Correlation between separate findings in 60 dogs suffering from CKD in IRIS stage 1

The International Renal Interest Society (IRIS) reports guidelines for classification and therapy for dogs suffering from chronic kidney disease (CKD). Dogs in stage 1 show no elevated serum creatinine, but some other sign of kidney disease like elevated Symmetric Dimethylarginin (SDMA), proteinuria, low urine specific gravity or abnormalities in the sonography of the kidneys. Objective of this study was to access the correlation between inclusion criteria and to give an estimation whether a more detailed staging or a substaging for patients with elevated SDMA levels might be useful regarding possible treatment recommendations. Sixty raised or In to and compare developed. Additionally, results from blood work, ultrasonography and in some cases glomerular filtration rate (GFR) were examined. Correlation analysis showed positive correlation between creatinine and urea and negative correlation between creatinine and urinary protein-to-creatinine ratio (Up/c). Between SDMA, phosphate, urine specific gravity and sonographic findings there is no dependency with any other examined parameter.


Abstract Background
The International Renal Interest Society (IRIS) reports guidelines for classification and therapy for dogs suffering from chronic kidney disease (CKD). Dogs in stage 1 show no elevated serum creatinine, but some other sign of kidney disease like elevated Symmetric Dimethylarginin (SDMA), proteinuria, low urine specific gravity or abnormalities in the sonography of the kidneys. Objective of this study was to access the correlation between inclusion criteria and to give an estimation whether a more detailed staging or a substaging for patients with elevated SDMA levels might be useful regarding possible treatment recommendations.

Results
Sixty patients in IRIS stage 1 were included into the study. Most of these dogs were included due to raised SDMA level (n=22) or sonographic abnormalities of the kidneys (n=16). In order to rank and compare results of ultrasonography, a sonography score was developed. Additionally, results from blood work, urinalysis, ultrasonography and in some cases glomerular filtration rate (GFR) were examined. Correlation analysis showed positive correlation between creatinine and urea and negative correlation between creatinine and urinary protein-to-creatinine ratio (Up/c). Between SDMA, phosphate, urine specific gravity and sonographic findings there is no dependency with any other examined parameter.

Conclusion
Results showed that patients in IRIS stage 1 are a heterogeneous group and giving precise treatment recommendation might be challenging. If future studies will suggest treatment in this stage of CKD, such as renal diet, a more detailed classification is needed. Background 3 Chronic kidney disease (CKD) is characterized by the progressive and irretrievable loss of functioning nephrons in one or both kidneys for at least three months or longer (1). The disease is the most common kidney disease and as various studies showed, most relevant in older dogs (2)(3)(4). The most recent study by O'Neill et al. reported a prevalence of 0.37% (5), older studies even referred to a prevalence up to 3.74% (6). While aetiology remains unknown in most cases, therapy concentrates on symptomatic treatment. Hypertension, nausea, polyuria and polydipsia are the most relevant signs shown by patients (7). Gold standard for the evaluation of kidney function is the measurement of glomerular filtration rate (GFR) (8). Direct measurement of GFR is costly and time-consuming, as it involves hospitalization and the collection and analysis of multiple blood samples (8,9). In everyday clinical practice, it is common to rely on endogenous serum biochemical markers like serum creatinine, which is known to be influenced by the patient's muscle mass and to be unable to detect early decrease of kidney function (10). The International Renal Interest Society (IRIS), a group of veterinarian experts, developed a four-stage system to classify dogs suffering from CKD (11). The four-stage system is based on the results of blood chemistry, urinalysis and systemic arterial blood pressure measurement (11). The initial staging is based on serum creatinine. While stages two to four are composed of patients with different levels of raised creatinine, stage one includes non-azotaemic patients, which exhibit other signs of CKD. Possible criteria for the inclusion into IRIS stage 1 are: Polyuria, polydipsia, proteinuria, poor urine concentration, abnormalities in from receiving a renal diet as early as IRIS stage 1 (determined by raised serum SDMA level) (15). SDMA is the methylated form of arginine and primarily eliminated by renal excretion. In previous studies, it showed a high correlation to glomerular filtration rate (GFR) in dogs suffering from kidney disease and was increased in average 9.4 months earlier compared to creatinine (16). Therefore, SDMA seems to be a promising biomarker for an early detection of CKD (16)(17)(18)(19). While correlation of serum SDMA and GFR as the Gold standard for the evaluation of renal function has already been established, correlation between abnormalities in sonography of the kidneys and GFR is still inconclusive. Regarding ultrasonography, a scoring system considering several sonographic parameters of the kidneys was developed, since single criteria of the sonographic appearance of the kidneys have shown little correlation to kidney function in past studies (20-23). Regarding the wide variety of symptoms that lead to inclusion into stage 1, it is obvious that giving precise treatment advice is challenging. Due to the possible positive effect of receiving a renal diet in patients with raised SDMA levels (24), further differentiation of IRIS stage 1 might be necessary to improve treatment recommendations for early CKD. In the current study, blood results, reports of abdominal sonography and results of urinalysis of 60 patients that were presented to the Small Animal Clinic of the University of Veterinary Medicine Hanover from 2017 to 2019 and who were classified as belonging to IRIS stage 1 were reviewed. The objectives of this study were to 1. record the different causes that lead to inclusion into IRIS Stage 1 in clinical practice 2. access the correlation between the numerous criteria associated with CKD and lead to inclusion into the study 3. give an estimation whether a more detailed differentiation between patients, which are classified as IRIS Stage 1 right now, might be necessary regarding possible treatment recommendations and

