Feline Renal Cortical Thickness–Aortic Diameter Ratio in Healthy Versus Diseased Kidneys: Comparative Ultrasonographic Evaluation

ABSTRACT

In this retrospective multicenter study, we aimed to establish the renal cortical thickness–aortic diameter (RCT:Ao) ratio as a diagnostic parameter for detecting feline acute kidney injury or chronic kidney disease (AKI or CKD). This study included bilateral kidneys of 152 normal, 171 CKD, 19 AKI, and 15 acute‐on‐chronic kidney disease (ACKD) cats. Ultrasonographic measurements were obtained in the midsagittal plane of the kidneys and aorta. Multiple linear regression analysis of RCT, body weight (BW), and body condition score (BCS) revealed a positive correlation of RCT with BW (p < 0.001), but not with BCS (p = 0.343). Multiple linear regression analysis of RCT:Ao ratio, BW, and BCS showed a poor model fit (F value: 0.119). There were significant intergroup differences among the normal, CKD, AKI, and ACKD sub‐cohorts (p < 0.001). Compared to normal cats, CKD and AKI cats each had lower and higher RCT:Ao ratio (both p < 0.001), respectively. The RCT:Ao ratio of the ACKD group significantly differed from that in normal and CKD groups (both p < 0.001), but not the AKI group (p = 0.159). Optimal RCT:Ao ratio cutoffs of 1.15 and 1.45 were used to distinguish between the normal and CKD groups (75% sensitivity, 80% specificity) and the normal and AKI groups (90% sensitivity, 89% specificity), respectively. The RCT:Ao ratio was unaffected by the BW and BCS and is a clinically useful diagnostic parameter for feline kidney disease.

Abbreviations

ACKD

acute‐on‐chronic kidney disease

AKI

acute kidney injury

ANOVA

analysis of variance

Ao

aorta

AUC

area under the curve

BCS

body condition score

BLKS

big kidney–little kidney syndrome

BSA

body surface area

BW

body weight

CI

confidence interval

CKD

chronic kidney disease

IRIS

International Renal Interest Society

M

mean

MRS

medullary rim sign

RCT

renal cortical thickness

ROC

receiver operating characteristic

sCREA

serum creatinine

SD

standard deviation

SDMA

serum dimethylarginine

1. Introduction

In dogs and cats, kidney disease is generally classified into acute kidney injury (AKI) and chronic kidney disease (CKD), both of which involve varying degrees of structural damage and functional impairment [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]. AKI refers to a sudden onset of renal parenchymal injury developing within approximately 48 h, whereas CKD is defined by persistent structural or functional abnormalities affecting one or both kidneys for more than 3 months [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]. Various influences, such as congenital factors, infections, inflammation, ischemia, and nephrotoxins, can cause kidney disease. Hypoxic injury is a major trigger of AKI and CKD, with ischemic damage to the renal tubules resulting in AKI [2, 7, 11, 12]. With sustained injury, the renal parenchyma will fibrose, facilitating the progression to CKD [13, 14]. During AKI, inflammation leads to edema and increased cortical thickness [4, 15]. The outer medulla is particularly affected by acute renal injury, and size changes secondary to cellular swelling in the outer medulla can be measured ultrasonographically in the cortex [4, 7, 14, 16]. Therefore, for diagnosing kidney disease, evaluation of the renal cortical thickness (RCT), rather than renal medullary thickness, may be more important.

The renal length increases above the normal range in AKI [2, 4, 17] and decreases below the normal range in CKD [18, 19, 20]. However, kidney length refers to the total size of the kidney, which includes both the cortex and medulla as well as the renal sinus fat, and may be an inaccurate parameter for assessing actual disease, as the affected intrarenal areas may vary depending on the disease stage [21, 22, 23, 24]. In clinical human studies among patients with CKD, the RCT strongly correlated with the estimated glomerular filtration rate, rather than with renal length [18, 19, 25, 26]. Specifically, in humans, cortical thinning preceded changes in kidney length [7].

In addition to the evaluation of renal size parameters such as renal length and RCT, ultrasonography is a useful and accurate diagnostic modality for assessing morphological changes, internal renal structure, and the surrounding perirenal tissues [7]. In feline CKD, renal volume typically decreases with irregular contours, and increased cortical echogenicity is a commonly reported ultrasonographic feature [7]. When echogenicity is elevated in both cortex and medulla, the corticomedullary junction becomes progressively indistinct [7]. In cases of AKI, structural abnormalities such as renomegaly and increased echogenicity of both the cortex and medulla are commonly observed [2]. Pylectasia may also be present, although it is a nonspecific finding and can occur in both AKI and CKD [2, 7]. Additionally, the presence of retroperitoneal fluid has been associated with oliguria and anuria, which are major clinical signs of AKI [2].

Recently, ultrasonographic RCT estimation has been used to help diagnose kidney disease in humans [18, 19, 26]. Particularly in early stage CKD, decreased RCT is significantly associated with reduced renal function and is used as a diagnostic tool for CKD [24, 28]. Similar to the renal length, the RCT tends to decrease in CKD [18, 19, 26]. In dogs, ultrasonographic RCT estimation has also been utilized as a diagnostic parameter for kidney disease [15]. RCT in dogs typically increases in AKI and decreases in CKD [15]. In cats, the RCT shows a strong positive correlation with the reciprocal of the serum creatinine (sCREA) concentration [29]. Thinning of the RCT is observed with higher CKD stage, indicating that with CKD progression, the RCT decreases and cortical echogenicity increases [7, 9, 17, 27, 29]. However, unlike humans and dogs, cats exhibit fat vacuole deposition in the proximal convoluted tubular epithelium, which contributes to cortical hyperechogenicity even in the normal kidney [2, 7, 27, 30]. Therefore, the echogenicity of the renal cortex is unsuitable for assessing and monitoring CKD; the RCT potentially constitutes a more useful parameter. Consequently, appropriate evaluation criteria are needed for determining feline RCT.

On the basis of several studies, reference ranges for normal sonographic RCT in cats have been proposed; however, a standardized range has not been established yet [17, 21, 22, 23, 29, 31, 32]. Although cats have less body‐type diversity than dogs [27], the effects of body conditions, such as body weight (BW) and the body condition score (BCS), on RCTs should be considered. To minimize the effect of BW and BCS on the RCT in dogs, the ratio of RCT to the aorta (RCT:Ao ratio) has been studied as a parameter that is unaffected by the body conformation [33]. In normal dogs, the RCT positively correlated with BW, whereas the RCT:Ao ratio was unaffected by either BW or BCS [15, 33]. The RCT:Ao ratio in dogs increased with AKI and decreased with CKD, similar to the changes in renal length and RCT, which makes it a clinically useful marker [15]. Several studies have suggested a normal reference range for the RCT as a cutoff for diagnosing CKD in cats; however, the RCT:Ao ratio has not yet been studied [17, 21, 22, 29].

On the basis of these facts, we formulated a hypothesis and secondary hypotheses as follows. First, there will be a correlation between the RCT, BW, and BCS in cats, as in dogs. Second, the RCT:Ao ratio will be unaffected by BW and BCS in cats. Third, the RCT in cats with CKD and AKI, considering the BCS and BW, will be outside the normal range. Fourth, in feline CKD and AKI, the RCT:Ao ratio will be less than and greater than the normal range, respectively. This study aimed to establish a normal reference range for the RCT:Ao ratio, which may be unaffected by the body conformation in clinically normal cats, and to establish a diagnostic cutoff for feline CKD or AKI.

