Although GFR estimation is considered the best overall index for precise assessment of kidney function in dogs,1,2 application of classic urinary clearance in canine medicine is labor-intensive and time-consuming. Glomerular filtration rate estimation relies on accurately timed, repeated blood and urine sample collections and usually involves bladder catheterization to accurately measure urine volume. Moreover, the standard tracer inulin has extremely low solubility with a complicated laboratory analysis. Although the nonionic, monomeric, radiographic contrast medium iohexol, instead of inulin, has been frequently used for GFR assessment in small animal practice,3–5 there is concern regarding the nephrotoxic potential of iohexol in animals with reduced kidney function.6,7
The isotonic, nonionic, dimeric iodine radiographic contrast medium iodixanol is of interest as a new tracer that can be used for GFR estimation in dogs. Iodixanol is rapidly excreted into urine without metabolic degradation and no protein binding, and it has a very short half-life, as shown in pharmacokinetic studies with experimental animals (rats and monkeys)8,9 and humans.9,10 Furthermore, iodixanol has been proven to be less nephrotoxic than other nonionic, nonisotonic contrast media, including iohexol, in extensive randomized, blinded, prospective, multicenter studies11–13 of humans with chronic renal diseases who are at high risk of contrast medium–associated nephropathy. In fact, the toxicity of a single IV dose of iodixanol (50% lethal dose, > 21 g of I/kg) in male Wistar rats is significantly lower than that of iohexol (11.7 g of I/kg), iopentol (11.7 g of I/kg), or iopamidol (10.8 g of I/kg) under the same experimental conditions.8
In humans, the concentration of the tracer chromium 51–labeled EDTA (51Cr-EDTA) in a single plasma sample obtained a few hours after tracer injection has been reported to correlate well with renal clearance.14 Subsequently, Jacobsson15 derived a formula from a simple 1-compartment model combined with the Vd and optimum time for obtaining a single plasma sample in relation to administration of technetium Tc 99m diethylene triamine pentaacetic acid and accurately determined the GFR in humans. Because the Vd is dependent on elimination kinetics of each tracer and animal size, it should be determined in each individual for estimating GFR and the reference value needs to be determined for each species.16
The purpose of the study reported here was to establish an SBSM involving iodixanol (based on the Jacobsson formula15) to estimate GFR in dogs and compare data provided by that procedure with data provided by a conventional MBSM involving inulin.
Materials and Methods
Study design
In the study protocol, GFR was first estimated by the MBSM involving iodixanola (320 mg of I/mL; 290 mOsm/kg H2O) in healthy dogs. Next, the equation for the SBSM was generated on the basis of the Jacobsson formula.15 Briefly, by substituting the GFR, serum iodixanol concentration, and sample collection time (relative to tracer injection) obtained from the MBSM involving iodixanol into the Jacobsson formula,15 the Vd was calculated for individual dogs. After confirming a relationship between the Vd values and serum iodixanol concentrations, a formula for calculating the estimated Vd was determined from a scatter diagram. The GFR estimated by the SBSM was calculated by substituting the dose of iodixanol injected, estimated Vd, serum iodixanol concentration, and sample collection time (relative to tracer injection) for each dog into the Jacobsson formula15 once again. In this study, the MBSM involving inulinb (100 mg/mL) was used as the standard reference procedure. Inulin is a semisynthetic, diagnostic drug for human use. All other chemicals and reagents were of the highest grade available from commercial sources, unless otherwise stated.
