Method Chosen to Estimate Pre-PCI Kidney Function Matters

Different approaches to calculating estimated glomerular filtration rate (eGFR) in PCI patients can result in widely variable numbers, which in turn can influence risk stratification and potentially medication dosing. These findings, published in the June 30, 2015, issue of the Journal of the American College of Cardiology, support use of the newer CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) equation, researchers say. 

Take Home:  Method Chosen to Estimate Pre-PCI Kidney Function Matters

“The message is that we need to have 1 standard approach to assessing renal function,” lead author Hitinder S. Gurm, MD, of the University of Michigan Health System (Ann Arbor, MI), told TCTMD in a telephone interview. “What’s happening is that there’s a huge amount of disagreement between these equations, as we found in this study, and this is something clinicians need to know,” he added.

For the all-comers study, Dr. Gurm and colleagues looked at eGFR data from a statewide Michigan registry comprising 128,805 patients (mean age 65 years; 33.5% women) who underwent PCI at 47 participating centers from 2010 to 2014. Methods used to calculate eGFR were:

  • Cockcroft-Gault 
  • Modification of Diet in Renal Disease (MDRD) 
  • CKD-EPI 

Median eGFR before PCI varied significantly from equation to equation: from 73.0 mL/min/1.73 m2 for MDRD to 76.7 for CKD-EPI and 89.0 for Cockcroft-Gault.

CKD-EPI Comes Out on Top

After PCI, 3.2% of patients developed acute kidney injury (AKI), while 0.35% had a new requirement for dialysis.

Receiver operating characteristic analysis showed that for AKI (primary endpoint) the prognostic accuracy of CKD-EPI was superior to those of MDRD and Cockcroft-Gault (both P < .001). For in-hospital mortality, CKD-EPI performed as well as Cockcroft-Gault (P = .711) and better than MDRD (P < .001). For dialysis, the accuracy of CKD-EPI did not differ from that of MDRD but exceeded that of Cockcroft-Gault.

Classification of CKD stage was most similar between MDRD and CKD-EPI, with agreement 84.69% of the time. Of the 19,720 patients who were reclassified by CKD-EPI compared with MDRD, 11.6% were reclassified to a lower CKD stage and 3.6% to a higher stage. The agreement was much lower when CKD-EPI was compared with Cockcroft-Gault (56%), with the vast majority of reclassifications by CKD-EPI placing patients in a higher-severity CKD stage.

Overall, reclassification to a higher or lower CKD stage with CKD-EPI was found to most appropriately predict risk of AKI and a new requirement for dialysis.

Additionally, agreement between the 3 equations decreased with declining eGFR cutoff values. Adjusting for body surface area again improved the agreement for Cockcroft-Gault, but the 3 equations still only agreed on drug-dose adjustment in 62.9%, 56.2%, and 45.0% of patients for eGFR cutoffs of < 60, < 50, and < 30 mL/min/1.73 m2, respectively.

Call for Standardization  

The 2012 Kidney Disease Improving Global Outcomes (KDIGO) guidelines recommend using CKD-EPI because of its superior accuracy, yet according to Dr. Gurm and colleagues, a recent survey conducted by the College of American Pathologists found that 83% of reporting laboratories use the MDRD, 4% use Cockcroft-Gault, 2% use another equation, and 12% are “unsure” what they use to obtain eGFR values.

Cockcroft-Gault was developed before standardization of creatinine assays, yet remains widely used by many, including pharmacists, who often employ it to recalculate renal function, the authors say.

“The [professional] societies are recommending [CKD-EPI] and our results show that it is the preferred method, so it makes sense that we should all stick with the same standard,” Dr. Gurm observed in the interview. “It’s the safest thing to do.”

Adding to the problem is that the FDA provides no guidance as to which of the equations pharmaceutical companies should use when conducting studies to support the safety of a given drug with regard to renal function, he stressed.

“We have spoken to the FDA about this, because we feel that we need some regulatory guidance to standardize these equations,” Dr. Gurm said.

Most companies use Cockcroft-Gault, which is significantly impacted by body weight, he reported. While the FDA has attempted in recent years to include information on which equation was used, it is not always readily available to them. Furthermore, Dr. Gurm added, it would be interesting to look at the pharmacokinetic data of supporting trials for recent drug approvals “and see if the dosing holds when we use the CKD-EPI equation.”

Clinical Implications

In an email with TCTMD, Eugenia Nikolsky, MD, PhD, of Rambam Medical Center (Haifa, Israel), said the study “clearly demonstrates to what degree outcomes depend on the chosen definition [of eGFR].” This has important clinical implications, she added, since “it is well known that inappropriate dosing in patients with chronic kidney disease is related to increased incidence of complications and mortality or to ineffective therapy, and the precise assessment of renal function may improve outcome.”

Dr. Nikolsky added that in practice, dosage recommendations for traditional antiplatelet or antithrombotic agents given in the cath lab are based on the calculation of preprocedural creatinine clearance as per prescribing information on the packaging.

“It would be interesting to see whether a reduction of complications (bleeding vs ischemia) may be achieved once the renal function is calculated based on CKD-EPI versus Cockcroft-Gault or MDRD formulas,” she said, echoing Dr. Gurm’s suggestion.

In an editorial accompanying the paper, Peter A. McCullough, MD, MPH, and colleagues from Baylor University Medical Center (Dallas, TX), write: “Bivalirudin and enoxaparin (contraindicated at creatinine clearance < 30 mL/min) would be most affected by a shift from [creatinine clearance] to [eGFR]-rate drug dosing in a relatively conservative manner, which would bias care toward a decreased risk of bleeding with both agents.”

The bottom line, asserts Dr. Gurm, is that eGFR data should be available for any patient coming into the cath lab. “In an elective PCI situation, there’s really no excuse for not knowing a patient’s eGFR,” he said. “And certainly you should know how it was calculated.”

Note: Coauthor Roxana Mehran, MD, is a faculty member of the Cardiovascular Research Foundation, which owns and operates TCTMD.


1. Parsh J, Seth M, Aronow H, et al. Choice of estimated glomerular filtration rate equation impacts drug-dosing recommendations and risk stratification in patients with chronic kidney disease undergoing percutaneous coronary interventions. J Am Coll Cardiol. 2015;65:2714-2723. 
2. McCullough PA, Patanker G, Stoler RC. Estimating renal filtration, drug dosing, and clinical outcomes [editorial]. J Am Coll Cardiol. 2015;65:2724-2725. 


  • Dr. Gurm reports having served as a consultant for Osprey Medical. 
  • Drs. McCullough and Nikolsky report no relevant conflicts of interest. 

Related Stories:


We Recommend