2. Materials and Methods

2.1. Data Collection and Extraction

Information from January 2021 to March 2024 on feline patients was collected retrospectively from five veterinary clinics (Bundang Leaders Animal Medical Center, Jamsil ON Animal Medical Center, Jeonbuk National University Animal Medical Center, The Care Animal Medical Center, and VIP Animal Medical Center). Medical records, including physical examinations, laboratory tests, and ultrasound examinations, were reviewed. Only cats with BCS assessed using a 9‐point scale were included. Blood tests and imaging examinations were performed within a 1‐week interval. Cats with a history of nephrectomy or unilateral kidney disease were excluded from the study if only one kidney was identifiable. On the basis of the International Renal Interest Society (IRIS) classification system [34], the cats were subdivided into the clinically normal (no evidence of renal disease), AKI, and CKD groups. Normal cats had no urinary system‐related clinical signs (e.g., polyuria, pollakiuria, dysuria, incontinence), no azotemia, and no increased sCREA or serum dimethylarginine (SDMA) on blood tests. The presence of any renal structural changes, such as changes in echogenicity and size, identification of internal calcifications or cysts, or pelvis dilation, was considered pathologic. Moreover, the medullary rim sign (MRS), defined as a thin hyperechoic line at the outer medulla that is parallel to the corticomedullary junction, was considered normal in the absence of other related abnormalities [35, 36]. Cats with an acute urinary clinical presentation (e.g., anorexia, vomiting, polyuria/polydipsia, oliguria/anuria), onset of azotemia within 3 days, no history of a diagnosis of CKD, and proteinuria if urinalysis were available were included in the AKI group. Cats with CKD were categorized according to the IRIS guidelines and further subdivided into two stages: Stages 1 and 2 were categorized as early stage CKD, and Stages 3 and 4 as late‐stage CKD. The urinary protein–creatinine ratio (UPCR), urine dipstick test, and urinary specific gravity were utilized, if available; however, blood pressure measurements were not used. As in previous studies [29, 37], Stage 1 disease was categorized on the basis of sCREA <1.6 mg/dL, SDMA <18 $\mu$ g/dL, urinary specific gravity <1.035, and proteinuria. In the CKD group, if the length difference between the kidneys of a cat was at least 7 mm, the cat was included under the “big kidney–little kidney syndrome (BLKS)” [38]. Cats with the previous serological or ultrasound evidence of CKD (e.g., decreased renal length, irregular contour, indistinct corticomedullary junction, increased echogenicity of the cortex and/or medulla, renal cysts), a ≥20% increase in the sCREA and clinical signs (e.g., lethargy, depression, vomiting, anorexia), azotemia, or urinalysis findings indicative of AKI were diagnosed with acute‐on‐chronic kidney disease (ACKD) using previously reported criteria [39]. Cats with renal diseases that can cause renal morphological deformities (e.g., polycystic kidney disease (PKD), feline infectious peritonitis (FIP), and renal neoplasia) were excluded because of the difficulty in obtaining consistent RCT measurements. This study was approved (approval no. JBNU 2023‐032) by the Institutional Animal Care and Use Committee of the Jeonbuk National University, Iksan‐si, Jeollabuk‐do, Republic of Korea.

2.2. Measurements and Analysis of Ultrasound Images

Ultrasound images were scanned using an Aplio 300 (Canon Medical Systems, Europe B.V., Zoetermeer, the Netherlands) with a 12‐MHz linear array 18L7 transducer, Aplio a (Canon Medical Systems, Tochigi, Japan) with a 12‐MHz linear array 18L7 transducer, Aplio i800 (Canon Medical Systems, Tokyo, Japan) with a 12‐MHz linear array i18LX5 transducer, or Aplio i700 (Canon Medical Systems, Tustin, CA) with a 12‐MHz linear i18LX5 probe. All ultrasound scans were performed with the cats in a dorsal recumbent state. Scanned sagittal plane images of the kidney and aorta were used for analysis.

All radiographic and sonographic images were obtained by veterinary radiologists with at least 5 years of clinical experience, and all measurements and analyses were performed by a single veterinary radiology resident (S.H.J.). The RadiAnt DICOM Viewer (version 2023.1, 64‐bit, Poznan, Poland) was used for image measurements. The methods for measuring the RCT and aortic diameter were similar to those used in previous studies on dogs [15, 33] and cats [17, 22]. The RCT of each kidney was measured in the midsagittal plane, where two parallel hyperechoic cross‐sectional pelvic diverticular lines (Figure 1A) were observed [17, 22]. The RCT was measured as the shortest perpendicular distance from the trailing edge of the hyperechoic renal capsule to the leading edge of the base of the renal pyramid (Figure 1A,B) [15, 17, 22, 33]. As in dogs, the presence of hyperechoic outer medulla in cats without evidence of renal dysfunction or structural changes was considered an incidental finding [40, 41]. Consequently, these areas were included in the cortical measurements. The three most clearly visible renal cortices in the ventral part of the kidney were measured, and the average values were used for statistical analysis. The RCT measurement sites were not necessarily limited to the cranial and caudal poles. The renal length in the BLKS group was measured as the straight‐line distance between the cranial and caudal poles of the kidney at the same location as the cross‐sectional RCT measurement. The abdominal aortic luminal diameter was measured from the trailing to the leading edges perpendicular to the vessel wall just posterior to the bifurcation of the renal artery during the systolic phase (when the abdominal aorta is maximally dilated; Figure 1C).

FIGURE 1.

(A) Measurements of the RCT in clinically normal kidneys without MRS and (B) with MRS; (C) the abdominal Ao diameter. Midsagittal plane (A) of the kidney showing two parallel hyperechoic cross‐sectional pelvic diverticular lines (asterisks) was used. The three different shortest perpendicular distances (bidirectional arrows) from the trailing edge of hyperechoic renal capsule to the leading edge of the base of the renal pyramid were measured. (B) Thin hyperechoic lines (arrow) at the medulla were defined as MRS. The same measurement sites and methods were used for the kidneys without MRS. (C) Sagittal plane of the abdominal Ao. The aortic luminal diameter was measured from the trailing edge to the leading edge of the distance perpendicular to the vessel wall (bidirectional arrows), just posterior to the renal artery bifurcation (arrowhead), in the systolic phase. Ao, aorta; MRS, medullary rim sign; RCT, renal cortical thickness.

2.3. Statistical Analysis

The statistical techniques were partially adapted from those used in previous studies in dogs [15, 33] and were performed by a single veterinary radiology resident (S.H.J.) with 3 years of veterinary imaging experience and verified by two veterinary imaging specialists (H.Y.Y. and K.C.L.). All statistical analyses were performed using IBM SPSS Statistics version 20.0 (Armonk, New York, United States). For all statistical procedures, p < 0.05 was considered significant.

