Published online Nov 19, 2024.
https://doi.org/10.4142/jvs.24183
Diagnostic validation of the urine albumin-to-creatinine ratio for early renal disease in healthy dogs and dogs with chronic kidney disease
Abstract
Importance
This paper highlights the urine albumin-to-creatinine ratio (UAC) as a valuable biomarker for the early detection of chronic kidney disease (CKD) in dogs. The UAC effectively distinguishes between healthy dogs and those with CKD, particularly in the early stages, and enhances diagnostic accuracy when used alongside other renal biomarkers.
Objective
To evaluate the utility of the UAC as a biomarker for early CKD diagnosis in dogs and examine its correlation with other renal biomarkers in a large-scale clinical study.
Methods
This study included 99 healthy dogs and 122 dogs with CKD. The UAC and other renal biomarkers were measured and evaluated in healthy dogs and those with CKD and categorized according to the staging criteria of the International Renal Interest Society (IRIS).
Results
Dogs with CKD had significantly higher UACs than healthy dogs (p < 0.05). The UAC correlated with the IRIS stages and other renal biomarkers (p < 0.05). Receiver operating characteristic curve analysis yielded an area under the curve of 0.817 (p < 0.05) for the UAC, with a cut-off value of 19.20 mg/g, showing 72% sensitivity and 71% specificity. A “grey zone” diagnostic window for early-stage CKD was introduced.
Conclusions and Relevance
The UAC is effective for the early diagnosis of renal disease in dogs. The UAC can differentiate between healthy dogs and those with CKD at IRIS stage 1. The diagnostic value is enhanced when used alongside other renal biomarkers, allowing for more specific guidelines for pet owners and veterinarians. This large-scale study addresses the limitations of previous research conducted on small clinical samples.
INTRODUCTION
Chronic kidney disease (CKD) in dogs is a condition marked by functional or structural kidney changes lasting for more than three months [1]. Its prevalence was reported to be 7%, increasing to 15% in dogs older than 10 years [2]. As the kidney function deteriorates, complications such as uremic gastroenteritis, anemia, and renal secondary hyperparathyroidism can occur [3]. In veterinary practice, the CKD stages in dogs are classified according to the International Renal Interest Society (IRIS) guidelines, with sub-staging based on blood pressure and proteinuria to evaluate the disease severity and prognosis [1, 3]. Early diagnosis is essential to prevent CKD progression. In human medicine, the glomerular filtration rate (GFR) and albumin-to-creatinine ratio are used for CKD management and monitoring [4]. Microalbuminuria, which is defined as urine albumin levels greater than 1 mg/dL but below 30 mg/dL on conventional dipstick tests, serves as an early CKD indicator [5]. An albumin-to-creatinine ratio over 30 mg/g is considered abnormal and is used for CKD monitoring, with early detection and treatment helping to slow disease progression [6].
In dogs, CKD diagnosis, according to IRIS guidelines, typically involves measuring the serum creatinine (CREA) concentration and observing clinical symptoms. Nevertheless, early-stage CKD might not show changes in CREA or symptoms concurrently [3]. In such cases, symmetric dimethylarginine (SDMA) can aid a diagnosis because its levels rise earlier than CREA [7, 8]. Other renal biomarkers, such as the urine protein-to-creatinine ratio (UPC), urine specific gravity (USG), and blood urea nitrogen (BUN), are also used to assess the risk and prognosis [9]. In addition, research into biomarkers, such as serum cystatin C, urinary cystatin B, neutrophil gelatinase-associated lipocalin, and retinol-binding protein, for early CKD diagnosis and disease progression assessment is ongoing [10, 11, 12, 13]. On the other hand, identifying reliable early CKD biomarkers remains a challenge.
Some studies suggest albuminuria detection can indicate early renal disease [14, 15, 16]. Research has indicated that the urine albumin-to-creatinine ratio (UAC) is significantly higher in dogs with CKD compared to healthy dogs [17]. Another study suggested a normal UAC range in dogs of ≤ 19 mg/g [18]. Various conditions, including inflammatory, infectious, metabolic, neoplastic, or cardiovascular diseases, can also cause albuminuria [14, 17, 19, 20].