Animals
Twenty-six healthy purpose-bred male Beagles were used in the study. Of those healthy dogs, 4 were owned by the Iwate University Veterinary Teaching Hospital and 22 were by owned by Medicinal Safety Research Laboratories, Daiichi-Sankyo, Tokyo. Dogs were regarded as healthy on the basis of the results of clinical observations, hematologic and serum biochemical analyses, and urinalysis. Thirty-six dogs with naturally occurring renal disease admitted to the Veterinary Teaching Hospital at Iwate University, Morioka, Japan, were also used in the study. Breeds of the dogs with renal disease included Papillon, Cavalier King Charles Spaniel, Miniature Dachshund, Shetland Sheepdog, Pembroke Welsh Corgi, and mixed. The dogs with renal disease were selected after obtaining the owner's consent for their dog's participation in this investigation. The dogs with renal disease had a history of kidney disease (proteinuria) for ≥ 3 months without heart disease or other organ failure and were grouped as having stage 1, 2, or 3 chronic kidney disease according to the guidelines of the International Renal Interest Society17 regarding serum creatinine concentration as follows: stage 1, < 1.4 mg/dL; stage 2, 1.4 to 2.0 mg/dL; stage 3, 2.1 to 5.0 mg/dL; and stage 4, > 5.0 mg/dL. Among the dogs with renal disease in the study, 31 had stage 1 disease, 4 had stage 2 disease, and 1 had stage 3 disease; no dog had stage 4 disease. All procedures were performed in accordance with the Guidelines for Animal Experimentation issued by the Japanese Association for Laboratory Animal Science18 and approved by the Animal Experimental Ethics Committee of Iwate University and Daiichi-Sankyo.
Healthy dogs and those with naturally occurring renal disease were used in 1 or more experiments in this study. Among the experiments, the number of healthy dogs and dogs with renal disease used varied (Table 1).
Number of healthy dogs and dogs with naturally occurring renal disease used in various parts of a study to establish an SBSM involving iodixanol (based on the Jacobsson formula15) for estimation of GFR in dogs and compare data provided by that procedure with data provided by a conventional MBSM involving inulin.
| Group | Experimental portion of study | No. of healthy dogs | No. of dogs with naturally occurring renal disease |
|---|---|---|---|
| A | Assessment of dose dependency of iodixanol (3 × 3 Latin square design) | 3 | — |
| B | Assessment of dose dependency of inulin | 4 | — |
| C | Serum concentration of iodixanol (40 mg of I/kg) or inulin (50 mg/kg) over time | 6 (6) | — |
| D | Comparison of 1- and 2-compartment models for iodixanol data | 3 (3) | — |
| E | Selection of blood sample collection times for assessment of serum iodixanol or inulin concentration | 6 (6) | — |
| F | MBSM vs SBSM involving iodixanol | 26 (19) | 36 |
| G | Calculation of estimated Vd of iodixanol | 26 (26) | 9 (9) |
| H | Assessment of coadministration of iodixanol and inulin | 22 (22) | 3 (3) |
| I | Assessment of GFR vs BUN or serum creatinine concentration | 26 (26) | 36 (36) |
Overall, the study included 26 healthy dogs and 36 dogs with naturally occurring renal disease. Numbers in parentheses indicate the number of dogs that had been used in other experimental portions of the study.
— = Not applicable.
GFR measurement in preliminary experiments
To select the appropriate dose of iodixanol to dogs for estimation of GFR, iodixanol was administered IV to 20 healthy dogs (1 female and 19 male Beagles; mean ± SD weight, 12.4 ± 1.0 kg) and 4 dogs with naturally occurring renal disease (a female Papillon, a male Shetland Sheepdog, and 2 male mixed-breed dogs; median weight, 6.8 kg; weight range, 2.7 to 11.5 kg) according to a 3 × 3 Latin square design at a dose of 20, 40, or 80 mg of I/kg (group A; Table 1). Inulin was administered IV to 4 other healthy Beagles at a dose of 50 or 100 mg/kg (group B). The time of tracer injection was designated as 0 minutes. For each dog, the tracer was injected into a cephalic vein (left or right vein selected randomly) with a 24-gauge indwelling catheter and a blood sample (0.8 mL) was collected from the contralateral cephalic vein immediately before and 30, 45, 60, 90, 120, or 150 minutes after tracer administration.
Next, to confirm the detailed disappearance of each tracer from serum, iodixanol (40 mg of I/kg) or inulin (50 mg/kg) was administered IV to 1 of 2 groups of 3 healthy dogs (group C), and a blood sample (0.8 mL) was collected immediately before and 5, 10, 15, 30, 45, 60, 90, 120, 150, and 180 minutes after iodixanol administration (n = 6; group E), and immediately before and 5, 10, 15, 30, 45, 60, 90, and 120 minutes after inulin administration (6; group E). In this protocol, a blood sample was collected 9 to 11 times from all dogs used. On the basis of these results, the GFRs from the 1- and 2-compartment models were calculated (group D).
