 |
||||||||||||||||||||||||||||
 |
||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||
Incidence of Chronic Kidney Disease Chronic kidney disease (CKD) is estimated to affect approximately 20 million patients in the United States (11% of the adult population) at a cost of $20 billion annually. CKD is defined as: persistent proteinuria (urine albumin/creatinine ratio of greater than 30 mg/g in an untimed urine) or a reduced glomerular filtration rate (GFR less than 60 mL/min/1.73 m2). The most common causes of CKD are diabetes mellitus, hypertension, and glomerulonephritis In addition, certain racial groups (African-American) and ethnic groups (Hispanics or Latinos regardless of race) are at increased risk. The elderly are also at increased risk; 17% of people over the age of 60 have a GFR less than 60 mL/min/1.73 m. With the aging population and the increasing incidence of diabetes, the medical impact of CKD is likely to increase significantly in the future. A variety of treatments are available to slow the progression of kidney disease. For type 1 diabetics, strict control of blood glucose concentrations has been shown to limit the development of microalbuminuria. Patients with hypertension benefit from treatment with angiotensin converting enzyme (ACE) inhibitors. Regardless of etiology, patients with kidney disease benefit from following a low-protein diet. Unfortunately, CKD is not being adequately recognized, and, even when recognized, it is not being adequately treated. Laboratory Screening for CKD Given the prevalence and societal costs of renal disease, as well as the fact that interventions are available to ameliorate disease, physicians would like a reliable method for detecting CKD in the general population. Gold standard tests for measurement of renal function include the inulin clearance assay and the 125I-iodothalamate clearance assay. Unfortunately, performing these tests is difficult, and they are not available in most labs. Instead, the focus has traditionally been on serum creatinine level and on the 24-hour creatinine clearance. Creatinine Physiology Creatinine is a by-product of muscle turnover, and human beings produce a relatively constant amount of creatinine per day, an amount correlated to each individual's muscle mass. In general, men and women produce roughly 20 (range 14-26) and 15 (range 11-20) mg/kg/day, respectively. The difference between the sexes is related to the lower proportion of muscle typically found in females. Thus, the prototypical 70 kg man would be expected to produce 70 kg × 20 mg/kg/day or 1400 mg creatinine/day, while a 70 kg woman would produce 70 × 15 or 1050 mg creatinine/day. Because creatinine is filtered and not appreciably re-absorbed or secreted by the kidney tubules, the serum creatinine level reflects the glomerular filtration rate (GFR). The more blood that is filtered through the kidneys, the more creatinine that is excreted from the body, and the lower the serum creatinine level gets. Conversely, the less blood that is filtered, the less creatinine excreted, and the higher the serum creatinine level becomes. In other words, assuming a constant production of creatinine, the lower the GFR, the higher the serum creatinine. As long as the GFR and the production of creatinine remain constant, the serum creatinine level remains constant. This phenomenon underlies the use of the equation for creatinine clearance as an estimate of GFR: Creatinine Clearance = GFR (in mL/min) = (Ucr × V) / (Pcr × 1440) Where: Ucr = urine creatinine concentration in a 24-hour urine (mg/dL) V = volume of the 24-hour collection (mL/day) Pcr= plasma (or serum) creatinine concentration (mg/dL) 1440 = minutes in a day (min/day) This formula helps to explain some of the difficulties associated in interpreting serum creatinine values. Consider the extreme case of a small woman whose weight is 95 pounds (43 kg). Her creatinine excretion would be expected to be 43 × 15 = 645 mg/day. Assuming a normal GFR of roughly 100 mL/min, her serum creatinine level would be estimated as follows: GFR = (Ucr × V) / (Pcr × 1440) 100 mL/min = 645 mg/day / (Pcr × 1440 min/day) Pcr = 645 mg / (100 mL × 1440) Pcr = 645 mg / 14400 mL = 0.0044 mg/mL = 0.44 mg/dL If her GFR were to decrease to 50 mL/min, her serum creatinine would double to 0.88 mg/dL. This creatinine level would be considered normal on most laboratory reports, giving no indication that a serious loss of renal function had occurred. In contrast, consider the other extreme, a large man in good shape whose weight is 260 pounds (118 kg). His expected creatinine excretion would be expected to be 118 × 20 = 2360 mg/day. At a normal GFR of roughly 100 mL/min, his serum creatinine level would be: GFR = (Ucr × V) / (Pcr × 1440) 100 mL/min = 2360 mg/day / (Pcr × 1440 min/day) Pcr = 2360 mg / (100 mL × 1440) Pcr = 2360 mg / 14400 mL = 0.0163 mg/mL = 1.63 mg/dL This creatinine level would be flagged as high on most laboratory reports. While a serum creatinine level of 1.63 reflects a