This study examined the utility of UAC as a biomarker for early CKD diagnosis in dogs and its correlation with other kidney diagnostic biomarkers such as UPC, USG, BUN, CREA, and SDMA.
METHODS
Study design
Two hundred and twenty-one client-owned dogs were included in this study. The dogs were categorized into two groups based on the data from their history, clinical examinations, blood and urine analyses, and abdominal ultrasonography: 99 healthy dogs and 122 dogs diagnosed with CKD. The diagnosis and classification of CKD followed the 2023 IRIS criteria for CKD staging [21]. Stage 1 comprised dogs that were non-azotemic but showed other renal abnormalities, such as inadequate urine concentrating ability without an identifiable non-renal cause, abnormal renal palpation, and abnormal renal imaging findings. Stage 2 included dogs with mild azotemia, defined by a CREA level of 1.4–2.8 mg/dL and an SDMA concentration of 18–35 μg/dL. Stage 3 included dogs with moderate azotemia, indicated by a CREA level of 2.9–5.0 mg/dL and an SDMA concentration of 36–54 μg/dL. Stage 4 consisted of dogs with severe azotemia, with a CREA level > 5.0 mg/dL and an SDMA concentration ≥ 54 μg/dL. The distribution of dogs across the IRIS stages was as follows: stage 1 (46 dogs), stage 2 (30 dogs), stage 3 (23 dogs), and stage 4 (23 dogs). Samples were collected from four local animal hospitals (Times Animal Hospital, Siso Animal Hospital, Songjeong Animal Hospital, and Gangnam Ani Animal Hospital) between April 2023 and June 2023. Urine samples were used for UAC, UPC, and USG analysis, while serum samples were used to measure the BUN, CREA, and SDMA concentrations. The Institutional Animal Care and Use Committee at CurePharmTech (CPT-23-002-R) approved all procedures in this study.
Sample collection
Blood samples were drawn via jugular venipuncture and centrifuged at 1,500 × g for 15 min to obtain serum, which was stored at −70°C until use. Urine samples were collected by cystocentesis or urinary catheterization, stored at 4°C, and equilibrated to room temperature (20–30°C) for 30 min before analysis.
Inclusion and exclusion criteria
The healthy group comprised clinically healthy dogs with no abnormalities detected in their complete blood count, biochemistry profile, radiography, urinalysis, or ultrasound examination. The CKD group included dogs diagnosed based on the abnormalities found in their biochemistry profile, urinalysis, and ultrasound examination. The exclusion criteria for both groups included dogs with concurrent urinary tract infections, systemic illnesses, or any conditions affecting the kidney function or urine composition. This study also excluded dogs on medications that could influence the kidney function or urine protein levels, as well as those with significant dehydration, overhydration, or recent strenuous exercise. In addition, urine samples were excluded if they were left at room temperature for over 24 h, had insufficient volume for testing, or were contaminated with debris, blood, or other foreign materials.
Retrieval of medical records
The medical records for each patient were retrieved from local animal hospitals and critically reviewed to evaluate the renal function. In addition, clinical data, including breed, age, sex, body weight, physical examination findings, and clinical signs, were obtained from these records. Serum samples were stored at −70°C until submission to the NEODIN Laboratory (Korea) to determine the BUN, CREA, and SDMA values.
Urinalysis
The USG was measured using a refractometer (ATAGO Inc., Japan). A Catalyst One Chemistry Analyzer (IDEXX Inc., USA) quantified the UPC according to the manufacturer’s protocols. UAC was determined using Care Sign-V (i-SENS Inc., Korea) using immunoturbidimetry and chemical colorimetry for albumin and creatinine concentrations in urine, respectively.
