ESTIMATING CREATININE CLEARANCE

Transcription

1 ESTIMATING CREATININE CLEARANCE Robert E. Pachorek CHAPTER The importance of accurate estimations of glomerular filtration rate (GFR), the principal measure of renal function, cannot be overemphasized. Many drugs or active metabolites are eliminated to some extent by renal excretion, creating the need for dosage adjustments as renal function deteriorates, particularly for drugs with a narrow therapeutic range (see Chapter 3, Renal Drug Dosing Concepts). In general, decreases in the rate of elimination of drugs primarily excreted unchanged by the kidneys are proportional to decreases in the GFR. Although methods to determine the clearance of inulin and other exogenous compounds are available for such assessment, the renal clearance of the endogenously produced amino acid creatinine is the most commonly used estimate of GFR in hospitalized patients. eatine, a product of protein metabolism, is primarily produced in the liver, pancreas, and kidneys and is actively transported into muscle tissue where it is stored. eatinine is produced from creatine in muscle tissue proportionally to muscle mass. It is released at a constant rate (~1.5% of the total pool per day) into the general circulation and is distributed to total body water. The glomerulus passively filters creatinine, although 10% to 40% of the total creatinine found in urine is a result of active renal tubular secretion. 1 4 The creatinine clearance (Cl), in milliliters per minute (ml/min), can be directly measured by collecting urine over a period of time (e.g., ) using the following relationship: U Cl= S V t where U is the urine creatinine concentration (mg/dl), V is the urine volume collected (ml), S is serum creatinine (mg/dl) measured at the midpoint of the urine collection, and t is the time interval in minutes of the urine collection (e.g., 1,440 min for ). 1, FACTORS AFFECTING ESTIMATES OF GLOMERULAR FILTRATION RATE Because some creatinine found in the urine is due to tubular secretion, Cl overestimates the GFR at all levels of renal function. Drugs such as amiloride, cimetidine, trimethoprim, salicylate, triamterene, and spironolactone, which inhibit this secretory function, may increase S and decrease the Cl estimate without actually affecting the GFR. 5 As GFR declines, there is a progressive relative increase in creatinine tubular secretion compared to filtration and a progressive disparity between actual GFR and Cl. In one study, it was noted that when the true GFR was around 100 ml/min, the Cl overestimates the GFR by about 0%; for a GFR of 60 ml/min, the Cl was overestimated by about 60%; and for a GFR of 0 ml/min, the Cl was overestimated by 100% or more. 5 In the past this error was partially offset if the laboratory was using an uncorrected Jaffe-based method assay to determine serum creatinine because endogenous serum substances added ~0. mg/dl to the creatinine value when measured by this common method. 5 The new assay reference standard negates the increase in S seen with the Jaffe-based method and thus reintroduces the error of GFR estimate caused by tubular secretion of creatinine (see eatinine Assay Standardization, below). 6 Disease states and aging effects on muscle mass affect creatinine production and turnover. The measured 4-hr Cl is less likely to be affected than estimates using patient demographics and S 9

2 10 CLINICAL PHARMACOKINETICS to estimate Cl. Drugs and exogenous or endogenous compounds that interact with the laboratory assays for creatinine may also affect estimates of Cl. The compounds in Table -1 may positively bias the creatinine concentration by interfering with creatinine assays (dependent on the assay system, methodology, and instrument used). These compounds all have the potential to interfere with the S assays although the extent of interference is dependent on the given instrument used. If a patient has a potential for a creatinine laboratory assay interaction, clinicians should check with the laboratory for the specific interference with the instrument used. Judicious timing of the drawing of blood for S determination (at minimal interfering drug concentration) or using an alternate assay method may be advisable.,3,5,7 TABLE -1. COMPOUNDS THAT INTERFERE WITH CREATININE ASSAYS Jaffe-based Assays Acetoacetate Bilirubin Cefoxitin, cephalothin Clavulanic acid Other cephalosporins (supratherapeutic) Furosemide (supratherapeutic) Lactulose Endogenous serum substances (+ 0. mg/dl creatinine prior to assay standardization to isotope dilution mass spectrometry [IDMS] reference method) Enzymatic Assays Dopamine Dobutamine Flucytosine Lidocaine Exogenous creatine supplementation could be thought to affect the creatinine serum concentration because creatine is converted to creatinine. One review of studies of acute and chronic creatine ingestion in young healthy populations found minimal impact on serum creatinine concentrations; however, other more recent studies found an increase in S with a buffered creatine supplement and a creatine ester supplement CREATININE ASSAY STANDARDIZATION The National Kidney Disease Education Program (NKDEP) of the U.S. National Institutes of Health (NIH) launched an international program to standardize the creatinine assay calibration in healthcare laboratories. This decreased the former positive bias (up to 0%) of the creatinine assay through laboratory instrument recalibration to agree with an IDMS reference method. 6,11 This calibration may lead many of the Cl estimating equations for adults and pediatrics to give higher Cl estimates than in the past, depending on the previous assay used These differences may not be as clinically important for dosing of a drug with a wide therapeutic index (e.g., cefazolin); however, for drug dosing using pharmacokinetic population estimates in drugs with a narrow therapeutic index (e.g., aminoglycosides) this change may have important clinical implications. It should be noted that most 4-hr measured Cl values from the past that were obtained using a non-standardized assay (non-idms) would be expected to yield the same value as today because the serum and urine creatinine were usually assayed using the same calibration scheme. 1 This is a result of the creatinine concentration being located in the numerator and the denominator of the 4-hr Cl equation noted above. 1

