Clinical trials in cats with naturally occurring chronic kidney disease (CKD) are critical to advance new diagnostic and therapeutic opportunities. Evaluation of client-owned cats presents challenges, including variable husbandry practices, comorbidities, patient and client recruitment, variations in concurrent medications, and owner compliance, among others. Additionally, the rate of CKD progression can vary substantially and is often slow in cats.1,2 This can prolong clinical trials and make optimal study duration, enrollment criteria, and enrollment numbers difficult to predict. For these reasons, the establishment and characterization of CKD models in cats are important to study species-specific mechanisms of disease initiation and progression, new biomarkers of renal function, and novel therapeutics in a controlled and more homogenous population. These models serve as crucial tools to understand CKD in cats and provide relevant translational information for other species.
Remnant kidney models have been developed and utilized in multiple species including cats,3–8 mice,9 rats,10–16 dogs,17–19 and pigs,20,21 and are commonly used to study the progression, clinical outcomes, and possible treatments for CKD.14,15,17 The model is achieved by partial amputation or infarction of 1 kidney (remnant) and simultaneous or delayed nephrectomy of the contralateral kidney. Creation of a remnant kidney can be performed by ligation of branches of the renal artery,5,6,10,18,22 partial nephrectomy,12,15,20,21 partial nephrectomy with wrapping7,19 or arterial embolization.21 A goal of the renal injury and contralateral nephrectomy is to reduce global renal function to below 25% of baseline, ideally to 8–17%, by causing a functional 11/12th to 5/6th nephrectomy.
In cats, remnant kidney models have been studied for durations ranging from 7 days to 1 year after injury.3–8 The main objective of this study was to describe acute and chronic outcomes after induction of 11/12th remnant kidney injury in otherwise healthy, purpose-bred cats. A secondary objective was to evaluate acute clinicopathologic variables as predictors of outcome. We hypothesized that cats with a greater acute reduction in renal function acutely after arterial ligation would have an increased risk of short-term mortality, and that chronically the remnant kidney injury model would mimic naturally occurring CKD in cats.
Materials and Methods
General design
This is a descriptive study of 32 cats with surgically induced CKD using a 2-stage remnant kidney model. Cats were enrolled consecutively in the study in 6 cohorts, including 1 pilot cohort of 4 cats (2 females, 2 males), and 5 cohorts of 2–7 cats. Cohorts 1 and 2 were all female, and cohorts 3–5 were all male cats. The pilot cohort was initiated in January 2019, with others following over several months. Approximately 7 months after the first cohort underwent vessel ligation, 23 cats were transported to a contract research facility, then returned to the home facility approximately 6 months later. Two cats remained at the home facility during this transfer. The last data reported here are from October and December 2022.
Detailed below, the first surgery involved selective ligation of renal vasculature in 1 kidney (day −28) followed by contralateral nephrectomy (day 0). Including baseline data collection, acute results reported here detail data from day −45 to day 29. During the chronic phase (day 30 to day >1,100, depending on the cohort), cats were enrolled in 3 independent, non-invasive, therapeutic studies with a treatment phase lasting 6 weeks or less. Details of those studies are not reported here and are not expected to have any meaningful impact on long-term outcome. Data obtained at the baseline of each study are used in this report as chronic results. Other data were gathered at time points before and after the transfer, as part of routine monitoring of the colony, or as part of a more intensive monitoring plan for individual cats, as medically indicated. A minimum of 3 months elapsed from the conclusion of 1 independent study to baseline data generation from another study.
Animals
Thirty-two intact, purpose-bred, mixed-breed cats, 15 females and 17 males, with a mean age of 9.1 months ± 0.13 (6 to 10 months) and a mean ± SD body weight of 4.3 kg ± 0.12 were obtained from a commercial source (Marshall Bioresources, North Rose, NY).
Cats were housed at the University of Georgia in an indoor vivarium with a controlled environment that maintained 12–15 air changes/hour, 12 hours of light/dark cycle, and room temperatures of 22.2 ± 2 °C. The cats were single-, or group-housed in USDA-approved cages with food and water bowls, an elevated shelf or hammock, a litter box, and toys for environmental enrichment. During the acute phase, cats were fed 1 cup (75 grams) per day of a maintenance adult diet (Cat Chow, Purina). During the chronic phase, cats transitioned to a prescription renal diet (Hills k/d or Royal Canin Multifunction Renal + Hydrolyzed). Over time, in response to general increases in body condition scores, the amount of food was adjusted to individual cat requirements. Municipal tap water was provided ad libitum. All procedures were approved by the Institutional Animal Care and Use Committee (Animal Use Protocol A2018 10-001).
Before induction of the model, general health at baseline was evaluated with a complete physical exam, complete blood count (CBC), serum biochemistry profile, including serum symmetric dimethylarginine (SDMA) concentration, urinalysis, urine protein: creatinine (UPC) ratio, and glomerular filtration rate (GFR) estimated by iohexol clearance.23
Anesthesia
Cats were fasted 12–18 hours before surgery. The water remained available. Cats were premedicated with ketamine (7 mg/kg, IM), acepromazine (0.01 mg/kg, IM), and buprenorphine (0.03 mg/kg, IM). An IV catheter was placed and perioperative IV fluids were administered (crystalloid isotonic solution, 3 mL/kg/hour). Anesthesia was induced by a facemask and maintained with a cuffed endotracheal tube with isoflurane and 100% oxygen.