Study design
A retrospective analytical study design was used. The medical record database at the Small Animal Clinic of the University of Veterinary Medicine Hanover was searched to identify dogs with chronic kidney disease in IRIS stage 1.

Patients
All the animals included in this study were presented to the clinic from January 2017 to December 2018. For each recorded case, the following data was analysed: age, breed, cause of presentation, presence of polyuria and polydipsia, feed of renal diet, outcome of ultrasonography of the urinary tract, lab work including blood count, clinical chemistry and urinalysis, and cause for inclusion into IRIS stage 1 (  Urinalysis was performed in house at the clinic for small animals by utilizing urine sticks, refractometer and cytology.

Ultrasound
Ultrasound of the urogenital tract was performed by ultrasonography specialists working at the Small Animal Clinic, using the standard ultrasonography machine of the clinic (Logiq 7, GE Healthcare, Solingen / Germany). Images were acquired with either a curvilinear or a linear array transducer with frequencies ranging from 8 to 14 MHz according to clinician preference. Results of the ultrasonographic examination were retrospectively converted into a score, considering the following criteria: kidney size in relation to aorta, shape, cortex thickness and echogenicity, proportion of cortex and medulla, kidney pelvis, medullary rim sign, calcifications, cysts and surface of each kidney (table 2). Scoring was performed for both kidneys independently and the results of both kidneys were summed up to build the finale score. To be included into the study due to sonographic abnormalities of the kidneys, patients had to meet at least two points of the score.

Glomerular filtration rate
In six dogs, GFR measurement was available in addition to the aforementioned parameters. Measurement was performed using endogenous creatinine clearance. 5%creatinine solution in a dosage of 5 gr creatinine/m 2 body surface was mixed and filtrated in an in house laboratory 24 hours prior to application. After measuring basal serum creatinine, 5% creatinine solution was applicated subcutaneously. Additional measurements of serum creatinine were performed three, five and eight hours after application. During the whole time, fluids were given intravenously in a dosage of 2 ml/kg/h.

