Chronic kidney disease is particularly prevalent in the aging feline population, with approximately 10% of cats aged 10 to 15 years affected.1 A primary goal in managing chronic kidney disease in cats is to identify those in the early stages of disease, when intervention may be directed at the underlying cause as well as at intrinsic and extrinsic factors contributing to its progression. This has led to an increased demand for precise methods of measuring GFR, which is generally accepted as the most sensitive index of functioning renal mass. Measurement of GFR is also useful clinically to evaluate patients with polyuria or polydipsia and patients with plasma creatinine concentrations that are at or near the upper reference limit to assess renal function prior to the use of renally excreted drugs, to predict the clearance of cancer drugs from the body,2 or to evaluate the response to therapeutic interventions aimed at improving renal function.
Glomerular filtration rate can be estimated from the plasma or urinary clearance of an endogenous or exogenously administered filtration marker. Many such markers are limited in their use in feline practice because of the need for accurately timed urine collections, urinary catheterization, or nuclear medicine facilities or because of lack of an available medical-grade injectable formulation. The nonionic radiographic contrast agent iohexol can be administered exogenously as a single bolus dose, and its plasma clearance can be used to estimate GFR. Iohexol is useful as a clearance marker because it is freely filtered by the kidneys, is not secreted into the renal tubules, has negligible protein binding, and is not metabolized within the body.3–5 Several studies6–17 have involved the use of this marker for estimation of GFR in cats.
Iohexol exhibits endoisomerism and exoisomerism.18 Analysis of iohexol concentrations in samples by use of high-performance liquid chromatography allows detection of the 2 types of isomers (EnIox and ExIox). The predominant isomer is quantitatively known to be ExIox. Previous studies12,13,16,17 have revealed a difference in the clearance of the 2 isomers in cats. Measurement of Tlox may therefore lead to inaccuracies in the estimation of clearance. To determine the doses of EnIox and ExIox that have been administered, the correct ratio of each isomer must be known. The stereoselective disposition of the isomers may change in different solutions as a result of their physicochemical properties, the solution temperature,18 or the storage duration.19 The ratio of plasma concentrations may therefore change after administration when compared with the ratio in the administered formulation.
Accurate methods of determining plasma clearance rely on the collection of blood samples at multiple points to create a plasma concentration-versus-time curve. Development of techniques that involve collection of a limited number of samples has become an important goal. It was first proposed by Brochner-Mortensen20 that the initial rate of redistribution of a marker is not only independent of GFR but is also relatively constant from subject to subject and that an estimate of GFR may, without any important loss of accuracy, be based on the second elimination exponential. Errors may nevertheless arise when calculating clearance from a limited number of samples; therefore, inclusion of both exponentials (redistribution and elimination exponentials) is recommended. Exclusion of the redistribution exponential leads to an overestimation of the GFR because the AUC is underestimated.21 The error becomes particularly important in patients with clinically normal or near-normal renal function, in which the initial rapid redistribution exponential contributes to a relatively larger proportion of the AUC, compared with patients with a low GFR.
Clearance calculated from a limited number of samples collected during the elimination phase is referred to as the slope-intercept technique. Formulae that correct for the distribution exponential when a slope-intercept method is used have been developed for human patients, the most commonly used being the Chantler22 and Brochner-Mortensen20,23 formulae. A dog-specific formula has also been developed.24 The primary purpose of the study reported here was to derive a correction formula for estimating GFR in cats by use of the slope-intercept technique. A secondary aim was to determine whether correction formulae derived in humans20,23 and dogs24 could substitute for the cat-specific formula. The final purpose was to compare the plasma clearance of EnIox, ExIox, and TIox.
Materials and Methods
Cats—Twenty client-owned healthy cats were enrolled in the study, which was conducted at the Referral Clinic De Kompaan, Ommen, The Netherlands. Informed consent was obtained from owners, and the study was conducted in accordance with national animal welfare regulations. Physical examinations and serum biochemical analyses were performed to confirm whether the cats were clinically normal. Cats with any concurrent medical disorder were excluded. Plasma iohexol clearance measurements performed in this study were components of breeding, rescue shelter rehoming programs, or geriatric screening programs.
Plasma clearance of iohexol—Food was withheld from cats for 12 hours prior to clearance measurement, although free access to water was allowed. In each cat, an IV catheter was placed in a cephalic vein. A bolus dose of iohexola (647 mg/kg [300 mg of iodine/kg]) was administered IV followed by 1 mL of saline (0.9% NaCl) solution. Completion of the iohexol was defined as injection time 0. At 5, 15, 30, 60, 120, 180, 240, and 360 minutes after iohexol injection, blood samples (1 mL) were collected via jugular venipuncture or from an IV catheter other than that used for iohexol administration and then transferred to serum separation tubes. The exact sample collection time was recorded for each cat and used in the iohexol clearance calculations. The serum was separated, harvested, and stored at −80°C until analysis.
