Symmetric dimethylarginine (SDMA)

Symmetric dimethylarginine (SDMA)


Andy Pachikerl, Ph.D



For over a millennium and a few centuries, urinalysis has given leads to medical diagnoses. It was until the repetitive use of clinical chemistry approximately 50 – 60 years that these data of renal biomarkers became commonplace in human and veterinary medicine. From here onwards, both an improved understanding of the renal system and ability to diagnose renal disease was updated. In the past, renal biomarkers have focused on kidney function testing, and this is the basis for current conventional test in blood (serum creatine [sCr], urea or urea nitrogen [UN] as endogenous indicators of glomerular filtration rate [GFR]). Recently, we are becoming more aware of the need to identify renal disease at an early stage when therapeutic options are most effective. Both sCr and UN both play vital role in diagnosis of kidney disease, their limitations create poor confidence for their use as early indicators for disease. New markers of renal function try to overcome these limitations. Additionally, there are now many urinary markers that can detect kidney damage and help localize that damage to the compartment of the kidney that is affected.


Endogenous markers of GFR


The most common endogenous marker for estimating GFR is serum creatinine and its metabolism. Measurement and diagnostic significance in dogs have previously been reviewed (Braun, et al., 2003). Recent reviews, however, suggested factors that can either enhance or limit the clinical use of sCr to optimize diagnostician and clinical pathologist to interpret the data of this conventional test. Particularly, accurate interpretation of published data, population-specific reference intervals, trending of sCr and consideration of muscle mass influence and analytic variability are all needed to best interpret sCr in dogs and cats. Of note, although creatinine is referred to as sCr throughout this manuscript, creatinine is also commonly measured in plasma.


Nephron mass vs nephron function. It is generally accepted that 75% of nephron mass must be lost before sCr increases above the reference limit (Braun, et al., 2003). The original source for this statement likely originates from partial nephrectomy studies in dogs. However, it is often mistaken for 75% loss of renal function vs mass. In partial nephrectomy studies, ¾ loss of renal mass related to about 50–60% or 35–45% reduction in renal function based on inulin clearance one month or 13 months post-surgery, respectively (Brown, et al., 1990; Bovee, et al., 1979). The much lower decrease in function as compared with the percentage of nephron loss is due to compensatory changes in remaining nephrons (ie, compensatory renal hypertrophy) (Brown, et al., 1990; Bricker, et al., 1964). Furthermore, using an age- and breed-specific reference limit (sCr ≥ 106 mmol/L or 1.2 mg/dL) along with frequent monitoring, adolescent dogs with rapidly progressive kidney disease due to X-linked hereditary nephropathy (XLHN) demonstrated increased sCr after GFR had decreased an average of 48% (range 39–68%).8 Based on these studies, sCr can be more sensitive for detecting decreased renal function than has been historically assumed.


Value of population-specific reference intervals. While sCr is not as poorly sensitive as generally believed, its inability to regularly detect < 50% decline in kidney function at least partly stems from reference intervals that are overly wide for patients with low baseline sCr. Since current methodologies are highly specific for creatinine, the wide reference intervals largely stem from biologic differences in sCr among individuals. Serum creatinine has relatively high individuality in dogs and cats (Baral, et al., 2014; Ruaux, et al., 2011), meaning that variability between individuals is much higher than the variability observed within a single animal. Serum creatinine is influenced by age (Rosset, et al., 2012; Rørtveit, et al., 2015) and particularly by breed in dogs (Misbach, et al., 2014; Zaldívar-López, et al., 2011) and, to a lesser extent, in cats.20 It might also be influenced by sex and the veterinary clinic evaluating the patient.21 Therefore, sCr would benefit from age- and breed-specific reference intervals, ideally (although not practically) for every individual instrument and laboratory.


