Blood urea nitrogen (BUN) and serum creatinine concentrations

Transcription

1 J Vet Intern Med 2001;15: Relationship between Plasma Iohexol Clearance and Urinary Exogenous Creatinine Clearance in Dogs Delmar R. Finco, W. Emmett Braselton, and Tanya A. Cooper The objective of this study was to determine if plasma iohexol clearance, computed by a 1-compartment model defined by 3 plasma samples, was an accurate measure of glomerular filtration rate (GFR) in dogs. Twenty-two adult Beagle dogs of both genders were studied. Ten dogs had intact kidneys, and 12 dogs had surgically reduced renal mass. A bolus injection of iohexol was made, and blood was obtained for plasma iohexol assay after 120, 180, and 240 minutes. Plasma was analyzed for iohexol concentration by means of 3 assay methods: chemical, high-performance liquid chromatography (HPLC), and inductively coupled plasma emission spectroscopy (ICP). Urinary clearance of exogenous creatinine was used to measure GFR for three 30-minute periods occurring between 150 and 240 minutes after iohexol injection. Plasma clearance of iohexol and renal clearance of creatinine were compared by linear regression analysis and by limits of agreement techniques. Plasma iohexol clearance and urinary exogenous creatinine clearance were significantly correlated (chemical R 2.90; HPLC R 2.96; and ICP R 2.96). The 1-compartment iohexol clearance : exogenous creatinine clearance ratios were , , and for the chemical, HPLC, and ICP methods of assay, respectively, indicating that plasma iohexol clearance slightly overestimated GFR. Assuming a 2 standard deviation interval for error, corrected plasma iohexol clearance measured GFR with 34% accuracy for the chemical, 26% accuracy for the HPLC, and 27% accuracy for the ICP method. These results indicate that plasma iohexol clearance should have utility for detection of renal dysfunction earlier in the course of progressive renal disease than is possible with measurement of plasma creatinine or urea concentrations. Key words: Glomerular filtration rate; 1-Compartment model; Pharmacokinetics; Renal disease. Blood urea nitrogen (BUN) and serum creatinine concentrations are measured to evaluate renal function in clinical patients, but these tests are relatively insensitive. 1 It is generally acknowledged that measurement of glomerular filtration rate (GFR) is the best single test of renal function. The classic method for measuring GFR utilizes inulin as a test substance and entails the determination of urinary clearance of inulin while plasma concentration is kept constant. Other test substances also have been used to measure GFR. In both normal dogs and those with reduced renal function, exogenous creatinine clearance has been found to be a valid test for measuring GFR. 2,3 Despite greater sensitivity than measurement of BUN and serum creatinine concentrations, urinary clearance methods seldom are used clinically because they are time-consuming and require carefully conducted urine collections. In theory, plasma clearance of a material can be used to measure GFR, provided that certain assumptions are fulfilled. The test material, in addition to meeting requirements for measuring GFR by renal clearance methods, must not be subjected to removal from plasma by extrarenal mechanisms. In addition, the validity of the pharmacokinetic model used for computations of clearance must be established. After a single intravenous injection of test substance, several blood samples are taken at timed intervals in order to characterize the decay curve accurately. If sample num- From the Department of Physiology and Pharmacology, College of Veterinary Medicine, The University of Georgia, Athens, GA (Finco, Cooper); and the Department of Pharmacology and Toxicology, Animal Health Diagnostic Laboratory, College of Veterinary Medicine, Michigan State University, East Lansing, MI (Braselton). Reprint requests: Dr Delmar R. Finco, University of Georgia CVM, Department of Physiology and Pharmacology, Athens, GA ; dfinco@vet.uga.edu. Submitted July 17, 2000; Revised December 28, 2000; Accepted February 6, Copyright 2001 by the American College of Veterinary Internal Medicine /01/ /$3.00/0 bers are limited for clinical expediency, clearance computations derived from limited sampling should be validated by comparison with urinary clearance of an appropriate test substance. Plasma clearance of several compounds has been advocated for measurement of GFR. Plasma clearance of iohexol, an iodine-containing agent used for contrast radiography, has been studied extensively in humans and pigs. 4,5 Several studies are reported on measurement of plasma iohexol clearance in dogs, 6 10 but validation by direct comparison with urinary clearance has been limited to 1 study, which utilized an apparatus for iohexol assay that is no longer available. 