Is GFR Calculation Reliable?
GFR calculation provides a reasonable initial assessment of kidney function for most clinical purposes, but it is fundamentally imprecise and subject to significant limitations that vary by formula, patient population, and clinical context. 1, 2
Understanding the Fundamental Limitations
No GFR estimation formula is sufficiently accurate and precise compared to measured GFR, and all are subject to multiple non-GFR determinants that compromise reliability. 2 The key issue is that these formulas estimate rather than measure kidney function, introducing inherent error that must be understood for proper clinical application.
General Performance Characteristics
eGFR based on serum creatinine (eGFRcr) is recommended as the initial test for routine kidney function evaluation because it is inexpensive, widely available, and easily repeatable. 1, 2
The 2021 CKD-EPI creatinine equation is currently the most accurate for general clinical use when evaluated in large diverse populations, and should replace older MDRD equations for routine reporting. 3
Accuracy is particularly poor when GFR is >60 mL/min/1.73 m², where formulas tend to underestimate true kidney function. 4
Specific Clinical Situations Where Reliability Breaks Down
Patient Characteristics That Compromise Accuracy
Extremes of muscle mass (very high or very low) alter creatinine generation independent of kidney function, making eGFRcr unreliable. 2, 4 This includes:
- Severe malnutrition or muscle wasting conditions 2
- Advanced cirrhosis or cancer with high cell turnover (high catabolism/inflammation) 2
- Class III obesity (BMI >40 kg/m²) reduces accuracy 2
- Very elderly patients where age-related muscle loss confounds interpretation 4
Dietary and Metabolic Factors
High protein or meat intake affects creatinine levels and requires consideration when interpreting results. 2
Recent or rapid changes in GFR reduce formula accuracy because steady-state creatinine levels have not been achieved. 4
Formula-Specific Reliability Issues
Cockcroft-Gault Formula
The C-G formula consistently underestimates GFR in patients with normal to moderately reduced renal function and overestimates in those with significantly impaired function. 5
- In elderly patients (≥65 years), the underestimation is enhanced, making it less reliable than MDRD in this population. 5
- The formula was derived only in men and assumes a 15% reduction for women without validation. 5
- It is not reliable for obese or edematous patients. 5
MDRD Formula
The MDRD formula may be the estimation of choice in elderly patients (≥65 years) where it blunts the underestimation seen with C-G. 5
- However, both C-G and MDRD formulas failed to reach K/DOQI accuracy standards when compared to inulin clearance, with 29-32% of subjects misclassified. 5
- No reliable estimation can be obtained with either formula when obesity is present. 5
Wright and Martin Formulas
The Wright formula shows improved precision primarily in patients with "normal" renal function (GFR 50-100 mL/min). 5
- It has significant positive bias for low GFR (overestimates) and negative bias for high GFR (underestimates). 5
- It cannot be recommended to provide reliable estimates across the full range of renal function. 5
Laboratory Method Impact on Reliability
The creatinine assay method significantly affects predictive accuracy. 5
Jaffe method overestimates serum creatinine by 5-15% due to chromogen interference, but this paradoxically may neutralize the overestimation from tubular creatinine secretion. 5
Enzymatic PAP method is more specific but yields results closer to actual creatinine levels, which are higher than true GFR and may lead to dose overestimation. 5
Formulas must be adjusted based on the assay method used - for PAP method, add 0.2 mg/dL to serum creatinine for C-G or Calvert formulas. 5
Laboratories must use standardized assays with calibration traceable to international standards (coefficient of variation <2.3%, bias <3.7%). 1
When to Use Alternative Methods
Cystatin C-Based Equations
Measure cystatin C and calculate eGFRcr-cys (combined creatinine-cystatin C equation) when eGFRcr is expected to be inaccurate and GFR affects clinical decision-making. 2
The 2012 CKD-EPI cystatin C equation is as accurate as the creatinine equation, does not require race specification, and may be more accurate in patients with decreased muscle mass. 3
The 2012 CKD-EPI creatinine-cystatin C equation is more accurate than either single-marker equation and serves as a useful confirmatory test. 3
Measured GFR
Use measured GFR with exogenous filtration markers (51Cr-EDTA, 125I-iothalamate, inulin clearance) for critical clinical decisions requiring precise GFR, such as kidney-cleared chemotherapy dosing. 2, 6
Measured creatinine clearance via timed urine collection can be considered if measured GFR is unavailable, though it requires lengthy collection and is prone to collection errors. 5, 2
24-hour creatinine clearance offers no increased precision over calculated estimates despite requiring substantially more time and effort. 7
Critical Pitfalls to Avoid
Never rely on serum creatinine alone - approximately 60% of cancer patients had abnormal renal function by eGFR but only 5% by serum creatinine alone. 5
Do not assume eGFRcr accuracy when assessing GFR changes over time without considering potential sources of error. 2
Ignoring clinical context suggesting altered creatinine generation (extremes of muscle mass, dietary intake, catabolic states) leads to inaccurate interpretation. 2
For medication dosing with renally-cleared drugs, adjust the standardized eGFR to the patient's actual body surface area by multiplying: eGFR × (patient's BSA / 1.73). 1