Impact factor 0,92
Ugeskrift for Læger
Ugeskrift for læger
  Print
Original article
Estimation of kidney function in cancer patients


Belén Redal-Baigorri1, Knud Heine Stokholm2, Knud Rasmussen1 & Nina Jeppesen3

1) Department of Medicine, Division of Nephrology, Roskilde Hospital,
2) Department of Clinical Physiology and Nuclear Medicine, Køge Hospital, and
3) Department of Oncology and Haematology, Roskilde Hospital


PDF

Abstract

Introduction: The Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formula has not been validated in patients with cancer. The present investigation was undertaken in order to study how well estimated glomerular filtration rate (eGFR) using the new CKD-EPI equation correlates with measured GFR (mGFR) by 51chrome ethylene diamine tetraacetic acid (51Cr-EDTA) clearance in a group of patients with cancer not known as having chronic kidney disease.

Material and methods: We investigated 185 patients with cancer who were referred for isotope measurement of GFR with 51Cr-EDTA before initiating chemotherapy treatment. The agreement between CKD-EPI and 51Cr-EDTA was assessed using a Bland-Altman plot. Test performance was analysed in a contingency table and bias, precision and the percentage of estimates within 30% of the mGFR (P30) were assessed.

Results: Bland-Altman plot analysis showed a limit of agreement in the range from −25.59 to 27.92 ml/min./ 1.73 m2. This formula was therefore not interchangeable with 51Cr-EDTA, as the above differences are of clinical importance. Bias was low: 1.16 ml/min./1.73 m2; P30 was high: 89.73%; and precision was 13.37 ml/min./1.73 m2. As a screening test, the CKD-EPI had a high specificity of 98% (95% confidence interval (CI): 96 to 100%) and a high negative predictive value 97% (95% CI: 95 to 100%). The accuracy of the validation test was 96% (95% CI: 93 to 99%).

Conclusion: The CKD-EPI may be used as a screening tool for CKD in the general population, but cannot replace isotope tests when a high GFR measurement accuracy is needed.

Chronic kidney disease (CKD) is a major public health issue. In oncology, it is mandatory to ascertain whether a patient has CKD before considering treatment with platinum-based compounds due to the risk of developing nephrotoxicity [1].

In the initial work-up, the patient’s glomerular filtration rate (GFR) should be measured, but GFR measurements are cumbersome and therefore not easily performed in clinical practice. Inulin clearance is the gold standard, but it is expensive and it is difficult and time-consuming to measure inulin clearance regardless of which technique is used [2].

Novel methods using radioactive tracers such as 51chrome ethylene diamine tetraacetic acid (51Cr-EDTA) [3, 4] and radio contrast agents such as iohexol [5] are also time-consuming since patients have to wait three to five hours or even 24 hours after injection of the tracer before blood samples can be drawn.

Because of these obstacles, prediction formulas based on serum creatinine have emerged.

One of the earliest such formulas is the Cockcroft and Gault formula [6] which estimates creatinine clearance.

In 1999, Levey et al [7] proposed the Modification of Diet in Renal Disease (MDRD) formula for estimation of the GFR (eGFR). The formula is based on 1,628 patients with a mean GFR of 39.8 ml/min./1.73 m2. Its performance has been validated for GFRs below 60 ml/min./1.73 m2, but the validity of the formula has been questioned in patients with a normal or near normal GFR [8].

In 2009 a new equation was presented: Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI), which is based on a sample size of 8,254 patients with a median measured GFR (mGFR) of 68 ± 40 ml/min./ 1.73 m2 [9].

Whether GFR estimation with creatinine-based methods such as MDRD or GFR may replace the GFR measurement has been widely investigated in a range of patient groups, but studies addressing this issue in oncology are lacking. The present study was undertaken to explore, using the new CKD-EPI equation, how well eGFR correlates with mGFR by 51Cr-EDTA in a group of patients not previously diagnosed with severe CKD and newly diagnosed with cancer. Furthermore, the validity of the CKD-EPI formula as a screening tool was evaluated.

Our Department of Oncology refers about 150-200 patients per year to GFR measurement by 51Cr-EDTA prior to platinum-based drug treatment. mGFR ex-pressed in absolute values is then used to calculate the carboplatin dose and to determine if there is indication for cisplatin treatment. If mGFR is below 50 ml/min, cisplatin is not administered.





MATERIAL AND METHODS
Study design

The study is a retrospective analysis based on data from cancer patients collected before they started chemotherapy with cisplatin or carboplatin. The study was approved by the Danish Data Protection Agency (j. no. 2007-41-1006). The study was conducted in agreement with the principles of the Declaration of Helsinki.

