Measurement of GFR is essential to screen for potential nephrotoxic effects during preclinical drug evaluation. Although technically complex, clearance techniques are preferred for this purpose.1,2 Two basic protocols are currently used. One method consists of a continuous rate IV infusion of a GFA until PGFA and UVGFA reach steady-state values. The GFR is then calculated as UVGFA/PGFA.2 For the other method, the rate of plasma disappearance of the GFA is analyzed after it is administered as a single IV injection and the GFR is calculated as the instantaneous rate of change of GFA, where the change (d) in quantity of the agent (q) at a given time (t) is represented by the derivative equation dq(t)/dt.3,4 In both methods, plasma concentrations are determined by collection of multiple samples of blood and urine over several hours followed by deferred assays, which increase the duration and complexity of the test. Attempts have been made to simplify the procedures5–7 (eg, by testing a single blood sample8); however, at best, an estimate of GFR by use of these techniques requires a minimum period of 4 to 8 hours. The inconvenience of clearance techniques further increases when they are used to determine GFR in animals that have a high acquisition cost; such animals are generally in limited supply and require special housing and handling. Studies in nonhuman primates have particularly raised concerns about animal well-being, especially with the use of more invasive techniques. Consequently, the alternative of a more convenient and less invasive means to measure renal function in these animals would be welcomed.
The purpose of the study reported here was to validate a method to assess GFR in conscious rhesus monkeys via transcutaneous radiation detection after IV injection of 99mTc-DTPA. This method is based on a single-injection technique5,9,10 in which continuous and instantaneous measurement of radiation is performed transcutaneously by use of an ARM instead of intermittent and deferred assays of blood and urine. Because the supply of rhesus monkeys was limited, the results of the present study were also compared with similar measurements reported in studies11,12 of humans. Although the ideal marker for evaluation of GFR is controversial,13 we used 99mTc-DTPA (a radiopharmaceutical agent with clearance characteristics similar to those of inulin and 125I-iothalamate14–18) to test the new method. We compared GFR values obtained using this method with those obtained in the same monkeys via a standard clearance method with a continuous rate infusion of 125I-iothalamate.
Materials and Methods
Animals—Four healthy adult rhesus monkeys (Macaca mulatto [2 males and 2 females; weight range, 2.6 to 3.3 kg])a were used in the study. Monkeys were determined to be healthy on the basis of physical examination by a veterinarian and were housed individually. Their maintenance, housing, and monitoring were in accordance with national guidelines for the use and care of laboratory animals,19 with feed and tap water available ad libitum. The monkeys were trained to sit quietly for 1 to 2 hours on a restraining primate chair but were not accustomed to wearing the ARM used to estimate GFR. The study protocol was approved by the Animal Care and Use Committee of Charles River Laboratories.
Study protocol—On day 1, anesthesia was induced with ketamine hydrochlorideb (10 mg/kg, IM) and maintained with 2.5% isofluranec in oxygen via a face mask. Sterile physiologic saline (0.9% NaCl) solution was administered IV (30 mL/kg/h) during the experiment, and the monkeys were monitored until recovery Simultaneous measurements for estimation of GFR were performed in anesthetized monkeys via 2 methods. A direct IV bolus of 99mTc-DTPAd was administered for transcutaneous radiation detection by use of the ARM secured at the brachium of each monkey, while 125I-iothalamatee was administered simultaneously via continuous IV infusion for evaluation of renal clearance by use of blood and urine samples as described elsewhere.20 On days 8 and 45, GFR estimation was repeated by use of the ARM in the same monkeys in the conscious state after IV injection of 99mTc-DTPA without other treatments.
Quality control of the radioactive pharmaceutical agents used in this study was performed by the manufacturer of the products. The agents were assessed via measurement of the labeling efficiency and determination of the radionuclide, chemical, and radiochemical purity of the compounds.
ARM preparation—An ARM11,f (intended for use on humans) was altered for use on the monkeys as previously described.21 The device was modified to fit the small size of a monkey's arm; the number and type of detectors used were also changed. The modified ARM basically consisted of 2 radiation detectors with a data collection and analysis system (Figure 1). Each detector comprised a CsI(Tl)-crystal-photodiode array connected to a low-noise, high-gain shaping amplifier and a low-level discriminator. This assembly was shielded by a 4-mm lead housing with the crystal surface exposed. Each detector was mounted on a printed circuit board that contained the amplifier and discriminator circuitry. Detector assemblies were surface mounted on a hemicylindrical metal armband with the exposed crystal surface oriented to the inside. A computer module (to record and store detector data and allow data streaming to a personal computer) and the power supply (four 9-V batteries) were also surface mounted on the arm band. To provide comfort and support to reduce variability in radiation counts caused by motion, the inside of the arm band was padded with high-density memory foam with 2.06-cm (0.8-inch) thickness. The combined weight of the unit was 0.34 kg. After placement on the monkey's brachium, the instrument was secured with a hook-and-loop fasteningg strap (imitating the placement of a blood pressure cuff). Compression from the metal hemicylinder and tension from the hook-and-loop fastening strap provided an almost motion-free assembly between the monkey's brachium and the ARM. A software program developed in our laboratory was used to start and stop data collection by the ARM and to adjust settings for plotting data during acquisition.