9 The purpose of the present study was to make simultaneous measurements of urinary clearance of exogenous creatinine and plasma clearance of iohexol in order to determine if iohexol clearance, as determined by a 1-compartment, 3-sample method of pharmacokinetic analysis, was a reliable marker of GFR. Materials and Methods Dogs Twenty-two Beagle dogs weighing kg and of both genders were procured from commercial suppliers and housed in USDA-approved facilities under conditions of controlled climate and lighting. Ten dogs were used for study without modification of renal function. Twelve dogs had renal mass surgically reduced at least 6 months before the present study, as part of unrelated studies. Dogs were fed commercial dog foods, had water available ad libitum, and were observed daily for clinical abnormalities. Iohexol Clearance Procedures Dogs were placed in Pavlov slings, where they rested quietly without use of chemical restraint. Each dog had a catheter placed in a jugular vein and a saphenous vein. After withdrawing a plasma sample (blank plasma required for chemical assay for iohexol I), iohexol a was administered via the saphenous vein catheter over a 30-second period. The completion of the injection represented time zero. Iohexol dosage

2 Iohexol and Urinary Exogenous Creatinine Clearance in Dogs 369 was measured precisely by volume and weight; dosage was approximately 1.0 ml/kg body weight in dogs with intact kidneys and 0.5 ml/kg in dogs with reduced renal mass. The same lot of injection solution was used for all clearance determinations and as a reference compound for all assay methods. Heparinized syringes were used to obtain blood samples from the jugular catheter at approximately 120, 180, and 240 minutes; precise times of withdrawal (minute, seconds) were recorded. Plasma was harvested and stored at 80 C until analyzed for iohexol I. Exogenous Creatinine Clearance Procedures At 110 minutes after injection of iohexol, a bolus injection of sterile creatinine solution (25 mg/ml) was made via the saphenous vein catheter. Dogs with intact kidneys received approximately 3 ml/kg body weight, and dogs with reduced renal mass received approximately 2.5 ml/kg body weight. Plasma creatinine concentration was sustained for the duration of urinary clearance procedures by injecting a creatininesaline solution at a rate of 0.5 ml/min with a constant infusion pump. The creatinine-saline solution used had a creatinine concentration of 9 mg/ml in renal-intact dogs and 6 mg/ml in dogs with reduced renal mass. A catheter was passed into the urinary bladder, and the bladder was emptied and rinsed with several 30-mL aliquots of sterile saline just before starting the 1st urine collection at 150 minutes. Thirty-minute urine collections were completed at 180, 210, and 240 minutes, with the urinary bladder rinsed just before the end of each period to ensure accurate collection of all creatinine filtered. The volumes of urine and rinses collected during each 30-minute period were measured and recorded. Blood was drawn from the jugular catheter into heparinized syringes at 150, 180, 210, and 240 minutes. The creatinine concentration of collected samples was determined with a semiautomated analyzer. b Serial Iohexol Clearance Determinations At least 2 weeks after iohexol-exogenous creatinine comparisons, 1 normal dog and 2 dogs with reduced renal mass had 3 serial iohexol clearance procedures performed without simultaneous urinary creatinine clearance procedures. Three or 4 days elapsed between serial determinations. These tests were performed to determine repeatability of results and to determine if any changes in iohexol clearance occurred with repeated injections of the test material. Iohexol Assay Iohexol in plasma was assayed by 3 methods. The chemical method entails deiodination of the iohexol by alkaline hydrolysis and measurement of iodine by the ceric arsenite method. This assay was performed as previously described 11 with minor modifications, which included use of standards representing 20, 80, and 160 g of I/mL, use of 25 L of plasma for assay, and use of 110 L of hydrolysate, 2.2 ml of working solution, and 165 L of ceric ammonium sulfate for the colorimetric assay. Plasma samples with iohexol iodine concentrations 160 g/ml were diluted with appropriate amounts of plasma obtained from the same animal before iohexol injection. Iohexol also was assayed by inductively coupled plasma emission spectroscopy (ICP) as previously described. 12 High-performance liquid chromatography (HPLC) was used to measure iohexol as previously described 13 with the following minor modifications. The diluted serum was filtered through an Acrodisc M Versapar filter c before application to a 125- by 4.6-mm 5- ODS Prodigy HPLC column. d This column separated iohexol into 2 isomers; peak area was determined on the largest (late-emerging) isomer. Computation of Clearances A 1-compartment model was used for computation of iohexol clearance by means of plasma iohexol concentrations obtained by each of Table 1. Repeatability of iohexol clearance values (ml/ min/kg) when serial determinations were done on 3 dogs. Clearance values were computed from results of 3 methods of plasma assay. Dog 1 Dog 2 Dog 3 Day Assay Method Chemical HPLC ICP HPLC, high-performance liquid chromatography; ICP, inductively