2.3.1. Normal Kidney in Cats

The values of the RCT, aorta, and RCT:Ao ratio were tested for normality using the Shapiro–Wilk test. All the mean values were expressed as the mean ± standard deviation (M ± SD). Independent‐sample t‐tests were used to assess whether the mean values of the RCTs in the left and right kidneys showed significant differences, and the mean value of the RCTs of the left and right kidneys was used in the study. Independent‐sample t‐tests were used to determine whether there was a significant difference in mean RCT values between cats with and without MRS.

Pearson's correlation analysis was performed to evaluate the association between the RCT and BW, and linear regression equations were obtained through simple linear regression analysis to determine the degree of correlation between BW (independent variable) and RCT (dependent variable). The same process was followed for the RCT (dependent variable) and BCS (independent variable). To assess whether BCS affected the correlation between BW and RCT, healthy cats were categorized into two groups. Group 1 included all cats with normal kidneys, and Group 2 included cats with normal kidneys with a BCS of 4 or 5. For each group, linear regression analysis was performed between the RCT and BW, and between the RCT and body surface area (BSA; ${\mathrm{m}}^2$). The coefficient of determination (R‐squared) for each relationship was compared. Multiple regression analysis was performed to determine the relationships among the RCT (dependent variable), BW (independent variable), and BCS (independent variable). Independent‐sample t‐tests were used to determine whether there was a significant difference in the mean RCT:Ao ratio between cats with and without MRS. Multiple regression analysis of the RCT:Ao ratio (dependent variable), BW (independent variable), and BCS (independent variable) was performed.

2.3.2. Cats With Pathologic Kidneys

In the kidney disease groups, except for the BLKS group, the average of the RCTs in the left and right kidneys was used. On the basis of the linear regression among the BW, BCS, and RCT in the normal population, the predicted RCT of the CKD and AKI populations was calculated and compared with the actual RCT measured in pathologic cats. Moderated regression analysis with the BW, RCT, and the presence of kidney disease as independent, dependent, and moderator variables, respectively, was performed to evaluate whether the presence of CKD or AKI affected the RCT values while considering BW. One‐way analysis of variance (ANOVA) was used to compare the means of the RCT values in normal cats and in CKD, AKI, and ACKD cats and to assess whether there was a difference in the RCT:Ao ratio between normal cats and cats with CKD, AKI, and ACKD. More specifically, one‐way ANOVA was used to determine the difference in the RCT:Ao ratio between each of the four stages of CKD in cats, to compare the RCT:Ao ratios among the normal, four CKD stages, and AKI groups, and among the normal, early stage CKD, late‐stage CKD, and AKI groups. The receiver operating characteristic (ROC) curve and Youden's Index were evaluated to set the RCT:Ao ratio cutoff criteria between the normal and CKD groups, and between the normal and AKI groups, respectively, and to calculate the sensitivity and specificity of the cutoff criteria. In the BLKS group, the mean differences in renal lengths, RCTs, and RCT:Ao ratios were compared using independent t‐tests between the larger and smaller bilateral kidneys. All post hoc one‐way ANOVAs were performed using Scheffé’s tests.

3. Results

3.1. Subject Characteristics

A total of 620 feline records were screened, and 263 were excluded for the following reasons: unusable ultrasound images of the kidneys or aorta, no blood test data within 1 week of the imaging study, no physical examination records, such as BW and BCS, presence of the aforementioned urinary system‐related clinical signs, or discrepancies between blood test results and imaging studies. Therefore, 152 healthy cats (38 with MRS and 114 without MRS), 171 with CKD, 19 with AKI, and 15 with ACKD were included in the study. Cats with CKD underwent CKD staging as follows: Stage 1 (N = 36, 21.1%), Stage 2 (N = 110, 64.3%), Stage 3 (N = 18, 10.5%), and Stage 4 (N = 7, 4.1%). A total of 18 cats were classified as having BLKS: 1 with CKD Stage 1, 11 with CKD Stage 2, 4 with CKD Stage 3, and 2 with CKD Stage 4.

Detailed information on normal cats and each kidney disease, and of normal cats with and without MRS, is described in Tables 1 and 2, respectively. Detailed information on each CKD stage and the BLKS group is presented in Tables 3 and 4, respectively.

TABLE 1.

Detailed information of normal cats and cats with CKD, AKI, and ACKD.

	Normal (N = 152)	CKD (N = 171)	AKI (N = 19)	ACKD (N = 15)
Age (years)	5.65 ± 3.83 (0.25–15.83)	8.65 ± 4.69 (0.25–20.50)	6.89 ± 3.60 (1.83–14.83)	11.19 ± 3.72 (5.50–17.50)
BW (kg)	4.78 ± 1.42 (1.51–9.00)	4.90 ± 1.66 (0.58–9.50)	5.03 ± 1.84 (2.87–11.70)	4.43 ± 1.41 (2.18–6.32)
BCS (/9)	1 (N = 1, 0.7%), 2 (N = 7, 4.6%), 3 (N = 7, 4.6%), 4 (N = 29, 19.1%), 5 (N = 41, 27.0%), 6 (N = 32, 21.1%), 7 (N = 24, 15.8%), 8 (N = 7, 4.6%), 9 (N = 4, 2.6%)	1 (N = 2, 1.2%), 2 (N = 10, 5.8%), 3 (N = 22, 12.9%), 4 (N = 29, 17.0%), 5 (N = 46, 26.9%), 6 (N = 34, 19.9%), 7 (N = 18, 10.5%), 8 (N = 9, 5.3%), 9 (N = 1, 0.6%)	3 (N = 2, 10.5%), 4 (N = 4, 21.1%), 5 (N = 8, 42.1%), 6 (N = 3, 15.8%), 7 (N = 1, 5.3%), 9 (N = 1, 5.3%)	1 (N = 4, 22.2%), 5 (N = 7, 38.9%), 6 (N = 3, 16.7%), 9 (N = 1, 5.6%)
Sex	Intact female (N = 10, 6.6%), spayed female (N = 64, 42.1%), castrated male (N = 71, 46.7%), intact male (N = 7, 4.6%)	Intact female (N = 5, 2.9%), spayed female (N = 74, 43.3%), castrated male (N = 86, 50.3%), intact male (N = 6, 3.5%)	Spayed female (N = 2, 10.5%), castrated male (N = 16, 84.2%), intact male (N = 1, 5.3%)	Intact female (N = 2, 13.3%), spayed female (N = 4, 26.7%), castrated male (N = 9, 60.0%)
Breed	Korean Shorthair (N = 84, 55.3%), Russian Blue (N = 12, 7.9%), Persian (N = 10, 6.6%), Mixed (N = 7, 4.6%), American Shorthair (N = 6, 3.9%), Turkish Angora (N = 5, 3.3%), British Shorthair (N = 5, 3.3%), Scottish Fold (N = 4, 2.6%), Munchkin (N = 4, 2.6%), Bengal (N = 4, 2.6%), Ragdoll (N = 3, 2.0%), Maine Coon (N = 2, 1.13%), Sphynx (N = 1, 0.7%), Norwegian Forest (N = 1, 0.7%), European Burmese (N = 1, 0.7%), Devon Rex (N = 1, 0.7%), American Curl (N = 1, 0.7%), Abyssinian (N = 1, 0.7%)	Korean Shorthair (N = 96, 56.1%), Russian Blue (N = 14, 8.2%), Persian (N = 10, 5.8%), Siamese (N = 9, 5.3%), Scottish Fold (N = 8, 4.7%), Abyssinian (N = 5, 2.9%), Turkish Angora (N = 5, 2.9%), Bengal (N = 4, 2.3%), Mixed (N = 3, 1.8%), Norwegian Forest (N = 3, 1.8%), Ragdoll (N = 3, 1.8%), American Shorthair (N = 2, 1.2%), British Longhair (N = 2, 1.2%), Maine Coon (N = 2, 1.2%), British Shorthair (N = 1, 0.6%), Devon Rex (N = 1, 0.6%), Domestic Shorthair (N = 1, 0.6%), Exotic Shorthair (N = 1, 0.6%), Munchkin (N = 1, 0.6%)	Korean Shorthair (N = 11, 57.9%), American Shorthair (N = 2, 10.5%), Scottish Fold (N = 2, 10.5%), British Shorthair (N = 1, 5.3%), Mixed (N = 1, 5.3%), Munchkin (N = 1, 5.3%), Siamese (N = 1, 5.3%)	Korean Shorthair (N = 4, 26.7%), Mixed (N = 3, 20.0%), Scottish Fold (N = 2, 13.3%), Abyssinian (N = 1, 6.7%), American Shorthair (N = 1, 6.7%), British Shorthair (N = 1, 6.7%), Norwegian Forest (N = 1, 6.7%), Russian Blue (N = 1, 6.7%), Turkish Angora (N = 1, 6.7%)