To determine appropriate blood sample collection times, GFR was calculated from the 1-compartment model data for various combinations of blood sample collection times after each tracer administration (group E). Plots of serum iodixanol and inulin disappearance were linear at 30 to 180 minutes and at 30 to 120 minutes, respectively, after tracer injection. The GFRs were calculated for the following combinations of blood sample collection times after iodixanol administration: 30, 60, 90, 120, 150, and 180 minutes (combination a); 30, 90, 120, 150, and 180 minutes (combination b); 60, 90, 120, 150, and 180 minutes (combination c); 30, 60, 90, and 120 minutes (combination d); 30, 60, and 90 minutes (combination e); 30, 60, and 120 minutes (combination f); and 60, 90, and 120 minutes (combination g). The GFRs were calculated for the following combinations of blood sample collection times after inulin administration: 30, 90, and 120 minutes (combination h); 30, 60, and 120 minutes (combination i); 60, 90, and 120 minutes (combination j); 30, 60, 90, and 120 minutes (combination k); and 30, 45, 60, 90, and 120 minutes (combination k).
Comparison of the SBSM and MBSM
To examine the relationship between GFRs obtained by the SBSM and the MBSM, iodixanol (40 mg of I/kg) was administered IV to 62 dogs (26 healthy dogs and 36 dogs with naturally occurring renal disease; group F). A blood sample (0.8 mL) was collected immediately before and 60, 90, and 120 minutes after tracer administration.
For the MBSM, clearance calculations were based on the 1- and 2-compartment models. The AUC for serum iodixanol or inulin concentration was calculated by the linear trapezoidal rule with extrapolation to infinity from the last 3 to 6 sample points (an elimination phase) in serum. Clearance was calculated from the following formula and was regarded as GFR:
To determine serum iodixanol clearance for the SBSM, the Vd of iodixanol in each dog in group G was back-calculated by substituting clearance and serum iodixanol concentrations as a function of time at 60, 90, or 120 minutes obtained by the MBSM into the following Jacobsson formula15:
where Ct is iodixanol concentration as a function of time and t is time.
This formula can be transformed to the following equation by the classic Newton method,19,20 and the variable b can also be solved by the same method:
The Vd obtained was then reconfirmed by use of commercially available spreadsheet software.c To examine the estimated Vd in each dog, an equation between the Vd and serum iodixanol concentration at 60, 90, or 120 minutes was determined from a scatter diagram. Finally, the GFR for the SBSM involving iodixanol was derived by substituting the dose of iodixanol (40 mg of I/kg), serum iodixanol concentration at 120 minutes, and estimated Vd (determined as t[Clearance/b]) calculated for each dog into the Jacobsson formula.15 The clearance term was regarded as the GFR for the present study. The GFR is represented as mL/min/kg or mL/min/m2 based on the body surface area (0.101 × [body weight in kg]0.71).21
To evaluate the relationship between GFR obtained by the SBSM involving iodixanol (40 mg of I/kg) and by the MBSM involving inulin (50 mg/kg), both tracers were coadministered IV to 25 dogs (22 healthy dogs and 3 dogs with renal disease; group H). A blood sample (1.2 mL) was collected immediately before and 30, 60, 90, and 120 minutes after administration of the tracers. In a preliminary study that used the same healthy dogs, no difference was noted between GFR estimated by iodixanol alone and that estimated by iodixanol and inulin in combination.