3 CHAPTER - Estimating eatinine Clearance 11 FORMULAS TO ESTIMATE CREATININE CLEARANCE IN ADULTS Collecting urine for for a creatinine measurement is tedious and must be done accurately with no missed collections and accurate measurement of urine volume to properly measure the Cl. Because this is difficult and time-consuming, formulas have been derived that use the serum creatinine and patient demographics to estimate Cl. The many equations for rapid estimation of Cl published over the last 40 yr generally produce similar values. Two well-validated and commonly used equations for adult patients when S is at steady state are the Cockcroft-Gault equation and the Jelliffe equation Cockcroft-Gault is the equation usually included in drug product labeling (package inserts) for individualizing drug dosages for renally cleared drugs when an equation is mentioned. 17 Other more recently developed equations are used to estimate GFR in adult patients with chronic kidney disease (Modification of Diet in Renal Disease [MDRD] Study equation, Chronic Kidney Disease Epidemiology Collaboration [CKD-EPI] equation; discussed later); however, at this time their use for drug dosing purposes has been questioned Cockcroft-Gault Equation: Cl ( males) (ml / min) = ( 140- Age)( W) ( 7)( S ) Cl ( females) (ml/min) = Cl( males) 0.85 where S is in milligrams per deciliter (mg/dl) and W is weight (kg). The International Systems conversion is: ( )( ) ( ) Age W Cl ( males) (ml/min) = S Cl ( females) (ml/min) = Cl( males) 0.85 where S is in micromoles per liter (µmol/l) and W is weight (kg). Jelliffe Equation: Cl ( males) (ml /min/ 1.73 m ) where S is in mg/dl. = ( ) ( S ) Age-0 The International Systems conversion is: ( ) ( ) Age-0 Cl ( males) (ml/min/1.73 m ) = S where S is in µmol/l. Cl (females) = Cl (males) 0.9 for both versions. Because creatinine is produced in muscle tissue, the weight used in the Cockcroft-Gault equation has commonly been ideal body weight (IBW) or actual body weight (ABW) if it is less than IBW. However,

4 1 CLINICAL PHARMACOKINETICS there are many important considerations in these generalizations (see section on body weight). The original authors used ABW in developing their equation, but other researchers have examined the impact of excessive weight on predictability. IBW is estimated by the following 0 : IBW (males) (kg) = 50 + [.3(H 60)] IBW (females) (kg) = [.3(H 60)] where H is height in inches, or, IBW (males) (kg) = 50 + [0.9(H 15)] IBW (females) (kg) = [0.9(H 15)] where H is height in centimeters (cm). In the Jelliffe equation, the Cl is normalized to an average adult body surface area (BSA) of 1.73 m and the units are ml/min/1.73 m. If the patient does not have a BSA of 1.73 m, then the result needs to be converted to their actual Cl in ml/min. To calculate the Cl (in ml/min), the result is multiplied by the patient s BSA and then divided by BSA may be determined from the following equation (other approaches to estimating BSA are shown in Chapter 4, Medication Dosing in Overweight and Obese Patients) 1 : BSA (m ) = W H where W is weight (kg) and H is height (cm). The Cockcroft-Gault and Jelliffe equations work reasonably well for most adults with S at steady state because they allow for declining muscle mass (and creatinine production) often associated with reduced weight and advancing age and are adjusted for the average smaller muscle mass of females. The Cockcroft-Gault equation is the most commonly used and recommended equation for Cl estimation for drug dosing and is discussed in greater detail in the following sections. The Jelliffe equation is still used by some clinicians and is reasonably accurate for use when height and weight are not available in average-sized patients. 16 The accuracy of these equations in predicting Cl is often limited in patients with various disease states or conditions. These include the elderly, the malnourished, the obese, patients with amputations or spinal cord injuries, those with chronic renal insufficiency, acutely changing renal function, those with liver disease, critically ill patients, and pediatric patients (see specific sections, below.). 1 3 Other, possibly more accurate, methods and equations for rapid prediction of renal function will continue to evolve as more patients and larger subgroups are studied.,3 NKDEP previously recommended using the MDRD Study equation or the CKD-EPI equation for estimation of GFR in adults as a screening tool for kidney disease and encouraged clinical laboratories to report this estimated GFR (egfr) along with the patient s serum creatinine. The CKD-EPI equation egfr result may now be reported in this situation, as it does not require the patient s weight. The NKDEP has suggested use of this egfr or the Cockcroft-Gault equation for drug dosing. 1 In the United States, most of the drugs that have the Food and Drug Administration approved product labeling for drug dosing in patients with renal dysfunction have dosing guidelines specifically indexed to Cl, and many note use of the Cockcroft-Gault equation For this reason, use of the egfr for drug dosing has been questioned, and it was reported in one survey that pharmacy clinicians did not substitute egfr for Cl, although this may evolve over time ,4 Abbreviated MDRD Study Equation: IDMS Laboratory-Calibrated eatinine 1 : GFR (ml/min/1.73 m ) = 175 (S ) (Age) (0.74 if female) (1.1 if African American)