Statistical analysis
All results are presented as means ± standard deviation. Statistical analysis was performed using IBM SPSS statistics version 27 (IBM Corp., USA). The age and body weight of the healthy dogs and CKD dogs were compared using a t-test. The differences in renal biomarkers were evaluated using a Mann–Whitney U test between healthy and CKD dogs and a Kruskal–Wallis test with Bonferroni correction for multiple comparisons across CKD stages. The Pearson’s correlation was used to evaluate the association between the UAC and other renal biomarkers. Receiver operating characteristic (ROC) curve analysis determined the UAC sensitivity, specificity, and optimal cut-off point, and a p value < 0.05 was considered statistically significant.
RESULTS
Signalments of the study population
This study included 99 healthy dogs and 122 with CKD. The mean age of the healthy dogs was 8.97 ± 2.79 years, whereas that of the dogs with CKD was significantly higher at 11.78 ± 3.16 years (p < 0.05). The healthy dogs comprised 27 (27.3%) intact males, 23 (23.2%) castrated males, 27 (27.3%) intact females, and 22 (22.2%) spayed females. Among the CKD dogs, there were 23 (18.9%) intact males, 43 (35.2%) castrated males, 18 (14.8%) intact females, and 38 (31.1%) spayed females. The mean body weight was 5.73 ± 5.04 kg and 4.97 ± 2.12 kg for healthy dogs and those with CKD. Maltese was the most common breed in both groups. No significant differences in breed or body weight were observed between the two groups (Table 1).
Table 1
Signalments of the study population
Comparison of renal biomarkers
Table 2 lists the mean values of the renal biomarkers. All renal biomarkers showed significant differences between healthy dogs and those with CKD (p < 0.05). The mean UAC, UPC, USG, BUN, CREA, and SDMA values were 20.52 ± 23.70 mg/g, 0.21 ± 0.30, 1.04 ± 0.02, 21.33 ± 7.91 mg/dL, 0.62 ± 0.23 mg/dL, and 9.76 ± 2.83 μg/dL, respectively, in healthy dogs, and 229.83 ± 283.46 mg/g, 3.51 ± 3.44, 1.02 ± 0.01, 53.27 ± 43.92 mg/dL, 2.81 ± 2.41 mg/dL, and 27.98 ± 15.81 μg/dL, respectively, in dogs with CKD. As CKD progressed, the UAC, UPC, BUN, CREA, and SDMA levels tended to increase, while USG showed a decreasing trend (Table 3).
Table 2
Comparison of the renal biomarkers between healthy dogs and dogs with CKD
Table 3
Comparison of renal biomarkers between healthy dogs and dogs with CKD based on IRIS staging
The UAC and UPC were significantly lower in healthy dogs than in those with CKD at every IRIS stage (p < 0.05). Similarly, the USG was significantly higher in healthy dogs than in those with CKD at every IRIS stage (p < 0.05). The serum BUN concentration was significantly lower in healthy dogs than in those with CKD at stages 2, 3, and 4 (p < 0.05). The CREA and SDMA concentrations were significantly lower in healthy dogs than in those with CKD at every IRIS stage (p < 0.05).
Correlation of UAC with other renal biomarkers
Fig. 1 shows the correlations between the UAC and other renal biomarkers. A strong positive correlation was observed between the UAC and UPC (r = 0.790, p < 0.05; Fig. 1A). Similarly, the UAC showed a significant moderate negative correlation with the USG (r = 0.548, p < 0.05; Fig. 1B). A weak-to-moderate positive correlation was observed between the UAC and BUN concentration (r = 0.401, p < 0.05; Fig. 1C). Furthermore, UAC exhibited a significant moderate positive correlation with CREA concentration (r = 0.644, p < 0.05; Fig. 1D). Lastly, the UAC showed a moderate positive correlation with serum SDMA concentration (r = 0.640, p < 0.05; Fig. 1E).
Fig. 1
*p < 0.05.
Scatter-plots with line regressions of the relationships between the UAC and UPC (A), USG (B), BUN (C), CREA (D), and SDMA (E).