Participants

A total of 362 patients were referred for determination of GFR by 51Cr-EDTA in the period from 1 January 2006 to 31 December 2007. All patients were examined at the same laboratory.

The inclusion criteria were: Age ≥ 18 years; a minimum of two serum creatinine values with less than 15% variation in range within a period of three months and a serum creatinine value measured a maximum of three weeks before the 51Cr-EDTA measurement. We only used the first value of 51Cr-EDTA in those cases where this test was performed more than once.

Exclusion criteria: Only one creatinine value determination; acute renal insufficiency; diabetic ketoacidosis; increased values of protein (plasma (P) protein > 90 g/l), glucose (P glucose > 17 mmol/l), bilirubin (P bilirubin > 65 micromol/l) or uric acid (P uric acid > 0.45 mmol/l); renal replacement therapy; pregnancy; amputation; treatment with cephalosporins, cimetidine, methyldopa or trimethoprim.

The study population had not been diagnosed with end-stage renal disease (ESRD), as renal replacement therapy was an exclusion criterion. None of the patients were undergoing treatment with erythropoietin (EPO) or active D vitamin at the time of inclusion.

A total of 189 consecutive patients fulfilled the inclusion criteria. Four were excluded due to a high serum level of uric acid, leaving 185 included patients for analysis.

Estimated glomerular filtration rate
Analysis of serum creatinine

Serum creatinine was measured with the Jaffé method (Abbott Architect C systems TMc8000, reagent 7D64). The Jaffé method was calibrated with Isotope Dilution Mass Spectrometry (IDMS) [10].

Calculation of estimated glomerular filtration rate

The MDRD and CKD-EPI formulas have previously been validated in patients with CKD stage 3 or lower-stage CKD, defined as GFR ≤ 59 ml/min./1.73 m2. Validation was set at this level because higher values are associated with increased imprecision. Even so, these formulas cannot be used alone to define CKD when eGFR is above 60 ml/min./1.73 m2 (CKD stages 1-2). In such cases, to establish CKD, other symptoms of kidney disease such as proteinuria or haematuria must be present alongside with eGFR.

CKD-EPI equation: eGFR = 141 × min (Scr/k, 1)a × max (Scr/k, 1)-1.209 × 0.993Age × 1.018 [if female] × 1.159 [if black].

Where Scr is serum creatinine, k is 0.7 for females and 0.9 for males, a is −0.329 for females and −0.411 for males, min indicates the minimum of Scr/k or 1 and max indicates the maximum of Scr/k or 1.

Modification of diet in renal disease formula (14) based on serum creatinine (Sc), age, gender and race:

eGFR (ml/min./1.73 m2 = 175 × (Sc/88.4)-1.154 × age -0.203 × 0.742 [if female] × 1.21 [if black].

Where Sc is in micromol/l; 88.4 is the molecular weight of Sc and age is stated in years.

Body surface area (BSA) was calculated from the DuBois & DuBois formula [11] with height in centimetres rather than metres, as BSA (m2) = 0.007184 × height (cm) 0.725 × weight (kg)0.425.

Glomerular filtration rate

GFR was determined from the total (renal and extra-renal) 51Cr-EDTA plasma clearance (ClEDTA) by a simpli- fied single-injection technique with a single-plasma sample [4]:

(ClEDTA) = −ln (C(t) × ECV/Q0) × ECV / (t × g)

Where C(t) are the counts per minute per ml plasma of 51Cr-EDTA at time t; ECV is the extracellular volume; Q0 is the total amount (counts per minute) of injected 51Cr-EDTA; g is a correcting factor for the difference between the actual t and the theoretical mean sojourn time of 51Cr-EDTA.

ECV = 10,800 × BSA - 5,578.6

g = 0.324 × e-0.0121 × t + 1.13 × e-0.000289 × t

The total 51Cr-EDTA plasma clearance was corrected for gender differences in plasma volume (PV) according to Brøchner-Mortensen [3]:

ClEDTA-PV = ClEDTA × (0.00002512 × PV + 0.9246)

where PV is determined by multiplying body weight with 41 in females and with 45 in males. The intravenous injection of 3.7 MBq 51Cr-EDTA was given in the morning and a blood sample was drawn after three to four hours.

Statistics

A Bland-Altman plot [12] was used to show the individual differences between eGFR and mGFR. Results are shown as mean ± standard deviation (SD). p values < 0.05 were considered significant.