Note: All continuous variables are expressed as M ± SD (range).

Abbreviations: ACKD, acute‐on‐chronic kidney disease; AKI, acute kidney injury; BCS, body condition score; BW, body weight; CKD, chronic kidney disease; M, mean; SD, standard deviation.

TABLE 2.

Detailed patient information of clinically normal cats based on the presence or absence of MRS.

Normal (N = 152)	Kidneys with MRS (N = 38)	Kidneys with no MRS (N = 114)
Age (years)	3.32 ± 2.81 (0.25–12.00)	6.42 ± 3.82 (0.30–15.83)
BW (kg)	4.39 ± 1.42 (1.51–7.20)	4.91 ± 1.40 (1.80–9.00)
BCS (/9)	2 (N = 3, 7.9%), 3 (N = 1, 2.6%), 4 (N = 11, 28.9%), 5 (N = 10, 26.3%), 6 (N = 6, 15.8%), 7 (N = 7, 18.4%)	1 (N = 1, 0.9%), 2 (N = 4, 3.5%), 3 (N = 6, 5.3%), 4 (N = 18, 15.8%), 5 (N = 31, 27.2%), 6 (N = 26, 22.8%), 7 (N = 17, 14.9%), 8 (N = 7, 6.1%), 9 (N = 4, 3.5%)
Sex	Intact female (N = 4, 10.5%), spayed female (N = 11, 28.9%), castrated male (N = 20, 52.6%), intact male (N = 3, 7.9%)	Intact female (N = 6, 5.3%), spayed female (N = 53, 46.5%), castrated male (N = 51, 44.7%), intact male (N = 4, 3.5%)
Breed	Korean Shorthair (N = 27, 71.1%), American Shorthair (N = 3, 7.9%), Munchkin (N = 2, 5.3%), British Shorthair (N = 1, 2.6%), European Burmese (N = 1, 2.6%), Maine Coon (N = 1, 2.6%), Persian (N = 1, 2.6%), Russian Blue (N = 1, 2.6%), Sphynx (N = 1, 2.6%)	Korean Shorthair (N = 57, 50.0%), Russian Blue (N = 11, 9.6%), Persian (N = 9, 7.9%), Mixed (N = 7, 6.1%), Turkish Angora (N = 5, 4.4%), Bengal (N = 4, 3.5%), British Shorthair (N = 4, 3.5%), Scottish Fold (N = 4, 3.5%), American Shorthair (N = 3, 2.6%), Ragdoll (N = 3, 2.6%), Munchkin (N = 2, 1.8%), American Curl (N = 1, 0.9%), Devon Rex (N = 1, 0.9%), Maine Coon (N = 1, 0.9%), Norwegian Forest (N = 1, 0.9%)

Note: All continuous variables are expressed as M ± SD (range).

Abbreviations: BCS, body condition score; BW, body weight; M, mean; MRS, medullary rim sign; SD, standard deviation.

TABLE 3.

Detailed information about cats stratified by the CKD stage.

CKD stage (N = 171)	1 (N = 36)	2 (N = 110)	3 (N = 18)	4 (N = 7)
Age (years)	7.01 ± 4.11 (0.25–16.00)	8.46 ± 4.59 (0.58–18.00)	12.70 ± 4.53 (3.91–20.75)	9.61 ± 3.88 (4.00–16.67)
BW (kg)	5.32 ± 1.69 (0.58–9.00)	5.02 ± 1.62 (2.10–9.50)	3.86 ± 1.45 (1.80–8.0)	3.49 ± 0.87 (2.40–5.10)
BCS (/9)	2 (N = 1, 2.8%), 3 (N = 2, 5.6%), 4 (N = 5, 13.9%), 5 (N = 14, 38.9%), 6 (N = 6, 16.7%), 7 (N = 5, 13.9%), 8 (N = 2, 5.6%), 9 (N = 1, 2.8%)	1 (N = 1, 0.9%), 2 (N = 2, 1.8%), 3 (N = 14, 12.7%), 4 (N = 17, 15.5%), 5 (N = 30, 27.3%), 6 (N = 27, 24.5%), 7 (N = 13, 11.8%), 8 (N = 6, 5.5%)	1 (N = 1, 5.6%), 2 (N = 4, 22.2%), 3 (N = 4, 22.2%), 4 (N = 6, 33.3%), 5 (N = 2, 11.1%), 8 (N = 1, 5.6%)	2 (N = 3, 42.9%), 3 (N = 2, 28.6%), 4 (N = 1, 14.3%), 6 (N = 1, 14.3%)
Sex	Intact female (N = 2, 5.6%), spayed female (N = 15, 41.7%), castrated male (N = 16, 44.4%), intact male (N = 3, 8.3%)	Intact female (N = 1, 0.9%), spayed female (N = 48, 43.6%), castrated male (N = 58, 52.7%), intact male (N = 3, 2.7%)	Intact female (N = 2, 11.1%), spayed female (N = 7, 38.9%), castrated male (N = 9, 50.0%)	Spayed female (N = 4, 57.1%), castrated male (N = 3, 42.9%)
Breed	Korean Shorthair (N = 20, 55.6%), Bengal (N = 3, 8.3%), Scottish Fold (N = 3, 8.3%), British Longhair (N = 1, 2.8%), Devon Rex (N = 1, 2.8%), Maine Coon (N = 1, 2.8%), Mixed (N = 1, 2.8%), Munchkin (N = 1, 2.8%), Norwegian (N = 1, 2.8%), Persian (N = 1, 2.8%), Ragdoll (N = 1, 2.8%), Siamese (N = 1, 2.8%), Turkish Angora (N = 1, 2.8%)	Korean Shorthair (N = 65, 59.1%), Russian Blue (N = 11, 10.0%), Siamese (N = 7, 6.4%), Persian (N = 5, 4.5%), Scottish Fold (N = 5, 4.5%), Turkish Angora (N = 4, 3.6%), Abyssinian (N = 3, 2.7%), Mixed (N = 2, 1.8%), Norwegian Forest (N = 2, 1.8%), Ragdoll (N = 2, 1.8%), American Shorthair (N = 1, 0.9%), Bengal (N = 1, 0.9%), British Longhair (N = 1, 0.9%), British Shorthair (N = 1, 0.9%)	Korean Shorthair (N = 6, 33.3%), Persian (N = 3, 16.7%), Russian Blue (N = 3, 16.7%), Abyssinian (N = 1, 5.6%), American Shorthair (N = 1, 5.6%), Domestic Shorthair (N = 1, 5.6%), Exotic Shorthair (N = 1, 5.6%), Maine Coon (N = 1, 5.6%), Siamese (N = 1, 5.6%)	Korean Shorthair (N = 5, 71.4%), Abyssinian (N = 1, 14.3%), Persian (N = 1, 14.3%)

Note: All continuous variables are expressed as M ± SD (range).