UAC, urine albumin-to-creatinine ratio; UPC, urine protein-to-creatinine ratio; USG, urine specific gravity; BUN, blood urea nitrogen; CREA, serum creatinine concentration; SDMA, symmetric dimethylarginine; r, Spearman correlation coefficient.
ROC curve analysis of UAC and SDMA
ROC curve analysis was performed to determine the sensitivity and specificity of the UAC and SDMA and identify the optimal cut-off points (Fig. 2). The area under the curve (AUC) values were also calculated. The UAC had an AUC of 0.817 (95% confidence interval [CI], 0.074–0.889; p < 0.05) with a cut-off value of 19.20 mg/g, showing a sensitivity and specificity of 72% and 71%, respectively, in distinguishing healthy dogs from those with CKD. The interval between this cut-off value and the higher cut-off value of 64.20 mg/g, as reported in previous studies [22], represents a “grey zone.” This “grey zone” indicates a diagnostic window where CKD progression may be detected at an early stage, offering a more detailed understanding of disease onset.
Fig. 2
ROC curve analysis for the UAC and SDMA in distinguishing healthy dogs from those with CKD. The ROC analysis of the UAC and SDMA revealed an AUC of 0.817 (95% CI, 0.746–0.889; p < 0.05) and 0.908 (95% CI, 0.853–0.962; p < 0.05).
ROC, receiver operating characteristic; UAC, urine albumin-to-creatinine ratio; SDMA, symmetric dimethylarginine; CKD, chronic kidney disease; AUC, area under the curve; CI, confidence interval.
SDMA exhibited a higher AUC of 0.908 (95% CI, 0.853–0.962; p < 0.05), with a cut-off value of 12.20 μg/dL, yielding a sensitivity and specificity of 84% and 83%, respectively, for distinguishing between healthy dogs and those with CKD.
Establishment of three-dimensional scatter plot
The distributions of UAC and CREA based on the IRIS stages are represented in a three-dimensional scatter plot (Fig. 3A). As the IRIS stage progressed, the CREA levels showed an increasing trend. On the other hand, the UAC values exhibited a wide range of distributions, even within the same stage. In addition, when depicting the distribution of UAC and SDMA based on the IRIS stages in a three-dimensional scatter plot, the SDMA levels increased with advancing stages, while UAC showed diverse distributions (Fig. 3B).
Fig. 3
Three-dimensional scatter plots of the UAC, CREA, and CKD IRIS stages (A) and the UAC, SDMA, and CKD IRIS stages (B). Stage 0 indicates a healthy dog group. This result suggests that UAC can exhibit varying values even within the same stage group.
UAC, urine albumin-to-creatinine ratio; CREA, serum creatinine concentration; CKD, chronic kidney disease; IRIS, International Renal Interest Society; SDMA, symmetric dimethylarginine.
DISCUSSION
The early detection of CKD is crucial to delaying its progression and minimizing clinical symptoms. Studies have suggested that high UAC and UPC are associated with rapid CKD progression [23]. Proteinuria is also closely linked to the patient survival rates [24]. This study revealed a significant difference between healthy dogs and those with CKD when comparing the UAC, UPC, USG, CREA, BUN, and SDMA values. Studies have shown that patients with CKD have significantly higher UAC values than healthy dogs [17], which is consistent with the present findings. In one study, the normal UAC range in dogs was 19 mg/g, which was narrower than the normal range in humans (30 mg/g) [18].
In addition, the UAC values between healthy dogs and those with CKD at each IRIS stage were significantly different. In particular, the difference between healthy dogs and those at IRIS stage 1 suggests that the UAC may be a potential early diagnostic biomarker of kidney disease. This finding is consistent with several studies suggesting that albuminuria can be used as an early diagnostic indicator of kidney disease [14, 15, 16].