For practical purposes, analysis was performed at 2 reference laboratoriesb,c by use of a high-performance liquid chromatography method. At the first laboratoryb (Belgium), EnIox and ExIox concentrations were determined, whereas at the second laboratoryc (United Kingdom), only TIox concentrations were measured. Duplicate analysis of samples at both laboratories was performed for 12 cats. Clearance was calculated by use of identical samples analyzed at both laboratories (TIoxBelg and TIoxUK), and agreement of results between the 2 laboratories was assessed.
Pharmacokinetc analysis—The multisample method was defined as estimation of GFR from the full-clearance curve, whereas the slope-intercept method was defined as estimation of GFR from samples collected at 120, 180, and 240 minutes after iohexol injection. Pharmacokinetic analysis was performed by use of a commercially available pharmacokinetic computer program.d Clearance was calculated as dose/AUC. Doses of the endoisomers and exoisomers were calculated for each cat from the total volume of iohexol administered and the mean plasma ratio. The plasma ratio was chosen rather than the ratio in the iohexol formulation because it is the plasma ratio of the isomers that is available for clearance by the kidneys. Dose of TIox was determined from the total volume of iohexol administered.
The AUC was determined by use of noncompartmental and compartmental approaches. The noncompartmental approach involved the trapezoidal rule with extrapolation to infinity.25 Clearance of EnIox and ExIox was determined by use of a noncompartmental model only. Compartmental modelling involved the application of a 1-compartment model to the slope-intercept method. The AUC was calculated as A/α, in which A is the iohexol concentration at the time 0 intercept and α is the elimination rate constant (slope) of the curve. For the multisample method, a 2-compartment model was applied and the AUC was calculated as A/α + B/β, in which A is the iohexol concentration at the time 0 intercept of the first (redistribution) exponential, α is the slope of this exponential, B is the iohexol concentration at the time 0 intercept of the second (elimination) exponential, and β is the slope of this exponential.26
Assessment of the best fit of a model was performed by application of the Akaike information criterion.27 In addition, calculation of clearance manually (without the use of computer software) was undertaken for the multisample data by resolving the exponential decay curve by use of a 2-stage, curve-stripping procedure. The slow elimination exponential (β) was constructed from the log-transformed iohexol concentrations in blood samples collected after the 60-minute point and extrapolated back to obtain the time 0 intercept. The antilog of the time 0 intercept was denoted as B. The fast redistribution exponential (α) was fitted by use of the equation y = B × e−-β(t) applied to the early time points. These values were then subtracted from the raw values, and the logarithmically transformed data were plotted to obtain α. The fast redistribution exponential was extrapolated back to obtain the time 0 intercept. The antilog of the time 0 intercept was denoted A. Clearance was calculated manually for the slope-intercept data by graphically plotting the logarithmically transformed values and extrapolating back to the time 0 intercept to derive B and β. The clearance values were standardized to body weight in kilograms.
Statistical analysis—Statistical analyses were performed by use of a statistical software package.e The data were assessed for normality through graphical plots and use of the Kolmogorov-Smirnov test. Because the data met the assumptions of Gaussian distribution, parametric testing was used. Correlations between the clearance of EnIox, ExIox, TIoxBelg, and TIoxUK were explored by calculating Pearson correlation coefficients.
Agreement between the 2 clearance methods was assessed by plotting the difference of the 2 clearances against the mean of the 2 clearances (Bland-Altman plots).28 Bias was defined as the mean difference between 2 clearances, and the absolute limits of agreement were expressed as mean difference ± 2 SD. In addition, the ExIox:EnIox ratio was determined for each plasma sample. The significance of the change in the ExIox:EnIox ratio over time was explored by use of repeated-measures ANOVA. The model assumption of sphericity was tested by use of the Mauchly test.
The correction formula was developed by regressing slope-intercept clearance calculated by use of a 1-compartment model against multisample clearance calculated with a noncompartmental model and exploring linear, quadratic, and cubic models. Forcing the regression line through the origin ensured a constant was not generated in the regression equation.
The correlation between results of the 2 methods was evaluated by calculating the Pearson correlation coefficient. Agreement was assessed by use of Bland-Altman plots as previously described. Percentage error was calculated as follows:
in which Cl represents clearance. In addition, multisample data were compared with slope-intercept clearance corrected by use of the Heiene canine24 and Brochner-Mortensen human (adult and child)20,23 formulae.
All data are reported as mean ± SD. Values of P < 0.05 were considered significant.
Results
Cats—The median age of the 20 enrolled cats was 4.6 years (range, 1.5 to 18.8 years), and the median weight of the cats was 4.18 kg (range, 2.10 to 6.25 kg). Breeds represented included the following: domestic shorthair (n = 14), domestic longhair (2), and Bengal (4). Physical examination and serum biochemical analysis performed before the study began indicated no abnormalities in any cat. The mean ± SD plasma creatinine concentration before iohexol administration was 1.30 ± 0.29 mg/dL. No adverse clinical signs were noticed in any of the cats during the study period.