coupled plasma emission spectroscopy. the assay methods. The area under the curve (AUC) was calculated by the formula AUC C 0 /k where C 0 plasma iohexol I concentration at zero time as determined by extrapolation of the line of best fit from 120-, 180-, and 240-minute samples, and k the elimination rate constant (slope) of the decay curve. Clearance was calculated as dose of iohexol I/AUC. An R 2 value of.98 or greater for the 3-point elimination phase was required for acceptance of the clearance value on each dog. The standard urinary clearance formula [C (U V U C )/P C ], where U V urine volume, U C urine creatinine concentration, and P C plasma creatinine concentration, was used to compute creatinine clearance for each of the 30-minute collection periods. The mean of plasma creatinine concentration determined at the beginning and end of each 30-minute clearance period was used in the denominator of the clearance equation. The mean of the 3 clearance values for each dog was used for comparison with iohexol clearance. For both plasma and urine clearance computations, clearance values were expressed as milliliter per minute per kilogram of body weight. Statistical Methods and Data Analysis The relationship between urinary creatinine clearance and plasma iohexol clearance was examined for each method of iohexol assay by computer modeling to determine the best fit of data. Linear, logarithmic, inverse, quadratic, cubic, compound, power, S, growth, exponential, and logistic models were examined. Nonlinear models gave lower or only slightly higher R 2 values than the linear model, and subsequent analyses were restricted to a linear model. The equation derived from linear regression of exogenous creatinine clearance versus iohexol clearance was used to correct individual iohexol clearance values (see Results). Corrected values were used to determine percentage difference between creatinine clearance and iohexol clearance, and percentage difference was plotted against exogenous creatinine clearance. Data are presented as mean standard deviation, and a significance level of P.05 was used when statistical evaluations were made. Results No adverse clinical signs were observed during or after injection or infusion of test substances. No decline in clearance values was observed on serial determinations of iohexol clearance (Table 1). Values for exogenous creatinine clearance ranged from 0.45 to 5.59 ml/min/kg body weight. After iohexol injec-

3 370 Finco, Braselton, and Cooper tion, plasma iohexol iodine concentration ranged from approximately 393 (120 minutes) to 12 g/ml (240 minutes). In comparing assay methods, values for iohexol iodine by the HPLC method were of values for the chemical method and of the ICP method; ICP values were of the chemical method. For the chemical method, the respective intra- and interassay coefficients of variation of assay standards were 20 g 2.8%, 0.4%; 80 g 2.4%, 2.0%; and 160 g %, 4.5%. Regression data computed from the 3-point elimination phase were acceptable (R 2.98) on data from all dogs when iohexol was measured by the ICP and HPLC methods and on 19 of 22 dogs when measured by the chemical method. A statistically significant linear correlation was found between exogenous creatinine clearance and plasma clearance of iohexol when iohexol was measured by the chemical (R 2.90), HPLC (R 2.96), and ICP (R 2.96) methods for iohexol assay (Fig 1). The iohexol clearance : exogenous creatinine clearance ratios were , , and for the chemical, HPLC, and ICP methods of assay, respectively. Linear regression equations describing the relationship between iohexol and creatinine clearance (GFR) values were Chemical (GFR ) HPLC (GFR ) ( ) ICP (GFR ) ( ) (1) To correct the iohexol values and equate them to GFR values, formulas given above (Equation 1) were used to derive the following conversions: Corrected chemical (observed chemical ) Corrected HPLC [observed HPLC ( )/1.0539] Corrected ICP [observed ICP ( )/1.1547] (2) The percentage difference between corrected iohexol clearance and urinary creatinine clearance (GFR) for each dog was computed as (corrected iohexol clearance exogenous creatinine clearance) exogenous creatinine clearance (3) A plot of percentage difference versus exogenous creatinine clearance resulted in a pattern in which the scatter of points for percentage difference was rather consistently distributed over the wide range of GFR values (Fig 2), indicating a similar agreement between the 2 tests over the wide range of GFR values tested. Assuming a 2 standard deviation error for iohexol in estimating GFR, the chemical, Fig 1. The relationship between urinary exogenous creatinine clearance and plasma iohexol clearance. Graphs depict results from analysis of the same plasma samples for iohexol by chemical (top), high-performance liquid chromatography (HPLC) (middle), and inductively coupled plasma emission spectroscopy (ICP) (bottom) methods of assay. Iohexol clearance and exogenous creatinine clearance were significantly correlated (R 2.90 for chemical,.96 for HPLC, and.96 for ICP assay methods).