Clinical experiments
Serum iodixanol concentrations were measured by reverse-phase high-performance liquid chromatography according to a previously reported procedure.16 The detection limit of serum iodixanol concentration was 5 μg of I/mL. Validation experiments revealed no significant difference between serum and plasma iodixanol concentrations. The intra- and interassay CVs in serum iodixanol concentration determination (26 μg of I/mL; n = 6) were 0.8% and 3%, respectively. Serum inulin concentrations were colorimetrically determined by an autoanalyzer method with a commercially available kit.d The assay was consigned to an independent testing service.e The detection limit of serum inulin concentrations was 20 μg/mL. It was confirmed beforehand that when both drugs were commingled, there was no effect on either iodixanol or inulin concentration in canine serum in the in vitro additive experiments. Serum BUN, creatinine, calcium, and inorganic phosphate concentrations were measured with the autoanalyzerf on the same days that GFR was estimated. Urine creatinine and protein concentrations were also measured quantitatively with the autoanalyzer.f The relationship between the GFR derived by the SBSM versus BUN or serum creatinine concentration was assessed for 62 dogs (26 healthy dogs and 36 dogs with renal disease; group I).
Statistical analysis
Tests were used to evaluate whether data were normally distributed. Quantitative data obtained in the present investigation were expressed as the mean ± SD. Results in dogs with naturally occurring renal disease were represented as the median and range because the data were not normally distributed. Comparison of GFR between the SBSM and MBSM was performed according to standard recommendations for comparing analytic techniques based on Deming regression.22 A Bland-Altman test with modifications for repeated measurements23,24 was performed with software.g Values of P ≤ 0.05 were considered significant.
Results
Clinicopathologic profiles of healthy dogs and dogs with naturally occurring renal disease used in the present study were summarized (Table 2). In 3 healthy dogs each given iodixanol (20, 40, or 80 mg of I/kg [group A]), mean iodixanol concentration in serum decreased with semilogarithmic linearity from 30 to 180 minutes after tracer injection at doses of 40 and 80 mg of I/kg (Figure 1). After administration of iodixanol providing a dose of 20 mg of I/kg, serum iodixanol concentration was below the detection limit 180 minutes later. Considering the detection sensitivity, a dose of iodixanol providing 40 mg of I/kg was chosen. At an iodixanol dose providing 40 mg of I/kg, no significant difference was noted between GFR estimated from the 1-compartment model with 6 blood sample collection times (4.40 ± 0.82 mL/min/kg [corresponding to 69.5 ± 14.3 mL/min/m2]) and that estimated from the 2-compartment model with 10 blood sample collection times (4.21 ± 0.88 mL/min/kg [corresponding to 64.5 ± 14.8 mL/min/m2]). There was no significant difference between GFR estimated from 5 or 6 blood sample collection times versus 3 blood sample collection times (4.40 ± 0.70 mL/min/kg) in the 1-compartment model with various combinations of sample collection times (Figure 2). For subsequent investigations, therefore, a combination of iodixanol (40 mg of I/kg) with blood sample collection times of 60, 90, and 120 minutes after tracer injection was selected for the MBSM.
Mean serum iodixanol concentration following IV administration of iodixanol at a dose providing 20 (black diamonds), 40 (white circles), or 80 (black triangles) mg of I/kg (black triangles) to 3 healthy dogs (group A [A]) or providing 40 mg of I/kg (white circles) to another 3 healthy dogs (group C [B]). Administration of the tracer was designated as 0 minutes. A blood sample (0.8 mL) was collected immediately before and 5, 10, 15, 30, 45, 60, 90, 120, and 150 or 180 minutes after tracer administration. The sample collection times in panel A are 30, 60, 90, 120, 150, and 180 minutes after iodixanol administration. For 3 dogs receiving 20 mg of I/kg, a blood sample was not collected at 180 minutes. The sample collection times in panel B are 5, 10, 15, 30, 45, 60, 90, 120, 150, and 180 minutes after iodixanol administration.
Citation: American Journal of Veterinary Research 76, 9; 10.2460/ajvr.76.9.828
Box-and-whisker plot of GFR estimated (by means of a 1-compartment model) from data obtained at various combinations of blood sample collection times after iodixanol injection in 2 groups of 3 dogs (group C): at 30, 60, 90, 120, 150, and 180 minutes (combination a); at 30, 90, 120, 150, and 180 minutes (combination b); at 60, 90, 120, 150, and 180 minutes (combination c); at 30, 60, 90, and 120 minutes (combination d); at 30, 60, and 90 minutes (combination e); at 30, 60, and 120 minutes (combination f); and at 60, 90, and 120 minutes (combination g). For each box, the horizontal line represents the median and the upper and lower boundaries represent the 75th and 25th percentiles, respectively. Upper and lower bars represent maximum and minimum values, respectively.