Trending of serum creatinine. Small increases in sCr even within the reference interval can reflect significant decreases in GFR in an individual patient8, particularly since variation in sCr within an individual healthy dog or cat is minimal over weeks to months and even years (Braun, et al., 2003; Baral, et al., 2014; Ruaux, et al., 2011). In fact, the critical difference or reference change value for detecting a significant increase or decrease in sCr is only 23–27 lM/L (0.3 mg/dL) in clinically healthy dogs10, and 17% (corresponding to similar absolute values as in dogs) in clinically healthy cats (Braun, et al., 2003). Thus, the sensitivity of sCr for detecting early kidney disease can be improved by evaluating serial fasted sCr measurements in an individual animal (trending) to look for increases that likely reflect worsening renal function. This concept of detecting small but clinically relevant increases in sCr is actively being adopted in cases of acute kidney injury (AKI), illustrated by the International Renal Interest Society (IRIS) Grading of AKI. In this grading scheme, an increase in sCr ≥ 26 lmol/L (0.3 mg/dL) within a 48- hour period is a criterion for identifying Grade I and Grade II AKI ( Furthermore, in adolescent dogs with XLHN, trending of sCr detected an average of 27% (range 5–49%) decrease in GFR (Nabity, et al., 2015). Despite heightened awareness of small, but clinically relevant increases in sCr over a short time frame, more recognition is needed with slowly progressive CKD, in which small increases might occur over many months or years.


Analytic challenges. Finally, sCr is plagued by inconsistencies in its measurement between instruments and laboratories, which can result in markedly different results. While most reference laboratory instruments have excellent precision and provide results of similar magnitude among instruments (Ulleberg, et al., 2011), recent studies illustrate the high imprecision and bias possible with some instruments and among different laboratories (Ulleberg, et al., 2011; Braun, et al., 2008). In normal to mildly azotaemia samples, one study using reference laboratories found differences of up to 40 lmol/L (0.45 mg/dL) in the same sample measured by the same laboratory and up to 50 lmol/L (0.57 mg/ dL) among different laboratories, even when excluding 2 laboratories (out of 10) with extreme outliers (Ulleberg, et al., 2011). Larger differences were noted in moderately to markedly azotaemia samples: excluding outlier results from 3 laboratories, up to a 68 lmol/L (0.77 mg/dL) difference was observed within the same laboratory, and 117 lmol/L (1.32 mg/dL) difference between laboratories (Ulleberg, et al., 2011). A still larger variation in sCr measurement was observed among 99 veterinary practices, where results ranged from 80 to 200 lmol/L (0.9–2.26 mg/ dL) for a single sample spiked with 112 lmol/L (1.25 mg/dL) creatinine (Braun, et al., 2008). These results highlight that many instruments/ laboratories are performing below the total allowable error (TEa) guideline for sCr (TEa ≤ 20%) set by the Quality Assurance and Laboratory Standards Committee of the American Society for Veterinary Clinical Pathology (ASVCP) (Harr, et al., 2013). Even with TEa ≤ 20%, analytic variability alone (instrument precision and bias) could account for an increase or decrease in sCr ≥ 18 mmol/ L (0.2 mg/dL) in normal to mildly azotaemia samples, which is approaching the critical difference for detecting a significant change in sCr in clinically healthy dogs and cats (Baral, et al., 2014). To minimize this analytic variability, it is particularly important that serial determinations of sCr are measured on a single instrument that is subjected to a strict quality assurance program.

In summary, sCr can be a sensitive marker of declining renal function if careful monitoring and/or appropriate reference intervals are used in animals with relatively stable muscle mass. However, because of both inherent biologic and current analytic limitations of sCr, additional endogenous markers of GFR would aid in the early diagnosis of renal disease. Two recently studied markers are cystatin C and symmetric dimethylarginine (SDMA). An excellent recent review of cystatin C is available (Ghys, et al., 2014). Therefore, only SDMA will be further discussed below.