Remnant kidney surgery model
Cats were positioned in right lateral recumbency and a left paracostal celiotomy was performed (day −28). The left renal artery was exposed and its branches were identified; 1 of the 2 major branches was ligated to stop the blood flow to the sagittal half of the kidney. The remaining branch was traced into the hilus, and additional branches were ligated using 4-0 silk, with the goal of ligating vessels associated with 1 or both poles and leaving the blood supply to the body of the kidney intact. Given the unique arterial branching pattern in each kidney and the surgical accessibility of smaller secondary and tertiary arterial branches, complete uniformity in arterial ligations between cats was not possible. A reduction of 5/6 of the renal blood supply was targeted, as previously described (Figure 1).7,8 Beginning with cohort 4, an ipsilateral nephropexy was performed to the adjacent body wall. This was initiated because of concerns about the potential for renal pedicle torsion after the dissection of perirenal tissues. During the same surgical episode, a flank ovariohysterectomy or an orchiectomy was performed to facilitate the cohousing of animals. Twenty-eight days after the ischemia surgery, a contralateral nephrectomy was performed through a right paracostal celiotomy (day 0). Immediately after each surgery, a fentanyl patch (3–5 µg/hour, transdermal) was placed, and cats were administered buprenorphine (0.03 mg/kg transmucosal q8h for 24 hours).
Intraoperative appearance of the left kidney during remnant kidney surgery. The kidney is being exteriorized through a lateral left flank incision. (A) Normal kidney before vessel ligation. (B) Kidney post-ischemia, notice the darker ischemic cranial pole and some sections of the caudal pole and the lighter more normal appearing island of renal parenchyma at the body and caudal pole.
Citation: American Journal of Veterinary Research 84, 8; 10.2460/ajvr.23.04.0081
Postoperative evaluation
Cats were evaluated at least twice daily for 5 days after either surgery. The general attitude, discomfort, food consumption (g), incision status, hydration, and presence of feces (normal vs. diarrhea), urine, and emesis were assessed. Cats were assigned a discomfort score (Supplementary Table S1). If any cat was assigned a total score >8, a 4 in any category or a 3 in the interactive category, then additional analgesia was administered with buprenorphine (0.03 mg/kg, transmucosal) or hydromorphone (0.05 mg/Kg, IM) if the fentanyl patch was still in place. All cats were administered amlodipine (1.25 mg/cat, PO, q24h) for 45 days after the ischemia surgery to prevent acute hypertensive encephalopathy previously described in this model.6–8
Outcome measurements
Cats were fasted from food for at least 6 hours before blood or urine collection; water was not restricted. Blood samples were routinely processed for CBC and biochemical analyses. Urine was collected via cystocentesis or from litter pans with non-absorbent plastic beads.
All hematologic, biochemical, and urine testing, other than GFR and serum SDMA concentration, were performed by the Clinical Pathology Laboratory in the College of Veterinary Medicine of the University of Georgia. For GFR and SDMA analyses, blood was centrifuged at 3000 rpm for 10 minutes, serum separated, and stored at −20 to −80 °C. Measurement of SDMA was performed by a commercial laboratory (IDEXX), and GFR analysis was performed by Michigan State University Diagnostic Center for Population and Animal Health.
Acute phase outcomes (days –45 to 29)
In the acute phase, all sampling was done at specific time points relative to day 0 for each cat. Serum creatinine and SDMA were measured at baseline (day −45 to −31), before nephrectomy (day −1), and on days 1, 5, 9, 15, 20, and 29 after nephrectomy. A CBC and serum biochemistry profile were performed at baseline and acute phase endpoint (day 29). Urine samples were collected at baseline, before nephrectomy (day −1), and on days 1, 2, 5, 9, 15, 20, and 29 after nephrectomy. Urine specific gravity (USG) was measured in all samples. UPC was determined at baseline and acute endpoint (day 29). Blood pressure (BP) was not measured during the acute phase of the study.
Glomerular filtration rate was estimated by using 3-point iohexol clearance (hours 2, 3, and 4 after IV administration) at baseline and the acute phase endpoint, as previously described.23
Chronic outcomes (days 30 to 1,337)
As described earlier, samples in the chronic phase were obtained irrespective of each cat’s day 0. Because enrollment of cats was done in cohorts over several months, data collection times relative to the individual cat’s day 0 varied over the duration of the study.
Symptomatic care was initiated as needed for clinical signs associated with CKD. Cats were evaluated more frequently if they showed signs of illness and laboratory analyses were performed as necessary to monitor their health status. Isotonic fluids (∼100 mL/cat) were administered subcutaneously for dehydration or during periods of inappetence. Cats with chronic, non-regenerative anemia (HCT <20%) were administered darbepoetin (1 µg/kg, SQ, weekly) and iron dextran (50 mg/kg, IM, monthly). In addition to fluids, mirtazapine (1.88 mg/cat, PO or transdermal, every other day), maropitant citrate (4 mg/cat, PO, once daily), or both were used to treat inappetence in cats. Potassium gluconate (1–2 mEq/kg, PO, twice daily) or potassium chloride (10–30 mEq/L in SQ fluids, once daily) were used for potassium supplementation in cats with moderate to severe hypokalemia (serum potassium concentration <3.0 mEq/L).