The correlation between UAC and other renal biomarkers was evaluated, and significant positive correlations were observed between the UAC and the UPC, BUN, CREA, and SDMA. Among these, the UPC and BUN showed the strongest and weakest correlation, respectively, suggesting that the UAC and UPC can be applied to early diagnosis of kidney diseases. The weak correlation with BUN may be due to its significant influence by factors such as a high protein diet, liver disease, or gastrointestinal bleeding. However, the correlation coefficients between UAC and CREA/SDMA were not high, suggesting that UAC may not necessarily increase as GFR, represented by CREA and SDMA, decreases. The UAC, CREA, SDMA, and IRIS stages were presented in a 3D scatter plot to evaluate the correlation between GFR and UAC. The graphs showed that decreases in GFR influenced the values of CREA and SDMA with advancing CKD stage. On the other hand, the UAC values exhibited a wide range of distributions within the same stage. Hence, the UAC should be monitored alongside CREA and SDMA because proteinuria may appear to decrease in progressive renal disease as the number of functional nephrons decreases [25]. In humans, the changes in UAC are dynamic, and the GFR changes are progressive, indicating a complementary rather than competitive relationship [26]. Therefore, measuring and monitoring the UAC appears to have significant value in managing patients with CKD.
ROC curve analysis revealed AUC values for UAC and SDMA of 0.817 and 0.908, respectively. This result suggests that UAC and SDMA are useful diagnostic biomarkers for diagnosing CKD. On the other hand, SDMA showed a higher AUC value, possibly because the CKD IRIS stage system is based on the CREA and SDMA concentrations [3]. The UAC had a sensitivity and specificity of 72% and 71%, respectively, with a cut-off value of 19.20 mg/g. This finding indicates a lower sensitivity and specificity, implying the difficulty of diagnosing CKD based solely on the UAC. On the other hand, because the mechanisms behind the increase in UAC and CREA or SDMA differ, it is essential for CKD diagnosis and management to comprehensively evaluate multiple biomarkers with different mechanisms. Therefore, a UAC assessment is considered to have clinical value.
Kang et al. [22] published foundational research on the albumin-to-creatinine ratio using small clinical samples of 50 healthy dogs and 50 dogs with CKD. Although their study established an optimal diagnostic cut-off value of 64.20 mg/g with high sensitivity and specificity (94% each), it was based on a limited number of dogs, particularly in the early stages of CKD (e.g., only six dogs in IRIS stage 1). In veterinary practice, this cut-off value might be somewhat high, given the small sample size.
In contrast, this study included 221 dogs, with a more balanced distribution across the CKD stages (e.g., 46 dogs in IRIS stage 1), providing a larger and more representative clinical sample. This larger scale allowed the establishment of a lower UAC cut-off value of 19.20 mg/g, reflecting improved diagnostic accuracy for early kidney disease detection. This study addresses the limitations of prior research and offers more effective differentiation between normal and abnormal conditions, particularly in an early CKD diagnosis.
This paper introduces the novel concept of a “grey zone” in UAC-based diagnosis, which was not addressed in previous research. The UAC cut-off value of 19.20 mg/g, identified in this study, differed from the higher cut-off of 64.20 mg/g reported elsewhere [22]. The interval between these two values is termed the “grey zone,” representing a critical diagnostic window where CKD progression may be detected in the early stages. This innovation allows for a more nuanced diagnostic approach, aiding clinicians in identifying an early renal dysfunction that may not fit neatly into “normal” or “abnormal” classifications. This grey zone concept enhances the diagnostic precision and highlights the importance of using refined cut-off values when monitoring CKD progression.
Although this study did not incorporate longitudinal data, the exploration of UAC in dogs with CKD can be significantly enriched by the insights gained from longitudinal studies conducted in human populations. A recent multicenter prospective cohort study showed that even within the normoalbuminuric range (UAC < 30 mg/g), higher baseline albuminuria levels were associated with an increased risk of CKD progression over time [27]. In particular, the findings suggested that the 10-year cumulative incidence of CKD progression in humans increased linearly as the albuminuria levels increased, highlighting the importance of monitoring the changes in renal biomarkers as potential indicators of disease advancement.