For the multisample data, the plasma concentration-versus-time curve could not be fitted to a 2-compartment model in 1 cat. In addition, the curve for that cat could not be resolved into 2 exponentials when attempted manually. There were therefore 20 clearance measurements available for noncompartmental analysis and 19 available for compartmental analysis. For the slope-intercept data, the goodness of fit of the data was not considered acceptable for 1 cat; therefore, only 19 sets of clearance measurements were used for analysis. Samples obtained at all 8 collection points (5, 15, 30, 60, 120, 180, 240, and 360 minutes after iohexol administration) were not available for 9 cats for the EnIox and ExIox evaluations as a result of insufficient sample volume. Data obtained at 6 to 8 time points were therefore used in the clearance calculations for EnIox and ExIox. In addition, the ExIox:EnIox ratio was not available at all time points for each cat; therefore, data for 17 cats at 7 time points were included in the repeated-measures ANOVA. The assumptions of the polynomial model were not met for data for 2 cats; therefore, the correction formula was derived by use of the clearance data for 17 cats. However, the formula was validated in 19 cats.
Plasma iohexol clearance—The mean ± SD clearance determined with the multisample method for EnIox, ExIox, TIoxBelg, and TIoxUK was 2.20 ± 0.34 mL/min/kg, 2.21 ± 0.34 mL/min/kg, 2.21 ± 0.34 mL/min/kg, and 2.17 ± 0.40 mL/min/kg, respectively. All clearances were in excellent agreement with each other (Figure 1; Table 1). Results of the Mauchly test for the repeated-measures ANOVA indicated that the assumption of sphericity had been violated (P < 0.001); therefore, the df were corrected by use of Greenhouse-Geisser estimates of sphericity (ϵ = 0.36). There was no significant change in the ExIox:EnIox ratio over time (P = 0.347; Figure 2).
Bland-Altman plot of agreement between clearance of EnIox and ExIox in healthy cats (n = 20) after IV administration of a bolus dose of iohexol (647 mg/kg). The solid line represents the bias, and dashed lines represent upper and lower 95% limits of agreement. Agreement was considered excellent.
Citation: American Journal of Veterinary Research 72, 12; 10.2460/ajvr.72.12.1652
Bland-Altman plot of agreement between clearance of EnIox and ExIox in healthy cats (n = 20) after IV administration of a bolus dose of iohexol (647 mg/kg). The solid line represents the bias, and dashed lines represent upper and lower 95% limits of agreement. Agreement was considered excellent.
Citation: American Journal of Veterinary Research 72, 12; 10.2460/ajvr.72.12.1652
Bland-Altman plot of agreement between clearance of EnIox and ExIox in healthy cats (n = 20) after IV administration of a bolus dose of iohexol (647 mg/kg). The solid line represents the bias, and dashed lines represent upper and lower 95% limits of agreement. Agreement was considered excellent.
Citation: American Journal of Veterinary Research 72, 12; 10.2460/ajvr.72.12.1652
The ExIox:EnIox concentration ratios over time in healthy cats (n = 17) after IV administration of a bolus dose of iohexol (647 mg/kg). Ratios did not change significantly (P = 0.347) over time. The 2 distinct groups of data represent different batches of iohexol formulation that were administered.
Citation: American Journal of Veterinary Research 72, 12; 10.2460/ajvr.72.12.1652
The ExIox:EnIox concentration ratios over time in healthy cats (n = 17) after IV administration of a bolus dose of iohexol (647 mg/kg). Ratios did not change significantly (P = 0.347) over time. The 2 distinct groups of data represent different batches of iohexol formulation that were administered.
Citation: American Journal of Veterinary Research 72, 12; 10.2460/ajvr.72.12.1652
The ExIox:EnIox concentration ratios over time in healthy cats (n = 17) after IV administration of a bolus dose of iohexol (647 mg/kg). Ratios did not change significantly (P = 0.347) over time. The 2 distinct groups of data represent different batches of iohexol formulation that were administered.
Citation: American Journal of Veterinary Research 72, 12; 10.2460/ajvr.72.12.1652
Limits of agreement and bias (95% CI) and results of correlation analysis for clearances of EnIox, ExIox, TIoxBelg and TIoxUK in healthy cats after IV administration of a bolus dose of iohexol (647 mg/kg).
| Limits of agreement (mL/min/kg) | |||||
|---|---|---|---|---|---|
| Comparison | Lower | Upper | Bias | r | P value |
| EnIox vs ExIox (n = 20) | −0.061 (−0.081 to −0.041) | 0.037 (0.017–0.056) | −0.012 (−0.024 to −0.001) | 0.998 | < 0.001 |
| EnIox vs TIox (n = 20) | −0.053 (−0.07 to −0.036) | 0.031 (0.016–0.048) | −0.011 (−0.021 to −0.001) | 0.998 | < 0.001 |
| ExIox vs TIox (n = 20) | −0.005 (−0.008 to −0.003) | 0.009 (0.006–0.011) | 0.002 (0 to 0.003) | 1.000 | < 0.001 |
| TIoxBelg vs TIoxUK (n = 12) | −0.309 (−0.504 to −0.114) | 0.401 (0.206–0.596) | 0.046 (−0.067 to 0.159) | 0.897 | < 0.001 |
CI = Confidence interval. TIoxBelg = Total iohexol as assessed at the Belgian laboratory. TIoxUK = Total iohexol as assessed at the United Kingdom laboratory.