Symmetric dimethylarginine (SDMA)

Symmetric dimethylarginine is a methylated amino acid (arginine) of similar size to creatinine (SDMA: 202 Daltons [Da]; creatinine: 113 Da). Symmetric dimethylarginine is formed by posttranslational methylation of arginine by type 2 protein arginine methyltransferases (Wei, et al., 2014), and it is released into circulation following proteolysis. Symmetric dimethylarginine was originally discovered 45 years ago in human urine (Kakimoto & Akazawa, 1970). Concentrations of Symmetric dimethylarginine in protein fractions from various organs in rats demonstrated SDMA to be highest in the brain and relatively high in the liver, lung, kidney, spleen, and small intestine as compared with heart, muscle, skin, and blood.29 The kidneys are the major source of SDMA excretion (Kakimoto & Akazawa, 1970; Nijveldt, et al., 2002; Nijveldt, et al., 2002), and SDMA does not appear to be reabsorbed by the tubules for reutilization (Kakimoto & Akazawa, 1970). Symmetric dimethylarginine was first shown to be increased in human patients with CKD over 20 years ago (Vallance, et al., 1992); however, SDMA has only recently gained prominence as an endogenous marker of GFR in people (Bode-Böger, et al., 2006), correlating strongly with renal function based on inulin clearance (Kielstein, et al., 2006). Symmetric dimethylarginine has largely been believed to be inert, reducing nitric oxide synthesis indirectly by competing with L-arginine uptake in cells (Bode-Böger, et al., 2006; Ei, et al., 1997). However, a recent study demonstrated a direct effect of SDMA, in which uncoupling of endothelial nitric oxide synthase resulted in superoxide anion production in glomerular endothelial cells (Feliers, et al., 2015). In addition, a recent review highlights several studies that support a proinflammatory role of SDMA (Schepers, et al., 2014). However, chronic infusion of SDMA in healthy mice had no evident effect on renal structure or function as well as cardiac function (Veldink, et al., 2013).


Measurement and stability. Mass spectrometry is the gold standard for SDMA measurement, as it uniquely and accurately detects the molecule (Schwedhelm, 2005). However, recent studies in people have used a commercially available ELISA with at least one study demonstrating good precision, although accuracy was not assessed (Tenderenda-Banasiuk, et al., 2013). An accurate and precise liquid chromatography–mass spectrometry (LC–MS/MS) assay has recently been developed for SDMA measurement in dogs and cats (Nabity, et al., 2015).


Reference limit. In clinically healthy animals, SDMA concentrations are similar across studies and species, despite differences in methodology and animal populations. Two studies in healthy human subjects determined the reference interval using LC–MS/MS as 0.32–0.65 lmol/L (6.5–13.1 lg/dL)46 and 0.225–0.533 lmol/L (4.5–10.8 lg/dL) (Schwedhelm, et al., 2011). Using serum from 120 healthy adult dogs of varying ages and breeds, a 95% reference interval for SDMA was calculated as 6–13 lg/dL (Rentko, et al., 2013), and in both dogs and cats, the upper reference limit using the LC–MS/MS methodology was set at < 14 lg/dl (0.69 lmol/L). Furthermore, a range of 7.3–12.4 lg/dL (0.36–0.61 lmol/L) was observed in 21 healthy geriatric cats in a recent study.43 However, in juvenile dogs, SDMA might occasionally reach or exceed the 14 lg/dL reference limit.8 In older studies using high-performance LC, SDMA in healthy dogs ranged from 0.22 to 0.61 lmol/L (4.4–12.3 lg/dL) (Moesgaard, et al., 2007; Moesgaard, et al., 2005; Tatematsu, et al., 2007), although values up to 14.1 lg/dL (0.7 lmol/L) were observed in a group consisting of both healthy dogs and dogs with mitral regurgitation (Pedersen, et al., 2006).


