To the Editor,

While we agree with many of the ideas expressed in the letter by Peral et al,1 we would like to expand on the following:

1. Clinical laboratories in Spain, according to national recommendations,2 generate analytical reports including the glomerular filtration rate (GFR) calculated by means of an equation. Unpublished data from a national survey carried out by the Kidney Function Commission of the Spanish Society of Clinical Biochemistry and Molecular Pathology (CFR-SEQC), show that out of 281 laboratories surveyed, 88% report GFR. Of these reporting laboratories, 32% calculate GFR using the MDRD-IDMS equation, 62% use the MDRD-4 equation and 4% use the Cockcroft-Gault (CG) or Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formulas. Standard procedures for measuring creatinine have become increasingly available in Spain, but it is true that some laboratories that introduced GFR calculated by MDRD-4 (factor 186) in past years have not yet made the necessary leap to the MDRD-IDMS method (factor 175). The CFR-SEQC is undertaking a series of actions in order to correct this situation.

2. We are grateful that the error in the description of the CKD-EPI equation in our article in Nefrología3 was reported. With a view to correcting this and other errors, we sent a list of errata to the journal.

3. The recommendations by pharmaceutical companies with regard to adjusting drug doses in patients with compromised renal function follow the Food and Drug Administration (FDA) guidelines and are based on creatinine clearance intervals obtained by using the CG equation.4 However, neither the methods for measuring creatinine nor the patient samples used to develop the equation are available, meaning that the equation cannot be reformulated for use with creatinine values obtained using current methods. Creatinine clearance values obtained using the CG equation are 10%-20% higher if standard procedures are followed, which overestimates renal function and therefore affects drug dose adjustments.

GFR values obtained by using the CG and MDRD methods are not interchangeable. Different studies that compare dosage adjustments based on the CG and MDRD methods report differences in between 10% and 40% of cases.5 Comparing these studies is difficult due to the variability of the creatinine measurement methods used when calculating the equations and the type of patients studied. In addition, their interpretation is complex, since they do not assess the clinical consequences of discrepancies between doses that result from using one equation or another. Only one study compares concordance between the assignment to an FDA-listed category, based on GFR measurement (iothalamate clearance), and 3 equations (MDRD-IDMS, CG using real weight, and CG using the ideal weight value), in addition to differences in recommended doses between the 3 equations with respect to 15 drugs that are excreted renally.6 Results from the comparison show that concordance with the recommended doses of the 15 drugs, based on the GFR measurement, was greater for MDRD-IDMS (88%) than for CG with ideal weight (82%) or CG with real weight (85%). Concordance between recommended doses was 89% between MDRD-IDMS and CG with ideal weight.

The American College of Clinical Pharmacy Nephrology Practice and Research Network recommends that neither CG nor MDRD be used as the only measurement to determine dosage-adjustment decisions. Other factors should also be considered, such as the way equations work in specific population groups, the therapeutic index, drug indication and toxicity profile, availability of other treatment agents, the possibility of monitoring drug concentrations in blood and more precise means of measuring creatinine clearance or glomerular filtration rate.5 The National Kidney Foundation Education Program recommends that both CG and MDRD-IDMS be used when estimating renal function in order to adjust drug doses.7 Likewise, the FDA recently proposed that MDRD-IDMS be used along with CG in future pharmacokinetic studies in patients with kidney disease.8

We believe that the value of GFR calculated based on the MDRD-IDMS equation is a valid tool for assessing renal function for purposes of adjusting drug doses for several reasons: 1) it is based on creatinine measurement procedures that have been standardised against the reference method; 2) it correlates better with measured GFR than the CG method for GFR values <60ml/min/1.73m2, which are the most susceptible to dosage adjustments; and 3) it is available in most clinical laboratory reports, unlike CG.

We agree with Peral et al that while the GFR value obtained by MDRD-IDMS is expressed in ml/min/1.73m2, absolute values (ml/min) should be used for this purpose in patients whose body surface area varies considerably from the standard area of 1.73m2

Conflicts of interest

The authors affirm that they have no conflicts of interest related to the content of this article.