Schematic representation of the modified ARM used to assess GFR in conscious monkeys via transcutaneous radiation detection after IV injection of 99mTc-DTPA. The device was secured on the left brachium, and transcutaneous radiation measurements were used to estimate GFR after IV administration of 99rnTc-DTPA (7.4 MBq/kg) in 4 healthy adult rhesus monkeys during anesthesia on day 1 and in the conscious state on days 8 and 45. Each detector (cross section, lower left) comprised a Csl(TI)-crystal-photodiode array and preamplifier connected to a low-noise, high-gain shaping amplifier; a low-level discriminator computer module; and 4 (9-V) batteries (because of space limitations, only 2 batteries are displayed). The components (depicted on the open inner aspect of the ARM, upper left) were contained and shielded in a 4-mm-thick lead housing assembly on the inside surface of a padded hemicylindrical arm band with the crystal surface exposed. Detectors were tested in vitro before use via measurement of the λ (reference value, 1.925 × 10–3/min) for various technetium 99m sources and were connected to a personal computer module (right) for data transfer and analysis.
Citation: American Journal of Veterinary Research 71, 12; 10.2460/ajvr.71.12.1492
Schematic representation of the modified ARM used to assess GFR in conscious monkeys via transcutaneous radiation detection after IV injection of 99mTc-DTPA. The device was secured on the left brachium, and transcutaneous radiation measurements were used to estimate GFR after IV administration of 99rnTc-DTPA (7.4 MBq/kg) in 4 healthy adult rhesus monkeys during anesthesia on day 1 and in the conscious state on days 8 and 45. Each detector (cross section, lower left) comprised a Csl(TI)-crystal-photodiode array and preamplifier connected to a low-noise, high-gain shaping amplifier; a low-level discriminator computer module; and 4 (9-V) batteries (because of space limitations, only 2 batteries are displayed). The components (depicted on the open inner aspect of the ARM, upper left) were contained and shielded in a 4-mm-thick lead housing assembly on the inside surface of a padded hemicylindrical arm band with the crystal surface exposed. Detectors were tested in vitro before use via measurement of the λ (reference value, 1.925 × 10–3/min) for various technetium 99m sources and were connected to a personal computer module (right) for data transfer and analysis.
Citation: American Journal of Veterinary Research 71, 12; 10.2460/ajvr.71.12.1492
Schematic representation of the modified ARM used to assess GFR in conscious monkeys via transcutaneous radiation detection after IV injection of 99mTc-DTPA. The device was secured on the left brachium, and transcutaneous radiation measurements were used to estimate GFR after IV administration of 99rnTc-DTPA (7.4 MBq/kg) in 4 healthy adult rhesus monkeys during anesthesia on day 1 and in the conscious state on days 8 and 45. Each detector (cross section, lower left) comprised a Csl(TI)-crystal-photodiode array and preamplifier connected to a low-noise, high-gain shaping amplifier; a low-level discriminator computer module; and 4 (9-V) batteries (because of space limitations, only 2 batteries are displayed). The components (depicted on the open inner aspect of the ARM, upper left) were contained and shielded in a 4-mm-thick lead housing assembly on the inside surface of a padded hemicylindrical arm band with the crystal surface exposed. Detectors were tested in vitro before use via measurement of the λ (reference value, 1.925 × 10–3/min) for various technetium 99m sources and were connected to a personal computer module (right) for data transfer and analysis.
Citation: American Journal of Veterinary Research 71, 12; 10.2460/ajvr.71.12.1492
Quality-control testing of the experimental technique—Radiation counts of 4 test sources (3 obtained by dilution of the same agent used in monkeys and 1 cobalt 57 from a commercial sourceh) were performed in vitro by use of a dose calibratori and subsequently analyzed by use of the ARM for quality control of the method. Source A contained 1.85 MBq of 99mTc-DTPA, which was accepted as having a radiation count rate similar to that measured transcutaneously at the brachium of monkeys immediately after IV administration of 7.4 MBq of 99mTc-DTPA/kg; source B contained 1.85 MBq of 125I-iothalamate; and source C (a sample of physiologic saline solution) contained no radiopharmaceutical agent. A radiation count (over a 10-second period) was obtained 5 consecutive times for each test sample. The signal-to-noise ratio was assessed as A:C, where A and C represented the radioactivity detected (counts/s) for sources A and C, respectively. Similarly, the interference of iodine 125 (125I) counts with technetium 99m (99mTc) counts was assessed as the ratio B:A, where B represented the radioactivity detected (counts/s) for source B. Radiation counting efficiency of the device was assessed as the ratio of A:D, where A represented the radioactivity detected (counts/s) for source A and D was the expected disintegration rate of the 99mTc-DTPA source (1.85 × 106 disintegrations/s) derived from measured activity. Sensitivity of the technique was defined by use of the limit of detection and limit of quantification22 from a source containing 0.3 MBq of cobalt 57 (terminal half-life, 271.79 days).
Quality-control testing of the ARM—Performance of the device was evaluated before its use on monkeys via measurement of λ (reference value, 1.925 × 10–3/min) for different dilutions of 99mTc-DTPA dissolved in 250 mL of physiologic saline solution. For these tests, the decay correction factor in the ARM (which automatically adjusts each measurement by λ) was inactivated.