No outliers were identified in any comparison.
Mean clearance for Tlox estimated by use of the multisample method (concentrations determined at the Belgian and United Kingdom laboratories) by use of a noncompartmental model was 2.16 ± 0.35 mL/min/kg (9.23 ± 2.57 mL/min). Multisample clearance determined from a 2-compartmental model by use of a computer and by resolving the 2 exponentials manually was in excellent agreement. Results for multisample clearance and corrected slope-intercept clearance determined by application of noncompartmental and compartmental models (calculated by computer program and manually) were summarized (Table 2). Mean clearance estimated by use of the uncorrected slope-intercept method was 2.44 ± 0.47 mL/min/kg. Mean percentage error between multisample clearance and uncorrected slope-intercept clearance was 12.48 ± 9.32% (range, 5.79% to 29.27%; Figure 3).
Relationship between clearance estimated by use of multisample and uncorrected slope-intercept methods for healthy cats (n = 19) after IV administration of a bolus dose of iohexol (647 mg/kg). The percentage error associated with the uncorrected slope-intercept method increases with increasing filtration function. The bold line represents the least squares polynomial regression line, and the dashed line is the line of equality.
Citation: American Journal of Veterinary Research 72, 12; 10.2460/ajvr.72.12.1652
Relationship between clearance estimated by use of multisample and uncorrected slope-intercept methods for healthy cats (n = 19) after IV administration of a bolus dose of iohexol (647 mg/kg). The percentage error associated with the uncorrected slope-intercept method increases with increasing filtration function. The bold line represents the least squares polynomial regression line, and the dashed line is the line of equality.
Citation: American Journal of Veterinary Research 72, 12; 10.2460/ajvr.72.12.1652
Relationship between clearance estimated by use of multisample and uncorrected slope-intercept methods for healthy cats (n = 19) after IV administration of a bolus dose of iohexol (647 mg/kg). The percentage error associated with the uncorrected slope-intercept method increases with increasing filtration function. The bold line represents the least squares polynomial regression line, and the dashed line is the line of equality.
Citation: American Journal of Veterinary Research 72, 12; 10.2460/ajvr.72.12.1652
Mean ± SD clearance (mL/min/kg) and results of agreement analyses for iohexol clearance in cats determined by use of multisample and corrected slope-intercept methods and calculated by use of noncompartmental and compartmental models.
| Limits of agreement | |||||
|---|---|---|---|---|---|
| Analysis | Clearance | Bias | Lower | Upper | r |
| Multisample NCA (computer calculation; n = 20) | 2.16 ± 0.35 | NA | NA | NA | NA |
| Multisample 2-compartment (computer calculation; n = 19) | 2.25 ± 0.36 | −0.076 | −0.158 | 0.006 | 0.955 |
| Multisample 2-compartment (manual calculation; n = 19)* | 2.16 ± 0.38 | −0.014 | −0.244 | 0.216 | 0.955 |
| Slope-intercept 1-compartment corrected (computer calculation; n = 19)* | 2.14 ± 0.34 | 0.019 | −0.285 | 0.323 | 0.905 |
| Slope-intercept NCA corrected (computer calculation; n = 19)* | 1.91 ± 0.24 | 0.177 | −0.311 | 0.807 | 0.601 |
| Slope-intercept 1-compartment corrected (manual calculation; n = 19) | 2.17 ± 0.30 | −0.012 | −0.236 | 0.212 | 0.955 |
Five percent of values obtained with this method were outliers.
NA = Not applicable. NCA = Noncompartmental analysis.
Clearance determined by use of the multisample method with noncompartmental analysis was considered the reference method for agreement analysis. Clearance determined by use of the multisample method was determined from 8 plasma samples, and clearance determined with the slope-intercept method was determined from 3 samples. Slope-intercept clearance was corrected by use of the formula derived in the present study.
A second-order polynomial regression line was found to provide the best fit to the data. The derived correction formula was as follows:
Mean corrected clearance was 2.14 ± 0.34 mL/min/kg. There was excellent agreement between multisample and corrected clearance (Figure 4). Mean clearance estimated via the slope-intercept method and corrected by use of the formula derived in the present study and also the Brochner-Mortensen human and Heiene canine formulae was summarized (Table 3). Clearance corrected via the feline-derived correction formula and the Heiene canine formula demonstrated a linear relationship with multisample clearance (Figure 5). However, the relationship between multisample clearance and slope-intercept clearance corrected by use of the Brochner-Mortensen formula was nonlinear. The feline-derived formula had the lowest mean percentage error relative to the multisample data (0.59 ± 7.05%). The Brochner-Mortensen formula had the highest mean percentage error (11.13% adult formula and 13.14% child formula). These values were similar to those calculated for uncorrected clearance (12.48%).