Renal disease in veterinary medicine. Several recently published studies of SDMA in dogs and cats have shown strong evidence for SDMA as an endogenous surrogate marker for GFR, particularly in animals with renal disease. In dogs, SDMA correlated strongly with inulin clearance in a partial nephrectomy model of CKD (r = -0.851)51 and with iohexol clearance using serial measurements in adolescent dogs with XLHN (r = -0.95) (Nabity, et al., 2015). In studies combining both healthy cats and cats with CKD, correlations of SDMA or 1/SDMA with iohexol clearance were r = -0.7943 and 0.8271, respectively. These were like correlations of sCr with clearance. Furthermore, in healthy geriatric cats, the correlation of SDMA with iohexol clearance was higher (r = -0.72) than that observed for sCr (r = -0.50) (Hall, et al., 2014). Symmetric dimethylarginine correlation with sCr was also strong in both dogs and cats when including those with kidney disease (r = 0.72–0.95) (Nabity, et al., 2015; Hall, et al., 2014) whereas in clinically healthy dogs and cats, the correlation between SDMA and sCr was much lower (r = 0.32–0.46) (Hall, et al., 2014; Pedersen, et al., 2006).

Importantly, SDMA appears to detect a decrease in GFR prior to sCr when based on reference limits in cats and dogs. In 21 geriatric laboratory cats with naturally occurring CKD, SDMA increased by the time azotaemia (sCr > 2.1 mg/dL) developed in all cats, and SDMA increased above its reference limit an average of 17 months earlier (up to 48 months earlier) than sCr (Hall, et al., 2014). Symmetric dimethylarginine also shows promise as a sensitive screening test for CKD in cats. Using iohexol clearance to estimate GFR and > 30% decrease from the median of clinically healthy controls as the gold standard for decreased GFR, SDMA had perfect sensitivity and negative predictive value (100%), indicating that cats with SDMA < 14 lg/dL did not have decreased GFR.43 Specificity and positive predictive value (PPV) were slightly lower (91% and 86%, respectively), due to 2 cases where SDMA was mildly increased with a GFR only 25% lower than the median (Hall, et al., 2014). However, this could indicate that SDMA can detect < 30% decrease in GFR in cats. In comparison, the upper reference limit of sCr provided perfect specificity and PPV (100%), indicating that cats with sCr above the reference interval (sCr > 2.1 mg/dL) all had > 30% decline in GFR. However, sensitivity was quite poor (17%), indicating that many cats with decreased GFR are undetected when using this reference limit (Hall, et al., 2014). Since specificity and sensitivity depend on the reference limit used, certainly use of a lower reference limit for sCr would have improved its sensitivity in this study, particularly since sCr in the clinically healthy cats was ≤ 1.6 mg/dL.


In dogs with rapidly progressive CKD due to XLHN, SDMA increased an average of 4–5 weeks prior to an increase in sCr and decrease in GFR, based on their respective reference limits (Nabity, et al., 2015). However, when trending both sCr and SDMA in individual dogs, SDMA increased an average of only 2 weeks prior to an increase in sCr (Nabity, et al., 2015). Notably, SDMA identified ≤ 34% (range, –6–34%) decrease in GFR when compared with the GFR of unaffected, age-matched dogs in this colony, regardless of whether the increase in SDMA was based on the reference limit, trending, or comparison with healthy controls. In some dogs, SDMA increased even before iohexol clearance was below that observed in unaffected, age-matched dogs (Nabity, et al., 2015). This contrasted with sCr, which detected a 5–68% decrease in GFR.

In summary, SDMA appears to be a useful endogenous marker of GFR and is particularly promising as a screening test for early detection of decreased renal function. Furthermore, because it is not influenced by lean body mass and is less variable among different dog breeds, SDMA could prove especially useful in those patients with poor muscle mass or ongoing muscle loss, where sCr would provide an unreliable estimate of GFR. Further studies are needed in veterinary medicine to determine possible extra-renal influences on SDMA concentration.