Intra-assay accuracy was determined by comparison of the 15-minute slopes (from 0 to 15, from 16 to 30, from 31 to 45, and from 46 to 60 minutes) of the correlation between the natural logarithm of radiation counts versus time of a single source containing 1.85 MBq of 99mTc with the expected λ of 99mTc. Intra-assay accuracy was defined as the range of differences between the mean ± 2 SDs of the measured rate constants and the expected λ for 99mTc expressed as a percentage of the mean. Interassay accuracy was calculated on the basis of the 15-minute slope (from 0 to 15 minutes) of the correlation between the natural logarithm of radiation counts versus time of sources containing 1.4, 1.5, 1.8, and 1.9 MBq of 99mTc. Interassay accuracy was defined as the range of differences between the mean ± 2 SDs of the measured rate constants in multiple sources and the expected λ for 99mTc expressed as a percentage of the mean. Precision was determined by comparison of the slope of the natural logarithm of radiation counts versus time measured for 8 consecutive periods of 15 minutes each in a single source containing 1.85 MBq of 99mTc-DTPA (intra-assay) or the sources containing 1.4 to 1.9 MBq of 99mTc-DTPA (interassay), each recorded for a total of 120 minutes. Precision was expressed as the coefficient of variation calculated as (SD/mean) × 100 at the 15- and 60-minute time points.
Estimation of GFR via use of the ARM—The same ARM protocol was used on day 1 in anesthetized monkeys and on days 8 and 45 in conscious monkeys. Before each use, the ARM was connected to a personal computer via a parallel port to load the operating program. The device was then disconnected and placed and secured on the monkey's left brachium in a manner similar to the placement of a blood pressure cuff. A single dose of 7.4 MBq of 99mTc-DTPA/kg was administered via direct IV injection, and transcutaneous measurements of radiation in the contralateral limb were performed over 10-second intervals without interruption for 30 minutes (a total of 180 measurements). Measurements were recorded as the natural logarithms of radiation count versus time. After the final measurement, the ARM was removed and reconnected to the personal computer and the data were transferred to a spreadsheetj for analysis.
Renal clearance of the radioactive GFA (99mTc-DTPA) was assessed via analysis of the slope of activity versus time (ie, κ values) determined from transcutaneous radiation measurements after automatic correction for λ. The κGFR was calculated as the product of κ × 16.74 minutes (ie, the slope of correlation between a previously reported GFR value11 and κ).
Estimation of GFR via 125I-iothalamate clearance—A continuous rate infusionk of 125I-iothalamate and timed blood and urine sample collections were used to determine GFR in the anesthetized monkeys. After dilution of the radiopharmaceutical agent (0.222 MBq/kg in 10 mL of saline solution), a priming injection of this diluted solution (3.0 mL, IV) was initiated, followed by the INFA at a rate of 0.0037 MBq/min. The urinary bladder was catheterized by use of a pediatricsized urinary catheter.
Samples of blood (2 mL obtained via an indwelling catheter placed in a cephalic vein contralateral to the infusion site) and urine (obtained via syringe attached to the urinary catheter; the urinary bladder was emptied and flushed with 10 mL of saline solution) were collected at 15- minute intervals after a 30-minute equiliberation period. The 125I radioactivity of samples was determined by use of a γ radiation counterl until UVGFA and PQFA each reached steady state (ie, similar count rates were determined in 2 consecutive samples). The UVGFA was determined as the product of UGFA (counts/min/mL) × VUP (mL/min; determined from sample volume vs time of collection). Plasma was obtained via centrifugation of blood samples in untreated glass tubes (15 minutes at 1,200 × g), and 0.5 mL of plasma was used to measure PGFA (counts/min/mL).
Once steady state was determined, 2.0-mL blood samples were collected via IV catheter at 15-minute intervals for 45 minutes (for a total of 3 samples). Urine samples were also obtained via urinary catheter (with some difficulty) by completely emptying and flushing the urinary bladder with 10 mL of saline solution every 15 minutes; plasma and urine samples were stored at 4°C until use. A 10-μL sample of the infusate was also obtained. The 125I radioactivity of samples was measured 7 days after the experiment to allow for complete decay of 99mTc (which has a terminal half-life of 6 hours and was administered concurrently for the comparison experiment), and GFR was calculated according to the formula INFA/PGFA for individual sample values at each time point. Values for UVGFA and PGFA were determined as previously described; the GFR was calculated again as UVGFA/PGFA. Values for GFR determined via both methods were normalized by body surface area calculated on the basis of body weight and height.23
Statistical analysis—The GFR data are expressed as mean ± SD. Comparisons between samples were performed via paired Student t test analysis, and linear regression analysis of natural logarithms of radiation counts versus time was used to calculate the rate constants via a commercially available statistical analysis software program.m A value of P < 0.05 was considered significant.