Chronically, cats determined to be persistently hypertensive were treated with amlodipine (0.625 −2.5 mg/cat PO, q24h). At the contract research facility, indirect BP measurement was performed using an oscillometric device. At UGA, BP was measured using Doppler sphygmomanometry following American College of Veterinary Internal Medicine guidelines.24 Indirect BP measurements were assumed to represent systolic BP (SBP). The decision to institute antihypertensive therapy was in accordance with the above guidelines, that is, in cats with repeatable systolic blood pressure over 160 mmHg.24
Renal ultrasound
A renal ultrasound examination was performed in all cats at the end of the chronic phase. Examinations were performed by the same operator (M.E.) using an Aloka Noblus unit with a high-frequency linear transducer (5–18 MHz; Hitachi Medical Corp.). All cats were examined in right lateral recumbency. Kidneys were assessed using B-mode imaging, and the following criteria were recorded: renal length, width, and height, renal shape, corticomedullary definition, subjective echogenicity of the renal cortex relative to the liver, and pelvic dilatation. The renal volume was subsequently manually calculated using the prolate ellipsoid method (volume = length X width X height X 0.523). Sonographic abnormalities were described.
Euthanasia and pathology
In both the acute and chronic phases, humane endpoints included serum creatinine concentration >12 mg/dL, surgical discomfort unable to be controlled by analgesic medication, not eating >48 hours, or other clinical signs unresponsive to medical therapy. Euthanasia was performed in all cats by administration of euthanasia solution (pentobarbital 390 mg/mL, 1 mL/4.5 kg, IV).
All euthanized animals underwent a gross necropsy examination. For cats in the early chronic phase euthanized outside of UGA, the data was obtained from the cat’s necropsy report. At the home facility, the remnant kidney and any grossly abnormal tissues were collected for histopathology. Tissue was processed routinely for histopathology, and sections (3 µm) were stained with hematoxylin and eosin, periodic acid-Schiff and hematoxylin, and Mason trichrome–PASH combination stains. Kidneys were grouped according to whether the cat died in the acute phase or in the chronic phase, and characteristic lesions were described for each group by a board-certified veterinary pathologist (D.R.R. or C.A.B.).
Statistical analysis
For the acute phase, continuous data variables are described with mean ± SD, and survival percentage is reported. A mixed model ANOVA was used to compare serum creatinine and SDMA concentrations, and USG over time during the acute phase. Other data commonly associated with kidney function or injury (GFR, Hct, serum urea nitrogen [SUN], potassium, phosphorus concentrations, and UPC) were compared between baseline and day 29 using a paired Student t test.
Generalized linear mixed models were used to test for the effects of serum creatinine and SDMA concentrations, and USG values on odds of survival with a binomial distribution and logit link as appropriate for binomial data. The model had a covariate for concentration, day and a concentration by day interaction, and a random intercept for each cat to account for repeated measurements for every cat. Satterthwaite degrees of freedom method and Residual Pseudo likelihood estimation were used. P-values testing for effects of concentrations/values at baseline and day −1, as well as the change between day −1 and baseline (represented by the interaction effect) were reported. A logistic regression was used to test for the effect of cat order on survival, and the log-likelihood ratio test P-value was reported. Fisher’s exact tests, due to small expected counts, were used to test for associations of cohort, sex, or nephropexy with survival.
In the chronic phase, cats were stratified based on the International Renal Interest Society (IRIS) acute kidney injury (AKI) grade on day 29, and median survival times were determined using Kaplan-Meier estimation. Curves were compared with the Long-rank (Mantel-Cox) test. Given the confounding effects of loss of more severely affected cats over time, symptomatic treatment to some cats, lack of uniform data in all cats, and the relatively small number of cats that were euthanized, statistical evaluation of data at other time points were not attempted. Changes in clinical parameters commonly abnormal in cats with CKD are described over time. These parameters included SBP, Hct, serum creatinine, phosphorus, potassium, SDMA concentrations, USG, and UPC. The correlation between the most recent creatinine or SDMA with USG was evaluated with the 2-tailed, Spearman correlation. All assumptions for statistical tests were met, including normality as appropriate. A P value <0.05 was considered significant.
Results
Acute outcomes
Survival—All cats survived both anesthetic episodes and surgeries without immediate complications. No cat showed clinical signs of renal dysfunction or was euthanized before the contralateral nephrectomy was performed. Mean post-operative discomfort scores after ischemia surgery were 3.78 ± 0.97 (day −27), 3.39 ± 0.50 (day −26), and 2.97 ± 0.33 (day −25), and after nephrectomy were 3.80 ± 1.03 (day 1), 3.30 ± 0.61 (day 2), and 3.25 ± 0.78 (day 3). No cat required rescue analgesia.
Measures of renal function—All cats had measurable deterioration of renal function after contralateral nephrectomy (Figure 2). The mean serum creatinine concentration at baseline and at the end of the acute phase (day 29) was 1.13 ± 0.23 mg/dL and 3.03 ± 1.2 mg/dL, respectively. The mean serum SDMA concentration at baseline and on day 29 was 12.44 ± 2.69 mg/dL and 41.04 ± 19.42 mg/dL, respectively. Compared with the baseline, serum creatinine and SDMA concentrations were significantly greater (P < .001) for all acute time points following the baseline. The mean baseline and day 29 USG values were 1.052 ± 0.0098 and 1.030 ± 0.0097, respectively. The USG was significantly reduced (P < .001) at all time points. The mean day 29 GFR was also significantly reduced from baseline (3.22 mL/min/kg ± 0.12 at baseline, 1.21 mL/min/kg ± 0.081 at day 29, P < .001).