  1. Baral, R. et al., 2014. Biological variation and referencechange values of feline plasma biochemistry analytes. J Feline Med Surg, Volume 16, pp. 317-325.
  2. Bode-Böger, S. M. et al., 2006. Symmetrical dimethylarginine: a new combined parameter for renal function and extent of coronary artery disease.. Journal of the American Society of Nephrology : JASN, 17(4), p. 1128–1134.
  3. Bovee, K., Kronfeld, D., Ramberg, C. & Goldschmidt, M., 1979. Long-term measurement of renal function in partiallynephrectomized dogs fed 56, 27, or 19% protein. Invest Urol, Volume 16, pp. 378-384.
  4. Braun, J., Lefebvre, H. & Watson, A., 2003. Creatinine in the dog: a review. Vet Clin Pathol, 32(4), p. 162‐179.
  5. Braun, J. P. et al., 2008. Comparison of plasma creatinine values measured by different veterinary practices. The Veterinary record, 162(7), p. 215–216.
  6. Bricker, N., Klahr, S. & Rieselbach, R., 1964. The functional adaptation of the diseased kidney I. Glomerular filtration rate. J Clin Invest, Volume 43, pp. 1915-1921.
  7. Brown, S. et al., 1990. Single-nephron adaptations to partial renal abla-tion in the dog. Am J Physiol, Volume 258, pp. 495-503.
  8. Ei, C., Basha, F., Habermeier, A. & F€orstermann, U., 1997. Interference of L-arginine analogues with L-argininetransport mediated by the y+carrier hCAT-2B. Nitric Oxide, Volume 1, pp. 65-73.
  9. Feliers, D., Lee, D., Gorin, Y. & Kasinath, B., 2015. Symmetricdimethylarginine alters endothelial nitric oxide activity in glomerular endothelial cells. Cell Signal, Volume 27, p. 1–5.
  10. Ghys, L. et al., 2014. Cystatin C: a new renal marker and its potential use in small animal medicine. Journal of veterinary internal medicine, 28(4), p. 1152–1164.
  11. Hall, J. A. et al., 2014. Comparison of serum concentrations of symmetric dimethylarginine and creatinine as kidney function biomarkers in cats with chronic kidney disease. Journal of veterinary internal medicine, 28(6), p. 1676–1683.
  12. Harr, K. E. et al., 2013. ASVCP guidelines: allowable total error guidelines for biochemistry. Veterinary clinical pathology, 42(4), p. 424–436.
  13. Kakimoto, Y. & Akazawa, S., 1970. Isolation and identification ofN-G, N-G- and N-G, N’-G-dimethyl-arginine, N-epsi-lon-mono-, di-, and trimethyllysine, and glucosyl-galactosyl- and galactosyl-delta-hydroxylysine fromhuman urine.. J Biol Chem., 245(57), p. 51–58.
  14. Kielstein, J. T. et al., 2006. Symmetric dimethylarginine (SDMA) as endogenous marker of renal function–a meta-analysis. Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association – European Renal Association, 21(9), p. 2446–2451.
  15. Misbach, C. et al., 2014. Basal plasma concentrations of routine variables and packed cell volume in clinically healthy adult small-sized dogs: effect of breed, body weight, age, and gender, and establishment of reference intervals.. Veterinary clinical pathology, 43(3), p. 371–380.
  16. Moesgaard, S., Holte, A. & Mogensen, T., 2007. Effects ofbreed, gender, exercise and white-coat effect on markers of endothelial function in dogs. Res Vet Sci., Volume 82, pp. 409-415.
  17. Moesgaard, S. et al., 2005. Neurohormonal and circulatory effectsof short-term treatment with enalapril and quinaprilin dogs with asymptomatic mitral regurgitation. J VetIntern Med., Volume 19, pp. 712-719.
  18. Nabity, M. B. et al., 2015. Symmetric Dimethylarginine Assay Validation, Stability, and Evaluation as a Marker for the Early Detection of Chronic Kidney Disease in Dogs. Journal of veterinary internal medicine, 29(4), p. 1036–1044.