Bland-Altman plot of the difference between clearance estimated by use of the multisample method and the slope-intercept method corrected by use of a derived formula for healthy cats (n = 19) after IV administration of a bolus dose of iohexol (647 mg/kg). The solid line represents the bias, and the dashed line represents the upper and lower limits of agreement. Agreement was considered excellent.
Citation: American Journal of Veterinary Research 72, 12; 10.2460/ajvr.72.12.1652
Bland-Altman plot of the difference between clearance estimated by use of the multisample method and the slope-intercept method corrected by use of a derived formula for healthy cats (n = 19) after IV administration of a bolus dose of iohexol (647 mg/kg). The solid line represents the bias, and the dashed line represents the upper and lower limits of agreement. Agreement was considered excellent.
Citation: American Journal of Veterinary Research 72, 12; 10.2460/ajvr.72.12.1652
Bland-Altman plot of the difference between clearance estimated by use of the multisample method and the slope-intercept method corrected by use of a derived formula for healthy cats (n = 19) after IV administration of a bolus dose of iohexol (647 mg/kg). The solid line represents the bias, and the dashed line represents the upper and lower limits of agreement. Agreement was considered excellent.
Citation: American Journal of Veterinary Research 72, 12; 10.2460/ajvr.72.12.1652
Relationship between clearance for the cats in Figure 4 estimated by use of the multisample method and the slope-intercept method corrected by use of the derived feline formula (A), the Heiene canine formula24 (B), and the Brochner-Mortensen adult human formula20 (C). The bold line represents the regression line, and the dashed line is the line of equality.
Citation: American Journal of Veterinary Research 72, 12; 10.2460/ajvr.72.12.1652
Relationship between clearance for the cats in Figure 4 estimated by use of the multisample method and the slope-intercept method corrected by use of the derived feline formula (A), the Heiene canine formula24 (B), and the Brochner-Mortensen adult human formula20 (C). The bold line represents the regression line, and the dashed line is the line of equality.
Citation: American Journal of Veterinary Research 72, 12; 10.2460/ajvr.72.12.1652
Relationship between clearance for the cats in Figure 4 estimated by use of the multisample method and the slope-intercept method corrected by use of the derived feline formula (A), the Heiene canine formula24 (B), and the Brochner-Mortensen adult human formula20 (C). The bold line represents the regression line, and the dashed line is the line of equality.
Citation: American Journal of Veterinary Research 72, 12; 10.2460/ajvr.72.12.1652
—Correction formulae for slope-intercept clearance and mean ± SD clearance and percentage error associated with each formula.
| Method | Formula | Clearance (mL/min/kg) | Percentage error |
|---|---|---|---|
| Multisample NCA | — | 2.16 ± 0.35 | — |
| Slope-intercept (uncorrected) | — | 2.44 ± 0.47 | 12.48 ± 9.32 |
| Present study | (1.036 × Cl)-(0.062 × Cl2) | 2.14 ± 0.34 | 0.59 ± 7.05 |
| Heiene and Moe24 | −0.03 + (1.06 × Cl)-(0.07012 × Cl2) | 2.12 ± 0.34 | 1.56 ± 6.95 |
| Brochner-Mortensen (adult)20 | (0.991 × Cl)-(0.00122 × Cl2) | 2.41 ± 0.46 | 11.13 ± 9.17 |
| Brochner-Mortensen (child)23 | (1.01 × Cl)-(0.0017 × Cl2) | 2.45 ± 0.47 | 13.14 ± 9.33 |
Discussion
In the present study, GFR in healthy adult and geriatric cats estimated by use of slope-intercept plasma iohexol clearance corrected via the formula derived in the present study was in excellent agreement with GFR estimated with the multisample clearance method. This correction formula was developed and validated in the same sample of cats; however, validation should have ideally been undertaken in a different group of cats, and this remains an area for future studies to examine.
Agreement between corrected slope-intercept clearance and multisample clearance methods was considered excellent on the basis of negligible bias and narrow limits of agreement. The maximum difference in 1 cat was approximately 0.3 mL/min/kg, which was < 14% of the mean clearance and therefore not considered to be clinically important. For most cats, however, the difference was < 10% of the mean.
Noncompartmental analysis was used as the reference method because unlike compartmental analysis, the noncompartmental method is free of assumptions regarding redistribution and exchange of the filtration marker.29 The advantage of compartmental analysis, however, is that it allows an accurate pharmacokinetic model to be developed. Clearance estimated by use of different modelling methods may yield different results. Nevertheless, we did not find the difference to be clinically important and there was good agreement between methods. Errors can occur in noncompartmental modelling when a limited number of samples are collected,29 and this was clearly demonstrated in the clearance results determined by use of a noncompartmental model applied to the slope-intercept technique (Table 2). Therefore, a noncompartmental model should only be applied to data obtained from a sufficient number of collected samples.