4 Iohexol and Urinary Exogenous Creatinine Clearance in Dogs 371 HPLC, and ICP methods measured GFR within approximately 35, 26, and 27% of the actual value, respectively. Serial determination of plasma iohexol clearance values in 3 dogs indicated good agreement between values obtained on different days (Table 1). Discussion Fig 2. Percentage difference between urinary exogenous creatinine clearance and corrected plasma iohexol clearance measurements over the range of exogenous creatinine clearance values tested. Solid lines represent mean, dotted lines represent 2 standard deviations for chemical (top), high-performance liquid chromatography (middle), and inductively coupled plasma emission spectroscopy (bottom) methods of iohexol assay. Exogenous creatinine clearance has been demonstrated to be a reliable method for measuring GFR in both normal dogs and those with reduced renal function. 2,3 Consequently, comparison of plasma iohexol clearance with urinary creatinine clearance should determine the utility of iohexol for estimating GFR. Results from this study indicate a high correlation between results from the 2 procedures. The percentage difference between methods was fairly uniform over the range of clearance values tested, indicating usefulness regardless of the level of renal function. The percentage difference between plasma iohexol clearance and urinary creatinine clearance that was encountered may preclude use of iohexol clearance for precise measures of GFR as required in some research protocols. However, the test may have considerable clinical utility, considering that alternative tests such as BUN and serum creatinine concentrations are incapable of detecting renal dysfunction until GFR is less than 25% of normal. 1 In making comparisons of 2 methods being tested to measure the same variable, the errors inherent in each method must be considered. Urinary clearance of a material freely filtered through the glomerular barrier but not influenced by tubular processes is considered a valid measure of GFR. 2 When plasma and urine concentrations of the material can be measured accurately, the only potential error lies in urine collection. Ideally, urine collections would be made from the renal pelvis to circumvent pooling of urine in the pelvis and ureters, but this preference is not important as long as the plasma concentration of the test substance is constant, and fluctuations in urine flow rate are avoided. With collections of urine from the bladder, procurement of all urine produced during a collection period may present a technical challenge to the neophyte, but the procedure is quickly mastered with appropriate instruction, practice, and attention to detail. Plasma clearance methods, by contrast, provide less of a technical challenge but have several important sources of potential error based on pharmacokinetic principles. 14 Technically, extravasation of test material at the time of injection or errors in timing subsequent blood samples can lead to error. Theoretically, clearance of the test substance must occur only by the process of glomerular filtration if plasma clearance is to be equated with GFR. If small but negligible extrarenal routes of clearance of the test substance occur during conditions of normal renal function, decreased GFR may lead to increased excretion by the extrarenal routes when blood concentrations are sustained during reduced renal function. In such circumstances, the plasma : urine clearance ratio would increase as GFR decreased, due to increased plasma clearance by the nonrenal route. Another potential source of error with plasma clearance methods is the lack of use of an appropriate pharmacokinetic model to estimate plasma clearance or the inaccurate

5 372 Finco, Braselton, and Cooper characterization of actual pharmacokinetics because of insufficient sample numbers. 14 Ideally, multiple plasma samples are obtained after injection of the test substance, usually beginning at 5 minutes and continuing for up to 6 hours, in order to accurately characterize the plasma decay curve for computation of the AUC. With fewer samples used to characterize the decay pattern, errors may occur because the limited data do not accurately depict the AUC. A limited number of samples also may preclude determining whether or not the pharmacokinetic model chosen for computations is appropriate. In this study, because of limited plasma sampling after injection of iohexol, inadequate information was available to differentiate causes of potential error from one another. It was assumed that the 1-compartment model was applicable to all dogs. Values for iohexol clearance were compared to exogenous creatinine clearance by computation of the iohexol : creatinine clearance ratios. These ratios were 1.0 for the 3 methods of iohexol assay, reflecting a slight overestimation of GFR by the iohexol method. This finding could be explained by choice of the 1-compartment model used for computing plasma iohexol clearance, because this model does not include AUC generated during the mixing phase of the plasma decay curve. Clearance is computed as iohexol I/AUC, and underestimation of AUC results in an overestimation of plasma clearance. Correction equations have been formulated for application to plasma iohexol clearance values determined in humans in order to correct errors inherent in use of the 1- compartment model. 