Results
Quality-control tests—Sources containing 1.85 MBq of 99mTc-DTPA (source A), 1.85 MBq of 125I-iothalamate (source B), or no radiopharmaceutical agent (source C) were analyzed by use of the ARM for quality control of the experimental method. Measured (mean ± SD) radioactivity was 1,530 ± 0.9 counts/s, 7 ± 0.2 counts/s, and 2 ± 0.5 counts/s for sources A, B, and C, respectively. Results of analysis of the signal-to-noise ratio of a source with a radiation count rate similar to that measured transcutaneously at the brachium of the monkeys immediately after injection of 7.4 MBq of 99mTc-DTPA/kg (A:C ratio, 765) and the ratio of interference of 125I counts over 99mTc counts (B:A ratio, 5 × 10–3) indicated that almost all of the radiation detected by the ARM when both radiopharmaceutical agents were administered was emitted from 99mTc-DTPA. The radiation counting efficiency of the ARM determined via comparison of actual radiation detection versus the expected disintegration rate of test source A was 0.083%. Limits of detection and quantification used to evaluate sensitivity were 0.2349 × 10–3/min and 0.4087 × 10–3/min, respectively.
The measured λ of a single 99mTc source obtained by use of the ARM in vitro (mean ± SD intra-assay values, 1.978 ± 0.105 × 10–3/min for 15 minutes and 1.910 ± 0.103 × 10–3/min for 30 minutes) was highly similar to the expected λ (1.925 × 10–3/min). The 15-minute intra-assay accuracy was calculated as 92% to 114%, and precision was 5.3%; values for 60 minutes (accuracy 89% to 110%; precision, 5.4%) were comparable. The measured 99mTc λ from multiple sources obtained with the ARM (1.992 ± 0.105) was also similar to the expected λ. The 15-minute interassay accuracy was calculated as 98% to 109%, and precision was 5.3%.
Estimation of GFR via use of the ARM—The natural logarithms of radiation count versus time were recorded by use of the ARM following administration of 99mTc-DTPA to monkeys (Figure 2). A rapid increase in count rate was detected during the first 5 minutes after injection; this was attributable to mixing of the radioactive GFA in the vascular space. After 5 minutes, the radiation count rates decreased versus time with first-order kinetics; this decrease represented the physical decay of the 99mTc as well as renal clearance of the radioactive GFA.
Plot of natural logarithm (In) values of transcutaneous radiation counts (obtained on day 8 of the study by use of the ARM depicted in Figure 1) versus time in a representative conscious adult rhesus monkey after IV injection of 74 MBq of 99mTc-DTPA/kg. The ARM was placed around the left brachium of the monkey, and transcutaneous measurements of radiation were performed over 10-second intervals without interruption for 30 minutes (for a total of 180 measurements) after 99mTc-DTPA was administered (at 0 minutes) in the contralateral limb. The progressive increase in signal obtained a few minutes after IV bolus injection of 99mTc-DTPA represented mixing of the radioactive agent in the vascular space. Following a short equilibration time, the radiation count rate decreased with time according to first-order kinetics.
Citation: American Journal of Veterinary Research 71, 12; 10.2460/ajvr.71.12.1492
Plot of natural logarithm (In) values of transcutaneous radiation counts (obtained on day 8 of the study by use of the ARM depicted in Figure 1) versus time in a representative conscious adult rhesus monkey after IV injection of 74 MBq of 99mTc-DTPA/kg. The ARM was placed around the left brachium of the monkey, and transcutaneous measurements of radiation were performed over 10-second intervals without interruption for 30 minutes (for a total of 180 measurements) after 99mTc-DTPA was administered (at 0 minutes) in the contralateral limb. The progressive increase in signal obtained a few minutes after IV bolus injection of 99mTc-DTPA represented mixing of the radioactive agent in the vascular space. Following a short equilibration time, the radiation count rate decreased with time according to first-order kinetics.
Citation: American Journal of Veterinary Research 71, 12; 10.2460/ajvr.71.12.1492
Plot of natural logarithm (In) values of transcutaneous radiation counts (obtained on day 8 of the study by use of the ARM depicted in Figure 1) versus time in a representative conscious adult rhesus monkey after IV injection of 74 MBq of 99mTc-DTPA/kg. The ARM was placed around the left brachium of the monkey, and transcutaneous measurements of radiation were performed over 10-second intervals without interruption for 30 minutes (for a total of 180 measurements) after 99mTc-DTPA was administered (at 0 minutes) in the contralateral limb. The progressive increase in signal obtained a few minutes after IV bolus injection of 99mTc-DTPA represented mixing of the radioactive agent in the vascular space. Following a short equilibration time, the radiation count rate decreased with time according to first-order kinetics.
Citation: American Journal of Veterinary Research 71, 12; 10.2460/ajvr.71.12.1492
Individual κ values obtained via analysis of the slope of 99mTc-DTPA activity versus time determined from radiation counts obtained via ARM showed less variability than the values of GFR determined via analysis of 125I-iothalamate clearance after blood and urine collections (Tables 1 and 2). The coefficient of variation of κGFR determined in monkeys by use of the ARM ranged from 6.5% to 22.7%. The mean K value determined on day 1 in anesthetized monkeys was 16% to 23% less than that measured on days 8 and 45 in conscious monkeys. The κGFR was also less on day 1 than on days 8 and 45.