Abbreviations: BCS, body condition score; BW, body weight; CKD, chronic kidney disease; M, mean; SD, standard deviation.

TABLE 4.

Detailed information on BLKS cats.

	BLKS (N = 18)
Age (years)	12.31 ± 4.11 (2.00–17.58)
BW (kg)	3.89 ± 1.32 (2.10–6.50)
BCS (/9)	2 (N = 2, 11.1%), 3 (N = 7, 38.9%), 4 (N = 2, 11.1%), 5 (N = 3, 16.7%), 6 (N = 2, 11.1%), 7 (N = 2, 11.1%)
Sex	Intact female (N = 2, 11.1%), spayed female (N = 11, 61.1%), castrated male (N = 4, 22.2%), intact male (N = 1, 5.6%)
Breed	Korean Shorthair (N = 6, 33.3%), Siamese (N = 3, 16.7%), Abyssinian (N = 2, 11.1%), Scottish Fold (N = 2, 11.1%), American Shorthair (N = 1, 5.6%), Exotic Shorthair (N = 1, 5.6%), Norwegian Forest (N = 1, 5.6%), Persian (N = 1, 5.6%), Russian Blue (N = 1, 5.6%)

Note: All continuous variables are expressed as M ± SD (range).

Abbreviations: BCS, body condition score; BLKS, big kidney–little kidney syndrome; BW, body weight; M, mean; SD, standard deviation.

3.2. Normal Kidney in Cats

The Shapiro–Wilk test confirmed that the RCT, aorta diameter, and RCT:Ao ratio satisfied normality (p > 0.05). The mean values of the RCT (mm) in the left and right kidneys were 3.71 ± 0.57 (2.13–5.52) and 3.71 ± 0.56 (1.98–5.17), respectively, and no significant difference was found (p = 0.631; Table 5). No significant differences were found in the RCT values in the groups with and without MRS (p = 0.677; Table 6). Thus, the RCT values (mm) of normal cats were 3.71 ± 0.53 (2.23–5.24; Table 5). The Pearson correlation coefficient (r) between the RCT and BW was 0.589, which confirmed a moderately positive correlation (p < 0.001). The Pearson correlation coefficient (r) between the RCT and BCS was 0.439 and confirmed a moderate positive correlation (p < 0.001). In Group 1, the RCT and BW showed a weak positive correlation ($R^2$= 0.346, p < 0.001; Figure 2A). The RCT and BSA showed a weak positive correlation ($R^2$= 0.365, p < 0.001; Figure 2B). In Group 2, the RCT and BW showed a weak positive correlation ($R^2$= 0.377, p < 0.001; Figure 2C). The RCT and BSA showed a weak positive correlation ($R^2$ = 0.390, p < 0.001; Figure 2D). The coefficients of determination in all correlations showed slightly higher values in Group 2, but with very small numerical differences. Multiple regression analysis of RCT, BW, and BCS showed a significant relationship between the RCT and BW (p < 0.001) but not between RCT and BCS (p = 0.343). Consequently, linear regression analysis of the RCT and BW in healthy cats resulted in the following equation:

TABLE 5.

The RCT and RCT:Ao ratio of the left and right kidneys in clinically normal cats.

	Left kidney (N = 152)	Right kidney (N = 152)	Both kidneys (N = 152)
RCT (mm)	3.71 ± 0.57 (2.13–5.52)	3.71 ± 0.56 (1.98–5.17)	3.71 ± 0.53 (2.23–5.24)
RCT:Ao ratio	1.27 ± 0.16 (0.74–1.63)	1.27 ± 0.16 (0.84–1.74)	1.27 ± 0.15 (0.83–1.62)

Note: No significant difference was found in the measurements of the RCT (p = 0.631) and RCT:Ao ratio (p = 0.325) between bilateral kidneys. All continuous variables are expressed as M ± SD (range).

Abbreviations: Ao, aorta; M, mean; RCT, renal cortical thickness; SD, standard deviation.

TABLE 6.

The RCT, Ao, and RCT:Ao ratio in clinically normal cats with and without MRS.

Normal cats (N = 152)	Kidneys with MRS (N = 38)	Kidneys without MRS (N = 114)
RCT (mm)	3.72 ± 0.51 (2.23–5.24)	3.68 ± 0.61 (2.36–5.08)
Ao (mm)	2.92 ± 0.36 (1.90–4.00)	2.97 ± 0.35 (1.90–3.80)
RCT:Ao ratio	1.28 ± 0.14 (0.83–1.62)	1.24 ± 0.17 (0.89–1.57)

Note: No significant difference, with and without MRS, was detected in the RCT (p = 0.677) and the RCT:Ao ratio (p = 0.185). All continuous variables are expressed as M ± SD (range).

Abbreviations: Ao, aorta; M, mean; MRS, medullary rim sign; RCT, renal cortical thickness; SD, standard deviation.

FIGURE 2.

(A and C) Linear regression graph and equations of the relationship between the RCT and BW and (B and D) RCT and BSA in Group 1 (blue dots) and Group 2 (red dots), respectively. Coefficient of determination (R‐squared) in each relation is shown in the corresponding graph. Slightly higher but similar coefficients of determination for the relationships were found in Group 2. BCS, body condition score; BSA, body surface area; BW, body weight; RCT, renal cortical thickness.

RCT (mm) = 0.220 × BW (kg) + 2.657 ($R^2$ = 0.346, p < 0.001).

No significant difference was found in the RCT:Ao ratios with and without MRS (p = 0.185; Table 6), and the mean RCT:Ao ratio was 1.27 ± 0.15 (0.83–1.62) in normal cats. The F value of the regression model for the multiple regression analysis between the RCT:Ao ratio, BW, and BCS was 0.119, indicating a poor model fit without any significant correlation between the variables.