The validity of eGFR as a screening test for GFR was calculated from 2 × 2 contingency tables. Overall agreement in terms of screening was expressed as accuracy, defined as the sum of true positive and true negative findings in percentages of the total number of patients [13].

Confidence intervals (CI) were calculated as Gaussian approximations for single proportions [14].

Bias, precision and accuracy of eGFR with CKD-EPI equation: Bias was defined as the mean difference between the eGFR (CKD-EPI) and mGFR (51Cr-EDTA), whereas precision was expressed as the SD of the mean difference between the eGFR (CKD-EPI) and the mGFR (51Cr-EDTA), or the SD of the bias. A large width equals a low precision. Both precision and bias were expressed as ml/min./1.73 m2.

Accuracy was defined as the percentage of patients who had an estimated kidney function within 30% limits of the mGFR (51Cr-EDTA).

RESULTS
Patient characteristics

As shown in Table 1, the mean patient age exceeded 60 years, and there was a higher proportion of females than males. The mean mGFR was 85.12 ± 20.31 ml/min./1.73 m2.

The types of cancer diagnosed in this population were lung cancer (77%), ovarian cancer (15%) and cancer of the testis (4%).

Validation of Chronic Kidney Disease Epidemiology
Collaboration as a laboratory parameter

The Bland-Altman plot is recommended for comparison of two methods with a view to ascertaining whether they agree sufficiently to allow one to replace the other. The present article therefore does not offer correlation coefficients which would be misleading [12].

The mean difference between eGFR and mGFR was 1.16 ml/min./1.73 m2 with a 95% confidence interval (CI) of −0.76 to 3.09.

The limits of agreement are defined as mean ± 2 SD: 27.92 (95% CI: 24.56 to 31.24) and −25.59 (95% CI: −22.24 to −28.92) ml/min./1.73 m2. This shows that the interval of both limits of agreement is not clinically acceptable because the variation in differences may indicate discrepancies between both methods. This may be of clinical relevance, especially in those cases where GFR is below 85ml/min./1.73 m2 as shown in the Bland-Altman plot (Figure 1).

Bias, precision and accuracy results of eGFR with CKD-EPI equation (Table 2): The present study reports a mean measured GFR of 85.12 ± 20.31 ml/min./1.73 m2 and a mean estimated GFR of 86.29 ± 17.04 ml/min./ 1.73 m2 (32.84 to 134.56 ml/min./1.73 m2.) based on the CKD-EPI formula. The bias based on mean difference is 1.16 (−0.76 to 3.09) ml/min./1.73 m2 and inter-quartile range (IQR) of the differences of 18.15. Precision was 13.37 and accuracy within 30% (P30) was 89.73% versus 84.1% in the original study.

Validation of Chronic Kidney Disease Epidemiology Collaboration as a diagnostic test: Only 17 patients had an mGFR < 60 ml/min./1.73 m2, which corresponds to a prevalence of 9%.

The positive likelihood ratio of CKD was 39.53 (95% CI: 12.36-26.43) and, consequently, a positive test increased the post-test probability from 9% to 79.63%. Conversely, the negative likelihood ratio was 0.3 (95% CI: 0.14-0.63) and a negative test thus reduced the post-test probability from 9% to 2.88%.

Accuracy was 96%, defined as the proportion of all tests that would have yielded the correct result in a contingency table (Table 3).

DISCUSSION

Whenever a new screening method is validated against a gold standard, it is important to ascertain whether the differences in measurements between the methods are of clinical relevance or not.

In this study, eGFR was not significantly different from mGFR by 51Cr-EDTA (p value: 0.81), but there was considerable variation in the Bland-Altman plot as the limits of agreement were above 25 ml/min. This is clinically important. These two measurement methods therefore cannot be used interchangeably, at least not at values below 85 ml/min./1.73 m2 because if no other parameters of kidney damage are used, a difference of −25.59 ml/min./1.73 m2 may imply a misclassification as CKD stage 3 [15].

The CKD stage determines which limits of agreement may be considered clinically relevant as a difference of only 15 ml/min./1.73 m2 should give rise to much more concern in the interval between a GFR of 10 and 25 ml/min./1.73 m2 than should a difference of 15 in the interval between a GFR of 100 and 115 ml/min./1.73 m2 [16].

In 2005 Froissart et al [17] investigated a cohort of 2,095 adult Europeans. They used 51Cr-EDTA as a reference method and demonstrated that the imprecision of MDRD eGFR increased considerably at values above 60 ml/min./1.73 m2 and for patients younger than 65 years. Although the sample size population of the two studies differed, they reported comparable levels of bias, precision and accuracy, even if the mean mGFR was highest in our study (85.12 versus 61.1 ml/min./1.73 m2).