5 CHAPTER - Estimating eatinine Clearance 13 where S is in milligrams per deciliter (mg/dl). For conversion to International Systems using S in µmol/l, change the S equation term to: S The rest of the equation remains the same. This equation provides egfr values that are automatically normalized to ml/min/1.73 m. When using this equation in very large or very small patients, the NKDEP recommends that the normalized result in ml/min/1.73 m should be converted to ml/min. 1 To calculate the actual egfr (in ml/min), the normalized result is multiplied by the patient s BSA and then divided by 1.73 m. The MDRD Study equation, although created from a large group of patients, had not been validated in many population groups including children, the elderly (>70 yr), patients with serious comorbid conditions, body size extremes, muscle mass extremes, severe malnutrition, pregnant patients, ill hospitalized patients, amputees, and patients with near normal renal function. 11,5 The equation is not weight-based but is affected by obesity and other factors that affect creatinine production. 6,11,5 As noted above, the NKDEP has recommended egfr (MDRD Study or CKD-EPI equation) reporting by laboratories along with serum creatinine to aid in the detection, evaluation, and management of patients with CKD. For the MDRD Study equation, they had recommended reporting exact actual values for egfrs of 60 ml/min/1.73 m and below, but for values above 60 ml/min/1.73 m, they recommend reporting it as > 60 ml/min/1.73 m as this equation was found to underestimate measured GFR at higher levels. 6,1 The CKD-EPI equation was developed as another option to estimate GFR in adult patients with kidney disease. This equation was derived from over 1,000 patients of various races with and without CKD, diabetes, and kidney transplant. 6 The CKD-EPI equation is considered more accurate than the MDRD Study equation for a GFR > 60 ml/min/1.73 m, and appears to be the better equation to estimate GFR for staging of CKD. 1,7,8 NKDEP suggested this GFR estimating equation for drug dosing, but recent studies have indicated serious limitations. 1,19,9 Again, when using this equation in very large or very small patients, the NKDEP recommends that the normalized result in ml/min/1.73 m be converted to ml/min. 1 To convert egfr to ml/min, the normalized result is multiplied by the patient s BSA and then divided by 1.73 m. CKD-EPI Equation 11 : GFR (ml/min/1.73 m ) = 141 min (S /κ, 1) α max (S /κ, 1) Age [if female] [if African American] where: S is serum creatinine in mg/dl, κ is 0.7 for females and 0.9 for males, α is 0.39 for females and for males, min indicates the minimum of S /κ or 1, and max indicates the maximum of S /κ or 1. In international units where S is in µmol/l, the equation would be the same except for the terms below: S is serum creatinine in µmol/l, κ is 61.9 for females and 79.6 for males. Adjustments in the weight and/or creatinine variables based on a patient s clinical condition are commonly made in the Cockcroft-Gault equation in attempts to improve predictive performance. Clinicians should carefully assess the patient s clinical status and importance of an accurate assessment of renal function and modify these variables or use another more suitable equation or a timed Cl measurement if necessary. In general, the following subgroup reviews pertain to adult patients (see section on pediatrics for Cl estimation in children)

6 14 CLINICAL PHARMACOKINETICS BODY WEIGHT eatinine production is dependent on muscle mass, and the use of IBW in the Cockcroft-Gault equation appears to produce reliable results in patients whose ABW is not far from IBW. For patients who are malnourished or cachectic with an ABW less than their IBW, the ABW should be used. 1 3,0,30,64 For adult patients less than 1.5 m (60 in) tall, use of the lesser of ABW or IBW (males = 50 kg, females = 45.5 kg) has been proposed. 31 Other methods of predicting IBW are provided in the Introduction and Chapter 4, Medication Dosing in Overweight and Obese Patients. Obesity (defined as >0% over IBW or BMI > 30) is another factor that affects the Cockcroft-Gault Cl estimation. Obese patients appear to have a larger muscle mass than would be predicted when using height in the IBW equation. Using IBW is still preferable to using ABW; however, using an adjusted body weight (BW adj ) between IBW and ABW may be more accurate. Use of a factor of 40%, 30%, or 0% of the difference between ABW and IBW has been proposed. 3,,31,3,64 BW adj = IBW (ABW IBW) BW adj = IBW (ABW IBW) BW adj = IBW + 0. (ABW IBW) A meta-analysis including 13 trials comparing a 4-hr measured Cl with a Cockcroft-Gault estimation found that use of IBW with Cockcroft-Gault had a lower mean difference (5. ml/min) compared to a 4-hr Cl than using ABW (15.9 ml/min). In a smaller number of trials, the use of no body weight (using 7 in the Cockcroft-Gault equation numerator) or adjusted body weight (0.3 factor) also had a low mean difference. 3 The notion of correct weight to use in the prediction equations is an interesting one. As mentioned earlier, Cockcroft and Gault used ABW to develop the equation. Many authors have studied and suggested the use of IBW or one of the BW adj for patients who weigh more than their IBW. Intuitively such approaches seem reasonable because creatinine is produced in muscle, not fat tissue. One suggestion that might help in determining a reasonable weight is to avoid using only standards and to visually examine the patient. For example, a 180 cm (5 11 ) body builder who weighs 100 kg (0 lb) would clearly be expected to produce more creatinine daily than a sedentary individual of the same height and weight. Both patients would be estimated to have the same IBW, and both are considered obese by definition (ABW > 0% above IBW). It would seem reasonable to anticipate that the former patient should have Cl estimated using ABW, since the additional weight will be creatinine producing, while the latter would be best estimated using BW adj (or even IBW if they were extremely sedentary with little additional muscle mass associated with the adiposity). Finally, it should be expected that the more the patient differs from the patients used in the studies to develop the equations, the greater the potential that predictions might not match actual measured Cl. The Salazar-Corcoran equation (see below) is another method of Cl estimation that has been useful in the obese patient but is a bit more complicated to use. 3,,33 Methods of adjusting body weight in the morbidly obese (BMI > 40) have recently been reviewed, and for this body weight group use of lean body weight (LBW; see equation below) in the Cockcroft-Gault equation may be the most appropriate; however, this practice has been questioned and still needs to be validated ,64 Salazar-Corcoran Equation: Cl ( males) (ml/min) = Cl ( females) (ml/min) = ( 137-Age) ( 0.85 W ) + ( 1.1 H ) ( 51)( S ) ( 146-Age) ( 0.87 W ) + ( 9.74 H ) ( 60)( S )