3.3. Pathologic Kidneys in Cats

To understand the RCT tendencies by kidney diseases, the predicted RCT values calculated by RCT (mm) = 0.220 × BW (kg) + 2.657 ($R^2$ = 0.346, p < 0.001) were compared with the actual RCT values for pathological cats. Cats with CKD generally had lower values than the predicted values, whereas cats with AKI had higher values. Specifically, in the moderated regression analysis, regression models 2 and 3 showed significant changes in F (p < 0.001 and p < 0.05, respectively). Consequently, kidney disease significantly affected RCT with BW consideration. One‐way ANOVA comparing RCT showed significant differences among normal, CKD, AKI, and ACKD cats (p < 0.001; Table 7). Post hoc analysis revealed significant differences between all groups (p < 0.001), but not between AKI and ACKD cats (p = 0.514; Table 7). One‐way ANOVA of RCT:Ao ratio between normal, CKD, AKI, and ACKD cats revealed a significant difference (p < 0.001; Table 7, Figure 3A). However, post hoc tests revealed significant differences between all groups (p < 0.001), except for cats with AKI and ACKD (p = 0.248; Table 7, Figure 3A). The mean RCT:Ao ratio tended to be lower than normal in the CKD population and higher in the AKI population (Table 7, Figure 3A). The mean RCT:Ao ratio of cats with ACKD tended to be higher than that of normal and CKD cats, but lower than that of cats with AKI (Table 7, Figure 3A). One‐way ANOVA of RCT:Ao ratio between the four CKD stages revealed a significant difference (p < 0.001), and as the stage increased, the RCT:Ao ratio tended to decrease (Table 8, Figure 3B). However, Scheffé’s post hoc analysis showed no significant intergroup differences between CKD Stages 3 and 4 (p = 0.478; Table 8, Figure 3B). One‐way ANOVA of the RCT:Ao ratio between the normal, four CKD stages, and AKI groups showed significant differences (p < 0.001; Figure 3C). Furthermore, Scheffé’s post hoc analysis showed no significant intergroup differences between normal and Stage 1 CKD cats (p = 0.119), between Stages 2 and 3 CKD cats (p = 0.108), or between Stages 3 and 4 CKD cats (p = 0.752) (Figure 3C). One‐way ANOVA of the RCT:Ao ratio between normal, early stage CKD, late‐stage CKD, and AKI cats showed significant differences (p < 0.001; Figure 3D). All intergroups revealed significant differences in Scheffé’s post hoc analysis (p < 0.001; Figure 3D).

TABLE 7.

One‐way ANOVA‐based intergroup comparison of the RCT and RCT:Ao ratio among clinically normal cats and cats with CKD, AKI, or ACKD.

	Normal (N = 152)	CKD (N = 171)	AKI (N = 19)	ACKD (N = 15)
RCT (mm)	$3.71\pm 0.53^{\mathrm{a}}$ (2.23–5.24)	$3.24\pm 0.68^{\mathrm{b}}$ (1.59–6.26)	$4.97\pm 0.79^{\mathrm{c}}$ (3.65–6.20)	$3.97\pm 0.86^{\mathrm{c}}$ (2.84–5.93)
RCT:Ao ratio	$1.27\pm 0.15^{\mathrm{a}}$ (0.83–1.62)	$1.05\pm 0.19^{\mathrm{b}}$ (0.61–1.96)	$1.70\pm 0.26^{\mathrm{c}}$ (1.30–2.09)	$1.56\pm 0.38^{\mathrm{c}}$ (0.91–2.37)

Note: The average of the left and right kidney RCTs was used across all groups. Significant intergroup differences in the RCT and RCT:Ao ratio (both p < 0.001) were detected, except in AKI and ACKD. Different lowercase letters in the same row indicate significant differences. All continuous variables are expressed as M ± SD (range).

Abbreviations: ACKD, acute‐on‐chronic kidney disease; AKI, acute kidney injury; ANOVA, analysis of variance; Ao, aorta; CKD, chronic kidney disease; M, mean; RCT, renal cortical thickness; SD, standard deviation.

FIGURE 3.

(A) Comparison of the RCT:Ao ratio in normal, CKD, AKI, and ACKD patients revealed significant intergroup differences (**p < 0.001), except between the AKI and ACKD cats (p = 0.248). (B) Comparison of the RCT:Ao ratio in every CKD stage revealed significant intergroup differences (*p < 0.05, **p < 0.001), except between CKD Stages 3 and 4 (p = 0.478). However, there was tendency for a decreasing RCT:Ao ratio with an increasing CKD stage. (C) Comparison of differences in the RCT:Ao ratio among the normal, four CKD stages, and AKI groups revealed significant intergroup differences (*p < 0.05, **p < 0.001), except between the normal and CKD Stage 1 (p = 0.119), between CKD Stages 2 and 3 (p = 0.108), and between CKD Stages 3 and 4 (p = 0.752). (D) Comparison of the differences in the RCT:Ao ratio between the normal, early stage CKD, late‐stage CKD, and AKI groups revealed significant intergroup differences in all comparisons (**p < 0.001). ACKD, acute‐on‐chronic kidney disease; AKI, acute kidney injury; Ao, aorta; CKD, chronic kidney disease; RCT, renal cortical thickness.

TABLE 8.

CKD stage‐stratified comparison of the RCT:Ao ratio in cats using one‐way ANOVA (p < 0.001).

CKD stage (N = 171)	1 (N = 36)	2 (N = 110)	3 (N = 18)	4 (N = 7)
RCT:Ao ratio	$1.18\pm 0.16^{\mathrm{a}}$ (0.96–1.67)	$1.05\pm 0.18^{\mathrm{b}}$ (0.61–1.96)	$0.92\pm 0.18^{\mathrm{c}}$ (0.67–1.27)	$0.80\pm 0.16^{\mathrm{c}}$ (0.61–1.02)

Note: All intergroup comparisons, except of CKD Stages 3 and 4 (p = 0.478), showed significant differences in the RCT:Ao ratio (p < 0.05). Different lowercase letters indicate significant differences. All continuous variables are expressed as M ± SD (range).

Abbreviations: ANOVA, analysis of variance; Ao, aorta; CKD, chronic kidney disease; M, mean; RCT, renal cortical thickness; SD, standard deviation.

The ROC curve of the RCT:Ao ratio in the normal and CKD groups showed an area under the curve (AUC) of 0.84 (95% CI, 0.792–0.881), indicating a good diagnostic tool (Figure 4A). In particular, the RCT:Ao ratio of 1.15 was considered to be the optimal cutoff value for CKD cats with a sensitivity of 75%, specificity of 80%, and Youden's Index of 0.55. The ROC curve of the RCT:Ao ratio in the normal and AKI populations showed an AUC of 0.94 (95% CI, 0.888–0.994), suggesting an excellent diagnostic tool (Figure 4B). An RCT:Ao ratio of 1.45 was considered the optimal cutoff value for AKI cats with a sensitivity of 90%, specificity of 89%, and Youden's Index of 0.78.

FIGURE 4.

The ROC curve and the optimal cutoff value between (A) normal and CKD cats and (B) normal and AKI cats. In CKD cats, the AUC was 0.84 (95% CI, 0.792–0.881) the RCT:Ao ratio of 1.15 was considered a diagnostic characteristic, with a sensitivity of 75% and specificity of 80%. In AKI cats, the AUC was 0.94 (95% CI, 0.888–0.994), and an RCT:Ao ratio of 1.45 was considered a diagnostic characteristic, with a sensitivity of 90% and specificity of 89%. AKI, acute kidney injury; AUC, area under the curve; CI, confidence interval; CKD, chronic kidney disease; ROC, receiver operating characteristic.