The κ and κGFR values calculated on the basis of transcutaneous radiation detection data (obtained by use of an ARM) following IV administration of 99mTc-DTPA (7.4 MBq/kg) in 4 healthy adult rhesus monkeys that were anesthetized and received concomitant IV administration of 125l-iothalamate (0.0037 MBq/min) via constant rate infusion on day 1; on days 8 and 45, the monkeys were conscious and received no other treatment.
| Study day | ||||
|---|---|---|---|---|
| Variable | Monkey | 1 | 8 | 45 |
| κ (×; 10−3/min) | 1 | 5.25 | 8.10 | 10.79 |
| 2 | 8.58 | 8.97 | 8.04 | |
| 3 | 5.89 | 8.09 | 9.88 | |
| 4 | 7.75 | 7.68 | 6.85 | |
| Mean ± SD κ value (× 10−3/min) | — | 6.87 ± 1.56 | 8.20 ± 0.53 | 8.89 ± 1.77 |
| Mean ± SD κGFR | ||||
| (mL/min × 1.73 m2) | — | 115 ±26 | 137 ± 9 | 149 ± 30 |
Transcutaneous measurement of radiation was performed by use of the ARM secured around the left brachium of each monkey. Values were recorded in 10-second increments without interruption for 30 minutes (for a total of 180 measurements) after 99mTc-DTPA was administered (at 0 minutes) in the contralateral limb. The rate of renal clearance of the radioactive GFA (99mTc-DTPA) was assessed via analysis of the slope of activity versus time (ie, κ values) determined from transcutaneous radiation measurements after automatic correction for physical decay (ie, λ values). The κGFR was calculated as the product of κ × 16.74 minutes (ie, the slope of the correlation between GFR [normalized by body surface area]11 and κ). Although the mean value of κGFR on day 1 was smaller than values on days 8 and 45, the differences were not significant (P > 0.05).
— = Not applicable.
Physical characteristics, 126l-iothalamate clearance variables, and estimated GFR values determined from analyses of blood and urine samples collected from the 4 healthy adult rhesus monkeys in Table 1 (during anesthesia on day 1) following IV administration of 125l-iothalamate (0.0037 MBq/min] via constant rate infusion.
| Monkey | ||||
|---|---|---|---|---|
| Variable | 1 | 2 | 3 | 4 |
| Sex | Male | Male | Female | Female |
| Weight (kg) | 2.6 | 2.7 | 3.3 | 3.0 |
| Body surface area (m2) | 0.22 | 0.22 | 0.25 | 0.24 |
| INFA/PGFA(mL/min) | ||||
| 15 min | 19.5 | 16.2 | 17.6 | 18.1 |
| 30 min | 17.4 | 15.9 | 17.6 | 17.4 |
| 45 min | 15.5 | 15.3 | 16.1 | 17.4 |
| Calculated GFR* (mean ± SD[mL/min × 1.73 m2]) | 137 ± 16 | 137 ± 4 | 118 ± 6 | 127 ± 3 |
| UVGFA/PGFA (mL/min) | ||||
| 15 min | 9.7 | 7.9 | 17.6 | 13.3 |
| 30 min | 8.7 | 10.4 | 20.6 | 14.0 |
| 45 min | 7.7 | 6.4 | 10.6 | 24.6 |
| Calculated GFR† (mean ± SD[mL/min × 1.73 m2]) | 68.4 ± 9.9 | 64.7 ± 20.4 | 108.2 ± 38.5 | 124.7 ± 60.8 |
The monkeys also received a bolus injection of 99mTc-DTPA (7.4 MBq, IV) for concurrent transcutaneous radiation detection testing. After UVGFA and PGFA reached steady state following commencement of 125l-iothalamate infusion (at 0 minutes), blood samples were collected at 15-minute intervals for 45 minutes to obtain plasma. The urinary bladder was emptied and flushed with 10 mL of saline (0.9% NaCl) solution via a urinary catheter every 15 minutes. Samples were stored for 7 days to allow for radioactive decay of concurrently administered 99mTc-DTPA. A 0.010-mL sample of the infusate was obtained to calculate GFR according to the formula INFA/PGFA; 0.5 mL of plasma was used to measure PGFA (counts/min/mL). The UGFA (counts/min/mL) was measured by use of a γ radiation counter, and VUP (mL/min) was determined from the sample volume; GFR was then recalculated as UVGFA/PGFA. Values for GFR were normalized by body surface area.23 There was no significant difference between GFR values for male and female monkeys measured via this method.
GFR determined as INFA/PGFA normalized by body surface area.
GFR determined as UVGFA/PGFA normalized for body surface area.
Estimation of GFR via 125I-iothalamate clearance—Measurements obtained from blood and urine samples collected after IV administration of 125I-iothalamate were used to calculate GFR values (Table 2). The mean GFR values obtained by use of UVGFA and PGFA measurements were smaller than obtained by use of values for INFA, particularly in the 2 male monkeys. This variability was likely attributable to the variability in volumes of urine produced between collection periods.7 The substantial reduction in variability observed when calculations of GFR were performed with INFA values instead of UVGFA measurements was independent of the sex of the animals and collection periods.