7 CHAPTER - Estimating eatinine Clearance 15 where W is weight in kg, H is height in meters, and S is in mg/dl. In international units where S is in µmol/l, the equations would be: Cl ( males) (ml/min) = Cl ( females) (ml/min) = ( 137-Age) ( 0.85 W ) + ( 1.1 H ) ( 0.58)( S ) ( 146-Age) ( 0.87 W ) + ( 9.74 H ) ( 0.68)( S ) Lean Body Weight Equations (see Chapter 3): LBW (males) (kg) = (970 ABW) / [ (16 BMI)] LBW (females) (kg) = (970 ABW) / [ (44 BMI)] where ABW is in kg, and body mass index (BMI) is in kg per meter. LOW SERUM CREATININE IN ELDERLY OR UNDERWEIGHT PATIENTS It is a fairly common practice for clinicians to round measured S concentrations that are less than 0.8 or 0.9 mg/dl to a higher value in elderly or underweight adult patients before using the estimating equations. The S is inversely proportional to Cl, so using an unrealistically low S value in an elderly or other patient with significantly decreased muscle mass or creatinine production may overestimate the Cl, leading to the use of higher drug doses. The use of a value of 1 mg/dl as the lower limit of S in these equations has been popular with some clinicians; however, underprediction of Cl may also occur, and this practice is not recommended. 19,35 Using 0.8 or 0.7 (or less) as the lower limit of S may be more appropriate. 37,38 Intuitively, it would seem to make sense that a patient s muscle mass would give some guide to how much fudging should be done in setting a lower limit of S. That is, in a patient with obviously limited muscle mass, it might be more reasonable to adjust the S upward than in a patient with average muscle mass. Also, it seems reasonable to assume that the larger the degree of fudging of the S upward to some minimum value, the greater the likelihood of poor prediction. That is, changing a measured S from 0.3 to 0.7 might result in a poorer prediction of actual Cl than changing from 0.6 to 0.7. Further, it also seems logical that the same conditions (elderly with decreased muscle mass) present in a patient with an S of 1 or more would lead to a need to increase the S arbitrarily to avoid overestimating Cl. However, this has not generally been recommended, and there are few data to support such approaches. The use of a percentage increase in S might be more logical and would be an area for study. AMPUTATIONS Estimation of Cl in patients with amputated limbs poses a dilemma for IBW calculation. A reasonable approach would be to determine the height-based IBW before the amputation, then subtract the percent of the missing limb based on data from a body segment percentage table. 39 The weight used would be the lesser of this adjusted IBW or the ABW. The average weight of body segments of a 68-kg (150-lb) man are: upper limb, 4.9%; entire lower limb, 15.6%; thigh, 9.7%; leg (below knee), 4.5%; and foot, 1.4%. This suggested method for IBW calculation in amputees has not been validated for its utility in predicting Cl in these patients.

8 16 CLINICAL PHARMACOKINETICS SPINAL CORD INJURY Cl estimation in patients with spinal cord injury appears to be unpredictable (overprediction) because of the loss in muscle mass that occurs over time after the injury. Accuracy has been reported in some patient populations (paraplegics with good renal function); however, the 4-hr Cl measurement should be used in patients with spinal cord injuries if accuracy is necessary. 40,41 CHRONIC RENAL INSUFFICIENCY As already noted, with declining renal function S may become a less accurate indicator of renal function (GFR) due to the increasing percentage of tubular secretion of creatinine in relation to the total urinary excretion. That is, as renal function decreases, tubular secretion becomes a larger part of creatinine elimination. In addition, there appears to be an extrarenal route of creatinine elimination via the gastrointestinal tract in uremic patients. These patients may have poor dietary intake and reduced muscle mass as well., Use of oral cimetidine (800 mg every for three doses) to inhibit tubular secretion of creatinine prior to urine collection or S measurement has been advocated to improve estimation of GFR Ranitidine and famotidine have not been shown to inhibit the tubular secretion of creatinine and should not be used for this purpose. 43 Again, if the Cl is used to determine doses, it may not be necessary to make such changes for estimating GFR, since drug dosing studies have traditionally used Cl estimates. DIALYSIS Estimating Cl in patients receiving dialysis is problematic and not recommended. A patient without functioning kidneys has no glomerular filtration. Thus, the S concentration becomes primarily a function of the dialysis procedure rather than the patient s kidney function. Because no urine output indicates no renal function, monitoring residual urine output gives some idea whether the patient s kidneys have any potential role in drug elimination. LIVER DISEASE Estimation of renal function in patients with cirrhosis presents certain dilemmas. Estimates of GFR using Cl estimations (Cockcroft-Gault) and 4-hr Cl measurements may be unreliable as shown by inulin clearance tests in these patients. The GFR estimating equations (CKD-EPI, MDRD) have also been found to be inaccurate. 45 These patients should have their drug therapy monitored more carefully, particularly for drugs with a narrow therapeutic index, where drug concentration monitoring is recommended whenever possible PEDIATRICS Cl estimations in children have been shown to reasonably accurately predict GFR. Because a measured 4-hr Cl is as difficult and time-consuming as in adults, equations have been developed for rapid estimation based on a patient s height and weight. These estimates appear most accurate for patients of average weight for their size. The equations use S and height (body length) to estimate the normalized Cl (as if BSA was 1.73 m ). The correlation of a child s muscle mass with his or her height helps factor in the relationship between muscle mass and creatinine production. However, these estimations may be less accurate for children who are significantly under- or overweight for their height and in the first week of life when the serum creatinine of an infant still reflects maternal serum creatinine and renal function is immature. 5 A commonly used equation (the first equation in Table -) for children 1 18 yr of age was developed by Traub and Johnson. 53 For infants, children, and adolescents, specific equations shown in Table - have also been used. 5 The IDMS creatinine calibration introduces a potentially unacceptable amount

