Correlation of gross urine color with diagnostic findings in male cats with naturally occurring urethral obstruction

Abstract

Seventy-five male cats with urethral obstruction were prospectively enrolled to evaluate gross urine color at urinary catheter placement for correlation with diagnostic findings. Cats with darker red urine were more likely to be azotemic (serum creatinine concentration >2.0 mg/dl [177 µmol/l]), and urine color correlated well with serum creatinine and serum potassium concentrations. Darker urine color was negatively correlated with urine specific gravity. Urine color was not associated with the presence or absence of lower urinary tract stones on radiographs or ultrasound. Cats with darker red urine at the time of urinary catheter placement are likely to have more significant metabolic derangements and may require more aggressive supportive care.

Introduction

Urethral obstruction (UO) is a common problem in male cats and requires rapid emergency care owing to the potentially life-threatening metabolic derangements that occur with prolonged obstruction.^1,2 The most common causes of UO include urethral plugs and uroliths, although in many cats the cause cannot be determined and the condition is considered to be due to idiopathic cystitis.^3–6 Neoplasia and bacterial infection are much less common causes of UO in cats.^3,5,6

While most cats with UO have microscopic hematuria on urinalysis,^2,7 gross urine color can vary widely at the time of relief of the obstruction. Some affected cats have relatively normal-colored urine, whereas others have urine that is dark red. To our knowledge, there are no studies evaluating the relationship of gross urine color with clinicopathological variables and cause of UO. Many non-obstructed cats with feline idiopathic cystitis have gross hematuria,^7–9 so cats obstructing due to idiopathic cystitis may have darker urine than cats obstructing due to urolithiasis. However, uroliths themselves can also result in hematuria.^9,10 The severity of metabolic derangement may also correlate with urine color, as the bladder mucosa becomes more damaged with prolonged obstruction.^11,12

The goal of this study was to correlate gross urine color with other diagnostic findings in male cats with naturally occurring UO.

Materials and methods

Case selection

Male cats presenting to the emergency service at Ocean State Veterinary Specialists (OSVS) for naturally occurring UO were prospectively enrolled between October 2012 and June 2013. Patients were eligible for inclusion if the initial urinary catheter was placed through the OSVS emergency service, they had an admission chemistry panel and urinalysis, and if they had both a lateral survey abdominal radiograph and a lower urinary tract ultrasound examination.

Urine color and laboratory evaluation

A color chart with five colors (yellow = 1, peach = 2, pink = 3, red = 4 and burgundy = 5) was developed to standardize determination of urine color and was available in the emergency room for immediate reference (Figure 1). Urine color was recorded on each patient at the time of urinary catheter placement, using a sample obtained at the midpoint of the initial bladder emptying to minimize contamination with saline flush that may have been used during catheter placement. No cat had a decompressive cystocentesis performed prior to urinary catheter placement. All cats had initial laboratory evaluation to include serum creatinine and serum potassium concentrations. Urinalysis, including refractometric specific gravity, dipstick and microscopic sediment evaluation, was performed on the same sample used for urine color evaluation. Uncentrifuged urine was used for the urine dipstick and specific gravity determination.

Figure 1

Urine color chart

Imaging

A lateral survey radiograph was obtained in all cats using digital radiography within 24 h of placement of the urinary catheter. The urinary bladder was lavaged with saline solution through the urinary catheter prior to the radiograph in all cats (the quantity of flush was at the discretion of the clinician). Radiographs included the bladder and the entire penile urethra. Sonographic evaluation of the urinary bladder and proximal urethra was performed at the earliest radiologist availability (within 48 h of presentation), while the urinary catheter was still in place. Prior to the ultrasound, the urinary catheter was temporarily occluded to allow sufficient bladder distension for evaluation. A single, board-certified veterinary radiologist interpreted all radiographs, and performed and interpreted all ultrasound examinations. The radiologist was not specifically blinded to the urine color. A report was generated for each imaging study, and cases were classified as lower urinary tract stones ‘present’ or ‘absent’ based on this report.

Statistical analysis

For the purposes of statistical analysis, urine color, serum creatinine concentration, serum potassium concentration, urine dipstick blood count, microscopic urine cell counts and urine specific gravity (USG) were treated as continuous variables. Counts within categories, such as presence or absence of lower urinary tract stones and azotemia, were treated as discrete variables. Cats with a serum creatinine concentration ⩽2.0 mg/dl (177 µmol/l) were considered non-azotemic, while cats with a serum creatinine concentration >2.0 mg/dl (177 µmol/l) were considered azotemic. When a range was given for cell count on urinalysis (eg, 21–50 white blood cells per high power field [HPF]), the mean was used. For statistical analysis, a USG above the level of detection on the refractometer was recorded as 1.050 for statistical analysis, and a serum potassium concentration above the level of detection of the chemistry analyzer was recorded as 9.0 mEq/l (9.0 mmol/l).