Mean ± SD acute serum creatinine (A) and symmetric dimethylarginine (SDMA; B) concentrations, urine specific gravity (C) after partial renal arterial ligation (day −28) and contralateral nephrectomy (day 0). * = significant change from baseline, P < .05. Paired glomerular filtration rates (GFR; D) estimated by iohexol clearance in cats before (Baseline) and 28 days after remnant kidney surgery and contralateral nephrectomy. The GFR significantly decreased (P < .001).
Citation: American Journal of Veterinary Research 84, 8; 10.2460/ajvr.23.04.0081
Survival and pathology—In the acute phase, mortality was 22% (7/32); six cats were euthanized on day 2, and 1 on day 6. In all cats, the cause of death was euthanasia triggered by serum creatinine concentrations over the humane endpoint.
In cats euthanized in the acute phase, abnormal findings at necropsy were pleural effusion (n = 2), peritoneal effusion (n = 2), and pericardial effusion (n = 2). Gross evidence of pulmonary edema was present in 3 cats. Gross evidence of necrosis of the remnant kidney was found in 3 cats, and the remaining 4 cats had very small remnant kidneys. The histologic findings in cats euthanized in the acute phase were subtotal (n = 4) or total (n = 2) renal infarct. Histology was not available for 1 cat. Cats with total infarcts had no viable renal parenchyma. Renal tubules in the collapsed infarcted parenchyma were lined by hypereosinophilic epithelial cells with karyorrhectic or karyolitic nuclei and were often occluded by granular casts or clusters of basophilic minerals. There was also glomerular necrosis in all affected kidneys and renal papillary necrosis in cats with total infarction. Affected tubules were often surrounded by layers of neutrophils with adjacent interstitial clusters of lymphocytes and plasma cells, macrophages with cytoplasmic lipid droplets, and mild to moderate fibrosis. Areas of vascular thrombosis were observed in 2 cats. One cat with no necrosis had chronic changes characterized by diffuse, marked interstitial fibrosis with thickening and wrinkling of the tubular basement membrane.
Risk factors for acute death—day −28, day −1, or difference (day −28 to day −1) in serum creatinine (P = .33, .75, and .66, respectively); SDMA (P = .58, .497, and .80, respectively); or USG (P = .55, 0.91, and .59, respectively) were not associated with euthanasia post nephrectomy. Cat order (P = .093), cohort (P = .057), sex (P = .40), and whether a nephropexy was performed (P = .15) were also not statistically significant predictors for acute phase mortality.
Chronic outcomes
Survival—Twenty-five cats survived the acute phase and were included in the chronic phase data. At the end of the acute phase (day 29), 1 cat had IRIS AKI grade I, 9 cats had grade II, 13 had grade III, and 2 cats had grade IV. Ten cats were euthanized during the chronic phase of the study at a median of 163 days (50–736) after nephrectomy. Median survival time was significantly different based on AKI grade at day 29; MST for cats with grade I (n = 1) and II (n = 9) AKI was not reached (n = 0 cats euthanized over the study for either grade I or II); MST for cats with grade III AKI (n = 13) was 709 days (n = 8 cats euthanized over the study period); and MST for cats with grade IV AKI (n = 2) was 134 days (n = 2 cats euthanized over the study period; Figure 3; P < .001). The most recent data represent 15 cats at a median of 1,259 days (1,141 to 1,337) from nephrectomy.
Kaplan-Meier survival curve of 25 cats that survived at least 29 days after remnant kidney surgery and contralateral nephrectomy. Survival curves were significantly different when cats were stratified by the International Renal Interest Society (IRIS) Acute Kidney Injury (AKI) grade at day 29 (P < .001). For cats with grade I (n = 1) and grade II (N = 9) AKI, median survival time (MST) was not reached; grade III (n = 13) AKI MST was 709 days; grade IV (n = 2) MST was 134 days.
Citation: American Journal of Veterinary Research 84, 8; 10.2460/ajvr.23.04.0081
Chronic renal function—Chronic renal function, as evidenced by serum creatinine and SDMA concentrations, was reduced from normal, but stable over time. Including the month of recovery with the contralateral kidney in place, renal function tended to improve for approximately 6 months before stabilizing. During and after this stabilization period, some cats had acute deterioration in renal function (Figures 4 and 5). Of the 15 cats alive at the end of the study, 13 of 15 (86.7%) were in IRIS stage 2 CKD and 2 of 15 (13.3%) were in IRIS stage 3. Regarding proteinuria substaging, 11 of 15 (73.3%) were non-proteinuric (UPC < 0.2), 3 of 15 (20%) were borderline proteinuric (UPC = 0.2–0.4), and 1 of 15 (6.7%) were proteinuric (UPC < 0.4).25 Regarding BP substaging, 8 of 15 (53.3%) were normotensive (systolic BP < 140 mmHg), 4 of 15 (26.7%) were prehypertensive (systolic BP 140–159 mmHg), 1 of 15 (6.7%) were hypertensive (systolic BP 160–179 mmHg), and 2 of 15 (13.3%) were severely hypertensive (systolic BP ≥ 180).25
Individual cat serum creatinine (A), SDMA (B), hematocrit (C), phosphorus (D), and potassium (E) concentrations of 25 cats from day 29 onward in a colony of cats after remnant kidney surgery and contralateral nephrectomy. The dashed lines represent the reference interval, except SDMA where it represents the upper limit of normal.