Discussion
Quality-control testing of GFR estimations obtained via standard clearance techniques that rely on collection of blood and urine samples is difficult and often unreliable.2,24 This is largely due to the technical complexity of the tests. Other factors contribute to reduced accuracy when values are obtained and recorded by use of transcutaneous radiation detection devices. For instance, a method for estimation of GFR by measurement of 99mTc-DTPA radiation with transcutaneous CdTe detectors was reported to be the least accurate of several methods tested in 1 study24 However, accuracy in that study was compromised by the large amount of noise obtained with CdTe detectors, compared with that of the CsI(Tl)-crystal-photodiode array detectors used in the study reported here. An additional factor that contributed to this difference was the substantial reduction in the frequency of measurements (1/120 seconds vs 1/10 seconds, respectively) between that study and the present study The long counting interval required by the CdTe detectors is likely a reflection of the reduced efficiency and increased noise of that type of detector. The accuracy and response time of each technique depend on the type of detectors and settings used in the instrument during data collection. Consequently it is critical that the performance of the instrument and settings for transcutaneous detection of radiation be tested before use in monitors of this type.
Although not entirely comparable to measurements obtained in animals, the determination of λ by use of the ARM in vitro provides a simple means to test the performance of the device in the low range of GFR values and independent of biological variability In effect, this is a suitable method to establish a standard to which the κ values obtained via transcutaneous measurements in monkeys could be compared. The test can be performed easily via measurement of λ of a 99mTc source with data collected without decay correction. This type of quality-control test may help to reduce variability among transcutaneous measurements obtained in monkeys. Because the λ of 99mTc happens to be similar to the κ value obtained in patients with as much as a 70% reduction in GFR, the measurement of λ is an excellent method to perform quality testing of ARM even in a range that corresponds to stage III or moderate renal insufficiency in the classification of chronic kidney disease in humans.25
The results of analysis of the signal-to-noise ratio and of the interference of 125I counts with 99mTc counts indicated that almost all (99.5%) of the radiation detected by the ARM when both agents were present was emitted from 99mTc-DTPA. The difference in cumulative doses (7.4 MBq/kg for 99mTc-DTPA vs 0.222 MBq/kg for 125I-iothalamate), difference in photon energy between radiopharmaceutical agents, and use of a correct setting for the lower γ energy discriminator were factors in achievement of this result. Additionally, because the dose of 125I-iothalamate was infused during a period of several hours, the interference from this agent when the transcutaneous measurements were obtained was expected to be even lower than that determined in the present study.
The radiation counting efficiency of the modified ARM was increased from 0.054% obtained with the device designed for use in humans11 to 0.083% obtained with the modified device. Moreover, because this improvement was achieved with minimal changes in background activity, there was a net improvement in the signal-to-noise ratio.
Despite device modifications to accommodate the size of the monkey's arm and the use of CsI(Tl) detectors instead of CdTe detectors, the signal-to-noise ratio and radiation counting efficiency of the ARM used in the present study were similar to, or better than, those determined for the original device.11 Moreover, the sensitivity, accuracy, and precision determined in vitro with a source of 99mTc were excellent.
Because the accuracy and precision were very similar between the 15- and 60-minute measurements, we conclude that the measurement of radioactivity for the determination of GFR or GFR-equivalent κ could be performed as frequently as once every 15 minutes while still maintaining good statistical power. This is in agreement with similar results obtained in humans11,12 and monkeys,21 which indicated that the data processing time was much shorter than the period of several hours required by standard clearance techniques to evaluate GFR. This greatly reduced data processing time represents a unique and important advantage of the technique, particularly for preclinical safety testing of multiple drugs, because potential nephrotoxic effects can be detected rapidly and with better assurance regarding the relationship between cause and effect.
The variability of GFR values obtained via 125I-iothalamate clearance analysis of blood and urine samples in the present study likely resulted from incomplete urine collection, as any residual urine volume would have accounted for a large fraction of the small urine volumes produced by the monkeys during the 15-minute sampling intervals. This possibility was supported by the substantial reduction in variability detected when estimates of GFR were calculated on the basis of values for INFA instead of UVGFA, independent of sex of the monkeys. This was also in agreement with results of other studies,7,26,27 which indicated that GFR values calculated on the basis of values for INFA were more precise and accurate than those calculated on the basis of the UVGFA. The GFR values obtained during INFA in the study reported here were consistent with values obtained in rhesus monkeys by use of other techniques and various GFAs.20,21
The rather simple kinetics detected by use of the ARM in the present study are in contrast to the more complex patterns of blood radiation counts versus time. The decrease with time in blood sample radioactivity that follows a single injection of a radiopharmaceutical agent is represented by 3 stages.4 The first stage is detected immediately after injection and corresponds to the mixing of the agent in the vascular compartment. The second stage is represented by a rapid decrease in blood sample radioactivity, which corresponds to diffusion of the agent from the vascular compartment to the interstitial compartment. Finally, the third stage is represented by a slow decrease in blood sample radioactivity that follows first-order kinetics and corresponds to clearance of the agent by the kidneys. As a result of these more complex kinetics, attempts to simplify and shorten the duration of the test by reducing the number of blood samples obtained have been partially successful8,27 because several hours are required for the renal clearance of the agent to reach the third stage. This is in contrast to the extremely short time (≤ 5 minutes) required to reach first-order kinetics with the ARM detection technique. This unique characteristic of transcutaneous whole-tissue recording is a consequence of the simultaneous detection of radioactive signals from the interstitial and vascular compartments.7 Because transcutaneous whole-tissue radiation detection cannot resolve the vascular and interstitial compartments as separate entities, the system operates as a single compartment that comprises only the extracellular element.
