From KSAP
A 67-year-old man is referred for decreased kidney function. He was recently diagnosed with squamous cell lung cancer of the head and neck, and cisplatin-based chemotherapy is planned.
His medical history is notable only for mild hypertension and a distant history of tobacco use. He works with a trainer three times a week and usually cycles on the weekends.
He takes chlorthalidone and aspirin.
On physical examination, he appears physically fit. His weight is 75 kg, his body mass index (BMI) is 25 kg/m2, his blood pressure is 134/82 mm Hg, and his heart rate is 78/min. The physical examination is unremarkable.
Laboratory data:
| Result | Repeat | Reference range | |
|---|---|---|---|
| Sodium | 142 mEq/L | 136-145 | |
| Potassium | 3.4 mEq/L | 3.5-5.0 | |
| Total CO2 | 28 mEq/L | 23-30 | |
| BUN | 28 mg/dL | 8-20 | |
| Creatinine | 1.45 mg/dL | 1.1 | 0.7-1.3 |
| eGFR (MDRD) | 48.5 ml/min/1.73 m2 | 71 | >90 |
| randomn urine protein/creatinine ratio | 0.2 | 0-0.2 | |
| 24 hour urine collection | |||
| volume | 750 mL | ||
| urine creatinine | 100 mg/dL | ||
| Measured Creatinine Clearance | 47.4 mL/min |
Urinalysis shows pH 5, trace blood, trace protein, specific gravity 1.030. Microscopy shows 0–2 isomorphic red blood cells/high-power field.
What is the next MOST appropriate step in the assessment of this patient’s renal function to guide dosing of cisplatin?
A Estimate the GFR with the CKD Epidemiology Collaboration (CKD-EPI) equation
B Calculate creatinine clearance with the Cockcroft-Gault equation
C Calculate the reciprocal of the creatinine
D Repeat the 24-hour urine collection
E Estimate the GFR utilizing cystatin C
Answer & Explanation
D. Repeat the 24-hour urine collection
The most appropriate next step in the assessment of this patient’s kidney function is to repeat the 24-hour urine collection. This patient’s urine collection is likely an “under collection” because there is only a total of 750 mg creatinine measured in the 24-hour period as calculated below:
Urine volume × urine creatinine concentration = 24-hour urine creatinine:
750 mL × 100 mg/dL creatinine × 1 dL/100mL = 750 mg of creatinine
On average, men excrete approximately 20–25 mg of creatinine/kg lean body weight, although this value declines with increasing age and the associated fall in muscle mass. This patient’s collection contains only 10 mg/kg of creatinine, which is low for a man and particularly low for a physically active man. An incomplete collection of urine can lead to an underestimation of kidney function. This is more likely to occur when the urine volume is low, because small errors in timing can have considerable ramifications for the interpretation of the collection.
In general, an estimate of kidney function is sufficient for the care of most patients, but there are certain circumstances when a more precise assessment of glomerular filtration rate (GFR) is important. This includes assessment of kidney function for dosing medications, such as chemotherapeutic agents that are renally cleared or nephrotoxic. In this case, the bland urine sediment and the improvement in the serum creatinine without intervention indicates that a hemodynamic effect rather than parenchymal kidney disease might explain the initial abnormally elevated serum creatinine. The high urine specific gravity at the time of the initial laboratory testing suggests that mild volume depletion may have been present in this patient who takes a thiazide diuretic. Indeed, it is worth noting that small changes in the serum creatinine when the creatinine is low have considerable effects on the estimated GFR regardless of the formula used; thus, the formulas tend to perform relatively poorly when kidney function is close to normal. Even if the reduced measured creatinine clearance could be explained, the measurement should be repeated as noted above.
Cystatin C is an endogenous low molecular weight protein that is filtered but not secreted, as it is normally metabolized by the proximal tubules. Cystatin C has been touted as a better marker for assessment of kidney function than creatinine because it was initially thought to be unaffected by age, sex, or muscle mass. Unfortunately, it is now clear that cystatin C is affected by many factors. In addition, assays for cystatin C exhibit considerable variation even for the same patient and despite laboratory standardization. Although there is some evidence that it may be a useful guide to drug-dosing compared to creatinine, its role in oncology patients is even less certain. Fluctuations in cystatin C have been documented both before and after chemotherapy that do not necessarily correlate with GFR. Therefore, use of the cystatin C to guide cisplatin dosing is not preferred. Utilizing both cystatin C and creatinine in GFR estimation equations improves the accuracy in test populations but the precision in oncology patients has not been established.
The Cockcroft-Gault formula (CG), published in 1972, was based on 249 pairs of measured creatinine clearances in men. This formula uses weight and age as variables, assuming that higher weight correlates with more muscle mass and those with greater age have less muscle and therefore less creatinine production. Despite these assumptions and the potential errors that result, the CG formula is the formula that was used for most drug dosing guidelines. Calculating creatinine clearance using the CG formula would not provide a better assessment of kidney function in this patient.
Because kidney function is proportional to the inverse of the serum creatinine, the reciprocal of the serum creatinine has been used to estimate the rate of progression of CKD by plotting this fraction versus time. The reciprocal of the creatinine, however, is relatively inaccurate at specifically quantifying kidney function.
The CKD-EPI formula was designed to provide a more accurate estimate of GFR when kidney function is close to normal. This formula was developed by pooling data from nearly 30 studies (10 served as the equation development data set and were used to create the new equation and an additional 16 served as the validation data set). This formula would also yield significantly different results with the two different creatinine values; when the creatinine is 1.45 mg/dL, the CKD-EPI equation estimates the GFR at 49.5 mL/min per 1.73 m2, whereas when the creatinine is 1.2 mg/dL, the CKD-EPI equation estimates the GFR at 62.2 mL/min per 1.73 m2. Both values are very close to the estimates provided by the MDRD equation.
Although fraught with collection challenges, an accurately collected 24-hour urine under steady state conditions provides a reasonable assessment of creatinine clearance, which has been the basis for cisplatin dosing.
References
Jones M, Denieffe S, Griffin C, Tinago W, Fitzgibbon MC: Evaluation of cystatin C in malignancy and comparability of estimates of GFR in oncology patients. Pract Lab Med. 8:95–104, 2017
Barreto EF, Rule AD, Murad MH, Kashani KB, Lieske JC, Erwin PJ, Steckelberg JM, Gajic O, Reid JM, Kane-Gill SL: Prediction of the Renal Elimination of Drugs With Cystatin C vs Creatinine: A Systematic Review. Mayo Clinic Proc 94(3): 500–515, 2019
Mitch WE, Walser M, Buffington GA, Lemann J: A simple method of estimating progression of chronic renal failure. Lancet 2(7999): 1326–1328, 1976
Stevens LA, Levey AS: Measured GFR as a confirmatory test for estimated GFR. J Am Soc Nephrol 20(11): 2305–2313, 2009