Citation: American Journal of Veterinary Research 76, 9; 10.2460/ajvr.76.9.828
Mean (range) weight, age, and selected clinicopathologic variables of the healthy dogs and dogs with naturally occurring renal disease used in the study outlined in Table 1.
| Variable | Healthy dogs (n = 26) | Dogs with naturally occurring renal disease (n = 36) |
|---|---|---|
| Body weight (kg) | 12.7 (10.2–17.0) | 6.8 (2.2–9.8) |
| Age (y) | 6.3 (3.5–10.9) | 10.1 (2.3–15.5) |
| BUN (mg/dL) | 25.4 (19.5–29.5) | 44.7 (27.7–61.6) |
| Serum creatinine (mg/dL) | 0.69 (0.52–0.79) | 2.67 (1.00–4.23) |
| Serum calcium (mg/dL) | 10.2 (9.6–11.0) | 10.3 (9.6–11.3) |
| Serum inorganic phosphate (mg/dL) | 3.3 (2.5–3.7) | 4.3 (3.4–6.0) |
| Urine protein (g/g of creatinine) | 0.04 (0.01–0.17) | 1.96 (1.30–2.6) |
In 4 healthy dogs given inulin (50 or 100 mg/kg), mean inulin concentrations in serum decreased linearly (as seen for iodixanol) from 30 to 90 minutes after tracer injection at either dose (Figure 3). Considering the detection sensitivity and minimum exposure of the body to inulin, a dose of 50 mg/kg was chosen. There was no difference between GFR estimated from 5 blood sample collection times (4.30 ± 0.54 mL/min/kg) versus that estimated from 4 (4.10 ± 0.31 mL/min/kg) or 3 (4.02 ± 0.23 mL/min/kg) blood sample collection times in the 1-compartment model (data not shown). In subsequent investigations, therefore, a combination of inulin (50 mg/kg) with blood sample collection times of 30, 60, and 90 minutes after tracer injection was selected for the MBSM. The equation for calculating the estimated Vd was determined from the scatter diagram (Figure 4) as follows:
where C is serum iodixanol concentration at 120 minutes and e is the base of the natural logarithm. Serum iodixanol concentration at 120 minutes after tracer injection (r = 0.91; P < 0.001; n = 35) was chosen because at that time, there was a somewhat high correlation between Vd and serum iodixanol concentration, compared with that at 60 (r = 0.86) or 90 (r = 0.88) minutes.
Mean serum inulin concentration following IV administration of inulin at a dose of 50 (white circles) or 100 (black circles) mg/kg to 4 healthy dogs (group B [A]) or at a dose of 50 mg/kg to 3 healthy dogs (group E [B]). Administration of the tracer was designated as 0 minutes. A blood sample (0.8 mL) was collected immediately before and 5, 10, 15, 30, 45, 60, 90, and 120 or 150 minutes after tracer administration. The sample collection times in panel A are 30, 45, 60, 90, 120, and 150 minutes after inulin administration. For 3 dogs receiving 100 mg of inulin/kg, a blood sample was also collected at 150 minutes. The sample collection times in panel B are 5, 10, 15, 30, 45, 60, 90, and 120 minutes after inulin administration.
Citation: American Journal of Veterinary Research 76, 9; 10.2460/ajvr.76.9.828
Plot of estimated Vd and serum iodixanol concentrations at 120 minutes after IV administration of iodixanol at a dose providing 40 mg of I/kg in 26 healthy dogs (black diamonds) and 9 dogs with naturally occurring renal disease (white diamonds). From the data collected (group G), estimated Vd was 340.19e−0.017C (r = 0.91; P < 0.001), where C is serum iodixanol concentration at 120 minutes and e is the base of the natural logarithm.
