Urine cortisol‐creatinine and protein‐creatinine ratios in urine samples from healthy dogs collected at home and in hospital

Abstract Background Recently, urine protein:creatinine ratios (UPC) were shown to be lower in urine samples from dogs collected at home (AH) as compared to those collected in hospital (IH). Stress‐inducing procedures and travel to the hospital have been hypothesized to cause prerenal proteinuria. Objectives Evaluate patient stress using urine cortisol:creatinine ratios (UCCr) and correlate UCCr to UPC in urine samples obtained AH and IH. Animals Thirty‐six healthy, client‐owned dogs. Methods Prospective, non‐masked study. Two voided urine samples were obtained (AH and IH). Complete urinalysis as well as UPC and UCCr were performed. Clients graded their dogs' stress level AH, in transport, and IH. Results The UCCr was significantly higher in IH samples than in AH samples (P < .0001), but UPC was not significantly different between AH and IH urine samples (P = .14). In all samples and in both collection settings, UCCr was not significantly correlated with UPC. Travel time and time IH were not correlated with change in UCCr or UPC. In 8 dogs with borderline or overt proteinuria, no significant difference was found in UPC between settings, but UCCr was significantly higher in IH samples. Conclusions and Clinical Importance The UPC was not higher when measured in urine samples collected IH compared to AH. Dogs had higher UCCr IH, but UCCr was not associated with UPC. Stress, as estimated by UCCr, did not affect proteinuria. Further evidence is needed to support the claim that stress may result in proteinuria in healthy dogs.

patient's evaluation in hospital (IH). Urinalyses are routinely used to evaluate kidney function, including semiquantitative and quantitative measurement of urine protein concentration. Proteinuria can be extrarenal or renal in origin, and can be due to prerenal, renal, or postrenal processes. Pathologic renal proteinuria often is the most severe type of proteinuria, especially when the cause is altered glomerular permselectivity. In particular, pathologic renal proteinuria is considered a marker for increased risk for progression of kidney disease and prompts the need for intervention by practitioners. However, other nonrenal factors, such as stress, have been hypothesized to cause transient, nonpathologic prerenal proteinuria in dogs. 3 Several publications have demonstrated that emotional stress may result in transient proteinuria in humans. Altered urine concentrations of neurohormones such as endothelin-1 and prostaglandins have been observed after stressful stimuli, which may affect glomerular permselectivity. 4 Altered permeability of the glomerular barrier typically results in the appearance of high molecular weight proteins, such as albumin, in the urine. Several qualitative metrics of emotional stress have been observed to correlate with urine total protein concentration and with both urine protein:creatinine (UPC) and albumin:creatinine ratios in people. 5,6 Stress also was associated with increased urine concentration of malondialdehyde, a marker of lipid oxidation, suggesting oxidative stress may play a role in the etiology of proteinuria. 5 Although no studies have determined that stress causes proteinuria in dogs, some veterinarians believe stress to be a cause of proteinuria. 7,8 Serum and urine cortisol concentrations can be used to assess stress in dogs. 9,10 The IH setting can induce a stress response in dogs, as previously indicated by increased serum cortisol concentrations compared to dogs that stayed outside the hospital. 11 Similarly, veterinary care causes stress in dogs and increases their urine cortisol:creatinine ratio (UCCr) while IH compared to AH. 12 Both naturally occurring and iatrogenic hypercortisolemia have resulted in increased UPC in dogs. [13][14][15][16][17] A recent study found that UPC ratios were higher when urine was collected IH compared to AH in dogs with positive urine dipstick protein results. 3 Stress was a hypothesis provided by the authors as to the cause of the higher UPC observed IH, but no assessment of stress was performed in that study. Hospital of the University of Pennsylvania by phone call at least 1 day before their appointment. Clients who gave verbal consent for enrollment in the study were asked to collect a mid-stream voided urine sample in a dry clean plastic cup or bag on the morning of their appointment (within 12 hours of the appointment time) and to store the sample in a refrigerator until leaving for the hospital. Clients were asked to use cups that had never been in the dishwasher to avoid detergents that could interfere with laboratory testing. Dogs were enrolled regardless of the presence of proteinuria on urinalysis examination.
Upon arrival at the hospital, clients submitted the urine samples to investigators and samples were time stamped. Clients were given a client consent form to sign, as well as a short survey to complete, which inquired about travel time from their home to the hospital.
The survey also asked clients to grade their dog's stress level from 0 to 10 (0 being nonstressed and 10 being severely stressed [the most stressed their pet could be]) during the AH urine collection, the car ride to their appointment, and at the time of survey administration in the hospital lobby. Study dogs completed their scheduled appointments and were taken to an enclosed space to obtain an additional voided urine sample. Dogs that required cystocentesis were excluded from the study. Hospital sample collection time was noted to determine the total duration of time IH, and additional data (age, weight, breed, reason for visit) were collected from the medical history and examination forms. A minimum urine volume of 3 mL was required for study eligibility. Dogs qualified for enrollment regardless of whether or not they were positive for proteinuria on a urine dipstick analysis.
Urine protein:creatinine ratios were interpreted according to the