GFR of 100 mL/min for him, it would represent a GFR of 27 mL/min for the small woman considered above! Thus, the serum creatinine level alone is not a reliable indicator of renal function. MDRD Equation Recent studies have shown that, although serum creatinine level alone is not a good predictor of GFR, when used in combination with other variables in a variety of predictive equations, it becomes an excellent predictor. Among the best known of these equations is the Cockroft-Gault equation, which has been in existence for many years: GFR = {[140 - Age(yrs)] × Weight(kg)} / (72 × Pcr) × (0.85 if female) Nonetheless, the equation is not widely used. Although laboratories could theoretically perform the calculation, they rarely know the patient's weight, a required element. And clinicians, who would readily know their patient's weight, simply don't have the time to do the calculation on every one of their patients. More recently, attention has centered on the predictive equations derived from the Modification of Diet in Renal Disease (MDRD) study. The term "MDRD equation" is somewhat confusing because it encompasses several different equations. The equation advocated by the National Kidney Foundation involves the use of four parameters: serum creatinine level, age, sex, and ethnic group (i.e., whether the patient is African-American or not). As shown below, the equation is quite complicated, involving two exponential terms: GFR (mL/min/1.73 m2) = 186 × (Pcr)-1.154 × (Age)-0.203 × 0.742 (if female) × 1.210 (if African-American) There are a number of things to notice about the four-parameter MDRD equation. First, given the importance of muscle mass in the daily production of creatinine, it may appear striking that body weight (or surface area or some other estimate of size) does not appear in the four-parameter MDRD equation. The reason the equation works despite this omission is that the values are normalized by being reported as "per 1.73 m2" (an average body surface area). Smaller patients will have lower absolute GFRs, and larger patients will have higher absolute GFRs, but, to the extent that the estimated GFR and the reference range are both reported in mL/min/1.73 m2, body weight is not needed for the calculation. A second thing to note about the MDRD equation is that the "186" multiplier at its beginning has no physiologic significance; the factor was derived empirically. Third, although it is difficult to appreciate the effects of the serum creatinine and the patient's age because of their associated exponentials, the final two terms are straightforward. In essence, for any given age and serum creatinine level, the estimated GFR is the same until corrections are applied for being female and/or for being African-American. For example, if a patient is 50 years old and has a serum creatinine of 1.8, the estimated GFR from the first 3 terms of this equation is 43 mL/min/1.73 m2. But consider these four scenarios for that patient and their associated estimated GFRs: 50-year old African-American man: 52 mL/min/1.73 m2 (i.e., 43 × 1.210) 50-year old non-African-American man: 43 mL/min/1.73 m2 (i.e., neither correction factor applied) 50-year old African-American woman: 39 mL/min/1.73 m2 (i.e., 43 × 0.742 × 1.210) 50-year old non-African-American woman: 32 mL/min/1.73 m2 (i.e., 43 × 0.742) In other words, for this (and any other) age/creatinine combination, there can be more than a 1.5-fold range in estimated GFR based on gender and race (in this case, from 32 to 52 mL/min/1.73 m2). Black men have the highest multiplier (1.210), and non-black women have the lowest multiplier (0.742). Practical Issues Associated with Reporting Estimated GFR Information on the patient's ethnic group may not be readily available. This can be overcome by including values for both African-American and non-African-American patients in a single report, with the clinician being responsible for choosing the correct value for the patient in question. Although laboratories may use any reporting format they choose, the National Kidney Foundation suggests something along these lines for a 63-year-old woman with a creatinine of 1.8 mg/dL:
Third, estimated GFRs above 60 should be reported without a specific number, simply as "above 60 mL/min/1.73 m2." Values below 60 mL/min/1.73 m2 are considered indicative of chronic kidney disease and should indicate the actual number. Population mean GFR varies with age as shown in Table 2.
Who Should Be Tested? Because of the high prevalence of early kidney disease in the general population and because of the evidence that early intervention can prevent or delay adverse outcomes, the recommendation of the National Kidney Foundation is that patients in high-risk groups for developing CKD be screened (i.e., tested in the absence of symptoms). Specifically, they recommend a spot urine for albumin/creatinine ratio and a serum creatinine level. These high-risk groups include the following:
|
||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||
|
||