Statistical analyses were conducted using SPSS version 18. Logistic regression was used to model the association between a continuous variable and a discrete variable. Spearman’s rank correlation was used to evaluate the association between continuous variables. For all analyses, P ⩽0.05 was considered statistically significant.

Results

Seventy-five castrated male cats met the inclusion criteria and all were included in the study. The mean age was 5.5 years (range 1.0–14.0 years). Cats included 57 domestic shorthairs, nine domestic longhairs, six domestic mediumhairs, two Himalayans and one Maine Coon. Nine cats (12%) had a history of a previous UO.

The mean urine color was 2.5 (range 1.0–5.0). A summary of urine color distribution is presented in Table 1. The mean USG was 1.039 (range 1.010 to >1.050). The mean urine dipstick blood test was 183 cells/µl (range 0–200 cells/µl). Only one cat did not have detectable blood on urine dipstick, and 67 cats (89%) had maximal detectable blood on dipstick (200 cells/µl). On microscopic urine sediment examination, 22/75 cats (29%) had no crystals and 53/75 (71%) had crystals − 52 of these cats had struvite crystals and one cat had calcium oxalate crystals. The mean urine sediment red blood cell (RBC) count was 44/HPF (range 0 to >50 RBC/HPF). Seventy-four cats (99%) had microscopic hematuria (>3 RBC/HPF) and 62 cats (82%) had maximal microscopic hematuria (>50 RBC/HPF). The mean urine sediment white blood cell (WBC) count was 2.8/HPF (range 0–15 WBC/HPF). Sixteen cats (21%) had microscopic pyuria (>3 WBC/HPF). Two cats (3%) had microscopic bacteriuria.

Table 1

Gross urine color of cats with naturally occurring urethral obstruction

Urine color (number on chart)	Number of cats (n = 75)	Percentage of cats
Yellow (1)	30	40
Peach (2)	14	19
Pink (3)	2	3
Red (4)	21	28
Burgundy (5)	8	11

Thirty-five cats (47%) were azotemic at presentation. The mean presenting serum creatinine concentration was 3.8 mg/dl (336 µmol/l; range 0.7–18.2 mg/dl [62–1609 µmol/l]; reference interval 0.6–2.0 mg/dl [53–177 µmol/l]). The mean presenting serum potassium concentration was 4.7 mEq/l (4.7 mmol/l; range 2.8 to >9 mEq/l [2.8 to >9 mmol/l]; reference interval 3.5–4.9 mEq/l [3.5–4.9 mmol/l]).

Fifteen of 75 cats (20%) had lower urinary tract stone(s) identified on radiographs. Fourteen of these had stones present in the bladder, while one was seen in the distal urethra. Thirty-one cats (41%) had lower urinary tract stone(s) identified with ultrasound. Thirty of these had stones visible in the bladder, while one had a stone in the proximal urethra.

A summary of the correlation of diagnostic findings with urine color is presented in Table 2. Urine color was a significant predictor of azotemia (P <0.01), and was significantly positively correlated with serum creatinine concentration (P <0.01; Figure 2) and serum potassium concentration (P <0.01; Figure 3). An increase in color by one category produced a 0.88-fold increase in the odds of azotemia. The strength of the relationship was modest (Nagelkerke R² = 0.38); the logistic regression model correctly classified 76% of cases, including 69% of those in which azotemia was present and 83% of those in which it was absent. Azotemia was prevalent in cats that had red urine (71%) or burgundy urine (100%). All eight cats with a serum potassium concentration >7.0 mEq/l (7.0 mmol/l) had red or burgundy urine. Urine color was significantly negatively correlated with USG (P <0.01), with darker urine more likely to be poorly concentrated. Urine color was significantly positively correlated with RBCs on sediment (P <0.01) and blood on urine dipstick (P <0.01). Urine color was not significantly associated with the presence of crystals on sediment (P = 0.17), WBCs on sediment (P = 0.32), or the presence of lower urinary tract stones on either radiographs (P = 0.08) or ultrasound (P = 0.20).

Table 2

Association of diagnostic findings with urine color

Diagnostic finding	P value
Serum creatinine concentration	<0.01*
Presence of azotemia	<0.01*
Serum potassium concentration	<0.01*
Urine specific gravity	<0.01* †
Red blood cells on urine sediment	<0.01*
Blood on urine dipstick	<0.01*
White blood cells on urine sediment	0.32
Crystals on urine sediment	0.17
Stones on radiograph	0.08
Stones on ultrasound	0.20

*

Statistically significant association

†

Negative correlation

Figure 2

Scatter plot of urine color vs serum creatinine concentration

Figure 3

Scatter plot of urine color vs serum potassium concentration

Discussion

Normal yellow urine color is due to two pigments: urochrome and urobilin. ¹³ The intensity of the color is related to the quantity of water (concentration of urine). ¹³ Abnormal urine color can be due to excess urobilin, bilirubin, hemoglobin, myoglobin, hematuria or various drugs. ¹³ Pink or red urine is most commonly caused by the presence of RBCs, free hemoglobin or myoglobin, all of which will cause a color change on the urine dipstick test strip. ¹³ Microscopic sediment evaluation for RBCs and evaluation of the color of the spun urine supernatant can be helpful in determining whether true hematuria or pigmenturia are present. ¹³

Hematuria can be due to a variety of causes, including inflammation, neoplasia, necrosis or trauma in the urinary tract. ¹³ Iatrogenic trauma leading to hematuria can occur from palpation, cystocentesis or urinary catheter placement. ¹³ Obstruction of the lower urinary tract leads to increased pressure in the bladder and urethra proximal to the obstruction.^12,14,15 With prolonged obstruction this pressure is transmitted to the upper urinary tract, decreasing glomerular filtration rate and leading to progressive azotemia and hyperkalemia.^12,14,15 Submucosal hemorrhage and necrosis of the bladder epithelium can occur within 10 h of obstruction.^11,12 As expected, urine color was found to correlate well with clinicopathological abnormalities common in cats with prolonged obstruction, with higher serum creatinine and serum potassium concentrations associated with darker urine. Darker urine may imply a greater degree of bladder damage due to a longer duration of obstruction and thus a higher degree of metabolic derangement. Urine color correlated well with the degree of microscopic hematuria, as evaluated with both the urine dipstick and the microscopic sediment examination. However, the majority of cats had maximal microscopic hematuria on both tests, including 17/30 cats with yellow urine. The variability in gross urine color was much greater and more likely to correlate with the degree of metabolic derangement.

Urine color was not associated with the presence or absence of lower urinary tract stones with either imaging modality. Both idiopathic cystitis and the presence of stones themselves can cause hematuria,^7–10 although the degree of gross hematuria caused by these conditions has not been evaluated. A recent study comparing different etiologies of feline lower urinary tract disease found a similar incidence of microscopic hematuria in cats with idiopathic cystitis, urethral plugs, bacterial cystitis and urolithiasis. ⁹

In this study, a significantly higher number of stones were detected by ultrasound compared with survey radiographs. Ultrasound may be more sensitive for detecting smaller stones and it is also possible that some of the stones were radiolucent, although radiolucent stone types are rare in cats. ¹⁶ The ultrasound reports of 10 cats reported a single calculus described as ‘linear’ (32% of the cats with stones found on ultrasound). These linear calculi could have represented a urethral plug pushed back into the bladder during catheterization, rather than a true urolith. Urethral plugs are composed primarily of matrix with only a small amount of mineral, and they can be radiolucent or radiopaque. ¹⁷ Future studies are needed to compare survey radiography with ultrasound in cats with UO.

Urine color was negatively correlated with USG. This was unexpected, as lighter urine would traditionally be expected to be less concentrated. This finding could be explained by the fact that the cats with darker urine were more likely to have significant renal injury or postobstructive diuresis, and were less able to concentrate their urine. It is also possible that it was more difficult to pass a urinary catheter in the cats with darker urine, requiring more saline flush and causing iatrogenic dilution of the sample.

This study had a number of limitations. Urine color, based on the provided chart, may not have been an exactly linear variable for statistical purposes. Delayed imaging and imaging after bladder flushing may have affected the ability to detect bladder stones in this patient population, as it is possible that some small stones may have been flushed into the collection system prior to imaging and the urinary catheter may have obscured stone visibility. Additionally, the radiologist was not specifically blinded to urine color when evaluating the bladder on radiographs and ultrasound; however, it seems unlikely that this would have biased the findings. Urine color may have been affected by dilution from the initial unblocking procedure, although 76% of cats had a USG >1.030, so significant dilution is unlikely. It is possible that there was iatrogenic hematuria from irritation of the urethra during urinary catheter placement, although cystocentesis can also cause iatrogenic hematuria. ¹³ There were too few cats with bacteriuria on sediment to allow meaningful statistical analysis, and bacterial cultures were not performed on the initial urine sample in most cats, so the contribution of urinary tract infection to urine color could not be evaluated.

Future studies are needed to evaluate the association between gross urine color and the presence of urinary tract infections, as well as to evaluate any association between initial urine color and outcome of treatment for UO.

Conclusions

Urine color was not predictive of the presence or absence of lower urinary tract stones in cats with UO, but did correlate with the presence of azotemia, hyperkalemia and disruption of normal urine concentrating ability. Cats with darker red urine at the time of urinary catheter placement are likely to have more significant metabolic derangements and may require more aggressive supportive care.