9 CHAPTER - Estimating eatinine Clearance 17 of error (Cl appears higher than it is) with use of the Jaffe method of analysis. The following equation can be used only if the laboratory is employing the enzymatic assay. 11,54-56 Using IDMS Calibrated eatinine in Children Enzymatic Assay Only 11,54-56 GFR (ml/min/1.73 m ) = 0.41 (H/S ) where S is in mg/dl and H in centimeters. In international units where S is in µmol/l, the equation would be: GFR (ml/min/1.73 m ) = 36. (H/S ) All of these equations provide values that are automatically normalized to 1.73 m. To calculate the actual Cl (in ml/min), the equation result is multiplied by the patient s BSA and divided by TABLE -. CREATININE CLEARANCE ESTIMATION IN CHILDREN a,11 Age Range (Years) Standard Units b International Standardized Units c 1 to Cl (ml / min/ 1.73 m ) = ( 0.48)( H) ( S ) Cl (ml/min/1.73 m ) = ( 4.4)( H) ( S ) <1 5 Preterm infant GFR (ml / min/ 1.73 m ) = ( 0.33)( H) ( S ) GFR (ml/min/1.73 m ) = ( 9.)( H) ( S ) Full-term infant Cl (ml / min/ 1.73 m ) = ( 0.45)( H) ( S ) Cl (ml/min/1.73 m ) = ( 39.8)( H) ( S ) 1 to 1 5 Cl (ml / min/ 1.73 m ) = ( 0.55)( H) ( S ) Cl (ml/min/1.73 m ) = ( 48.6)( H) ( S ) 13 to 1 (girls) 5 GFR (ml / min/ 1.73 m ) = ( 0.55)( H) ( S ) GFR (ml/min/1.73 m ) = ( 48.6)( H) ( S ) ( )( ) 13 to 1 (boys) H GFR (ml / min/ 1.73 m ) = ( S ) GFR (ml/min/1.73 m ) = ( 61.9)( H) ( S ) a Developed using non-idms calibrated creatinine assay; results now likely to have a positive bias. b S is in mg/dl and H in cm. c S is in µmol/l and H in cm.

10 18 CLINICAL PHARMACOKINETICS PATIENTS WITH UNSTABLE RENAL FUNCTION The discussed equations for Cl estimation are based on patients with stable renal function with S at steady state. In patients with changing renal function, S may not reflect the current function for several days. Using a value of S that is not at steady state to estimate Cl may significantly over- or underestimate the patient s renal function and result in inappropriate drug dosing. ESTIMATING TIME TO STEADY STATE SERUM CREATININE CONCENTRATION The time to a steady state S value increases as the patient s renal function declines. Using the calculation below, the time to ~90% of steady state (3.3 half-lives) is estimated to be 6 hr, 53 hr, 5.6 days as a patient s renal function rapidly declines to 50%, 5%, and 10% of normal (70 kg, Cl = 10 ml/ min), respectively. It is estimated that if the S increases by more than 0. mg/dl in 8, steady state probably has not been reached. 57 Determining a rough estimate of the half-life and time to steady state of creatinine in a specific adult patient may be made by assuming that the Cockcroft-Gault equation is accurate in estimating Cl, that the patient s volume of distribution (V) of creatinine is ~0.6 L/kg of IBW 57,58 and making the following calculations: 1. Estimate the patient s Cl (ml/min) based on the Cockcroft-Gault equation.. Convert the patient s Cl in milliliters per minute (ml/min) to liters per hour (L/hr) by multiplying by 0.06 (because of the following conversion): 60 min 1L = 0.06 (units will cancel appropriately below) 1 hr 1000 ml Thus, ml/min 0.06 = L/hr. 3. Use the relationship K = CL/V to calculate K, the elimination rate constant, where CL is the creatinine clearance in liters per hour (L/hr), and V is the volume of distribution of creatinine in liters (0.6 L/ kg IBW). 4. Use t 1/ = 0.693/K to estimate the half-life. 5. Approximately 3.3 times the estimated half-life equals time to steady state (90% achieved). CREATININE CLEARANCE ESTIMATION IN UNSTABLE RENAL FUNCTION Several equations for use in patients with unstable renal function are available and may be useful for initial drug dosing; however, estimating Cl in a patient with changing renal function can be problematic and drug concentration monitoring is recommended for drugs with a narrow therapeutic index. 1 3,57,58 For 4-hr Cl measurements in patients with changing renal function, use the midpoint (1th hr) S or the average of the S at the beginning and the end of the 4-hr urine collection. 58 The following equation has been used to estimate Cl in patients with unstable renal function 59 : ( ) ( ) ( ) Age S 1+ S + 49 S1- S / t Cl ( males) (ml/min) = ( S 1+ S ) where t is the change in time in number of days between measurement of S 1 and S. Cl (females) = Cl (males) 0.86 Other methods of Cl estimation in unstable renal function using mass balance (estimated creatinine production versus creatinine excretion) may be useful and have been reviewed Various internet