Citation: American Journal of Veterinary Research 76, 9; 10.2460/ajvr.76.9.828
When the relationship between GFRs estimated by the MBSM involving inulin and the SBSM involving iodixanol was examined on the basis of Deming regression, close correlation was noted between the methods (r = 0.89; P < 0.001). On the basis of Bland-Altman bias analysis, 24 of 25 (96%) dogs were within the agreement plots, although 2 data points were outliers (Figure 5). With the SBSM involving iodixanol, the basal reference GFR in healthy dogs was 4.13 ± 0.99 mL/min/kg (corresponding to 63.2 ± 14.9 mL/min/m2; n = 26).
Bland-Altman plot of agreement between GFR estimated by the MBSM involving inulin (GFR inulin-MSBM) and the SBSM involving iodixanol (GFR iodixanol-SBSM) in 22 healthy dogs (black diamonds) and 3 dogs with naturally occurring renal disease (white diamonds). The solid line represents the mean bias (0.41 mL/min/Kg). The 95% confidence limits (dotted lines) of the agreement plot are from −1.04 to 1.87 mL/min/Kg.
Citation: American Journal of Veterinary Research 76, 9; 10.2460/ajvr.76.9.828
When the relationship between GFRs estimated by the MBSM and SBSM involving iodixanol was examined by means of Deming regression, close correlation was noted between the methods (r = 0.99; P < 0.001). On the basis of Bland-Altman analysis, 60 of 62 (97%) dogs were within the agreement plots, although 2 data points were outliers (Figure 6).
Bland-Altman plot of agreement between GFR estimated by the MBSM involving iodixanol (GFR iodixanol-MBSM) and the SBSM involving iodixanol (GFR iodixanol-SBSM) in 26 healthy dogs (black diamonds) and 36 dogs with naturally occurring renal disease (white diamonds). The solid line represents the mean bias (0.07 mL/min/Kg). The 95% confidence limits (dotted lines) of the agreement plot are from −0.71 to 0.86 mL/min/Kg.
Citation: American Journal of Veterinary Research 76, 9; 10.2460/ajvr.76.9.828
The relationship between GFR and BUN or serum creatinine concentration was assessed in 62 dogs. Plots (Figure 7) revealed that the correlation (r = 0.74; P < 0.01) between the GFR and serum BUN concentration (y = −27.3 ln x + 49.5) was low, compared with the correlation (r = 0.85; P < 0.01) between the GFR and serum creatinine concentration (y = −0.74 ln x + 1.54). No adverse clinical signs associated with iodixanol or inulin treatment were observed in any dog throughout the experiments.
Relationship between GFR and BUN (A) or serum creatinine (B) concentration derived from data collected from 26 healthy dogs (black diamonds) and 36 dogs with naturally occurring renal disease (white diamonds).
Citation: American Journal of Veterinary Research 76, 9; 10.2460/ajvr.76.9.828
Discussion
In the present investigation, we assessed the validity of the SBSM involving a bolus injection of iodixanol for estimation of GFR in dogs and compared that method with the conventional MBSM involving inulin. The advantages of the SBSM are that catheterization of the urinary bladder can be omitted, fewer analytic procedures are required, and precise timing of blood sample collection is unnecessary. Furthermore, the Jacobsson formula15 includes the dose, Vd, serum concentration of iodixanol, and sample collection time as variable factors. For veterinary practitioners, the merit of this procedure is that they may be able to explain the current patient situation quantitatively (as a percentage of the basal reference GFR) to the owner, whereas serum creatinine concentrations in the International Renal Interest Society renal disease stage definitions have somewhat wide ranges.
In healthy dogs given iodixanol (40 mg of I/kg) in the present study, a linear semilogarithmic plot of serum iodixanol concentrations versus time demonstrated the suitability of the use of a 1-compartment model for GFR calculation (MBSM). Although the 1-compartment model was a simplification and only applies after an equilibration period (30 to 120 minutes after iodixanol injection), it was thought to underestimate the AUC, compared with the 2-compartment model. Briefly, the AUC from the 1-compartment model was also approximately 5% to 10% lower than that from the 2-compartment model, indicative of higher GFR in the former. In addition, no significant difference was detected between the GFRs estimated from 5 versus 3 blood sample collection times in the 1-compartment model. Given these results, the approach involving 3 blood sample collection times (60, 90, and 120 minutes after tracer administration) was chosen because of a minimum SD of difference in GFR. From the same experiments as the aforementioned iodixanol investigations, the appropriate dose (50 mg/kg) and blood sample collection times (30, 60, and 90 minutes after tracer administration) of inulin were determined.
