Clinical impact of reporting estimated glomerular filtration rates.

When clinicians speak of kidney function, they almost always are talking about the glomerular filtration rate (GFR). The kidney has many other functions, however. Most excretory functions are related in some way to glomerular filtration, at least as it is modified by the variable reabsorption of water and some solutes and the addition of yet other solutes via secretion. The kidney also has important endocrine functions (such as the production of erythropoietin and active vitamin D) and metabolic functions (such as the metabolism of insulin and other low molecular weight proteins) that are not related to excretory function; however, as the GFR declines because of chronic kidney disease (CKD), all of these functions also tend to decline, roughly in parallel. Thus, taking the overall kidney function as equivalent to the GFR is a generally useful approximation. Clinicians’ equating of the GFR with kidney function has an additional practical utility. Estimating GFR relies on simple tests, which are based mainly on serum creatinine, whereas many other kidney functions are more difficult to measure. For example, the extensive machinery for secreting organic solutes has become increasingly well described in molecular detail, but the major natural or xenobiotic solutes that are secreted are still largely unidentified. Hence, estimating the residual secretory function by measuring the serum concentrations of such solutes is not feasible, regardless of how important to health this excretory mode might be. Creatinine was found to be a good marker of the GFR more than 70 years ago. Measurements of the urinary clearance of endogenous creatinine are a reasonable approximation of the GFR, as measured by the infusion of such exogenous markers as inulin or iothalamate. Measuring creatinine clearance, however, requires the timed collection of urine, which is cumbersome and notoriously inaccurate. To address this problem, researchers have developed several estimating equations for GFR or creatinine over the last 45 years. These equations use measured serum creatinine concentrations and such patient characteristics as age, sex, race, and sometimes body weight, which are factors related to muscle mass, the major determinant of creatinine excretion. They thus obviate urine collections. The best iteration of these equations is the CKDEPI (Chronic Kidney Disease Epidemiology Collaboration) equation (1 ). It was developed by relating the GFR, as measured by iothalamate clearance, to the serum creatinine concentration, coupled with the individual’s age, sex, and race—all normalized to a standard body surface area. This equation has been validated in nearly 4000 individuals. Most laboratories in the US and many laboratories around the world use the serum creatinine concentration along with the CKD-EPI equation [or its forerunner, the Modification of Diet in Renal Disease (MDRD) Study equation] to report the estimated GFR (eGFR). Although such automatic reporting of the eGFR provides a useful estimate of kidney function, does such reporting influence clinical practice? Researchers in Canada have begun to address this question (2 ). They used a laboratory registry in Alberta to track nephrology consultations before and after the implementation of eGFR reporting and found reporting to be associated with an increase in nephrology referrals. The increase was particularly notable in people with advanced CKD (eGFR 30 mL min 1 (1.73 m) ), in women, and in the elderly (in whom a reduced GFR is often not detected via an increased serum creatinine concentration alone because of low creatinine production). Although most laboratories have different serum creatinine reference intervals for men and women and although the decrease in creatinine production with age is known, clinicians tend to disregard these points and base CKD diagnoses solely on the serum creatinine concentration without consideration of the differences in production rates among the various subgroups. This tendency underscores the value of applying the equations for estimating GFR, as opposed to simply reporting the creatinine concentration. The Canadian study also noted an increase in referrals for patients with diabetes and hypertension, conditions known to be associated with CKD and with a high risk for adverse out1 Tufts University School of Medicine, Tufts New England Medical Center, Boston, MA; 2 Albert Einstein College of Medicine, Bronx, NY. * Address correspondence to this author at: Albert Einstein College of Medicine, Rm. 615, Ullmann Bldg., 1300 Morris Park Ave., Bronx, NY 10461. Fax 718-430-8963; e-mail [email protected]. Received May 24, 2010; accepted June 24, 2010. Previously published online at DOI: 10.1373/clinchem.2010.147827 3 Nonstandard abbreviations: GFR, glomerular filtration rate; CKD, chronic kidney disease; CKD-EPI, Chronic Kidney Disease Epidemiology Collaboration; MDRD, Modification of Diet in Renal Disease; eGFR, estimated GFR; ACE, angiotensinconverting enzyme. Clinical Chemistry 56:9 000 – 000 (2010) Perspectives