11 CHAPTER - Estimating eatinine Clearance 19 Cl and GFR calculators are available on line. General principles from the literature noted above and/ or institutional protocols, along with documentation of the specific equations and patient factors utilized, should guide the usage of these calculators. REFERENCES 1. Lam YW, Banerji S, Hatfield C, et al. Principles of drug administration in renal insufficiency. Clin Pharmacokinet. 1997;3: Duarte CG, Preuss HG. Assessment of renal function glomerular and tubular. Clin Lab Med. 1993;13(1): Robert S, Zarowitz BJ. Is there a reliable index of glomerular filtration rate in critically ill patients? DICP Ann Pharmacother. 1991;5: Stevens LA, Levey AS. Measurement of kidney function. Med Clin N Am. 005;89: Ducharme MP, Smythe M, Strohs G. Drug-induced alterations in serum creatinine concentrations. Ann Pharmacother. 1993;7: Myers GL, Miller WG, Coresh J, et al. Recommendations for improving serum creatinine measurement: a report from the Laboratory Working Group of the National Kidney Disease Education Program. Clin Chem. 006;5: Hanser AM, Parent X. Interference of clavulanic acid in the assay of creatinine using the Jaffe method on the Dade/RXL analyzer. Ann Biol Clin. 000;58(6): Pline KA, Smith CL. The effect of creatine intake on renal function. Ann Pharmacother. 005;39: Jagim AR, Oliver JM, Sanchez A, et al. A buffered form of creatine does not promote greater changes in muscle creatine content, body composition, or training adaptations than creatine monohydrate. J Int Soc Sports Nutr. 01;9(1): Spillane M, Schoch R, Cooke M, et al. The effects of creatine ethyl ester supplementation combined with heavy resistance training on body composition, muscle performance, and serum and muscle creatine levels. J Int Soc Sports Nutr. 009;6: National Institutes of Health. National Institute of Diabetes and Digestive and Kidney Diseases. Health Information. Health Communication Programs. National Kidney Disease Education Program. Laboratory Evaluation. Glomerular Filtration Rate (GFR). eatinine Standardization. eatinine Standardization Recommendations. Available at: aspx. Accessed November 30, National Institutes of Health. National Institute of Diabetes and Digestive and Kidney Diseases. Health Information. Health Communication Programs. National Kidney Disease Education Program. NKDEP Health Topics A-Z. Chronic Kidney Disease and Drug Dosing: Information for Providers. Available at: niddk.nih.gov/health-information/health-communication-programs/nkdep/a-z/ckd-drug-dosing/pages/ckddrug-dosing.aspx. Accessed November 30, Levey AS, Coresh J, Greene T, et al. Using standardized serum creatinine values in the Modification of Diet in Renal Disease Study equation for estimating glomerular filtration rate. Ann Intern Med. 006;145: Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976;16: Lott RS, Hayton WL. Estimation of creatinine clearance from serum creatinine concentration a review. Drug Intell Clin Pharm. 1978;1: Jelliffe RW. eatinine clearance: a bedside estimate. Ann Intern Med. 1973;79: Dowling TC, Matzke GR, Murphy JE, et al. Evaluation of renal drug dosing: Prescribing information and clinical pharmacist approaches. Pharmacotherapy. 010;30: Park EJ, Wu K, Mi Z, et al. A systematic comparison of Cockcroft-Gault and Modification of Diet in Renal Disease equations for classification of kidney dysfunction and dosage adjustment. Ann Pharmacother. 01;46: Dowling TC, Wang ES, Ferrucci L, et al. Glomerular filtration rate equations overestimate creatinine clearance in older individuals enrolled in the Baltimore longitudinal study on aging: impact on renal drug dosing. Pharmacotherapy. 013; 33(9): Devine BJ. Gentamicin therapy. Drug Intell Clin Pharm. 1974;7: Haycock GB, Schwartz GJ, Wisotsky DH. Geometric method for measuring body surface area: a height-weight formula validated in infants, children, and adults. J Pediatr. 1978;93:6-6.. Spinler SA, Nawarskas JJ, Boyce EG, et al. Predictive performance of ten equations for estimating creatinine clearance in cardiac patients. Iohexol Cooperative Study Group. Ann Pharmacother. 1998;3: Inker LA, Schmid CH, Tighiouart H, et al. Estimating glomerular filtration rate from serum creatinine and cystatin C. N Engl J Med. 01;367: Hudson JQ, Bean JR, Burger CF, et al. Estimated glomerular filtration rate leads to higher drug dose recommendations in the elderly compared with creatinine clearance. Int J Clin Pract. 015; 69: National Institutes of Health. National Institute of Diabetes and Digestive and Kidney Diseases. Health Information. Health Communication Programs. National Kidney Disease Education Program. Laboratory Evaluation. Estimating Glomerular Filtration Rate (GFR). Available at:

12 0 CLINICAL PHARMACOKINETICS information/health-communication-programs/nkdep/lab-evaluation/gfr/estimating/pages/estimating. aspx#when-not-to-use-the-mdrd. Accessed November 30, Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 009;150: Matzke, GR, Aronoff, GR; Atkinson, AJ, et al. Drug dosing consideration in patients with acute and chronic kidney disease: a clinical update from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int. 011;80: Earley A, Miskulin D, Lamb EJ, et al. Estimating equations for glomerular filtration rate in the era of creatinine standardization: a systematic review. Ann Intern Med. 01;156: Udy AA, Morton FJA, Nguyen-Pham S, et al. A comparison of CKD-EPI estimated glomerular filtration rate and measured creatinine clearance in recently admitted critically ill patients with normal plasma creatinine concentrations. BMC Nephrol. 013;14: Boyce EG, Dickerson RN, Cooney GF, et al. eatinine clearance estimation in protein-malnourished patients. Clin Pharm. 1989;8: Sawyer WT, Canaday BR, Poe TE, et al. Variables affecting creatinine clearance prediction. Am J Hosp Pharm. 1983;40: Wilhelm SM, Kale Pradhan PB. Estimating eatinine Clearance: A Meta analysis. Pharmacotherapy. 011;31: Salazar DE, Corcoran GB. Predicting creatinine clearance and renal drug clearance in obese patients from estimated fat-free body mass. Am J Med. 1988;84: Demirovic JA, Pai AB, Pai MP. Estimation of creatinine clearance in morbidly obese patients. Am J Health Syst Pharm. 009;66: Nyman HA, Dowling TC, Hudson JQ, et al. Comparative evaluation of the Cockcroft Gault equation and the Modification of Diet in Renal Disease (MDRD) Study equation for drug dosing: An Opinion of the Nephrology Practice and Research Network of the American College of Clinical Pharmacy. Pharmacotherapy. 011;31: Park EJ, Pai MP, Dong T, et al. The influence of body size descriptors on the estimation of kidney function in normal weight, overweight, obese, and morbidly obese adults. Ann Pharmacother. 01;46: Reichley RM, Ritchie DJ, Bailey TC. Analysis of various creatinine clearance formulas in predicting gentamicin elimination in patients with low serum creatinine. Pharmacotherapy. 1995;15: Smythe M, Hoffman J, Kizy K, et al. Estimating creatinine clearance in elderly patients with low serum creatinine concentrations. Am J Hosp Pharm. 1994;51: Brunnstrom S. Clinical Kinesiology. 4th ed. Philadelphia, PA: F.A. Davis Co.;1983: Mohler JL, Ellison MF, Flanigan RC. eatinine clearance prediction in spinal cord injury patients: comparison of 6 prediction equations. J Urol. 1988;139: Thakur V, Reisin E, Solomonow M, et al. Accuracy of formula-derived creatinine clearance in paraplegic subjects. Clin Nephrol. 1997;47: Walser M. Assessing renal function from creatinine measurements in adults with chronic renal failure. Am J Kidney Dis. 1998;3: Kemperman FAW, Krediet RT, Arisz, L. Formula-derived prediction of the glomerular filtration rate from plasma creatinine concentration. Nephron. 00;91: Ixkes MCJ, Koopman MG, van Acker AC, et al. Cimetidine improves GFR-estimation by the Cockcroft and Gault formula. Clin Nephrol. 1997;47: Gerhardt T, Pöge U, Stoffel-Wagner B, et al. eatinine-based glomerular filtration rate estimation in patients with liver disease: the new Chronic Kidney Disease Epidemiology Collaboration equation is not better. Eur J Gastroenterol Hepatol. 011;3(11): DeSanto NG, Anastasio P, Loguercio C, et al. eatinine clearance: an inadequate marker of renal filtration in patients with early posthepatitic cirrhosis (Child A) without fluid retention and muscle wasting. Nephron. 1995;70: Papadakis MA, Arieff AI. Unpredictability of clinical evaluation of renal function in cirrhosis. Am J Med. 1987;8: Hull JH, Hak LJ, Koch GG, et al. Influence of range of renal function and liver disease on predictability of creatinine clearance. Clin Pharmacol Ther. 1981;9: Pachorek RE, Wood F. Vancomycin half-life in a patient with hepatic and renal dysfunction. Clin Pharm. 1991;10: Lam NP, Sperelakis R, Kuk J, et al. Rapid estimation of creatinine clearances in patients with liver dysfunction. Dig Dis Sci. 1999;44: Orlando R, Floreani M, Padrini R, et al. Evaluation of measured and calculated creatinine clearances as glomerular filtration markers in different stages of liver cirrhosis. Clin Nephrol. 1999;51: Schwartz GJ, Brion LP, Spitzer A. The use of plasma creatinine concentration for estimating glomerular filtration rate in infants, children, and adolescents. Pediatr Clin North Am. 1987;34: Traub SL, Johnson CE. Comparison of methods of estimating creatinine clearance in children. Am J Hosp Pharm. 1980;37:

13 CHAPTER - Estimating eatinine Clearance Schwartz GJ, Munoz A, Schneider MF, et al. New equations to estimate GFR in children with CKD. J Am Soc Nephrol. 009;0: Staples A, LeBlond R, Watkins S, et al. Validation of the revised Schwartz estimating equation in a predominantly non-ckd population. J Am Soc Nephrol. 010;5: Lee CK, Swinford RD, Cerda RD, et al. Evaluation of serum creatinine concentration-based glomerular filtration rate equations in pediatric patients with chronic kidney disease. Pharmacotherapy. 01;3(7): Winter ME. eatinine Clearance. In: Winter ME, ed. Basic Clinical Pharmacokinetics. 5th ed. Baltimore, MD: Lippincott Williams and Wilkins; 010: Chow MS, Schweizer R. Estimation of renal creatinine clearance in patients with unstable serum creatinine concentrations: comparison of multiple methods. Drug Intell Clin Pharm. 1985;19: Brater DC. Drug Use in Renal Disease. Balgowlah, Australia: ADIS Health Science Press; 1983: Jelliffe R. Estimation of creatinine clearance in patients with unstable renal function, without a urine specimen. Am J Nephrol. 00;: Jelliffe R. Optimal methodology is important for optimal pharmacokinetic studies, therapeutic drug monitoring and patient care. Clinical Pharmacokinet. 015;54: Bouchard J, Macedo E, Soroko S, et al. Comparison of methods for estimating glomerular filtration rate in critically ill patients with acute kidney injury. Nephrol Dial Transplant. 010;5: Chen S. Retooling the creatinine clearance equation to estimate kinetic GFR when the plasma creatinine is changing acutely. J Am Soc Nephrol. 013;4: Winter MA, Guhr KN, Berg GM. Impact of various body weights and serum creatinine concentrations on the bias and accuracy of the Cockcroft-Gault equation. Pharmacotherapy. 01;3:604-1.