American College of Veterinary Internal Medicine and International
Renal Interest Society (IRIS) guidelines on proteinuria: UPC <0.2 = normal, 0.2-0.5 = borderline, and >0.5 = overtly proteinuric. 18,19 These standards were used to evaluate discrepancies among UPC results and assess clinical relevance.
Urine protein and creatinine concentrations were measured using pyrogallol red-molybdate and creatinase methods, respectively, on an automated chemistry analyzer (Ortho Clinical Vitros 4600) within 24 hours of collection. If submitted after hours, samples were refrigerated at 4 C until testing was performed the next day. Urine cortisol concentration was measured at an external reference laboratory using a chemiluminescent bead-based assay (Siemens Immulite 2000). A conventional urinalysis also was performed and included evaluation of macroscopic characteristics, specific gravity, chemical analysis by dipstick, and microscopic sediment examination.

| Statistical methods
The small sample size was presumed not to be normally distributed, and therefore nonparametric tests were utilized. A Wilcoxon signed rank test was used to evaluate AH versus IH variables including UPC and UCCr. Owner perceived stress scores for their dogs AH, during travel, and IH were analyzed using Friedman's test. Spearman's analysis was used to assess the strength of correlation between UPC and UCCr as well as the effect of travel time and duration of visit on UPC.
A P value <.05 was considered significant. A commercial statistical software program was used for all analyses (Prism 7 for Mac, GraphPad Software Inc, La Jolla, California).

| RESULTS
Thirty-six dogs were enrolled in the study: 33 dogs presented for annual physical examinations or vaccinations to the primary care service and 3 dogs presented for ophthalmologic reevaluations. The initial aim of the study was to enroll 40 dogs, but enrollment was limited to 36 because of funding limitations. Mixed breed dogs were most common (13), followed by German Shepherds (3)  and consistent with overt proteinuria. One of these proteinuric patients was an intact male.
No significant difference was identified in UPC between urine samples obtained AH versus IH, which also was true in the subpopulations of dogs that were borderline (UPC >0.2) or overtly proteinuric (UPC >0.5) in at least 1 sample ( Table 1).
The UCCr results were significantly higher for IH samples compared to those collected AH (P = .0001; Table 2  changes in UPC (P = .14 and P = .45, respectively) or UCCr (P = .50 and P = .73, respectively). The observed change in AH and IH UPC was not associated with the change between AH and IH UCCr (P = .14). Identification of proteinuria in azotemic and non-azotemic dogs will influence treatment recommendations, based on the IRIS staging protocol. Consistency when monitoring a proteinuric dog is recommended. 18,19 This includes minimizing the analytical variability such as using the same instrument and laboratory for repeated measurements as well as limiting preanalytical contributions such as time between specimen collection and analysis, the potential influence of sample collection timing, and variation between single and pooled specimens. [23][24][25] Twenty-four-hour urine collection for protein quantification is considered the gold standard method for assessing proteinuria. 20 Although the UPC does correlate with 24-hour urine protein excretion, variability can occur in spot UPC results. 26 has not been studies. Another limitation of our study is the low number of overtly proteinuric dogs that were enrolled. Because urine protein results were not known before enrollment, the study population was predominantly non-proteinuric dogs. The study population did experience a change in stress level, but this did not have any impact on UPC. Further study should be performed in proteinuric dogs to evaluate for any correlation between UCCr and UPC.

| DISCUSSION
In conclusion, no difference was found in UPC in urine samples collected AH compared to IH in the current population. Dogs did have higher UCCr results in urine collected IH compared to AH, suggesting the studied population did experience stress by the completion of their examination IH. Owners' assessment of their dogs' stress level agreed with the UCCr results, suggesting that the AH environment was less stressful than travel and stay at the veterinary hospital.
Potential stressors such as travel time, time spent in a potentially stressful hospital setting, and total time before urine collection were not correlated with changes in UPC or UCCr. We failed to find evidence that stress or increased UCCr causes a corresponding increase in UPC in healthy dogs. This relationship requires further investigation in dogs with glomerular proteinuria. We found that, in the case of true proteinuria in an otherwise healthy dog, stress alone may not fully explain the finding of proteinuria.