The commercial pharmacokinetic program used in the present study may not be available to all clinicians and researchers. For this reason, clearance was additionally determined manually. Clearance calculated manually by use of 2-compartment and corrected 1-compartment models was in excellent agreement. This demonstrates that clearance can be accurately estimated by use of the corrected slope-intercept technique without the need for expensive pharmacokinetic software.
A 1-compartment simplification overestimates true clearance because the AUC is underestimated. The error introduced by ignoring the initial redistribution exponential in healthy humans is approximately 15%.30 In a study24 of plasma iohexol clearance in dogs, dogs with clinically normal renal function were found to have a mean overestimation in uncorrected clearance of 20%. The Chantler correction22 method used in human patients is simply a single multiplying factor. That technique does not take into account the variation in percentage error that can exist at different filtration functions; therefore, polynomial correction factors are generally recognized as being more accurate. The mean overestimation in the present study was 12.48% and ranged from 5.79% to 29.27%, with the error increasing with increasing filtration function (Figure 3).
Forcing the regression line through the origin in our study ensured a constant was not generated in the development of the polynomial equation; therefore, the formula could also be appropriate in cats with very low renal clearance. The clinical indication for GFR estimation is for patients with polyuria or polydipsia or plasma creatinine concentrations at the upper limit of the reference interval and for assessing renal function prior to the use of renally excreted drugs. Such patients may have clinically normal or near normal renal function; therefore, application of the correction formula in this situation would be more important because there would be a larger AUC, which would be underestimated.
The Brochner-Mortensen correction formula was validated in humans by use of chromium isotope (51Cr)—labelled EDTA20 and has been shown to be valid for iohexol.4 Moreover, good agreement between the plasma clearance of iohexol with the Brochner-Mortensen correction applied and urinary clearance of inulin exists in human patients.31 In a canine study,32 the Brochner-Mortensen correction formula was accurate for large but not small-breed dogs. The deviation in small-breed dogs suggested the need for the development of a cat-specific correction formula.
The Brochner-Mortensen formula has been used to correct the plasma clearance of iohexol in cats with intact and partially nephrectomized kidneys, and a high degree of agreement with urinary clearance of creatinine has been demonstrated.7 However, an earlier study6 revealed only moderate agreement between the 2 methods of estimating GFR. The relationship between plasma iohexol clearance determined by use of the multisample method and slope-intercept method corrected by use of the Brochner-Mortensen adult formula appeared to be nonlinear in the present study, with errors increasing with increasing clearances (Figure 5). This may explain the conflicting results from previous studies6,7 in which the formula was used in cats. The nonlinearity of the relationship was not improved by applying the Brochner-Mortensen correction formula derived for children.23 Thus, clearance corrected by use of the Brochner-Mortensen formula appears to offer no greater accuracy than uncorrected clearance in cats with healthy renal function. Although the canine correction formula generated results similar to those of the cat-specific correction formula, the mean percentage error was smaller when the cat-specific correction formula was applied. This may have been because the cat formula was developed and validated in the same sample of cats, or it may be that clearance obtained by use of the slope-intercept method requires correction with a species-specific formula to improve accuracy. Additional studies are required to test whether species-specific correction formulae are required.
The mean clearance of EnIox and ExIox was in excellent agreement in the present study. In addition, statistical testing to explore the change in the ExIox:EnIox ratio over time revealed no significant change, suggesting the 2 stereoisomers were cleared at similar rates and that following iohexol administration, no important conversion of one isomer to the other takes place. Taken together, these results provide evidence that the clearance of EnIox and ExIox is identical. A previous study33 in which the pharmacokinetics and biotransformation of iohexol in dogs and rats were evaluated also found that the proportion of isomers remained unchanged.
The present study also revealed that the clearance of the 2 isomers did not differ from the clearance of TIox. These results agree with those of human4,31 and canine34 studies but differ from results of previous feline studies,12,13,16,17 in which clearance of the stereoisomers was found to differ. An early study18 conducted to investigate the isomerism of iohexol found the isomers to be interconvertible and that rotational conversion may be temperature dependent. Investigators in that study concluded that the isomers should be considered a single pharmacological entity. The results of the present study also support this conclusion.
