A 3-month-old boy was seen for routine follow-up at the pediatric nephrology outpatient clinic. He had been diagnosed as having Sotos syndrome manifesting with craniofacial dysmorphism, feeding difficulties, pulmonary artery stenosis, and atrial septal defect, as well as complex urological abnormalities. He had bilateral hydronephrosis with megaureter and grade V vesicoureteral reflux to the left and grade I to the right kidney. At the age of 6 weeks, static renal scintigraphy using DMSA (99mTc-dimercaptosuccinic acid) to assess renal morphology, structure, and function had demonstrated almost symmetrical kidney function (split kidney function left 44% vs right 56%) without cortical scarring.
While his baseline serum creatinine had been 40 μmol/L (0.45 mg/dL), a sudden rise to 69 μmol/L (0.79 mg/dL) was noted. Urinary tract infection was ruled out, as was dehydration. On renal ultrasound, dilation of the right collecting system and ureter had increased significantly and a novel fluid collection at the upper pole was noted, which prompted an MRI study (Fig. 1). In addition to serum creatinine, cystatin C measurement was ordered and was within the reference interval for age (1.13 mg/L).
Creatinine, the most commonly used endogenous marker of glomerular filtration rate (GFR),4 has in recent years been supplemented by cystatin C. In fact, today's most reliable equations for estimating GFR both in adults (1) and in children (2) make use of a combination of the 2. Still, in certain conditions the 2 markers can give very discordant findings, which may be diagnostic, as in the case presented here. Discordant findings may occur via 5 mechanisms, i.e., production, distribution, nonglomerular elimination, elimination by dialysis, and recirculation.
QUESTIONS TO CONSIDER
Does the rise in creatinine indicate deterioration of kidney function?
How do you explain the discrepancy between the 2 markers of kidney function?
What is the nature of the fluid collection at the right upper pole?
What test could be used to determine the nature of the fluid collection?
Serum creatinine (molecular weight 113 Da) is derived from muscle metabolism and strongly influenced by muscle mass, diet, age, and sex. By contrast, cystatin C, a low molecular weight protein (13.3 kDa), is produced by all nucleated cells at a constant rate, making it particularly useful in children and patients with body mass outside the reference interval (3). Therefore, low creatinine values in the presence of cystatin C concentrations at or above the reference interval are suggestive of muscle wasting, while the opposite may indicate very high muscle mass or excessive intake of cooked meat, fish, or creatine supplements (e.g., body builders) (4). Cystatin C is not an acute-phase protein. However, cystatin C production is influenced by disturbances in thyroid metabolism. Untreated hypothyroidism leads to decreased cystatin C concentrations, while the opposite is observed in hyperthyroidism (3). High-dose corticosteroids stimulate cystatin C production (3). Neither thyroid disease (5) nor corticosteroid therapy (Bökenkamp, unpublished data) affect creatinine metabolism.
There are large differences in the volume of distribution between creatinine and cystatin C. While creatinine equilibrates in total body water, cystatin C is restricted to the extracellular space. This is reflected in more rapid changes in serum cystatin C as compared to creatinine during acute changes in kidney function (6).
Both creatinine and cystatin C are eliminated almost exclusively via glomerular filtration. However, creatinine also undergoes renal tubular secretion. This process increases with declining glomerular filtration leading to the so-called “creatinine-blind range” between 50 and 90 mL/min/1.73 m2, in which moderately decreased glomerular filtration is concealed by tubular secretion (7). This is particularly worrisome in patients with low creatinine production (4). Although cystatin C is predominantly excreted via the kidney, there is some hepatic elimination accounting for about 5% at normal GFR (6). The relative contribution of extrarenal elimination increases with declining GFR and underlies the observation that cystatin C concentrations do not exceed 10 mg/L even in anuric patients (8). Thus, compared to creatinine, the rise in cystatin C is stronger with mildly to moderately impaired renal function (chronic kidney disease stage 2–3), whereas the opposite applies to severe and end-stage renal failure (chronic kidney disease stage 4–5).
Although there is considerable clearance of creatinine by dialysis, elimination of cystatin C by conventional hemodialysis and peritoneal dialysis is poor owing to its larger molecular weight (creatinine 113 Da, cystatin C 13250 Da). Only hemodiafiltration using convective elimination through large-pore dialysis membranes allows for significant elimination of low molecular weight proteins, which can be used for the assessment of residual renal function in dialysis patients (8).
A special situation is recirculation of urine due to leakage of urine into the abdominal cavity or the perirenal space. Being a low molecular weight protein, cystatin C is reabsorbed and metabolized in the proximal tubule. Therefore, unlike creatinine, cystatin C concentrations in the urine are negligible (9). Therefore, recirculation of creatinine (and urea as well) may falsely suggest renal failure while cystatin C concentrations remain within the reference interval. This constellation is an important clue to suspect a urinoma, i.e., extravasation of urine. Schreuder et al. reported a similar course in a child after renal transplantation (10), where the discrepancy between creatinine-based and cystatin C–based estimated GFR (eGFR) led to the diagnosis of urine leakage. If the fluid collection is easily accessible for needle aspiration, measurement of creatinine in the fluid can be used for differentiation. Still, this method does not answer the question if kidney function has been affected by urinary obstruction.
The patient presented herein had developed complete obstruction of his right ureter, which had prolapsed into an inguinal hernia. Acutely increased intrapelvic pressure had caused rupture of the renal calyces and extravasation of urine around the upper pole (11), which was visualized on MRI. The patient went for surgery to repair the hernia and perform a ureteral reimplantation. On follow up, the dilation improved considerably and the urinoma disappeared without surgical drainage. Creatinine normalized rapidly while cystatin C concentrations remained unchanged confirming recirculation of creatinine as the cause of the discrepancy between the 2 markers (Fig. 2).
As the combination of creatinine and cystatin C in a single estimated GFR (eGFR) equation has proven superior to GFR estimation based on the individual parameters in large cohorts (1, 2), some laboratories have adopted a policy of reporting only the combined eGFR. This method is not without risk as it obscures situations in which the discrepancy between both markers is diagnostic. Therefore, Grubb et al. advocate reporting cystatin C and creatinine-based eGFR separately, which will alert clinicians to start further investigations (12).
POINTS TO REMEMBER
Discrepant results of serum creatinine and cystatin C reflect differences in production, distribution, nonglomerular elimination, elimination by dialysis, and recirculation.
Low creatinine in the setting of increased cystatin C is observed with (a) stage 2 chronic kidney disease, (b) low muscle mass, (c) increased cystatin C production (high-dose corticosteroid therapy, untreated hyperthyroidism), or (d) creatinine elimination by dialysis.
Increased creatinine in the setting of cystatin C concentrations within reference intervals is observed with (a) increased muscle mass/creatine supplements, (b) decreased cystatin C production (untreated hypothyroidism), or (c) recirculation of urinary creatinine.
eGFR equations based on the combination of cystatin C and creatinine obscure a discrepancy between the 2 markers, which may be diagnostic.
↵4 Nonstandard abbreviations:
- glomerular filtration rate;
- estimated GFR.
Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.
Authors' Disclosures or Potential Conflicts of Interest: No authors declared any potential conflicts of interest.
- Received for publication June 6, 2016.
- Accepted for publication September 8, 2016.
- © 2016 American Association for Clinical Chemistry