Statistics
The Shapiro-Wilk-test was used to test for standard distribution of the data. Seeing only data for creatinine showed standard distribution, Spearman rang correlation was used to analyze correlation between results. The minimal level of significance was set to <0.05.
Results 60 dogs met the inclusion criteria for the study. The average age was 9.5 (± 3.3) years Results of blood analysis were available in all dogs. Urinalysis was performed in 56 cases.
Ultrasonographic examination was available in 52 patients. Retrospectively converted into a scoring system, 25 patients achieved zero to two scoring points and 30 patients three to six points in the ultrasonography score. Only three patients had a score of 7 or more points. The most common feature seemed to be reduced corticomedullary demarcation (n=41).
At the time of presentation, seven dogs were fed a renal diet by the owners, following their veterinarians' advice. Three of these dogs were included in the study because of increased urea, two dogs were presented with elevated SDMA. Proteinuria and low urinary specific weight were detected in one dog each.
Most dogs were originally included into the study due to increased SDMA (n=22). Initially, 16 dogs were included in the study because of sonographic changes. Nine dogs showed increased urea in fasted trials, seven dogs showed decreased urinary specific weight.
Proteinuria was present in four dogs and only two dogs were included because of polyuria and polydipsia ( fig. 3). There were 25 patients who showed two of the examined signs to be included into IRIS stage 1, eight patients showed three signs, three patients showed four signs and one patient showed five signs.
Overall, SDMA was measured in 51 patients and showed a mean value of 14  CKD is reported to be a disease mostly occurring in older dogs, which was confirmed in the present study with a medium age of 9.5 years. In total, 24 different breeds but mostly mongrels were seen, which fits the results of prior studies and shows that mixed breeds are as likely as pure breeds to develop CKD (5). Regarding patients' neuter status, a striking number of patients in this study were neutered. This partly meets the findings of other studies which demonstrated a high number of neutered patients suffering from CKD, but could not find a significant correlation between neuter status and the prevalence of kidney disease (5). Still, further studies might be needed to evaluate whether a connection between development of CKD and neuter status exists.
One purpose of this study was to identify the context between the numerous criteria that lead to inclusion into IRIS stage 1. No significant relationship between serum SDMA concentrations and serum creatinine, urea, phosphate, specific gravity, urine protein/creatinine or sonographic findings was observed. In the data of this study there was only a correlation between serum creatinine and urea and creatinine and Up/c. This is a discrepancy to the results of a former study that revealed a significant association between SDMA and creatinine in dogs affected with CKD caused by X-linked hereditary nephropathy (17), as well as in cats suffering from CKD and humans (31, 32). This might be due to the heterogenous inclusion criteria of the patients participating in the study, many of which did not show elevated SDMA levels. Additionally, only patients in a very early stage of kidney disease were analysed, whereas the dogs in the study of Nabity et al. affected with X-linked hereditary nephropathy developed severe renal insufficiency in the course of the study (17). In the healthy control group, no correlation between SMDA and GFR was found in the study by Nabity et al. either.
Correlation between serum creatinine and urea was shown in this study population.
Although urea is influenced by patients' food composition as well as liver function, it is known to be a marker for kidney function and therefore was expected to correlate with creatinine (5). As patients were in an early state of CKD, it is obligatory that creatinine was within the reference range and in most cases, urea was not increased either. The  Not applicable.

Consent to publish:
Not applicable.

Availability of data and material:
All data generated or analyzed during this study are included in this published article.

Competing interests:
The authors declare that they have no competing interests.

Funding:
No funding.

Authors' contributions:
MGS made substantial contribution to acquisition of data, analysis and interpretation. MGS             Most patients showed SDMA between 10 µg/dl and 20 µg/dl and sonographic score points between 2 and 6. SDMA and sonographic score points were not significantly correlated (p = 0.64). Figure 1 Age of patients, graded into three groups. Five Patients were included into the group of younger patients (6 months-3 years), six patients were 4-6 years old.

Figures
The majority of patients (n=49) were 7 years of age or older.      Scatter plot displaying the relation between measured creatinine and sonography.
Most patients showed serum creatinine between 0.7 mg/dl and 1.4 mg/dl and sonographic score points between 2 and 6. Creatinine did not correlate significantly to sonographic score points (p= 0.33).

Figure 8
Scatter plot displaying the relation between measured SDMA and sonography.
Most patients showed SDMA between 10 µg/dl and 20 µg/dl and sonographic score points between 2 and 6. SDMA and sonographic score points were not significantly correlated (p = 0.64).