15 For some data analysis performed in this study, iohexol values were corrected based on equations derived from linear regression analysis. The corrections that were applied probably represent a combination of the error imposed by the 1-compartment model and the error associated with experimental procedures and a relatively small number of dogs being tested. Some experimental error is exposed by the finding that different equations were required to correct iohexol clearance values, dependent on the method of iohexol assay. Although we have applied correction equations to data generated in this study to facilitate its interpretation, it would be inappropriate to put these equations into general use without further study. In particular, the number of dogs studied should be increased, and specific attention should be given to plasma : urine clearance ratios in dogs with markedly impaired renal function to determine if extrarenal clearance is a factor in plasma iohexol clearance values. Equations for correction of iohexol clearance were derived in another study performed on dogs with pyometra. 8 In that study, urinary clearance procedures were not performed as a basis for comparison, but multiple blood samples were obtained after iohexol injection to allow computation of plasma clearance by use of several pharmacokinetic models. Results from the 1-compartment model were compared to noncompartmental analysis in which AUC was derived by the trapezoid method, the latter providing a more accurate measurement of AUC. The 1-compartment : noncompartment clearance ratio was about 1.2:1, which was substantially greater than that found between the 1-compartment model for iohexol clearance and urinary clearance of exogenous creatinine. The meaning of these interstudy differences is unclear because experimental design was quite different. However, considerable variations in values for plasma iohexol clearance are obtained from the same data set when different pharmacokinetic models or different numbers of sampling points are used for computations. 14 These results attest to the potential for variation in results from plasma clearance measurement of GFR. Results from this study demonstrated reasonably good agreement between absolute values for plasma iohexol clearance and urinary exogenous creatinine clearance without application of correction factors. The R 2 values indicate good correlation between the methods. These findings suggest that uncorrected values may have clinical value as relative indices of renal function when serial determinations are performed in the same dog. Considering the importance of the slope of the plasma decay curve in computational results, we consider 3 plasma samples a minimum, not only for defining the curve but also as a safeguard for detecting spurious results that could arise from injection errors, analytical errors, or inappropriate sample times for the 1-compartment model. It is essential that a test substance for measuring renal function cause no toxicity when it is administered to the patient. Iodine-containing contrast agents have been reported to cause adverse reactions, and potential causes have been investigated in dogs. 16 No adverse reactions were clinically evident in the dogs we studied. In addition, limited numbers of serial iohexol clearance determinations showed no decrement in clearance suggestive of nephrotoxicity. Whether plasma iohexol clearance or urinary clearance is used to evaluate renal function in dogs, the question of normal values must be addressed. A review of values for GFR in normal dogs determined by urinary clearance procedures spans a wide range, 14 and several reasons probably account for this diversity. The GFR in dogs is known to be influenced by several factors, including hydration state 17 and dietary protein intake, 2,18 and body size is a factor even when GFR is factored by body weight or body surface area. 14 Decrements in GFR occur in humans with aging, 2 but 1 study of aging Cocker Spaniels and Miniature Schnauzers demonstrated no change in GFR between 7.5 and 1 years of age. 19 Studies of dogs beyond that age have not been reported. Although not studied to date, breed differences in normal GFR should be considered a possibility. Endogenous urinary creatinine clearance has been used as a measure of GFR, but values cannot be equated to GFR unless enzyme-specific methods of creatinine analysis are used to circumvent presence of noncreatinine reactants in plasma. 