14

15 RENAL DRUG DOSING CONCEPTS CHAPTER 3 Dean A. Van Loo and Thomas C. Dowling Chronic kidney disease (CKD) is a progressive consequence of systemic diseases such as diabetes and hypertension as well as localized kidney injury as the result of glomerulonephritis. Over 500,000 patients in the United States have stage 5 CKD, which is also categorized as end-stage renal disease (ESRD). Each year, for the last several decades, up to 100,000 patients have developed ESRD and over 80,000 have died. 1 Chronic renal replacement therapy, whether peritoneal or hemodialysis (HD), was life-sustaining for over 600,000 patients in 011 at a total cost of over $49 billion USD. A significant portion of patients who receive a kidney transplant continue on to develop CKD. Most stage 1 to 4 CKD patients are initially identified in primary care clinics, while others are identified in acute care environments. Populationbased studies, such as NHANES, report that the prevalence of CKD is increasing dramatically, with more than 50% of U.S. adults aged 30 to 64 expected to develop CKD in their lifetime. Kidney failure can also appear abruptly, with some patients presenting with acute kidney injury (AKI) in emergency departments, clinical wards, or intensive care units. 3 The majority of AKI cases are attributed to drug therapy or renal hypoperfusion in hospitalized patients, which often requires continuous renal replacement therapies (CRRT). Regardless of the cause of acute or chronic renal impairment, these patients are at increased risk of accumulating drugs, toxic metabolites, and other nephrotoxins. For any drug that relies extensively on the kidney for elimination from the body (i.e., renal clearance > 30% of total clearance) and drug concentrations in blood or plasma are clearly associated with a pharmacodynamic effect (success, failure, or toxicity), dose adjustments are necessary when renal function is considerably reduced. The aim of this chapter is to describe dosing strategies for patients with CKD, AKI, and those receiving renal replacement therapies on an intermittent and/or continuous basis. CLINICAL ASSESSMENT OF KIDNEY FUNCTION The indices of glomerular and tubular function most widely utilized clinically include daily urinary protein excretion rate (glomerular), urine albumin-creatinine ratio (glomerular), fractional excretion of sodium (tubular), and serum creatinine concentration (glomerular and tubular). eatinine is excreted by glomerular filtration and tubular secretion, making creatinine clearance (Cl) a composite index of renal function that has been strongly associated with the total and renal clearance of many drugs that are eliminated by the kidney and is the primary index of renal drug dosing in FDA product labeling. In patients with CKD stages 1 through 5 (pre-dialysis), the Cockcroft-Gault (CG) equation (see Chapter ) is commonly used to estimate Cl in the presence of stable kidney function. Newer equations that estimate GFR (egfr), such as the CKD-EPI equations, are most appropriately used for identifying CKD and staging their degree of CKD severity. 4 Although the Modification of Diet in Renal Disease (MDRD) equation was initially adopted into automated systems for reporting GFR in clinical settings, it has been shown to be largely inaccurate at GFR > 60 ml/min and has since been replaced by the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation. Neither of these egfr equations has been consistently demonstrated to be equivalent to CG or measured Cl when adjusting drug doses for renal impairment. 5,6 Recent studies by the Food and Drug Administration (FDA) and others showed that egfr equations yield significantly higher estimates of kidney function, and significantly different dose calculations, when compared to CG equation, particularly in elderly individuals and those receiving narrow therapeutic index drugs such as enoxaparin. 5-9 Thus, renal dosing practices should remain consistent with the original pharmacokinetic studies of a particular drug in CKD, which to date generally involves estimation of Cl. 3

16 4 CLINICAL PHARMACOKINETICS Quantification of renal function in patients with AKI, where renal function and serum creatinine values are rapidly changing, is a challenging situation. Here, numerous equations for estimating Cl based on two non-steady-state serum creatinine values have been proposed. See Chapter for further discussions of appropriate use of equations to quantify renal function in various situations and patient populations. For critically ill patients with AKI receiving CRRT, estimation of both residual renal function (Cl) and CRRT clearance are required for dose individualization (see section on dosing strategies). 10,11 MECHANISMS OF DRUG CLEARANCE Renal elimination The process of renal drug elimination is a composite of glomerular and tubular functions, with the amount of drug cleared by the kidney (A c ) described by the following equation: A c = A filt + A sec A reabs (Eq. 1) Initially, unbound drug is filtered through the glomerulus (A filt ) into the proximal tubular fluid. When in the tubule, filtered drug may then be passively or actively reabsorbed (A reabs ) back into the bloodstream. This reabsorptive process is rare and occurs primarily in distal segments for unionized drugs at low urine flow rates. Drugs may also undergo active tubular secretion (A sec ), where unbound drug in plasma is transported into the tubular cell. This process of secreting drugs into the urine is mediated by transporters such as the organic anionic transporter (OAT), organic cationic transporter (OCT), or p-glycoprotein (P-GP). These transporters act in an efflux and uptake manner and are located along the basolateral and apical membranes of the proximal tubule The pathways work together to form an extremely efficient process of detoxification, resulting in renal clearance values that can exceed GFR, and in some cases approach renal plasma flow, which can be observed with para-aminohippurate and several penicillins. As filtration capacity (measured as GFR) progressively diminishes in CKD, some experimental data suggest that tubular secretory mechanisms may maintain their functionality, thereby providing significant renal clearance for some drugs even in the presence of severe glomerular damage. 15 Kidney diseases can affect both glomerular and tubular function, leading to reduced overall drug elimination. As destruction of nephrons progresses, it has traditionally been believed that the function of all segments of the remaining nephrons is affected equally. 16 Based on this assumption, the rate of drug excretion in the normal or diseased kidney can be estimated by GFR or Cl, which are predominantly measures of glomerular function. 17 The total renal clearance of a drug from the body also depends on (1) the fraction of the drug eliminated unchanged by the normal kidney, () the renal mechanisms involved in drug elimination, and (3) the degree of functional impairment of each of these pathways. The fraction of unchanged drug eliminated renally (f e ) and an assessment of the relationship between renal function and the drug s parameters, such as half-life (t½), total clearance (CL), and renal clearance (CL Renal ), can be used to individualize drug therapy. Ideally, renal drug clearance is determined by quantifying the amount of drug excreted in urine relative to the area under plasma drug concentration versus time curve (AUC) of drug in plasma, and renal function is measured using a GFR method such as iohexol or iothalamate clearance. 18 More commonly, the relationship between Cl and drug clearance (CL) is evaluated in a large patient population with varying renal function, as follows: CL = (A Cl) + B (Eq. ) k = (A Cl) + B (Eq. 3) where A is the slope of the linear relationship between Cl and either CL or k (the elimination rate constant), and B is the nonrenal CL (CL NR ) or nonrenal k (k NR ), respectively. This drug-specific information can then be used to design dose adjustment strategies in patients with renal insufficiency to minimize drug toxicity and optimize therapeutic efficacy.