Under the conditions of the present study, the Vd for iodixanol ranged from 150 to 340 mL/kg and 10 to 200 mL/kg in healthy dogs (n = 26) and dogs with naturally occurring renal disease (9), respectively; however, published iodixanol Vd values for diseased dogs were not available for comparison. Generally, a compound that has a Vd of 50 to 60 mL/kg is considered to be in the bloodstream and a compound that has a Vd of < 600 mL/kg is considered to be in the extracellular fluid, when the compound distributes to blood and organs immediately after injection.25 The difference in Vd for iodixanol between the healthy dogs and dogs with renal disease in the present study implies that in dogs with reduced kidney functional mass, iodixanol was retained much more in the bloodstream as a consequence of decreased iodixanol filtration from the glomeruli. By use of the SBSM involving iodixanol, the basal reference GFR obtained in healthy dogs closely resembled previously reported GFR data,1–4,26 although the conditions and procedures of the present and previous studies were very different.
On the basis of data collected from the healthy dogs and dogs with naturally occurring renal disease in the present study, BUN and serum creatinine concentrations increased once GFR had decreased to approximately 70% of the mean basal level (4.13 mL/min/kg). These changes were consistent with data reported for dogs26 and other species.16,27 However, the shape of the scatter diagram of GFR versus serum creatinine concentration somewhat differed from that of the scatter diagram of GFR versus BUN concentration. This difference may be attributable to different renal handling of BUN and creatinine. Generally, BUN is reabsorbed in the proximal tubule, and decreased flow rate in this nephron segment substantially affects BUN reabsorption, whereas creatinine is mainly filtered and secreted only slightly. Moreover, BUN concentration is more influenced by high-protein meals.
The formula derived for calculating GFR with a single blood sample has been reported to require a known Vd, and the precision of that value seems to determine the accuracy of the method.14 Likewise, when the Vd of the tracer is known, the plasma disappearance curve can be closely approximated from a single, timed plasma measurement.28 In the present study, a relationship between the estimated Vd and serum iodixanol concentrations was needed to formulate an equation to calculate the estimated Vd. On the basis of Bland-Altman bias analysis, GFRs obtained from the study dogs were within or near the agreement plots. The Jacobsson formula15 can therefore be applied to dogs, and the SBSM may be used for GFR estimation as an alternative to use of the MBSM. However, further studies are necessary to elucidate the relationship between body weight and administered tracer volume because results of the present study may not apply to dogs of different weights with naturally occurring renal disease. In conclusion, findings of the present study indicated that the SBSM involving iodixanol can be applied to dogs in clinical and pharmacological settings.
Acknowledgments
The authors received no financial support for the research, authorship, or publication of this article.
The authors thank Dr. Anna Miyano for support with blood sample collection and Dr. Yoji Furuhama for providing information on the Newton method.
The authors declare no potential conflict of interest with respect to the research, authorship, or publication of this article.
ABBREVIATIONS
| AUC | Area under the blood concentration–time curve GFR Glomerular filtration rate |
| MBSM | Multiple-blood-sample method |
| SBSM | Single-blood-sample method |
| Vd | Volume of distribution |
Footnotes
Visipaque 320, Daiichi-Sankyo, Tokyo, Japan.
Inulead, Fuji Yakuhin, Saitama, Japan.
Goal-Seek function, Excel 2007, Microsoft Japan, Tokyo, Japan.
Dia-color-inulin, Toyobo, Osaka, Japan.
SRL, Tokyo, Japan.
Toshiba Medical Systems, Tochigi, Japan.
Prism, version 5, GraphPad Software, San Diego, Calif.
References
1. 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.
2. Von Hendy-Willson VE, Pressler BM. An overview of glomerular filtration rate testing in dogs and cats. Vet J 2011; 188: 156–165.
3. 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.
4. 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.
5. 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 dogs. Am J Vet Res 2009; 70: 176–185.