20 Although many biologic processes are known to contribute to the range of values reported for GFR in normal dogs, serial values obtained by exogenous creatinine urinary clearance procedures are very constant over time when conducted under standardized conditions. 21 Thus, the range of values reported in the literature for GFR in dogs is not a condemnation of urinary clearance procedures but rather, a reflection of the biologic variability of GFR in the variety of conditions under which it has been measured. Application of any clearance procedure performed to evaluate renal function should make standardization of conditions of testing a prerequisite. Until conditions are standardized and

6 Iohexol and Urinary Exogenous Creatinine Clearance in Dogs 373 data are acquired on subset populations of dogs based on age, breed, size, and diet, the range of values considered normal for GFR in dogs probably will remain large, regardless of whether plasma or urine clearance methods are used. Footnotes a Omnipaque, iohexol injection, 240 mg I/mL, Nycomed Inc, Princeton, NJ b Beckman CX3, Beckman Coulter Inc, Brea, CA c Versapar filter, Pall Gelman Laboratory, Ann Arbor, MI d ODS Prodigy HPLC column, Phenomenex, Torrance, CA References 1. Finco DR. Kidney function. In: Kaneko JJ, Harvey JW, Bruss ML, eds. Clinical Biochemistry of Domestic Animals, 5th ed. New York, NY: Academic Press; 1997: Smith HW. The Kidney. Structure and Function in Health and Disease. Oxford: Oxford Press; 1951: Finco DR, Brown SA, Crowell WA, Barsanti JA. Exogenous creatinine clearance as a measure of glomerular filtration rate in dogs with reduced renal mass. Am J Vet Res 1991;52: Frennby B. Use of iohexol to determine glomerular filtration rate. A comparison between different clearance techniques in man and animals. Scand J Urol Nephrol 1997;S-182: Gaspari F, Perico N, Matalone M, et al. Precision of plasma clearance of iohexol for estimation of GFR in patients with renal disease. J Am Soc Nephrol 1998;9: Moe L, Heiene R. Estimation of glomerular filtration rate in dogs with 99M-Tc-DTPA and iohexol. Res Vet Sci 1995;58: Gleadhill A, Michell AR. Evaluation of iohexol as a marker for clinical measurement of glomerular filtration rate in dogs. Res Vet Sci 1996;60: Heinne R, Moe L. The relationship between some plasma clearance methods for estimation of glomerular filtration rate in dogs with pyometra. J Vet Intern Med 1999;13: Brown SA, Finco DR, Boudinot D, et al. Evaluation of a single injection method, using iohexol, for estimating glomerular filtration rate in cats and dogs. Am J Vet Res 1996;57: Laroute V, Lefebvre HP, Costes G, et al. Measurement of glomerular filtration rate and effective renal plasma flow in the conscious beagle dog by single intravenous bolus of iohexol and p-aminohippuric acid. J Pharmacol Toxicol Methods 1999;41: Back SE, Masson P, Nilsson-Ehle P. A simple method for the quantification of the contrast agent iohexol, applicable to glomerular filtration rate measurements. Scand J Lab Clin Invest 1988;48: Braselton WE, Stuart KJ, Kruger JM. Measurement of serum iohexol by determination of iodine with inductively coupled plasmaatomic emission spectroscopy. Clin Chem 1997;43: Shihabi ZK, Thompson EN, Constantinescu MS. Iohexol determination by direct injection of serum on the HPLC column. J Liquid Chromatogr 1993;16: Heiene R, Moe L. Pharmacokinetic aspects of measurement of glomerular filtration rate in the dog: A review. J Vet Intern Med 1998; 12: Brochner-Mortensen J. A simple method for determining glomerular filtration rate. Scand J Clin Lab Invest 1972;30: Tanaka T, Katayama H, Shirakata A, Takahasi H. Biochemical aspects on adverse reactions to contrast media. Changes of kininogen levels in dog plasma after intravenous injections of iohexol, iopamidol, and iothalamate. Invest Radiol 1988;23(Suppl 1):S195 S Tabaru H, Finco DR, Brown SA, Cooper T. Influence of hydration state on renal functions of dogs. Am J Vet Res 1993;54: Finco DR. Effects of dietary protein intake on renal functions. Compend Cont Educ Pract Vet 1999;21(K): Finco DR, Brown SA, Crowell WA, et al. Effects of aging and dietary protein intake on uninephrectomized geriatric dogs. Am J Vet Res 1994;55: Finco DR, Tabaru H, Brown SA, Barsanti JA. Endogenous creatinine clearance measurement of glomerular filtration rate in dogs. Am J Vet Res 1993;54: Finco DR, Brown SA, Brown CA, et al. Progression of chronic renal disease in the dog. J Vet Intern Med 1999;13: