The Effect of Experimental Cystitis and Iatrogenic Blood Contamination on the Urine Protein/Creatinine Ratio in the Dog Rodney S. Bagley, DVM, Sharon A. Center, DVM, Robert M. Lewis, DVM, PhD, Sang Shin, DVM, Susan A. Dougherty, DVM, John F. Randolph, DVM, and Hollis Erb, DVM, PhD
The effect of experimentally induced cystitis and iatrogenic blood contamination on the urine protein/ creatinine ratio (U P/C) was evaluated in 17 dogs. Before they were included in the study, all dogs were judged to be healthy on the basis of physical examination, serum concentrations of urea nitrogen and creatinine, complete urinalysis, and a U P/C less than 0.4. A single urine sample was contaminated with increasing quantities of canine fresh whole blood (PCV = 42%;total protein = 6.2 g/dl). When added blood was equal to or greater than 25%of the total urine sample volume, the U P/C exceeded 3.5, a finding consistent with nephrotic range proteinuria. When added blond was 10%of the total urine sample volume, the U P/C was less than 1.8. Eleven Beagles underwent routine laparotomy during which a cystotomy was done. The median U P/Cs on postoperative days 1 and 2 were significantly increased compared with preoperative values (P < 0.05); no U P/C exceeded 2.0. Renal biopsies performed on postoperative day 3 eliminated renal proteinuria as a source of urine protein. Five dogs had bacterial cystitis experimentally induced. At 72 and 96 hours after bacterial inoculation, the median U P/Cs were significantly increased (P< 0.05); individual values ranged from 1.5 to 40.8. Renal biopsies performed between 5 and 6 days after inoculation eliminated renal proteinuria as a source of urine protein. Cytologic evaluation of urine sediment in each group did not correlate with the magnitude of the increase in the U P/C. The U P/C significantly increased in each model of lower urinary tract inflammation. Apparently, an increased U P/C in a urine sample having cytologic evidence of inflammation or heavy blood contamination cannot be used to distinguish pathologic renal from postrenal proteinuria. (Journal of Veterinary Internal Medicine 1991; 566-70)
genital disease. In these circumstances, the value of the U P/C in predicting the presence of pathologic renal proteinuria is questionable due to its inability to distinguish renal from postrenal proteinuria. Recently, quantitative tests for urine protein and the U P/C have been laudedas useful screening tests for the detection of clinically important urinary diseases6 In this retrospective review, clinical diagnoses were garnered from medical records. Renal tissue was not histologically examined and, therefore, the presence or absence of renal disease was inferred from other information. Proteinuria was not detected in 4.4% (22/500)of the specimens in which clinically important numbers of leukocytes or bacteria were observed. When proteinuria was detected, the contribution from nonrenal sources was not discernable. It is not currently known to what extent lower urinary tract hemorrhage, inflammation, or infection influences
THE URINE PROTEIN/creatinine ratio (U P/C) is a useful means of estimating proteinuria in the When this ratio is used in dogs lacking inflammatory or active urine sediment, it can quantitatively predict the magnitude of daily renal protein 1 0 ~ sUnfortunately, .~ proteinuria is often detected in urine samples having evidence of inflammatory, infectious, or hemorrhagic uroFrom the Departments of Clinical Sciences (Bagley, Center, Dougherty, Erb, Randolph) and Pathology (Lewis), and the Diagnostic Laboratory (Shin), New York State College of Veterinary Medicine, Cornell University, Ithaca, New York. Supported in part by the Alumni Unrestricted Gift Funds, Alumni Association of the New York State College of Veterinary Medicine, Ithaca, New York. Reprint requests: Rodney S. Bagley, DVM, North Carolina State University, College of Veterinary Medicine, Department of Companion Animal and Special Species Medicine, 4700 Hillsborough Street, Raleigh, NC 27606.
66
Vol. 5
. NO.2, 1991
URINE PROTEIN/CREATININE RATIO IN THE DOG
the U P/C. The three-part study described here was conducted to investigate the influence of each of these variables on the U P/C in the dog. In addition, the association between the magnitude of hematuria and pyuria (cells per high power field [HPF]) and the U P/C was explored to determine whether or npt a significant correlation exists that can be used to predict changes in the U P/C. Materials and Methods
Study A A specimen of urine from a healthy adult male dog with
a U P/C less than 0.4 was contaminated with canine fresh whole blood (total protein, 6.2 g/dl; PCV, 42%). Blood was added to obtain a final urine sample that was lo%, 25%, 50%, or 75% total volume of fresh whole blood; in each case, urine sediment contained too-numerous-to-count (TNTC) red blood cells per HPF (magnification, X400). Urine sediment was prepared by centrifuging urine in conical tipped tubes for 10 minutes at approximately 300 G, decanting the supernatant, and suspending the sediment in less than 0.2 ml of residual urine. Dilutions of blood and urine were arbitrarily selected to simulate grossly evident hematuria and maximal protein contribution from iatrogenic hemorrhage. These samples were mixed and evaluated within 2 hours. Mean U P/Cs were calculated from duplicate measurements made on the supernatant of each sample. Urine protein was measured using trichloracetic acid precipitation.' Urine creatinine was determined using the alkaline picrate method.' Study B
Eleven female Beagles were assessed as normal on the basis of physical examination, serum urea nitrogen and creatinine concentrations, and complete urinalysis (specific gravity, dipstick, and centrifuged sediment examination). A grading scale was established to represent the degree of hematuria and pyuria using RBCs and WBCs per HPF (Table 1). All dogs initially had U P/Cs less than 0.4. Each dog underwent general inhalant anesthesia, routine laparotomy, and cystotomy. The bladder was incised either dorsally or ventrally and was closed with a TABLEI. Grading Scale for the Magnitude of Hematuria or Pyuria in Cells/HPF Grade
HPF: high power field.
67
double inverting pattern (Lembert overlying a Cushing's) with 3/0 catgut suture material. No other urinary tract structures were manipulated. The abdominal incision was closed routinely. All dogs recovered uneventfully. Each dog was monitored twice daily for signs of systemic illness, fever, and pollakiuria. Urine samples were collected on postoperative days 1 and 2 by urethral catheterization. The samples were evaluated by complete urinalysis and quantitative urine protein and creatinine concentrations. Urine protein and creatinine were measured as in study A. Renal tissue was collected on the third postoperative day by laparotomy wedge biopsy. Tissue was preserved for light microscopy in 10%buffered formalin, embedded in paraffin, and stained with hematoxylin and eosin. Tissues were reviewed by a single pathologist without prior knowledge of the study design.
Study C Five Beagles (3 female, 2 male) were judged as normal on the basis of physical examination, serum urea nitrogen and creatinine concentrations, and complete urinalysis. All dogs had initial U P/Cs less than 0.4. Experimental cystitis was induced using a previously established protoc0L9 After the dogs were anesthetized with a thiobarbiturate, 10 mls of 0.1% sulfosalicylic acid in ethanol solution were instilled into the urinary bladder. This solution was retained for 10 minutes and then removed. The urinary bladder was subsequently rinsed with three successive 50 ml aliquots of sterile saline solution before instilling a 5-ml saline suspension of Escherichia cofi containing 2 X lo5 organisms/ml. One Jog had sulfosalicylic acid solution only instilled into the urinary bladder. Each dog was monitored twice daily for signs of systemic illness, fever, or pollakiuria. Urine samples were collected by urethral catheterization or cystocentesis on the third and fourth days after inoculation. Samples were evaluated using complete urinalysis, including sediment examination and U P/C determination on the supernatant. Quantitative urine protein and creatinine concentrations were determined by the methods described in study A. Urine was collected for aerobic bacterial culture on the third day after inoculation to document persistent infection. At the end of the study, on postoperative days 5 and 6, renal tissue was collected and prepared for light microscopy as in study B.
Red or White Blood Cells/HPF
Statistical Analysis
100
The data were analyzed using nonparametric procedures. The Wilcoxon signed rank test was used to test for significant (P < 0.05) differences between the median precystotomy and postcystotomy U P/Cs and between the median preinoculation and postinoculation U P/ Cs.'* Spearman's rank correlation was used to investi-
68
TABLE2. Effect of Cystotomy and Experimental Infectious/ Inflammatory Cystitis on the U P/C
gate whether urine red blood cell (RBC) and white blood cell (WBC) counts per HPF correlated with the magnitude of increase in the U P/C.'' Results
Cystotomy* Pre
Post I
Post 2
Pre
Post 3
Post4
1 2 3 4 5 6 7
0.02 0.03 0.00
0.50
1.01 1.00
0.10 0.10 0.10 0.20 0.30
25.4 14.6 20.3 6.5 I .5
40.8 12.0 4.2 2.0
0.10
14.6
9.7
9 10 I1
0.11 0.00 0.16 0.08 0.15 0.18 0.25
1.05 0.44 0.86 0.12 0.57 0.52 1.01 0.44 0.52 1.80
Median
0.1 1
0.52
All urine samples showed evidence of gross hematuria with RBC counts per HPF TNTC in the urine sediment. An average U P/C was calculated from the duplicate measurements made on each urine specimen. Urine P/ Cs ranged from 1.53 in the 10% blood contaminated sample to 30.6 in the 75% blood contaminated sample (Fig. 1). Study B
No dog developed systemic signs of infection, although several had mildly increased rectal temperatures. Most dogs were pollakiuric, which necessitated urine collection by catheterization. The median U P/C was significantly increased on postoperative days 1 and 2 ( P < 0.05) (Table 2). No dog's U P/C was greater than 2.0. Urine sediments revealed increases in both RBC and WBC counts, with cells ranging from 5 to greater than 50 per HPF. Renal histopathology was normal in each dog. Study C
All dogs developed gross hematuria, stranguria, and pollakiuria. The median U P/C was significantly increased on days 3 and 4 after inoculation ( P < 0.05).The U P/C ranged from 1.5 to 40.8 (Table 2). Aerobic bacterial urine cultures on day 3 after inoculation revealed greater than lo5E. coli organisms/ml, confirming persistent urinary tract infection. Urine sediments revealed bacteriuria, pyuria ( 5 to >50 WBCs/HPF), and hematuria ( 5 to >50 RBCs/HPF). Renal histopathology was normal in each dog. When results of studies B and C were combined, no correlation could be made between the magnitude of in-
W
z z
1.65 1.85
35 30 25
,A
15
10-
i
1.02 1.02 0.77
1.12 1.05 1.02
U P / C urine protein to creatinine ratio; Pre: precystotomy or preinoculation; Post 1: postcystotomy day I ; Post 2: postcystotomy day 2; Post 3: postinoculation day 3; Post 4: postinoculation day 4. * Figures refer to U P/C.
crease in either the urine RBC counts ( P > 0.1) (Fig. 2) or the urine WBC counts per HPF ( P > 0.1) (Fig. 3) and the magnitude of increase in the U P/C. Discussion
The U P/C is useful in estimating quantitative proteinuria in a single urine sample in dogs whose only urinalysis abnormalities are proteinuria and occasional casts.'-' The method of urine collection does not influence the U P/C ifthe sample is cytologicallyn ~ r m a l . ' ,Urine ~ , ~ specimens in this study were collected by both cystocentesis and urethral catheterization. Although it would have been preferable to collect all urine specimens using the same technique, this could not be accomplished owing to dysuria, pollakiuria, and behavior of individual dogs. In clinical veterinary practice, proteinuria is commonly detected in urine having cytologic evidence of urinary tract inflammation. In children with lower urinary
d
z
z
K
o
2&
2o 15 10
(L
a L
5-
25
W
w
W
5
(L
/A
9.7
0.80
30
z
3
1.78
Q +
8.20
cc
0.1 1
8
40
\
6 E
Experimental Cystitis*
Dog
Study A
5
Journal of Veterinary Internal Medicine
BAGLEY ET AL.
3
0
Vol. 5
. NO. 2, 1991
URINE PROTEIN/CREA TlNlNE RATIO IN THE DOG
3 0 ~
d
25 20
a
~
1510-
5 0 0
a
* 1
a
3 2
3
mild
5
4
severe
URINE SEDIMENT: WBC
FIG. 3. Grade of urine WBC counts (Table 1) versus U P/C.
1
tract disease, the U P/C can exceed values associated with nephrotic range proteinuria. l 1 This finding invalidates the use of the U P/C in humans as an indicator of renal proteinuria when urinary tract inflammation is detected on urine sediment examination. Our studies found similar results in the dog. The trichloroacetic acid precipitation method of protein quantification used in this study is slightly less sensitive for detecting low urine protein concentrations than techniques using Coomassie Brilliant Blue or Ponceau-S dye.” Nevertheless, our results indicated that substantial and confusing proteinuria develops in dogs with lower urinary tract inflammation. Our study demonstrated that gross blood contamination of a urine sample increasesthe U P/C. It was interesting that, despite the gross hematuria evident in the 10% blood contaminated sample (RBC per HPF TNTC), the U P/C only increased to 1.8. It is possible, therefore, that urine contamination with a small amount of blood (i.e., 5-20 RBC/HPF) may not substantially affect the diagnostic usefulness of the U P/C observed in severe protein-losing nephropathies, but this remains to be proven. Based on our observations, we speculate that a nephrotic range U P/C in such a minimally contaminated sample would suggest serious renal proteinuria. When greater than 25% of an initially normal urine sample was composed of whole blood, the U P/C increased to values observed with nephrotic range proteinuria, a finding that invalidated application of the ratio. The models used in this study were chosen to simulate iatrogenic hematuria, hematuria associated with lower urinary tract inflammation due to mechanical/surgical trauma, and septic cystitis. Renal proteinuria was ruled out in each dog on the basis of histopathologic examination of the kidneys. Transient changes in glomerular filtration rate were not evaluated. Induced increases in the urine protein concentration, therefore, were assumed to be postrenal in origin. In dogs undergoing cystitis induction, the magnitude of increase in the U P/C varied
69
greatly among individual animals and may have been associated with the experimental methods used. We were unable to separate the effect of mucosal irritation resulting from intravesical sulfosalicylic acid instillation from the effect of infectious cystitis. The only dog that received sulfosalicylic acid solution but not bacteria developed an increased U P/C on days 3 and 4 after inoculation, which was comparable to that invoked by septic cystitis. The sulfosalicylic acid solution was used to remove the protective glycosaminoglycans layer from the bladder wall and is a known mucosal irritant.’ Irritation created by this solution probably contributed to the lower urinary tract inflammation. It would have been preferable to evaluate urine from dogs after chronic urinary tract infections (two weeks) had been established, but ethical and financial constraints prohibited such a study design. The amount of protein exuded from the lower urinary tract is influenced by the type and severity of bladder insult. This was evidenced by the marked difference in the magnitude of the U P/C in the two models of lower urinary tract inflammation used. The temporal relationship of sample collection and cystitis induction also may have influenced our results. These models evaluated acute bladder injury. Chronic or established lower urinary tract infection or inflammation may be associated with less protein exudation as compared with an acute process. It is possible that differences in individual dogs, such as bleeding tendencies resulting in excessive hematuna, may have influenced the magnitude of proteinuria. After cystotomy or cystitis induction, all the dogs in our studies would have been classified as having pathologic proteinuria based on their U P/C. Four of the five dogs in the second study would have been classified as having nephrotic range renal proteinuria (U P/C > 3.5) if results from the urine sediment examination had not been considered. Unfortunately, the magnitude of hematuria or pyuria estimated on the basis of cell count per HPF did not correlate with the magnitude of increase in the U P/C and, therefore, we could not predict changes in the U P/C on the basis of urine sediment examination. We concluded that, in urine grossly contaminated with blood, an increased U P/C cannot be used to diagnose renal proteinuria. Similarly, an increased U P/C in a urine sample with cytologic evidence of inflammation or infection cannot be used to diagnose renal proteinuria. Any inflammatory or hemorrhagic disorder of the urogenital system may significantly alter the U P/C, invalidating its use for the accurate estimation of renal protein excretion. It seems possible, however, that, despite minor blood contamination of a urine specimen during collection (sediment showing 5-20 RBCs per HPF), the U P/C may still be used to indicate serious nephrotic range proteinuria. Further studies are needed to substantiate this contention.
70
BAGLEY ET AL.
References 1. White JV, Olivier NB, Reimann K, et al. Use of the protein-to-
2.
3.
4. 5.
6.
creatinine ratio in a single urine specimen for quantitative estimation of canine proteinuria. J Am Vet Med Assoc 1984; 1851882485. McCaw DL, Knapp DW, Hewett JE. Effect of collection time and exercise restriction on the prediction of urine protein excretion, using urine proteinlcreatinine ratio in dogs. Am J Vet Res 1985; 46:1665-1669. Grauer GF, Thomas CB, Eicker SW. Estimation of quantitative proteinuria in the dog, using the urine protein-to-creatinine ratio from a random, voided sample. Am J Vet Res 1985; 46:2116-2119. Jergens AE, McCaw DL, Hewett JE. Effects of collection time and food consumption on the urine protein/creatinine ratio in the dog. Am J Vet Res 1987; 48:1106-1109. Center SA, Wilkinson E, Smith CA, et al. 24-Hour urine protein/ creatinine ratio in dogs with protein-losing nephropathies. J Am Vet Med Assoc 1985; 187:820-824. Fettman MJ. Comparison of urinary protein concentration and
7. 8.
9.
10. 11.
12.
Journal of Veterinary Internal Medicine
protein/creatinine ratio vs routine microscopy in urinalysis of dogs: 500 cases (1987-1988). J Am Vet Med Assoc-1989; 195~972-976. Henry RJ, Sobel C, Segalove M. Turbidimetric determination of proteins with sulfosalicylic and trichloracetic acids. Proc SOC Exp Biol Med 1956; 92:748-751. Bonsnes RW, Taussky HH. On the colorimetric determination of creatinine by the Jaffe reaction. J Biol Chem 1945; 158581591. Roge1.sKS, Lees GE, Simpson RB. Effectsofsingle-dose and threeday trimethoprim-sulfadiazine and amikacin treatment of induced Escherichia coli urinary tract infections in dogs. Am J Vet Res 1988; 49:345-349. Conover WJ. Practical Nonparametric Statistics, 2nd ed. New York: John Wiley and Sons, 1980; 2 13-243. Doehring E, Ehrich JHH, Vester U, et al. Proteinuria, hematuria, and leukocyturia in children with mixed urinary and intestinal schistosomiasis. Kid Internal 1985; 28:520-525. Dilena BA, Penberthy LA, Fraser CG. Six methods for determining urinary protein compared. Clin Chem 1988; 29:553-557.
5th Annual Meeting of the European Society for Veterinary Neurology October 2, 1991 Vienna, Austria
Those interested in presenting a paper at the meeting should submit an abstract to Dr. Andrea Tipold, Institut for Tierneurologie, Bremgartenstrasse 109a, CH-300 1, Bern, Switzerland. Deadline: April 30, 1991.
The effect of experimentally induced cystitis and iatrogenic blood contamination on the urine protein/ creatinine ratio (U P/C) was evaluated in 17 dogs. Before they were included in the study, all dogs were judged to be healthy on the basis of physical examination, serum concentrations of urea nitrogen and creatinine, complete urinalysis, and a U P/C less than 0.4. A single urine sample was contaminated with increasing quantities of canine fresh whole blood (PCV = 42%;total protein = 6.2 g/dl). When added blood was equal to or greater than 25%of the total urine sample volume, the U P/C exceeded 3.5, a finding consistent with nephrotic range proteinuria. When added blond was 10%of the total urine sample volume, the U P/C was less than 1.8. Eleven Beagles underwent routine laparotomy during which a cystotomy was done. The median U P/Cs on postoperative days 1 and 2 were significantly increased compared with preoperative values (P < 0.05); no U P/C exceeded 2.0. Renal biopsies performed on postoperative day 3 eliminated renal proteinuria as a source of urine protein. Five dogs had bacterial cystitis experimentally induced. At 72 and 96 hours after bacterial inoculation, the median U P/Cs were significantly increased (P< 0.05); individual values ranged from 1.5 to 40.8. Renal biopsies performed between 5 and 6 days after inoculation eliminated renal proteinuria as a source of urine protein. Cytologic evaluation of urine sediment in each group did not correlate with the magnitude of the increase in the U P/C. The U P/C significantly increased in each model of lower urinary tract inflammation. Apparently, an increased U P/C in a urine sample having cytologic evidence of inflammation or heavy blood contamination cannot be used to distinguish pathologic renal from postrenal proteinuria. (Journal of Veterinary Internal Medicine 1991; 566-70)
genital disease. In these circumstances, the value of the U P/C in predicting the presence of pathologic renal proteinuria is questionable due to its inability to distinguish renal from postrenal proteinuria. Recently, quantitative tests for urine protein and the U P/C have been laudedas useful screening tests for the detection of clinically important urinary diseases6 In this retrospective review, clinical diagnoses were garnered from medical records. Renal tissue was not histologically examined and, therefore, the presence or absence of renal disease was inferred from other information. Proteinuria was not detected in 4.4% (22/500)of the specimens in which clinically important numbers of leukocytes or bacteria were observed. When proteinuria was detected, the contribution from nonrenal sources was not discernable. It is not currently known to what extent lower urinary tract hemorrhage, inflammation, or infection influences
THE URINE PROTEIN/creatinine ratio (U P/C) is a useful means of estimating proteinuria in the When this ratio is used in dogs lacking inflammatory or active urine sediment, it can quantitatively predict the magnitude of daily renal protein 1 0 ~ sUnfortunately, .~ proteinuria is often detected in urine samples having evidence of inflammatory, infectious, or hemorrhagic uroFrom the Departments of Clinical Sciences (Bagley, Center, Dougherty, Erb, Randolph) and Pathology (Lewis), and the Diagnostic Laboratory (Shin), New York State College of Veterinary Medicine, Cornell University, Ithaca, New York. Supported in part by the Alumni Unrestricted Gift Funds, Alumni Association of the New York State College of Veterinary Medicine, Ithaca, New York. Reprint requests: Rodney S. Bagley, DVM, North Carolina State University, College of Veterinary Medicine, Department of Companion Animal and Special Species Medicine, 4700 Hillsborough Street, Raleigh, NC 27606.
66
Vol. 5
. NO.2, 1991
URINE PROTEIN/CREATININE RATIO IN THE DOG
the U P/C. The three-part study described here was conducted to investigate the influence of each of these variables on the U P/C in the dog. In addition, the association between the magnitude of hematuria and pyuria (cells per high power field [HPF]) and the U P/C was explored to determine whether or npt a significant correlation exists that can be used to predict changes in the U P/C. Materials and Methods
Study A A specimen of urine from a healthy adult male dog with
a U P/C less than 0.4 was contaminated with canine fresh whole blood (total protein, 6.2 g/dl; PCV, 42%). Blood was added to obtain a final urine sample that was lo%, 25%, 50%, or 75% total volume of fresh whole blood; in each case, urine sediment contained too-numerous-to-count (TNTC) red blood cells per HPF (magnification, X400). Urine sediment was prepared by centrifuging urine in conical tipped tubes for 10 minutes at approximately 300 G, decanting the supernatant, and suspending the sediment in less than 0.2 ml of residual urine. Dilutions of blood and urine were arbitrarily selected to simulate grossly evident hematuria and maximal protein contribution from iatrogenic hemorrhage. These samples were mixed and evaluated within 2 hours. Mean U P/Cs were calculated from duplicate measurements made on the supernatant of each sample. Urine protein was measured using trichloracetic acid precipitation.' Urine creatinine was determined using the alkaline picrate method.' Study B
Eleven female Beagles were assessed as normal on the basis of physical examination, serum urea nitrogen and creatinine concentrations, and complete urinalysis (specific gravity, dipstick, and centrifuged sediment examination). A grading scale was established to represent the degree of hematuria and pyuria using RBCs and WBCs per HPF (Table 1). All dogs initially had U P/Cs less than 0.4. Each dog underwent general inhalant anesthesia, routine laparotomy, and cystotomy. The bladder was incised either dorsally or ventrally and was closed with a TABLEI. Grading Scale for the Magnitude of Hematuria or Pyuria in Cells/HPF Grade
HPF: high power field.
67
double inverting pattern (Lembert overlying a Cushing's) with 3/0 catgut suture material. No other urinary tract structures were manipulated. The abdominal incision was closed routinely. All dogs recovered uneventfully. Each dog was monitored twice daily for signs of systemic illness, fever, and pollakiuria. Urine samples were collected on postoperative days 1 and 2 by urethral catheterization. The samples were evaluated by complete urinalysis and quantitative urine protein and creatinine concentrations. Urine protein and creatinine were measured as in study A. Renal tissue was collected on the third postoperative day by laparotomy wedge biopsy. Tissue was preserved for light microscopy in 10%buffered formalin, embedded in paraffin, and stained with hematoxylin and eosin. Tissues were reviewed by a single pathologist without prior knowledge of the study design.
Study C Five Beagles (3 female, 2 male) were judged as normal on the basis of physical examination, serum urea nitrogen and creatinine concentrations, and complete urinalysis. All dogs had initial U P/Cs less than 0.4. Experimental cystitis was induced using a previously established protoc0L9 After the dogs were anesthetized with a thiobarbiturate, 10 mls of 0.1% sulfosalicylic acid in ethanol solution were instilled into the urinary bladder. This solution was retained for 10 minutes and then removed. The urinary bladder was subsequently rinsed with three successive 50 ml aliquots of sterile saline solution before instilling a 5-ml saline suspension of Escherichia cofi containing 2 X lo5 organisms/ml. One Jog had sulfosalicylic acid solution only instilled into the urinary bladder. Each dog was monitored twice daily for signs of systemic illness, fever, or pollakiuria. Urine samples were collected by urethral catheterization or cystocentesis on the third and fourth days after inoculation. Samples were evaluated using complete urinalysis, including sediment examination and U P/C determination on the supernatant. Quantitative urine protein and creatinine concentrations were determined by the methods described in study A. Urine was collected for aerobic bacterial culture on the third day after inoculation to document persistent infection. At the end of the study, on postoperative days 5 and 6, renal tissue was collected and prepared for light microscopy as in study B.
Red or White Blood Cells/HPF
Statistical Analysis
100
The data were analyzed using nonparametric procedures. The Wilcoxon signed rank test was used to test for significant (P < 0.05) differences between the median precystotomy and postcystotomy U P/Cs and between the median preinoculation and postinoculation U P/ Cs.'* Spearman's rank correlation was used to investi-
68
TABLE2. Effect of Cystotomy and Experimental Infectious/ Inflammatory Cystitis on the U P/C
gate whether urine red blood cell (RBC) and white blood cell (WBC) counts per HPF correlated with the magnitude of increase in the U P/C.'' Results
Cystotomy* Pre
Post I
Post 2
Pre
Post 3
Post4
1 2 3 4 5 6 7
0.02 0.03 0.00
0.50
1.01 1.00
0.10 0.10 0.10 0.20 0.30
25.4 14.6 20.3 6.5 I .5
40.8 12.0 4.2 2.0
0.10
14.6
9.7
9 10 I1
0.11 0.00 0.16 0.08 0.15 0.18 0.25
1.05 0.44 0.86 0.12 0.57 0.52 1.01 0.44 0.52 1.80
Median
0.1 1
0.52
All urine samples showed evidence of gross hematuria with RBC counts per HPF TNTC in the urine sediment. An average U P/C was calculated from the duplicate measurements made on each urine specimen. Urine P/ Cs ranged from 1.53 in the 10% blood contaminated sample to 30.6 in the 75% blood contaminated sample (Fig. 1). Study B
No dog developed systemic signs of infection, although several had mildly increased rectal temperatures. Most dogs were pollakiuric, which necessitated urine collection by catheterization. The median U P/C was significantly increased on postoperative days 1 and 2 ( P < 0.05) (Table 2). No dog's U P/C was greater than 2.0. Urine sediments revealed increases in both RBC and WBC counts, with cells ranging from 5 to greater than 50 per HPF. Renal histopathology was normal in each dog. Study C
All dogs developed gross hematuria, stranguria, and pollakiuria. The median U P/C was significantly increased on days 3 and 4 after inoculation ( P < 0.05).The U P/C ranged from 1.5 to 40.8 (Table 2). Aerobic bacterial urine cultures on day 3 after inoculation revealed greater than lo5E. coli organisms/ml, confirming persistent urinary tract infection. Urine sediments revealed bacteriuria, pyuria ( 5 to >50 WBCs/HPF), and hematuria ( 5 to >50 RBCs/HPF). Renal histopathology was normal in each dog. When results of studies B and C were combined, no correlation could be made between the magnitude of in-
W
z z
1.65 1.85
35 30 25
,A
15
10-
i
1.02 1.02 0.77
1.12 1.05 1.02
U P / C urine protein to creatinine ratio; Pre: precystotomy or preinoculation; Post 1: postcystotomy day I ; Post 2: postcystotomy day 2; Post 3: postinoculation day 3; Post 4: postinoculation day 4. * Figures refer to U P/C.
crease in either the urine RBC counts ( P > 0.1) (Fig. 2) or the urine WBC counts per HPF ( P > 0.1) (Fig. 3) and the magnitude of increase in the U P/C. Discussion
The U P/C is useful in estimating quantitative proteinuria in a single urine sample in dogs whose only urinalysis abnormalities are proteinuria and occasional casts.'-' The method of urine collection does not influence the U P/C ifthe sample is cytologicallyn ~ r m a l . ' ,Urine ~ , ~ specimens in this study were collected by both cystocentesis and urethral catheterization. Although it would have been preferable to collect all urine specimens using the same technique, this could not be accomplished owing to dysuria, pollakiuria, and behavior of individual dogs. In clinical veterinary practice, proteinuria is commonly detected in urine having cytologic evidence of urinary tract inflammation. In children with lower urinary
d
z
z
K
o
2&
2o 15 10
(L
a L
5-
25
W
w
W
5
(L
/A
9.7
0.80
30
z
3
1.78
Q +
8.20
cc
0.1 1
8
40
\
6 E
Experimental Cystitis*
Dog
Study A
5
Journal of Veterinary Internal Medicine
BAGLEY ET AL.
3
0
Vol. 5
. NO. 2, 1991
URINE PROTEIN/CREA TlNlNE RATIO IN THE DOG
3 0 ~
d
25 20
a
~
1510-
5 0 0
a
* 1
a
3 2
3
mild
5
4
severe
URINE SEDIMENT: WBC
FIG. 3. Grade of urine WBC counts (Table 1) versus U P/C.
1
tract disease, the U P/C can exceed values associated with nephrotic range proteinuria. l 1 This finding invalidates the use of the U P/C in humans as an indicator of renal proteinuria when urinary tract inflammation is detected on urine sediment examination. Our studies found similar results in the dog. The trichloroacetic acid precipitation method of protein quantification used in this study is slightly less sensitive for detecting low urine protein concentrations than techniques using Coomassie Brilliant Blue or Ponceau-S dye.” Nevertheless, our results indicated that substantial and confusing proteinuria develops in dogs with lower urinary tract inflammation. Our study demonstrated that gross blood contamination of a urine sample increasesthe U P/C. It was interesting that, despite the gross hematuria evident in the 10% blood contaminated sample (RBC per HPF TNTC), the U P/C only increased to 1.8. It is possible, therefore, that urine contamination with a small amount of blood (i.e., 5-20 RBC/HPF) may not substantially affect the diagnostic usefulness of the U P/C observed in severe protein-losing nephropathies, but this remains to be proven. Based on our observations, we speculate that a nephrotic range U P/C in such a minimally contaminated sample would suggest serious renal proteinuria. When greater than 25% of an initially normal urine sample was composed of whole blood, the U P/C increased to values observed with nephrotic range proteinuria, a finding that invalidated application of the ratio. The models used in this study were chosen to simulate iatrogenic hematuria, hematuria associated with lower urinary tract inflammation due to mechanical/surgical trauma, and septic cystitis. Renal proteinuria was ruled out in each dog on the basis of histopathologic examination of the kidneys. Transient changes in glomerular filtration rate were not evaluated. Induced increases in the urine protein concentration, therefore, were assumed to be postrenal in origin. In dogs undergoing cystitis induction, the magnitude of increase in the U P/C varied
69
greatly among individual animals and may have been associated with the experimental methods used. We were unable to separate the effect of mucosal irritation resulting from intravesical sulfosalicylic acid instillation from the effect of infectious cystitis. The only dog that received sulfosalicylic acid solution but not bacteria developed an increased U P/C on days 3 and 4 after inoculation, which was comparable to that invoked by septic cystitis. The sulfosalicylic acid solution was used to remove the protective glycosaminoglycans layer from the bladder wall and is a known mucosal irritant.’ Irritation created by this solution probably contributed to the lower urinary tract inflammation. It would have been preferable to evaluate urine from dogs after chronic urinary tract infections (two weeks) had been established, but ethical and financial constraints prohibited such a study design. The amount of protein exuded from the lower urinary tract is influenced by the type and severity of bladder insult. This was evidenced by the marked difference in the magnitude of the U P/C in the two models of lower urinary tract inflammation used. The temporal relationship of sample collection and cystitis induction also may have influenced our results. These models evaluated acute bladder injury. Chronic or established lower urinary tract infection or inflammation may be associated with less protein exudation as compared with an acute process. It is possible that differences in individual dogs, such as bleeding tendencies resulting in excessive hematuna, may have influenced the magnitude of proteinuria. After cystotomy or cystitis induction, all the dogs in our studies would have been classified as having pathologic proteinuria based on their U P/C. Four of the five dogs in the second study would have been classified as having nephrotic range renal proteinuria (U P/C > 3.5) if results from the urine sediment examination had not been considered. Unfortunately, the magnitude of hematuria or pyuria estimated on the basis of cell count per HPF did not correlate with the magnitude of increase in the U P/C and, therefore, we could not predict changes in the U P/C on the basis of urine sediment examination. We concluded that, in urine grossly contaminated with blood, an increased U P/C cannot be used to diagnose renal proteinuria. Similarly, an increased U P/C in a urine sample with cytologic evidence of inflammation or infection cannot be used to diagnose renal proteinuria. Any inflammatory or hemorrhagic disorder of the urogenital system may significantly alter the U P/C, invalidating its use for the accurate estimation of renal protein excretion. It seems possible, however, that, despite minor blood contamination of a urine specimen during collection (sediment showing 5-20 RBCs per HPF), the U P/C may still be used to indicate serious nephrotic range proteinuria. Further studies are needed to substantiate this contention.
70
BAGLEY ET AL.
References 1. White JV, Olivier NB, Reimann K, et al. Use of the protein-to-
2.
3.
4. 5.
6.
creatinine ratio in a single urine specimen for quantitative estimation of canine proteinuria. J Am Vet Med Assoc 1984; 1851882485. McCaw DL, Knapp DW, Hewett JE. Effect of collection time and exercise restriction on the prediction of urine protein excretion, using urine proteinlcreatinine ratio in dogs. Am J Vet Res 1985; 46:1665-1669. Grauer GF, Thomas CB, Eicker SW. Estimation of quantitative proteinuria in the dog, using the urine protein-to-creatinine ratio from a random, voided sample. Am J Vet Res 1985; 46:2116-2119. Jergens AE, McCaw DL, Hewett JE. Effects of collection time and food consumption on the urine protein/creatinine ratio in the dog. Am J Vet Res 1987; 48:1106-1109. Center SA, Wilkinson E, Smith CA, et al. 24-Hour urine protein/ creatinine ratio in dogs with protein-losing nephropathies. J Am Vet Med Assoc 1985; 187:820-824. Fettman MJ. Comparison of urinary protein concentration and
7. 8.
9.
10. 11.
12.
Journal of Veterinary Internal Medicine
protein/creatinine ratio vs routine microscopy in urinalysis of dogs: 500 cases (1987-1988). J Am Vet Med Assoc-1989; 195~972-976. Henry RJ, Sobel C, Segalove M. Turbidimetric determination of proteins with sulfosalicylic and trichloracetic acids. Proc SOC Exp Biol Med 1956; 92:748-751. Bonsnes RW, Taussky HH. On the colorimetric determination of creatinine by the Jaffe reaction. J Biol Chem 1945; 158581591. Roge1.sKS, Lees GE, Simpson RB. Effectsofsingle-dose and threeday trimethoprim-sulfadiazine and amikacin treatment of induced Escherichia coli urinary tract infections in dogs. Am J Vet Res 1988; 49:345-349. Conover WJ. Practical Nonparametric Statistics, 2nd ed. New York: John Wiley and Sons, 1980; 2 13-243. Doehring E, Ehrich JHH, Vester U, et al. Proteinuria, hematuria, and leukocyturia in children with mixed urinary and intestinal schistosomiasis. Kid Internal 1985; 28:520-525. Dilena BA, Penberthy LA, Fraser CG. Six methods for determining urinary protein compared. Clin Chem 1988; 29:553-557.
5th Annual Meeting of the European Society for Veterinary Neurology October 2, 1991 Vienna, Austria
Those interested in presenting a paper at the meeting should submit an abstract to Dr. Andrea Tipold, Institut for Tierneurologie, Bremgartenstrasse 109a, CH-300 1, Bern, Switzerland. Deadline: April 30, 1991.