  19. Nijveldt, R. J. et al., 2002. Handling of asymmetrical dimethylarginine and symmetrical dimethylarginine by the rat kidney under basal conditions and during endotoxaemia. Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association – European Renal Association, 18(12), p. 2542–2550.
  20. Nijveldt, R. J. et al., 2002. Net renal extraction of asymmetrical (ADMA) and symmetrical (SDMA) dimethylarginine in fasting humans. Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association – European Renal Association, 17(11), p. 1999–2002.
  21. Patch, D., Obare, E. & Xie, H., 2015. High throughput immunoas-say that correlates to gold standard liquid chromatog-raphy mass spectrometry (LC-MS) assay for thechronic kidney disease (CKD) marker symmetricdimethylarginine (SDMA). J Vet Intern Med, Volume 29, p. 1216.
  22. Pedersen, L. et al., 2006. Body size, but neither age nor asymptomaticmitral regurgitation, influences plasma concentrationsof dimethylarginines in dogs.. Res Vet Sci., Volume 80, pp. 336-342.
  23. Rentko, V., Nabity, M. & Yerramilli, M., 2013. Determination ofserum symmetric dimethylarginine reference limit inclinically healthy dogs. J Vet Intern Med, Volume 27, p. 750.
  24. Rørtveit, R. et al., 2015. Age-related changes in hematologic and serum biochemical variables in dogs aged 16-60 days. Veterinary clinical pathology, 44(1), p. 47–57.
  25. Rosset, E., Rannou, B., Casseleux, G. & Chalvet-Monfray, K., 2012. Age-related changes in biochemical and hema-tologic variables in Borzoi and Beagle puppies frombirth to 8 weeks. Vet Clin Pathol, Volume 41, pp. 277-282.
  26. Ruaux, C., Carney, P., Suchodolski, J. & Steiner, J., 2011. Estimates of biological variation in routinely measuredbiochemical analytes in clinically healthy dogs. Vet Clin Pathol, Volume 41, pp. 551-547.
  27. Schepers, E. et al., 2014. Dimethylarginines ADMA and SDMA: thereal water-soluble small toxins?. Semin Nephrol, Volume 34, p. 97–105.
  28. Schwedhelm, E., 2005. Quantification of ADMA: analytical approaches. Vasc Med, 10(Suppl 1), p. S89–S95.
  29. Schwedhelm, E., Xanthakis, V. & Maas, R., 2011. Plasmasymmetric dimethylarginine reference limits from theFramingham offspring cohort. Clin Chem Lab Med, Volume 49, p. 1907–1910.
  30. Tatematsu, S., Wakino, S. & Kanda, T., 2007. Role of nitricoxide-producing and -degrading pathways in coronaryendothelial dysfunction in chronic kidney disease. JAm Soc Nephrol, Volume 18, pp. 741-749.
  31. Tenderenda-Banasiuk, E., Wasilewska, A., Taranta-Janusz, K. & Korzeniecka-Kozerska, A., 2013. Asymmetric andsymmetric dimethylarginine in adolescents with hyperuricemia. Dis Markers, Volume 35, p. 407–412.
  32. Ulleberg, T. et al., 2011. Plasma creatinine in dogs: intra- and inter-laboratory variation in 10 European veterinary laboratories.. Acta veterinaria Scandinavica, 53(1), p. 25.
  33. Vallance, P. et al., 1992. Accumulation of an endogenous inhibitor of nitricoxide synthesis in chronic renal failure. Lancet, 339(5), p. 72–75.
  34. Veldink, H. et al., 2013. Effects of chronic SDMA infusion on glomerular filtration rate, blood pressure, myocardial function and renal histology in C57BL6/J mice. Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association – European Renal Association, 28(6), p. 1434–1439.
  35. Wei, H., Mundade, R., Lange, K. C. & Lu, T., 2014. Protein arginine methylation of non-histone proteins and its role in diseases.. Cell cycle, 13(1), p. 32–41.
  36. Zaldívar-López, S. et al., 2011. Clinical pathology of Greyhounds and other sighthounds. Veterinary clinical pathology, 40(4), p. 414–425.