Because of the variability of the data, it is unclear whether the difference detected between day 1 and days 8 and 45 represented changes in GFR induced by general anesthesia, as has been described in reports of anesthesia and surgery in rats28 and humans.29 Nevertheless, the ability to determine GFR in conscious monkeys would eliminate the concern for any effect of anesthesia as a confounding factor in nephrotoxicity tests.
The coefficient of variation of κGFR determined in monkeys by use of the ARM ranged from 6.5% to 22.7% with a mean value of 14.7%, which is similar to the value obtained in human studies that involved the same technique12 and approximately half the coefficient of variation obtained with standard clearance techniques.1 Although the measurement of κ is less accurate and precise than the measurement of λ, results are at least similar to measurements in multiple blood and urine samples in this and a previous study11 Moreover, the use of the λ determination as a standard for sample analysis should help to determine at what step in the measurement of κ a decrease in accuracy or precision may occur. With this approach, we were previously able to identify motion as one of the most important factors in κ variability11 Consequently, we expect the accuracy and precision of κ values to reach a similar quality to that determined for the measurement of λ with better training and improved immobilization of the ARM.
Body size is indirectly measured via determination of variables such as weight and body surface area. The κ values obtained in the study reported here were highly similar to rate constants obtained in rhesus monkeys with twice the body weight of the monkeys used in the present study21 and similar to the rate constant determined in humans with normal renal function and body size approximately 20 times as great as that of monkeys in the present study11,12 This indicates, as contended in another report11 by our group, that no adjustment of κ is necessary to correct for differences in body size as is typically done for clearance techniques. The similarity of κ values between humans and monkeys indicates that the rate constant for the clearance of 99mTc-DTPA measured transcutaneously with the ARM likely represents a measurement of the relative efficiency of the kidneys to clear the GFA from body compartments independent of body size, as was reported30,31 for 125I-iothalamate after normalization on the basis of body surface area. In other terms, the measurement of κ likely represents a more pertinent way to measure renal function than to measure the absolute value of GFR by use of standard clearance techniques. A meaningful comparison of renal function can be performed independent of age, sex, and size of the subjects by use of this alternative method.
Measurements of GFR by use of the ARM are indirect. In the authors' experience in a human intensive care unit setting, the correlation detected between the rate of excretion of 99mTc-DTPA and GFR is observed only in hemodynamically stable patients under steady-state circumstances. Space or volume variation imposed by fluid resuscitation or extreme local blood flow changes at the site of detection could affect the values for GFR. The correlation detected via this method is valid only when the rate-limiting step in the excretion of 99mTc-DTPA is attributable to kidney function.
Because the ARM radiation detection method requires minimal restraint, has a brief measurement time of only 15 to 30 minutes, and is minimally invasive, we conclude that it represents a fast, sufficiently accurate, convenient, and less-invasive way to measure GFR in conscious nonhuman primates.
Abbreviations
| 125l-iothalamate | lothalamate sodium I 125 |
| 99mTc-DTPA | Technetium Tc 99m pentatate |
| ARM | Ambulatory renal monitor |
| CdTe | Cadmium telluride |
| Csl | Cesium iodide |
| GFA | Glomerular filtration agent |
| GFR | Glomerular filtration rate |
| INFA | Infusion of glomerular filtration agent |
| κ | Renal rate constant |
| κGFR | Glomerular filtration rate determined by use of the renal rate constant |
| λ | Physical decay constant |
| PGFA | Plasma concentration of glomerular filtration agent |
| TI | Thallium |
| UGFA | Urine concentration of glomerular filtration agent |
| UVGFA | Rate of urinary excretion of glomerular filtration agent |
| VUP | Rate of urine production |
Three Springs Scientific, Perkasie, Pa.
Ketamine hydrochloride, S. S. Pharmaceuticals, Morris, Okla.
Isoflurane, Abbott Laboratories, Abbott Park, Ill.
Technetium Tc 99m–labeled diethylenetriamine pentaacetic acid, Cardinal Health, Woburn, Mass.
Sodium iodide I 125-labeled iothalamate, Cardinal Health, Woburn, Mass.
Ambulatory renal monitor, Cuyo Tech Inc, Chelsea, Mass.
Velcro, Velcro USA Inc, Manchester, NH.
Cobalt-57, Cardinal Health, Woburn, Mass.
Capintec CRC 15B Dose Calibrator, Capintec Inc, NJ.
Excel, Microsoft Corp, Redmond, Wash,
AS40A infusion pump, Baxter, Deerfield, Ill.
Wizard 1470, PerkinElmer, Waltham, Mass.
GraphPad Prism, version 5, Graph Pad Software Inc, San Diego, Calif.
References
- 2↑
Smith HW. Measurement of the filtration rate. In: The kidney: structure and function in health and disease. New York: Oxford University Press, 1951;39–62.
- 3
Sapirstein LA, Vidt DC, Mandel MJ, et al. Volumes of distribution and clearances of intravenously injected creatinine in the dog. Am J Physiol 1955; 181:330–336.