Citation: American Journal of Veterinary Research 84, 8; 10.2460/ajvr.23.04.0081
Individual cat urine specific gravity (A) and urine protein: creatinine (B) of 25 cats from day 29 onward in a colony of cats after remnant kidney surgery.
Citation: American Journal of Veterinary Research 84, 8; 10.2460/ajvr.23.04.0081
The first BP measurement was performed with an oscillometric device between 34 and 188 days post-nephrectomy. At that time, 7 of 21 (33.3%) were normotensive, 3 of 21 (14.3%) were prehypertensive, 2 of 21 (9.5%) were hypertensive, and 9 of 21 (42.9%) were severely hypertensive. Subsequently, Doppler BP was measured regularly. During the chronic phase, 6 of 15 (40%) of the cats developed severe hypertension. Treatment with oral amlodipine resulted in the reduction of BP to prehypertensive levels.
Cats were variably anemic over time. The anemia (hematocrit < 33.6%) was commonly non-regenerative or very weakly regenerative. Within the last year, almost all cats were no longer anemic, which corresponds to a period of minimal blood sampling within the colony. Except for several cats with acute hyperphosphatemic episodes, phosphorus concentrations stayed within normal limits and stable over time. Cats were hyper- or hypokalemic depending on their state of renal function and food intake. Generally, potassium concentrations normalized over the last year of the study.
Urine-specific gravity was variable in this model, and some cats maintained USG > 1.035 despite having persistent azotemia and no clinical evidence of dehydration (Figure 5). Using the most current data, there was no correlation between USG and serum creatinine (r = 0.08, P = .76) or SDMA (r = 0.27, P = .34) concentrations.
Throughout the chronic course of the study, proteinuria occurred in a subset of cats. Ten cats were non-proteinuric (UPC < 0.2) at all time points for which UPC was evaluated, 7 cats were borderline proteinuric (UPC = 0.2 to 0.4), and 8 cats were proteinuric (UPC > 0.4) on at least 1 sample over the course of the study. The median highest UPC for each proteinuric cat was 0.68 (0.42 to 5.06). Four cats with UPC values over 0.4 at some point in the study were borderline proteinuric on subsequent readings, and 3 cats with the highest UPC values (5.06, 1.56, and 0.98, respectively) died in the early chronic phase with only 1 urine sample being evaluated during the chronic phase.
Currently, medical therapy is limited to a renal diet in all cats, antihypertensive therapy in 6 cats, and subcutaneous fluid therapy and potassium supplementation (100 mL SC 3 days a week) in 1 cat.
Ultrasound examination—Uniformly, all remnant kidneys were irregular in shape and had overall increased echogenicity, with reduced corticomedullary definition (Figure 6). While the specific individual renal morphology was variable, kidneys frequently lost the normal prolate spheroid shape and became more globoid with a less pronounced difference between length and width. Two kidneys had large, well-defined hypoechoic areas within the body or pole of the kidneys, attributed to atypical cysts, hematomas, abscesses, or necrotic areas. Two kidneys had ultrasonographic evidence of chronic infarcts. Remnant kidneys were small with mean length = 3.45 ± 0.84 cm, width = 2.63 ± 0.48 cm, height = 2.63 ± 0.40 cm, and volume = 12.74 ± 4.66 cm3. The renal pelvis was generally small (<0.1 cm) and often difficult to observe. Two cats had mild pelvic dilation at 0.11 and 0.2 cm, without evidence of ureteral dilatation.
Sagittal ultrasound images of the left kidney of 4 different cats from the colony. Cranial is to the left on all images. (A–D) All the kidneys had irregular margins, overall increased echogenicity, and reduced corticomedullary definition. (B) A small, well-defined, rounded, anechoic structure is present at the corticomedullary junction along the caudal pole of the kidney, consistent with a cyst (arrowhead). (C) A large, wedge-shaped, cortical defect is present along the caudal pole of the kidney, consistent with a chronic infarct (arrow). (D) The kidney is overall reduced in size compared with the other cats. A large, well-defined, primarily anechoic structure containing mobile, punctate echogenic foci is present within the caudal pole of the kidney (asterisk).
Citation: American Journal of Veterinary Research 84, 8; 10.2460/ajvr.23.04.0081
Cats euthanized in the chronic period—Six of the 10 cats euthanized in the chronic period were euthanized during the temporary relocation of the colony in the early chronic phase between days 50–164 (Supplementary Table S2). Before euthanasia, clinical signs included depression (n = 6), anorexia (6), pale mucous membranes (2), and poor body condition (2). These cats were hyperphosphatemic and azotemic; 1 cat had markedly elevated liver enzymes and elevated creatine kinase. Another cat developed acute blindness after a brief episode of depression. That cat had bilaterally intact pupillary light reflexes with no menace reflexes. A BP or retinal examination was not reported, although the clinical suspicion was that the cat suffered hypertensive encephalopathy.
After the return of the cats to the home facility, four additional cats were euthanized between days 350 to 736 because of the progression of CKD characterized by azotemia and hyperphosphatemia, as well as clinical signs associated with uremia. Cats were treated with subcutaneous fluids for variable durations before euthanasia. Three cats showed clinical signs compatible with acute on chronic kidney disease with creatinine values of 23.7, 13.0, and 8.6 mg/dL immediately before euthanasia. One cat developed chronic progressive anemia and hypokalemia that did not resolve with iron and darbepoetin therapy.