Ratios of Exlox to Enlox have varied in previous feline studies, with reported ratios of 84:16 (5.25),13 78:22 (3.55),17 and 85:15 (5.68).16 Furthermore, differences in the clearance of the 2 isomers have been reported.12,13,16,17 However, it is unclear from these studies how the ratio needed for determination of the dose of administered isomer was derived. In the present study, the mean ExIox:EnIox ratio in plasma was 4.7:1; however, there was large variation among cats (3.8:1 to 5.6:1) and among batches of product (4.6:1 to 9:1). For these reasons, we determined the administered dose of each isomer for each cat on the basis of its specific plasma ratio. This resulted in identical calculation of ExIox, EnIox, and TIox clearance, which is biologically more plausible than the glomerular filtration barrier reacting differently depending on the isomer. It is possible the ExIox:EnIox ratio differed in the previous studies12,13,16,17 because of the combined administration of iohexol and exogenous creatinine. However, that appears unlikely given that creatinine is produced endogenously and interference of creatinine with iohexol measurements in azotemic patients is not known to occur.13
Healthy adult and geriatric cats were used in the present study, and exclusion criteria included hyperthyroidism or any other identified disease. The population that this sample was intended to represent is similar to that of previous studies; however, in 2 previous studies,16,17 hyperthyroid cats were included. Results from previous studies are inconsistent, with ExIox clearance reportedly higher than EnIox clearance in hyperthyroid cats in 1 study,17 yet no identified in a different study.16 Furthermore, there are discrepancies in the clearance of iohexol isomers in healthy adult and geriatric cats, with 2 studies12,16 finding EnIox clearance to be higher than that of Exlox and another study13 finding the opposite. Differences in the relative clearance of EnIox and ExIox between studies suggest inaccuracies in the calculation of clearance. Because GFR is calculated from the dose of the biomarker (in this case, the isomer of iohexol) divided by the area under the plasma clearance curve, inaccurate clearance measurement would result from inaccurate dosage calculation. We propose this is likely to have occurred if a standard ExIox:EnIox ratio was assumed or if the mean ratio determined in a group of cats was used rather than the ratio measured for each cat in the calculation of dose for each cat.
One limitation of the present study is that the repeated-measures ANOVA used to determine the change in the ExIox:EnIox ratio over time did not include the 5-minute point. This was because an ExIox:EnIox ratio was not available at 5 minutes for all of the cats. It could be argued that this may be the most important time point, as it is the time point closest to the point of EnIox and ExIox (iohexol) administration. However, when the analysis was repeated by use of data for 11 cats in which ratios were available at all time points, the results were similar and the change over time was not significant (data not shown; P = 0.604). The same conclusion can therefore be drawn that the ratio of the isomers does not appear to change over time in cats, indicating that both isomers are cleared by the kidneys at the same rate.
Several studies have involved calculation of clearance from the ExIox and EnIox peak only in cats12,13,15–17 and dogs.32,34,35 The present study revealed that clearance of ExIox, EnIox, and TIox calculated accurately by use of the correct dose administration, based on the ratio of the different stereoisomers in plasma, does not differ. In addition, TIox results from 2 laboratories were in good agreement. It can therefore be concluded that the clearance of EnIox, ExIox, TIoxBelg, and TIoxUK can be used interchangeably when estimating GFR in cats. In addition, the cat-specific correction formula derived in our study can be applied to a slope-intercept method of estimating plasma iohexol clearance. Clearance corrected by use of the formula was in excellent agreement with that obtained with the multisample method. Use of the slope-intercept technique with the feline formula offers an important advantage in reducing the stress associated with repeated blood collection in cats and the financial cost of multiple sample analysis.
Abbreviations
| AUC | Area under the curve |
| EnIox | Endo-iohexol |
| ExIox | Exo-iohexol |
| GFR | Glomerular filtration rate |
| TIox | Total iohexol |
Omnipaque300, Amersham Health, Piscataway, NJ.
Department of Pharmacology, Toxicology and Biochemistry, Faculty of Veterinary Medicine, Ghent University, Belgium.
Chemical Pathology Department, Epsom and St Helier University Hospital NHS trust, Epsom, England.
Phoenix WinNonlin 6.0, Pharsight, Sunnyvale, Calif.
SPSS, version 17.0, SPSS Inc, Chicago, Ill.
References
- 1.↑
Lulich JP, Osborne CA, Obrien TD, et al. Feline renal failure—questions, answers, questions. Compend Contin Educ Pract Vet 1992; 14: 127–152.
- 2.↑
Bailey DB, Rassnick KM, Prey JD, et al. Evaluation of serum iohexol clearance for use in predicting carboplatin clearance in cats. Am J Vet Res 2009; 70: 1135–1140.
- 3.
Olsson B, Aulie A, Sveen K, et al. Human pharmacokinetics of iohexol. A new nonionic contrast medium. Invest Radiol 1983; 18: 177–182.
- 4.↑
Krutzen E, Back SE, Nilsson-Ehle I, et al. Plasma clearance of a new contrast agent, iohexol: a method for the assessment of glomerular filtration rate. J Lab Clin Med 1984; 104: 955–961.
- 5.
Edelson J, Shaw D, Palace G. Pharmacokinetics of iohexol, a new nonionic radiocontrast agent, in humans. J Pharm Sci 1984; 73: 993–995.
- 6.↑
Brown SA, Finco DR, Boudinot FD, 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: 105–110.
- 7.↑
Miyamoto K. Use of plasma clearance of iohexol for estimating glomerular filtration rate in cats. Am J Vet Res 2001; 62: 572–575.
- 8.
Miyamoto K. Clinical application of plasma clearance of iohexol on feline patients. J Feline Med Surg 2001; 3: 143–147.
- 9.
Haller M, Rohner K, Muller W, et al. Single-injection inulin clearance for routine measurement of glomerular filtration rate in cats. J Feline Med Surg 2003; 5: 175–181.
- 10.