- 4↑
Hall JE, Guyton AC, Farr BM. A single-injection method for measuring glomerular filtration rate. Am J Physiol 1977; 232:F72–F76.
- 5
Swinkels DW, Hendriks JCM, Nauta J, et al. Glomerular filtration rate by single injection inulin clearance: definition of a workable protocol for children. Ann Clin Biochem 2000; 37:60–66.
- 6
Thirimurthi K, Casey M, Barker F, et al. Glomerular filtration rate measurements with a CdTe detector system in patients undergoing renal scintigraphy Nucl Med Allied Sci 1984; 28:245–250.
- 7↑
Earle DP, Berliner RW. A simplified clinical procedure for measurement of glomerular filtration rate and renal plasma flow Proc Soc Exp Biol Med 1946; 62:262–264.
- 8↑
Tepe PG, Tauxe WN, Bagchi A, et al. Comparison of measurement of glomerular filtration rate by single sample, plasma disappearance slope/intercept and other methods. Eur J Nucl Med 1987; 13:28–31.
- 9
Donath A. The simultaneous determination in children of glomerular filtration rate and effective renal plasma flow by the single injection clearance technique. Acta Paediatr Scand 1971; 60:512–520.
- 10
Jung K, Henke W, Schulze BD, et al. Practical approach for determining glomerular filtration rate by single-injection inulin clearance. Clin Chem 1992; 38:403–407.
- 11↑
Rabito CA, Moore RH, Bougas C, et al. Noninvasive, real-time monitoring of renal function: the ambulatory renal monitor. J Nucl Med 1993; 34:199–207.
- 12↑
Rabito CA, Scott JA & Rubin-Tolkoff N. Effortless and accurate measurement of glomerular filtration rate in individual kidney of prospective donors. Transplantation 2010;in press.
- 14
Barbour GL, Crumb CK, Boyd CM, et al. Comparison of inulin, iothalamate, and 99mTc-DTPA for measurement of glomerular filtration rate. J Nucl Med 1976; 17:317–320.
- 15
Blaufox MD, Potchen EJ, Merrill JP. Measurement of effective renal plasma flow in man by external counting methods. J Nucl Med 1967; 8:77–85.
- 16
Klopper JF, Hauser W, Atkins HL, et al. Evaluation of 99mTc-DTPA for the measurement of glomerular filtration rate. J Nucl Med 1972; 13:107–110.
- 17
Rabito CA, Fang LST, Waltman AC. Renal function in patients at risk of contrast material-induced acute renal failure: noninvasive, real-time monitoring. Radiology 1993; 186:851–854.
- 18
Rabito CA, Panico F, Rubin R, et al. Noninvasive, real-time monitoring of renal function during critical care. J Am Soc Nephrol 1994; 4:1421–1428.
- 19↑
Institute of Laboratory Animal Research, Commission on Life Sciences, National Research Council. Guide for the care and use of laboratory animals. Washington, DC: National Academies Press, 1996;21–25.
- 20↑
Timchalk C, Finco DR, Quast JF. Evaluation of renal function in rhesus monkeys and comparison to Beagle dogs following oral administration of the organic acid trilopyr (3,5,6-trichloro-2-pyridinyloxyacetic acid). Fundam Appl Toxicol 1997; 36:47–53.
- 21↑
Haug CE, Lopez IA, Moore RH, et al. Real-time monitoring of renal function during ischemic injury in the rhesus monkey. Ren Fail 1995; 17:489–502.
- 23↑
Liu CT, Higbee GA. Determination of body surface area in the rhesus monkey. J Appl Physiol 1976; 40:101–104.
- 24↑
Waller DG, Keast CM, Fleming JS, et al. Measurement of glomerular filtration rate with technetium-99m DTPA: comparison of plasma clearance techniques. J Nucl Med 1987; 28:372–377.
- 25↑
Levey AS, Coresh J, Balk E, et al. National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med 2003; 139:137–147.
- 26
Ficsher PA, Bogoliuk CB, Ramirez AJ, et al. Technical note. A new procedure for evaluation of renal function without urine collection in rat. Kidney Int 2000; 58:1336–1341.
- 27
Schnurr E, Lahme W & Küppers H. Measurement of renal clearance of inulin and PAH in the steady state without urine collection. Clin Nephrol 1980; 13:26–29.
- 28↑
Walker LA, Buscemi-Bergin M & Gellai M. Renal hemodynamics in conscious rats: effects of anesthesia, surgery, and recovery Am J Physiol 1983; 245:F67–F74.
- 29↑
Mazze RI, Callan CM, Galvez ST, et al. The effects of sevoflurane on serum creatinine and blood urea nitrogen concentrations: a retrospective, twenty-two-center, comparative evaluation of renal function in adult surgical patients. Anesth Analg 2000; 90:683–688.
- 30
Peters AM, Allison H, Ussov WYu. Simultaneous measurement of extracellular fluid distribution and renal function with a single injection of 99mTc DTPA. Nephrol Dial Transplant 1995; 10:1829–1833.
- 31
Visser FW, Muntinga JHJ, Dierckx RA, et al. Feasibility and impact of the measurement of extracellular fluid volume simultaneous with GFR by 125I-iothalamate. J Am Soc Nephrol 2008; 3:1308–1315.
