end-stage kidney disease.

However, whether incorporating SDMA into existing CKD risk models could enhance the early detection of vascular dysfunction and enable clinicians to tailor interventions more effectively for patients at high CVD risk remains an open question. Additional research is needed to determine whether SDMA monitoring provides additive predictive value beyond traditional risk factors such as hypertension, proteinuria, and eGFR. Since the kidneys almost exclusively clear SDMA, therapeutic strategies to directly reduce SDMA levels remain challenging. However, longitudinal studies could help establish whether improving renal clearance of SDMA through dialysis, pharmacological, or lifestyle interventions would improve cardiovascular outcomes. Moreover, interventions targeting SDMA’s downstream effects, such as endothelial NO bioavailability, oxidative stress, and inflammation, may provide alternative strategies to mitigate its vascular impact. For example, L-arginine supplementation has been proposed as a potential therapy to counteract reduced NO bioavailability from elevated uremic toxins such as SDMA, though its effectiveness in CKD remains unclear (40–42). Moreover, antioxidant therapies to reduce SDMA-mediated oxidative stress may help preserve endothelial function in patients with CKD (43). Thus, while our findings highlight a potential role of SDMA in vascular dysfunction in pre-dialysis CKD, further research is needed to explore underlying mechanisms, its role in vascular disease progression, cardiovascular events, and response to therapeutic interventions.

Conclusion

In conclusion, our findings suggest that SDMA, compared to ADMA, is more consistently elevated in patients with stage 3–4 CKD and exhibits stronger relationships with vascular dysfunction and renal function.