Goy-Thollot I, Besse S, Garnier F, et al. Simplified methods for estimation of plasma clearance of iohexol in dogs and cats. J Vet Intern Med 2006; 20: 52–56.
- 11.
Goy-Thollot I, Chafotte C, Besse S, et al. Iohexol plasma clearance in healthy dogs and cats. Vet Radiol Ultrasound 2006; 47: 168–173.
- 12.
van Hoek I, Vandermeulen E, Duchateau L, et al. Comparison and reproducibility of plasma clearance of exogenous creatinine, exo-iohexol, endo-iohexol, and 51Cr-EDTA in young adult and aged healthy cats. J Vet Intern Med 2007; 21: 950–958.
- 13.↑
Le Garreres A, Laroute V, De La Farge F, et al. Disposition of plasma creatinine in non-azotaemic and moderately azotaemic cats. J Feline Med Surg 2007; 9: 89–96.
- 14.
Heiene R, Reynolds BS, Bexfield NH, et al. Estimation of glomerular filtration rate via 2- and 4-sample plasma clearance of iohexol and creatinine in clinically normal cats. Am J Vet Res 2009; 70: 176–185.
- 15.
Goodman LA, Brown SA, Torres BT, et al. Effects of meloxicam on plasma iohexol clearance as a marker of glomerular filtration rate in conscious healthy cats. Am J Vet Res 2009; 70: 826–830.
- 16.↑
van Hoek IM, Lefebvre HP, Paepe D, et al. Comparison of plasma clearance of exogenous creatinine, exo-iohexol, and endoiohexol over a range of glomerular filtration rates expected in cats. J Feline Med Surg 2009; 11: 1028–1030.
- 17.↑
van Hoek I, Lefebvre HP, Kooistra HS, et al. Plasma clearance of exogenous creatinine, exo-iohexol, and endo-iohexol in hyper-thyroid cats before and after treatment with radioiodine. J Vet Intern Med 2008; 22: 879–885.
- 18.↑
Foster SJ, Sovak M. Isomerism in iohexol and ioxilan. Analysis and implications. Invest Radiol 1988; 23 (suppl 1):S106–S109.
- 19.↑
Jacobsen PB, Larsen A, Konarboland R, et al. Biotransformation of nonionic x-ray contrast agents in vivo and in vitro. Drug Metab Dispos 1999; 27: 1205–1213.
- 20.↑
Brochner-Mortensen J. A simple method for the determination of glomerular filtration rate. Scand J Clin Lab Invest 1972; 30: 271–274.
- 21.↑
Heiene R, Moe L. Pharmacokinetic aspects of measurement of glomerular filtration rate in the dog: a review. J Vet Intern Med 1998; 12: 401–414.
- 22.↑
Chantler C, Barratt TM. Estimation of glomerular filtration rate from plasma clearance of 51-chromium edetic acid. Arch Dis Child 1972; 47: 613–617.
- 23.↑
Brochner-Mortensen J, Haahr J, Christoffersen J. A simple method for accurate assessment of the glomerular filtration rate in children. Scand J Clin Lab Invest 1974; 33: 140–143.
- 24.↑
Heiene 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: 587–596.
- 25.↑
Watson AD, Lefebvre HP, Concordet D, et al. Plasma exogenous creatinine clearance test in dogs: comparison with other methods and proposed limited sampling strategy. J Vet Intern Med 2002; 16: 22–33.
- 26.↑
Sapirstein LA, Vidt DG, Mandel MJ, et al. Volumes of distribution and clearances of intravenously injected creatinine in the dog. Am J Physiol 1955; 181: 330–336.
- 27.↑
Yamaoka K, Nakagawa T, Uno T. Application of Akaike's information criterion (AIC) in the evaluation of linear pharmacokinetic equations. J Pharmacokinet Biopharm 1978; 6: 165–175.
- 28.↑
Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1: 307–310.
- 29.↑
DiStefano JJ, 3rd. Noncompartmental vs. compartmental analysis: some bases for choice. Am J Physiol 1982; 243:R1–R6.
- 30.↑
Peters AM. The kinetic basis of glomerular filtration rate measurement and new concepts of indexation to body size. Eur J Nucl Med Mol Imaging 2004; 31: 137–149.
- 31.↑
Gaspari F, Perico N, Ruggenenti P, et al. Plasma clearance of nonradioactive iohexol as a measure of glomerular filtration rate. J Am Soc Nephrol 1995; 6: 257–263.
- 32.↑
Bexfield NH, Heiene R, Gerritsen RJ, et al. Glomerular filtration rate estimated by 3-sample plasma clearance of iohexol in 118 healthy dogs. J Vet Intern Med 2008; 22: 66–73.
- 33.↑
Mutzel W, Speck U. Pharmacokinetics and biotransformation of iohexol in the rat and the dog. Acta Radiol Suppl 1980; 362: 87–92.
- 34.↑
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: 17–25.
- 35.
Finco DR, Braselton WE, Cooper TA. Relationship between plasma iohexol clearance and urinary exogenous creatinine clearance in dogs. J Vet Intern Med 2001; 15: 368–373.