These findings underscore the necessity of conducting longitudinal studies in canine populations to evaluate how the albumin-to-creatinine ratios change over time in dogs with CKD. Such research could provide critical insights into the natural history of CKD in dogs, potentially allowing for earlier intervention and management strategies as the disease progresses. Longitudinal studies would facilitate the identification of threshold values for albumin-to-creatinine ratios that could indicate an increased risk of deterioration in the kidney function, paralleling the findings observed in humans.
This study evaluated the renal biomarkers in dogs with CKD compared to healthy controls. One notable limitation of this study is the age difference between the two groups. The CKD group comprised predominantly older dogs. Ideally, age-matched groups would provide a more precise comparison of the renal markers. On the other hand, the natural progression of CKD predominantly affects older dogs, leading to this age disparity in the study population. Furthermore, the effects of sex and breed on the renal biomarkers were not assessed in this study. Previous studies suggested that age alone does not significantly affect renal biomarkers. One study showed that markers such as urinary albumin, retinol-binding protein, and N-acetyl-β-D-glucosaminidase were significantly elevated in CKD dogs compared to healthy dogs, while no significant differences were observed between healthy young and older dogs for these markers [13]. This finding reinforces the assertion that the primary factor influencing the renal markers in the present study is the presence of CKD rather than age.
Despite these limitations, more meaningful results could be generated through further studies by obtaining samples from age-matched dogs at higher stages of CKD through follow-up research. In addition, future longitudinal studies are vital to validating the early diagnostic trends observed in this research and understanding the implications of changing albumin-to-creatinine ratios in the context of CKD progression in dogs.
Funding:This research was financially supported by i-SENS Inc (Seoul, Republic of Korea).
Conflict of Interest:The authors declare that the research was funded by by i-SENS Inc (Seoul, Republic of Korea). The funding source had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Author Contributions:
Conceptualization: Kang MH, Park HM.
Data curation: Lee SY, Cha YE, Kang HM, Kang DJ.
Formal analysis: Lee SY, Kang MH.
Funding acquisition: Park HM.
Methodology: Park HM.
Supervision: Kang HM, Kang MH, Park HM.
Validation: Kang MH, Kang DJ, Kang HM.
Writing - original draft: Lee SY.
Writing - review & editing: Lee SY, Cha YE, Kang HM, Kang DJ, Kang MH, Park HM.
ACKNOWLEDGMENTS
This work was presented as a poster at the WSAVA 2023 Congress and was utilized for the author’s master’s thesis.
References
-
Grauer GF, Oberhauser EB, Basaraba RJ, Lappin MR, Simpson DF, Jensen WA. Development of microalbuminuria in dogs with heartworm disease (abstract). J Vet Intern Med 2002;16:352.
-
-
Lees GE, Jensen WA, Simpson DF, Kashtan CE. Persistent albuminuria precedes onset of overt proteinuria in male dogs with X-linked hereditary nephropathy (abstract). J Vet Intern Med 2002;16:353.
-
-
Vaden SL, Jensen WA, Longhofer S, Simpson D. Longitudinal study of microalbuminuria in soft-coated wheaten terriers (abstract). J Vet Intern Med 2001;15:300.
-
-
Pressler BM, Proulx DA, Williams LE, Jensen WA, Vaden SL. Urine albumin concentration is increased in dogs with lymphoma or osteosarcoma (abstract). J Vet Intern Med 2003;17:404.
-
-
Kang HM, Kim HS, Kang MH, Kim JW, Kang DJ, Ro WB, et al. Evaluation of albumin creatinine ratio as an early urinary biomarker for chronic kidney disease in dogs. J Vet Clin 2023;40(6):399–407.
-
-
International Renal Interest Society (IRIS). IRIS staging of CKD (modified 2023) [Internet]. IRIS; Published 2023 [Accessed 2024 June 2].http://www.iris-
kidney.com/pdf/2_ IRIS_ Staging_ of_ CKD_ 2023.pdf.
-
Publication Types
MeSH Terms
Figures
1 / 3
Tables
1 / 3