Pathology—Necropsy reports were reviewed for 6 cats euthanized at the contract research facility. One cat had a focally extensive renal abscess with fibrosis, although no bacteria were observed or isolated on tissue culture. One cat had a focally extensive renal infarct with severe fibrosis and lymphoplasmacytic inflammation, and random, hepatocellular necrosis suggestive of septicemia of unknown origin. Two cats had ischemic renal necrosis with fibrosis, 1 was diffuse and the other was regionally extensive; both had severe cardiac hypertrophy. Two cats had chronic lymphoplasmacytic, fibrosing interstitial nephritis with marked cortical atrophy, 1 had extensive cardiac myocyte degeneration and necrosis within the left ventricle, and the other had left ventricular hypertrophy and stenotic left ventricular lumen suspicious of cardiac hypertrophy.
Of the four cats euthanized at the home facility, no abnormalities were identified at necropsy, with the exception of small, pale remnant kidneys. Histologically, changes consisted of extensive (n = 2) or diffuse (1) parenchymal collapse due to tubular loss, mild to moderate interstitial fibrosis, and clusters of interstitial lymphocytes and plasma cells. Less often, there was a chronic subtotal polar infarct with tubular and glomerular loss and free lipid (n = 2), thickening and wrinkling of the tubular basement membrane (1), subtotal medullary tubular necrosis (1), and a mineralized renolith (1).
Discussion
The remnant kidney model described here is effective in inducing CKD in cats, mimics naturally occurring CKD in many ways, and provides a valuable tool for research of this disease. Injury to the remnant kidney was significant in most cats, and renal function required approximately 6 months to stabilize. No predictors of survival were identified in the acute phase, but animals with more significant injury did have reduced long-term survival. Chronically, many of the characteristics of the colony reported here mirror the biochemical features and clinical progression of cats with naturally occurring CKD. Thus, we accept our hypotheses.
Acute mortality has previously not been reported in this model; whether this is a result of mortality not occurring or not being reported, is unknown. The necrosis reported at necropsy in cats euthanized acutely was likely a consequence of thromboembolism of non-ligated vessels or, in cats that did not have a nephropexy, renal pedicle torsion. This is supported by microscopic evidence of intravascular thrombus in several cats. Although nephropexy was not identified as a protective factor, given the degree of dissection around the renal pedicle required to identify and ligate renal arterial branches and the potential for type II error, it may still be an important maneuver to consider intraoperatively. This is particularly true if the cranial and caudal poles are freed from retroperitoneal attachments.
Chronically, the variable progression of renal disease for the cats reported here mimics observations for cats with naturally occurring CKD.1 Azotemia occurred uniformly, but its degree was variable. When cats were classified by IRIS AKI grades on day 29, the degree of azotemia was predictive of MST. Although the scales are different, the survival times reported here relative to AKI grades are similar to those reported in cats based on IRIS CKD staging, where cats in stage 3 and 4 CKD had MSTs of 770 and 103 days, respectively.2
Chronically, as most cats in the colony were in IRIS CKD stage 2, hyperphosphatemia was not common, except in those cats with progressive decline in renal function. This is similar to cats with naturally occurring CKD where hyperphosphatemia is observed significantly more in cats in IRIS stage 3 and stage 4 compared with cats in stage 2.26 Forty percent of cats alive at the time of reporting were hypertensive or receiving antihypertensive therapy, which is in the range of what is reported for hypertension in cats with naturally occurring CKD.27–29 Mild to moderate non-regenerative anemia was also common in the colony. This seemed to be exacerbated by blood sampling despite conservative sampling protocols, and Hct gradually normalized when the cats were not sampled for long periods of time. This suggests the regenerative capability was still present, but markedly reduced.
A difference between the remnant kidney model and spontaneous CKD in cats is that unlike naturally occurring CKD,30 the remnant kidney model did not frequently reduce USG to the degree, one would expect relative to the magnitude of concurrent azotemia. While there was a significant acute reduction in USG in the acute phase, in chronically affected cats, there was no inverse correlation between creatinine or SDMA and USG. Nonetheless, although frequently used as a diagnostic criterion for renal azotemia and CKD, USG < 1.035 may not be seen in all cats with spontaneous CKD, which has been acknowledged by other authors.31,32
Other major limitations between this cohort of cats and those with naturally occurring CKD are their youth and the lack of comorbidities. Naturally occurring CKD is typically a disease of older cats and endocrine, cardiac, gastrointestinal, or other comorbid conditions are frequently diagnosed in these cats. Limitations specific to the technical aspects of the model include inherent individual variation in renal arterial supply, a subjective determination of which and how many vessels to ligate, and potential post-dissection thromboembolism of neighboring, non-ligated small arterial branches. The potential for this may be reduced by limiting perivascular inflammation with very careful dissection, use of less reactive sutures for ligation, or administration of systemic anticoagulants in the acute post-operative period. While nephropexy was not shown to reduce acute mortality in this population of cats, its use prevents renal torsion, which has catastrophic consequences, so its continued use is logical.