Due to the large variation in the estimates, it is important to consider measuring the GFR with a more reliable method whenever an accurate value of GFR is needed, as in patients for whom platinum-based treatment is planned.

We also acknowledge that a cancer population has pitfalls such as low muscular mass which will result in low creatinine levels. This may explain the differences between eGFR and mGFR, as plasma creatinine was measured up to three weeks before GFR. Patients with severe illness like cancer may display significant fluctuations in plasma creatinine in the interval between measurement of creatinine and 51Cr-EDTA. Still, the difference between the two plasma creatinine measurements was less than 15%. This is a potential source of error. The sample size was small, but sufficient power was obtained.

The CKD-EPI formula had a high specificity and a low sensitivity, but a low negative likelihood ratio of 0.30. This supports the claim that the test may be used as a screening tool in the general population whose risk of CKD is about 9-11%. However, the precise prevalence of CKD is unknown in cancer patients, wherefore MDRD may be a better choice because its negative predictive value is higher (98% versus 97%) and its negative likelihood ratio lower (0.25 versus 0.3). This may help us avoid overlooking patients with reduced GFR who may require a more accurate GFR measurement.

It remains unknown whether eGFR by means of the CKD-EPI or MDRD equation can detect minor changes in GFR during chemotherapy. Although a prospective longitudinal study would be required to settle this point, we are inclined to believe that eGFR would not be able to detect minor changes in accordance with the conclusions of a recent review [18].

Correspondence: Belén Redal-Baigorri, Department of Nephrology, Roskilde Hospital, 4000 Roskilde, Denmark. E-mail: belenredal@hotmail.com

Accepted: 19 November 2010

Conflicts of interest: None



Literature
  1. Launay-Vacher V. Prevention of Cisplatin nephrotoxicity: state of the art and recommendations from the European Society of Clinical Pharmacy Special Interest Group on Cancer Care. Cancer Chemother Pharmacol 2008;61:903-9.
  2. Ladegaard-Pedersen HJ: Measurement of extracellular volume and renal clearance by a single injection of inulin. Scand J Clin Lab Invest 1972;29:145-53.
  3. Brochner-Mortensen J. A simple single injection method for determination of the extracellular fluid volume. Scand J Clin Lab Invest 1980;40:567-73.
  4. Martensson J, Groth S, Rehling M et al. Chromium-51-EDTA clearance in adults with a single-plasma sample. J Nucl Med 1998;39:2131-7.
  5. Gaspari F, Perico N, Ruggenenti P et al: Plasma clearance of nonradioactive iohexol as a measure of glomerular filtration rate. J Am Soc Nephrol 1995;6:257-63.
  6. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976;16:31-41.
  7. Levey AS, Bosch JP, Lewis JB et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med 1999;130:461-70.
  8. Traynor J, Mactier R, Geddes CC et al. How to measure renal function in clinical practice. BMJ 2006;333:733-7.
  9. Levey AS, Stevens LA, Schmid CH et al: A new equation to estimate glomerular filtration rate. Ann Intern Med 2009;150:604-12.
  10. Levey AS, Coresh J, Greene T et al: Expressing the modification of diet in renal disease study equation for estimating glomerular filtration rate with standardized serum creatinine values. Clin Chem 2007;53:766-72.
  11. Du BD, Du Bois EF. A formula to estimate the approximate surface area if height and weight be known. Nutrition 1989;5:303-11.
  12. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307-10.
  13. Greenhalgh T. How to read a paper. Papers that report diagnostic or screening tests. BMJ 1997;315:540-3.
  14. Newcombe RG. Two-sided confidence intervals for the single proportion: comparison of seven methods. Stat Med 1998;17:857-72.
  15. Hallan SI, Coresh J, Astor BC et al. International comparison of the relationship of chronic kidney disease prevalence and ESRD risk. J Am Soc Nephrol 2006;17:2275-84.
  16. Rule AD. The CKD-EPI equation for estimating GFR from serum creatinine: Real improvement or more of the same? Clin J Am Soc Nephrol 2010;5:951-3.
  17. Froissart M, Rossert J, Jacquot C et al. Predictive performance of the modification of diet in renal disease and Cockcroft-Gault equations for estimating renal function. J Am Soc Nephrol 2005;16:763-73.
  18. Prigent A. Monitoring renal function and limitations of renal function tests. Semin Nucl Med 2008;38:32-4.












* Required
Name*
E-mail*
Headline*