In the BLKS group, 8/18 (44.4%) cats had a larger right kidney, and 10/18 (55.6%) had a larger left kidney. The renal lengths and RCT of the larger and smaller kidneys differed significantly (both p < 0.001; Table 9). The RCT:Ao ratio between the larger and smaller kidneys differed significantly (p < 0.001; Table 9).

TABLE 9.

Renal length, the RCT, and the RCT:Ao ratio of the bilateral kidneys in the BLKS group.

N = 18	Larger kidney (N = 18)	Smaller kidney (N = 18)
Renal length (mm)	36.86 ± 5.12 ** (29.25–47.60)	23.66 ± 6.32 ** (12.90–39.25)
RCT (mm)	3.44 ± 0.93 ** (1.87–5.73)	2.28 ± 0.86 ** (1.15–4.50)
RCT:Ao ratio	1.10 ± 0.27 ** (0.64–1.54)	0.73 ± 0.25 ** (0.38–1.09)

Note: In all three measurements, there were significant differences between the larger kidneys and the smaller kidneys.

Abbreviations: Ao, aorta; BLKS, big kidney–little kidney syndrome; M, mean; RCT, renal cortical thickness; SD, standard deviation.

^** p < 0.001. All continuous variables are expressed as M ± SD (range).

4. Discussion

This study was designed to determine the impact of BW and BCS on the RCT in cats and to establish a new parameter for evaluating the RCT that is independent of diverse body conformations and can be meaningfully utilized in CKD and AKI. In the present study, a moderate positive correlation was found between the RCT and BW, which is consistent with the results of previous studies on cats [17, 22]. The Pearson correlation coefficient (r) between the RCT and BCS and between the RCT and BSA was slightly higher in Group 2 than Group 1, but the difference was too small to be considered significant. Multiple regression analysis, including RCT, BCS, and BW, demonstrated a significant correlation between RCT and BW, but not between RCT and BCS. Therefore, the effect of BCS on the RCT was considered low, which led to the partial rejection of the first hypothesis. This is in contrast to the negative correlation between the BCS and RCT in dogs [33]. Thus, the importance of BCS in evaluating the RCTs in cats is likely to be low, indicating that obesity in cats has less impact on the RCT evaluations. This may be due to the relatively low variation in body shape in cats compared to that in dogs [27, 30], which reduces the effect of the BCS. However, cats were not evenly recruited to all the BCS sections in this study, which may have conferred selection bias. Further studies with more balanced populations at each BCS level are required.

The normal range of the RCT in feline kidneys has been studied previously. The average RCT (mm) of normal cats in this study was 3.71 ± 0.53, which is smaller than what was reported previously [17, 21, 22, 23, 29]. Possible reasons for this discrepancy are as follows: First, in some studies, only the Korean Shorthair [22] or Ragdoll [21] breeds were measured as a group, whereas in this study, approximately 18 different breeds were included. However, in previous studies, the dorsal plane of the kidney was used to measure the RCT, and the measurement site of the defined RCT was different from that in the present study [21]. Therefore, it is difficult to determine whether an RCT will reveal breed variations. Second, although the same cross‐sectional planes of the kidney were used in both this study and previous studies, the location of the estimated RCT within the cross section differed [17, 22]. Previous studies measured the RCTs at three sites in the kidney: the cranial pole, mid‐area, and caudal pole, and then used the average value for analysis [17, 22]. However, because the renal capsule margin can be obscured by edge‐shadowing artifacts at the cranial and caudal poles [32, 42], we selected three cortical sites in the ventral region of the kidney where the margin was most clearly distinguished and then used the average value in our analysis. Considering the elliptical shape of the kidney, the cranial and caudal poles correspond to the long axis of the ellipse, which is the longest of its multiple axes. Thus, it is expected that values larger than those measured in this study can be obtained.

The clinical significance of MRS has been discussed in several studies and has been observed in normal cats who do not have kidney disease, with a prevalence of approximately 36.8% [36]. The MRS is believed to be caused by mineral deposition within the renal tubular lumen of normal kidneys [7, 35, 36, 43]. Therefore, cats showing MRS without concurrent evidence of renal or systemic disease were classified as having normal kidneys. Previous reports indicate a thicker medulla in cats with MRS; however, it is unclear whether this affects the RCT [21, 23]. In this study, the RCT did not significantly differ between kidneys with and without MRS in the normal population. Consequently, the presence or absence of an MRS is unlikely to significantly impact the RCT measurements in individuals who do not have conditions that would cause an MRS.

In normal cats, the RCT:Ao ratio, unlike the RCT, showed no significant correlation with BW and BCS. This is consistent with the conclusions of a previous study on dogs [33]. This confirms our second hypothesis. In addition, similar to RCT, the RCT:Ao ratio showed no significant difference between normal cats with MRS and those without MRS. Therefore, to indirectly assess renal disease without considering the body conformation in cats, the RCT:Ao ratio may be a more useful diagnostic parameter than the RCT.

Compared to healthy cats, RCT values were lower in cats with CKD and higher in those with AKI. This supports previous findings of different tendencies in RCTs based on kidney disease [15, 18, 19, 26]. In addition, the mean RCT values between normal, CKD, AKI, and ACKD cats showed significant differences, which is consistent with our third hypothesis. Similar to the RCTs, the RCT:Ao ratio tended to be lower in the CKD group and higher in the AKI group than in the normal group, which confirmed the fourth hypothesis. This is consistent with the results of previous canine studies [15]. The RCT:Ao ratio in ACKD cats significantly differed from that in normal and CKD, but not from that in AKI. The mean RCT:Ao ratio in cats with ACKD was higher than that in normal and CKD cats but lower than that in cats with AKI. This suggests that cats with ACKD have values similar to those of cats with AKI. However, the absolute RCT:Ao ratio values of ACKD cats overlapped with those of CKD, normal, and AKI cats, indicating limited diagnostic reliability of this parameter. Thus, a comprehensive evaluation of clinical signs, blood tests, and other non‐imaging tests will be more important in cats with ACKD. However, owing to the small number of ACKD cats included in this study, the severity of ACKD was not assessed. Further studies with larger sample sizes of ACKD cats are needed to draw more specific and accurate conclusions.

One‐way ANOVA of the RCT:Ao ratio in the four CKD stages showed significant differences (p < 0.001), but not between CKD Stages 3 and 4 (p = 0.478). A significant difference was identified in the RCT:Ao ratio between normal, four stages of CKD, and AKI cats (p < 0.001), but not between normal and CKD Stage 1 (p = 0.119), between CKD Stages 2 and 3 (p = 0.108), or between CKD Stages 3 and 4 (p = 0.752). These results may be attributable to the possibility that cats with early stage kidney disease who have not yet developed azotemia may have been categorized as normal because blood tests, such as SDMA, or urinalysis were not performed. Furthermore, cats with azotemia but having a normal kidney appearance on ultrasonography can also explain the overlapping range of the RCT:Ao ratios between normal and early stage CKD cats. The lack of a significant difference in the RCT:Ao ratio for CKD Stages 3 and 4 is possibly due to the smaller number of cats who were evaluated in these stages, especially Stage 4, compared to Stages 1 and 2. Moreover, owing to the increased echogenicity of the cortex and medulla in the kidneys of CKD Stages 3 or 4, the corticomedullary junction was sometimes difficult to distinguish during the RCT measurements. Therefore, it is possible that the RCT was overestimated in cats with higher CKD stages. Overall, although no significant relationship was identified between some CKD stages, the absolute RCT:Ao ratio values tended to decrease with the increase in CKD stage. Consequently, if a sufficiently large number of cats had been recruited evenly, significant differences between all groups of normal, four CKD stages, and AKI, similar to those observed in dogs, might have been found.