The remnant model creates a dramatic acute reduction in normal, functional renal tissue. Over time, nephrons adapt to this “overwork” scenario by hyperfiltration and glomerular hypertrophy, which leads to further damage, ongoing nephron loss, and eventual organ failure.31 After remnant kidney surgery, the degree to which cats resemble those with naturally occurring CKD likely depends on the amount of functional kidney tissue remaining, the ability of that functional mass to adapt to hyperfiltration, the progression of maladaptive responses to hyperfiltration, and the time since injury. In cats, acute changes in single nephron GFR and renal morphology after a less severe three-fourths remnant model have been described.6 In that study, increases in single nephron GFR and glomerular hypertension were documented, and glomerular hypertrophy and expansion of the mesangial matrix were also present.6 Although focused on dogs, a recent review discusses in detail the pathophysiology of the remnant model.33 In the model described here and distinct from “polectomy” remnant models where partial nephrectomy is performed, the presence of hypoxic and necrotic tissue, consequential activation of the renin-angiotensin-aldosterone system, and inflammation likely also play a role in the eventual outcome.34 A review of these two different approaches in rodents has been recently published.34 Briefly, nephrectomy and infarction models both produce similar uremia, fibrosis, and hyperfiltration, among other changes, but a major difference is expected postoperative hypertension.34 Because of the activation of the renin-angiotensin systems from the remaining hypoxic tissue, infarction models are characterized by early moderate to severe hypertension, whereas partial nephrectomy models are typically normotensive initially and develop mild to moderate hypertension over time.34 In the study reported here, at least in part, the mortality of cats in the early chronic phase might be related to hypertensive disease. More aggressive monitoring and treatment of blood pressure may have reduced mortality in these cats. In conclusion, the feline remnant kidney model provides a critically valuable tool for more rapid investigation of CKD in a homogenous population of cats. Acute morbidity may occur and cannot be predicted by standard measures of renal function before contralateral nephrectomy. The chronic course of cats with remnant kidney-induced CKD is similar to cats with naturally occurring CKD and is predicted by AKI grades at 29 days after contralateral nephrectomy.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org
Acknowledgments
Funding for the creation and maintenance of the colony described here was provided in part by Elanco Animal Health. Funding sources did not have any involvement in the study design, data analysis and interpretation, or writing and publication of the manuscript.
The authors have nothing to declare.
References
- 1.↑
Brown CA, Elliott J, Schmiedt CW, et al. Chronic kidney disease in aged cats: clinical features, morphology, and proposed pathogeneses. Vet Pathol. 2016;53:309–326. doi:10.1177/0300985815622975
- 2.↑
Boyd LM, Langston C, Thompson K, et al. Survival in cats with naturally occurring chronic kidney disease (2000–2002). J Vet Intern Med. 2008;22:1111–1117. doi:10.1111/j.1939-1676.2008.0163.x
- 3.↑
Adams LG, Polzin DJ, Osborne CA, et al. Influence of dietary protein/calorie intake on renal morphology and function in cats with 5/6 nephrectomy. Lab Invest. 1994;70:347–357.
- 4.
Buranakarl C, Mathur S, Brown SA. Effects of dietary sodium chloride intake on renal function and blood pressure in cats with normal and reduced renal function. Am J Vet Res. 2004;65:620–627. doi:10.2460/ajvr.2004.65.620
- 5.↑
Brown SA, Brown CA, Jacobs G, et al. Effects of the angiotensin converting enzyme inhibitor benazepril in cats with induced renal insufficiency. Am J Vet Res. 2001;62:375–383. doi:10.2460/ajvr.2001.62.375
- 6.↑
Brown SA, Brown CA. Single-nephron adaptations to partial renal ablation in cats. Am J Physiol. 1995;269:R1002–1008. doi:10.1152/ajpregu.1995.269.5.R1002
- 7.↑
Mathur S, Brown CA, Dietrich UM, et al. Evaluation of a technique of inducing hypertensive renal insufficiency in cats. Am J Vet Res. 2004;65:1006–1013. doi:10.2460/ajvr.2004.65.1006
- 8.↑
Mathur S, Syme H, Brown CA, et al. Effects of the calcium channel antagonist amlodipine in cats with surgically induced hypertensive renal insufficiency. Am J Vet Res. 2002;63:833–839. doi:10.2460/ajvr.2002.63.833
- 9.↑
Kren S, Hostetter TH. The course of the remnant kidney model in mice. Kidney Int. 1999;56:333–337. doi:10.1046/j.1523-1755.1999.00527.x
- 10.↑
Garrido P, Ribeiro S, Fernandes J, et al. Iron-hepcidin dysmetabolism, anemia and renal hypoxia, inflammation and fibrosis in the remnant kidney rat model. PLoS ONE. 2015;10:e0124048. doi:10.1371/journal.pone.0124048
- 11.
Nankivell BJ, Chen J, Boadle RA, et al. The role of tubular iron accumulation in the remnant kidney. J Am Soc Nephrol. 1994;4:1598–1607. doi:10.1681/ASN.V481598
- 12.↑
Ghanem S, Lesznyak T, Fazekas L, et al. Microrheology, microcirculation and structural compensatory mechanisms of a chronic kidney disease rat model. A preliminary study. Clin Hemorheol Microcirc. 2020;75:47–56. doi:10.3233/CH-190763
- 13.