6. Miyamoto K. Use of plasma clearance of iohexol for estimating glomerular filtration rate in cats. Am J Vet Res 2001; 62: 572–575.
7. Tanaka T, Katayama H, Shirakata A, et al. Biochemical aspects on adverse reactions to contrast media. Changes of kininongen levels in dog plasma after intravenous injections of iohexol, and iothalamate. Invest Radiol 1992; 23(suppl 1): 195–196.
8. Heglund IF, Michelet ÅA, Blazak WF, et al. Preclinical pharmacokinetics and general toxicity of iodixanol. Acta Radiol Suppl 1995; 399: 69–82.
9. Jacobsen PB, Blindheim L, Skotland T. Bioanalytical methods for iodixanol and their application to studies on metabolism and protein binding. Acta Radiol Suppl 1995; 399: 61–66.
10. Svaland MG, Haider T, Langseth-Manrique K, et al. Human pharmacokinetics of iodixanol. Invest Radiol 1992; 27: 130–133.
11. Aspelin P, Aubry P, Fransson SG, et al. Nephrotoxic effects in high-risk patients undergoing angiography. N Engl J Med 2003; 348: 491–499.
12. Karlsberg RP, Dohad SY, Sheng R. Contrast-induced acute kidney injury (CI-AKI) following intra-arterial administration of iodinated contrast media. J Nephrol 2010; 23: 658–666.
13. McCullough PA, Bertrand ME, Brinker JA, et al. A meta-analysis of the renal safety of isosmolar iodixanol, compared with lowosmolar contrast media. J Am Coll Cardiol 2006; 48: 692–699.
14. Groth S, Aasted M. 51Cr-EDTA clearance determined by one plasma sample. Clin Physiol 1981; 1: 417–425.
15. Jacobsson L. A method for the calculation of renal clearance based on a single plasma sample. Clin Physiol 1983; 3: 297–305.
16. Katayama R, Yamaguchi N, Yamashita T, et al. Calculation of glomerular filtration rate in conscious rats by the use of a bolus injection of iodixanol and a single blood sample. J Pharmacol Toxicol Methods 2010; 61: 59–64.
17. Polzin DJ, Osborne CA, Ross SJ. Chronic kidney disease. In: Ettinger SJ, Feldman EC, eds. Textbook of veterinary internal medicine. 6th ed. St Louis: WB Saunders Co, 2005;1756–1785.
18. Japanese Association for Laboratory Animal Science. Guidelines for animal experimentation. Exp Anim 1987; 3: 285–288.
19. Smale S. On the efficiency of algorithms of analysis. Bull Am Math Soc 1985; 13: 87–121.
20. Varona JL. Graphic and numerical comparison between iterative methods. Math Intell 2002; 24: 37–46.
21. Price GS, Frazier DL. Use of body surface area (BSA)-based dosages to calculate chemotherapeutic drug dose in dogs: I. Potential problems with current BSA formulae. J Vet Intern Med 1998; 12: 267–271.
22. Deming WE. Statistical adjustment of data. Mineola, NY: Dover Publication, 1964.
23. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1: 307–310.
24. Bland JM, Altman DG. Measuring agreement in method comparison studies. Stat Methods Med Res 1999; 8: 135–160.
25. Ritschel WA. Handbook of basic pharmacokinetics. 3rd ed. Washington, DC: Drug Intelligence Publications Inc, 1986;193.
26. Finco DR, Brown SA, Vaden SL, et al. Relationship between plasma creatinine concentration and glomerular filtration rate in dogs. J Vet Pharmacol Ther 1995; 18: 418–421.
27. Michigoshi Y, Yamagishi N, Satoh H, et al. Using a single blood sample and inulin to estimate glomerular filtration rate in rabbits. J Am Assoc Lab Anim Sci 2011; 50: 702–707.
28. Harvey JN, Jaffa AA, Loadholt CB, et al. Measurement of glomerular filtration rate and renal plasma flow in the diabetic rat by the single-injection isotopic technique: effects of altered distribution volumes of 51Cr-EDTA and 125I-hippuran. Diabetes Res 1988; 9: 67–72.