The RCT:Ao ratio significantly differed among all groups of normal, early CKD stage, late‐CKD stage, and AKI cats. This indicates that the RCT:Ao ratio is more useful for diagnosing CKD as compared to the ratio of the renal length to the aortic diameter, given that the ratio of the renal length to aortic diameter did not significantly differ between early stage and late‐stage CKD in a previous feline study [44]. Furthermore, the RCT:Ao ratio may serve as a practical and noninvasive parameter for identifying early or subclinical renal changes, particularly in cats without definitive clinical signs or with inconclusive laboratory results. Early recognition of such changes could facilitate closer monitoring, earlier diagnostic workup, or more cautious use of nephrotoxic medications. Consequently, an RCT:Ao ratio of ≤1.15 can be used to predict the presence of CKD with a sensitivity of approximately 75% and a specificity of 80%, whereas an RCT:Ao ratio of ≥1.45 can be used to predict the presence of AKI with a sensitivity of approximately 90% and a specificity of 89%.

Kidney disease may affect one or both kidneys. If the function of one kidney is reduced, the other kidney compensates, and, despite the presence of kidney disease, blood tests may appear normal [38]. However, compensatory hypertrophy can occur when the contralateral kidney is overworked [38]. In the CKD group in this study, if the difference between the lengths of the two kidneys was >7 mm, cats were classified as having BLKS. Ureteral obstruction (e.g., ureteral calculi, mucus plugs, strictures) is the most common cause of BLKS, and non‐ureteral obstruction is considered a possible cause of CKD [38]. However, the cats with BLKS in this study were simply classified as having a difference in kidney length >7 mm, and notably, only two had ureteral calculi. We did not analyze whether the length of each kidney was within the normal range, whether only one side was normal (and if so, whether the larger side was normal or the smaller side was normal), or whether both sides were outside the normal range. Therefore, it is unlikely that all cats with BLKS in the present study had compensatory hypertrophy; thus, the possibility of unilateral CKD should be considered. In BLKS groups, the mean RCT:Ao ratio of the larger kidney was 1.10 ± 0.27, which was between the mean values recorded for cats with CKD Stages 1 and 2, and the mean RCT:Ao ratio of the smaller kidney was 0.73 ± 0.25, which was less than the mean value for cats with CKD Stage 4. As such, staging CKD by simply calculating the RCT:Ao ratio by using the average value of the RCTs of bilateral kidneys may lead to evaluation errors in cats with a difference of >7 mm in length between the two kidneys. Consequently, the RCT:Ao ratio criteria presented in this study should be used cautiously to predict CKD staging in cats with BLKS.

This study had some limitations. First, owing to the retrospective and multi‐institutional nature of the study, the ultrasound protocol may not have been uniform across many animal hospitals. For example, the use of sedation during the ultrasound examinations was not evaluated. However, the use of sedatives, such as butorphanol and propofol, in cats does not affect renal perfusion and is unlikely to have had a significant impact on the results of this study [45]. Although the concentration of sCREA should be measured in a fasted, well‐hydrated state [46, 47, 48], because of the retrospective study design, it was difficult to accurately confirm the status of all cats. Second, there was some uncertainty in the cat classification as cats with CKD and AKI were stratified on the basis of the IRIS criteria; however, the sub‐staging of each disease was not considered. Blood pressure was not considered in all cats, and urinalysis was only assessed in some cats. In addition, it is not possible to rule out errors caused by discrepancies between blood tests and ultrasonography. As a result, this may have caused overlapping ranges in the groups. However, although these measurements are not intended as a standalone diagnostic tool, the RCT:Ao ratio could complement current diagnostic protocols and may be useful in future screening procedures. Further studies are needed to establish specific thresholds and to determine which patient populations may benefit most from routine assessment of renal measurements. Third, in the kidneys of cats with end‐stage CKD, the renal cortex may be difficult to distinguish due to severe renal morphologic deformities, which makes the measurement of RCT difficult. However, cats with end‐stage CKD who were included in the present study were usually already undergoing long‐term management for CKD, suggesting that the RCT:Ao ratio may be more clinically useful for diagnosing early CKD and AKI. These findings also highlight the potential value of these measurements in future longitudinal studies to assess their predictive utility for the onset or progression of renal disease. Such studies may help develop standardized imaging protocols and establish reference ranges for various clinical settings.

In conclusion, unlike RCTs, the RCT:Ao ratio in cats was not influenced by body conformation and varied depending on the type of kidney disease. Therefore, the RCT:Ao ratio may serve as a practical, noninvasive, and clinically useful sonographic parameter for diagnosing feline kidney disease.

List of Author Contributions

Category 1

• (a)Conception and Design: Hyeonji Sim, Yoojin An, Sung‐Soo Kim, Danbee Kwon, Jeongmin Lee, Hakyoung Yoon
• (b)Data acquisition: Hyeonji Sim, Yoojin An, Sung‐Soo Kim, Danbee Kwon, Jeongmin Lee, Hakyoung Yoon
• (c)Analysis and interpretation: Hyeonji Sim, Yoojin An, Sung‐Soo Kim, Danbee Kwon, Jeongmin Lee, Hakyoung Yoon, Kichang Lee

Category 2

• (a)Drafting the article: Hyeonji Sim, Yoojin An, Sung‐Soo Kim, Danbee Kwon, Jeongmin Lee, Hakyoung Yoon
• (b)Review article for intellectual content: Hyeonji Sim, Yoojin An, Sung‐Soo Kim, Danbee Kwon, Jeongmin Lee, Hakyoung Yoon, Kichang Lee

Category 3

• (a) Final approval of the completed article: Hyeonji Sim, Yoojin An, Sung‐Soo Kim, Danbee Kwon, Jeongmin Lee, Hakyoung Yoon, Kichang Lee

Category 4

• (a)Agreement to be accountable for all aspects of the work, ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved: Hakyoung Yoon

Disclosure

The EQUATOR network was not used.

Conflicts of Interest

The authors declare no conflicts of interest.

Sim H., An Y., Kim S.‐S., et al. “Feline Renal Cortical Thickness–Aortic Diameter Ratio in Healthy Versus Diseased Kidneys: Comparative Ultrasonographic Evaluation.” Veterinary Radiology & Ultrasound 66, no. 5 (2025): e70090. 10.1111/vru.70090

Data Availability Statement

Raw data were generated at the Department of Veterinary Medical Imaging, College of Veterinary Medicine, Jeonbuk National University, VIP Animal Medical Center, Bundang Leaders Animal Medical Center, and The Care Animal Medical Center. Derived data supporting the findings of this study are available from the corresponding author (Hakyoung Yoon) and Hyeonji Sim upon reasonable request.