Uchida L, Tanaka T, Saito H, et al. Effects of a prolyl hydroxylase inhibitor on kidney and cardiovascular complications in a rat model of chronic kidney disease. Am J Physiol Renal Physiol. 2020;318:F388–F401. doi:10.1152/ajprenal.00419.2019
- 14.↑
Manotham K, Tanaka T, Matsumoto M, et al. Evidence of tubular hypoxia in the early phase in the remnant kidney model. J Am Soc Nephrol. 2004;15:1277–1288. doi:10.1097/01.ASN.0000125614.35046.10
- 15.↑
Song YR, You SJ, Lee YM, et al. Activation of hypoxia-inducible factor attenuates renal injury in rat remnant kidney. Nephrol Dial Transplant. 2010;25:77–85. doi:10.1093/ndt/gfp454
- 16.↑
Imamura R, Isaka Y, Sandoval RM, et al. A nonerythropoietic derivative of erythropoietin inhibits tubulointerstitial fibrosis in remnant kidney. Clin Exp Nephrol. 2012;16:852–862. doi:10.1007/s10157-012-0647-x
- 17.↑
Kang JG, Yu MY, Lee H, et al. Blood pressure management and progression of chronic kidney disease in a canine remnant kidney model. Gen Physiol Biophys. 2018;37:243–252. doi:10.4149/gpb_2017055
- 18.↑
Fine A. Remnant kidney metabolism in the dog. J Am Soc Nephrol. 1991;2:70–76. doi:10.1681/ASN.V2170
- 19.↑
Finco DR. Association of systemic hypertension with renal injury in dogs with induced renal failure. J Vet Intern Med. 2004;18:289–294. doi:10.1111/j.1939-1676.2004.tb02547.x
- 20.↑
Lubanda JC, Chochola M, Mlcek M, et al. The effect of renal denervation in an experimental model of chronic renal insufficiency, The REmnant kidney Denervation In Pigs study (REDIP study). J Transl Med. 2017;15:215. doi:10.1186/s12967-017-1319-0
- 21.↑
Misra S, Gordon JD, Fu AA, et al. The porcine remnant kidney model of chronic renal insufficiency. J Surg Res. 2006;135:370–379. doi:10.1016/j.jss.2006.04.001
- 22.↑
Rempel LCT, Faustino VD, Foresto-Neto O, et al. Chronic exposure to hypoxia attenuates renal injury and innate immunity activation in the remnant kidney model. Am J Physiol Renal Physiol. 2019;317:F1285–F1292. doi:10.1152/ajprenal.00367.2018
- 23.↑
Schmiedt CW, Brainard BM, Hinson W, et al. Unilateral renal ischemia as a model of acute kidney injury and renal fibrosis in cats. Vet Pathol. 2016;53:87–101. doi:10.1177/0300985815600500
- 24.↑
Acierno MJ, Brown S, Coleman AE, et al. ACVIM consensus statement: guidelines for the identification, evaluation, and management of systemic hypertension in dogs and cats. J Vet Intern Med. 2018;32:1803–1822. doi:10.1111/jvim.15331
- 25.↑
Cowgill L. IRIS Staging of CKD (modified 2023): International Renal Interest Society. Modified in 2023. Accessed May 18, 2023. http://www.iris-kidney.com/pdf/2_IRIS_Staging_of_CKD_2023.pdf
- 26.↑
Lin J, Lin L, Chen S, et al. Serum fibroblast growth factor 23 (FGF-23): associations with hyperphosphatemia and clinical staging of feline chronic kidney disease. J Vet Diagn Invest. 2021;33:288–293. doi:10.1177/1040638720985563
- 27.↑
Kobayashi DL, Peterson ME, Graves TK, et al. Hypertension in cats with chronic renal failure or hyperthyroidism. J Vet Intern Med. 1990;4:58–62. doi:10.1111/j.1939-1676.1990.tb03104.x
- 28.
Hori Y, Heishima Y, Yamashita Y, et al. Relationship between indirect blood pressure and various stages of chronic kidney disease in cats. J Vet Med Sci. 2018;80:447–452. doi:10.1292/jvms.17-0620
- 29.↑
Syme HM, Barber PJ, Markwell PJ, et al. Prevalence of systolic hypertension in cats with chronic renal failure at initial evaluation. J Am Vet Med Assoc. 2002;220:1799–1804. doi:10.2460/javma.2002.220.1799
- 30.↑
Elliott J, Barber PJ. Feline chronic renal failure: clinical findings in 80 cases diagnosed between 1992 and 1995. J Small Anim Pract. 1998;39:78–85. doi:10.1111/j.1748-5827.1998.tb03598.x
- 31.↑
Williams TL, Archer J. Evaluation of urinary biomarkers for azotaemic chronic kidney disease in cats. J Small Anim Pract. 2016;57:122–129. doi:10.1111/jsap.12439
- 32.↑
Jepson ER, Brodbelt D, Vallance C, et al. Evaluation of predictors of the development of azotemia in cats. J Vet Intern Med. 2009; 23:806–813. doi:10.1111/j.1939-1676.2009.0339.x
- 33.↑
Brown SA. Renal pathophysiology: lessons learned from the canine remnant kidney model. J Vet Emerg Crit Care (San Antonio). 2013;23:115–121. doi:10.1111/vec.12030
- 34.↑
Adam RJ, Williams AC, Kriegel AJ. Comparison of the surgical resection and infarct 5/6 nephrectomy rat models of chronic kidney disease. Am J Physiol Renal Physiol. 2022;322:F639–F654. doi:10.1152/ajprenal.00398.2021
