Urethral obstruction in cats: Predisposing factors, clinical, clinicopathological characteristics and prognosis

Abstract

Feline lower urinary tract diseases in general, and urethral obstruction (UO) in particular, are common clinical conditions in cats. The aims of this study were to identify risk factors for UO, to characterise clinical and clinicopathological signs, outcome and recurrence, as well as risk factors for mortality and recurrence. Eighty-two cats with UO were compared to 82 sex and time matched controls. The mean age of cats with UO was significantly lower compared to controls, while the mean body weight was higher. The proportion of indoors–outdoors cats was significantly lower in the study group compared to the control group, and the proportion of cats consuming only dry food was higher. Overall mortality was 8.5%. Ionised calcium was significantly higher in survivors compared to non-survivors, and the prevalence of hypocalcaemia was lower. Recurrence in 6 months and 2 years were 22% and 24%, respectively. Cats with recurrence had significantly lower urine pH at presentation.

Feline lower urinary tract diseases (FLUTDs) are common, and accounted in one study for a mean of 8% (range 2–13%) of the proportional morbidity rate of cats presented to veterinary teaching hospitals in the United States and Canada. ¹ FLUTDs include urolithiasis, urinary tract infections (UTI), urethral obstruction (UO), idiopathic diseases (ie, feline idiopathic cystitis), and less frequently neoplasia, urethral stricture, anatomic malformations and foreign bodies. ^1–3

UO is reported to occur more commonly in young cats, and almost exclusively in male cats due to their relatively long and narrow urethra. ^1,4,5 Risk factors for FLUTDs have been identified in various studies, but the veterinary literature has limited information regarding risk factors for UO. ^1,6 The most common aetiology of UO in cats is urethral plugs, which accounted for 60% of the cases in one study, while uroliths and idiopathic causes accounted for 10% and 30%, respectively. ³ Urethral plugs are usually composed of a combination of proteinaceous material and crystals, but occasionally plugs are composed primarily of matrix, organic material (eg, tissue), blood cells, or aggregates of crystalline minerals. ⁷ Despite the dramatic changes in the prevalence of calcium oxalate and struvite uroliths in the last three decades in cats, the dominant crystal found in urethral plugs is still struvite. ^2,8 It has been hypothesised that inflammation of the urinary bladder promotes urine protein leakage that consequently combines to urinary crystals to form urethral plug formation, and thus increases the risk for UO. ^7,9 Bacterial involvement is only rarely reported in UO in cats. ³

Clinical signs of UO include stranguria, dysuria and pain, and varying degrees of systemic signs, which correlate with the severity of the obstruction. Systemic signs may be life threatening, and occur mainly due to accumulation of uraemic toxins accompanied by electrolyte and acid–base imbalances, all of which result from a sharp decrease in the glomerular filtration rate (GFR). ⁴ The short-term mortality rate of UO in three studies was 5.8%, 5.9% and 8.9%, ^4,10,11 and recurrence was reported in third of the cats. ^10,11

The aims of this study were to identify epidemiological risk factors for UO, to characterise the clinical signs, laboratory abnormalities, outcome and recurrence rates, and to identify risk factors for mortality and recurrence among cats with UO.

Materials and methods

Selection of cases and collection of data

The study was a retrospective, case-control study including cats diagnosed with UO (study group) and sex and time matched controls (control group). All cats were presented to the Hebrew University Veterinary Teaching Hospital (HUVTH) between the years 2006–2008. Cats were included in the study group based on typical clinical signs of UO (eg, dysuria, stranguria), large and firm urinary bladder at presentation that could not be expressed manually, and a clinical evaluation of the attending clinician that the cat was obstructed. Cats were excluded if they did not have a complete medical history and physical examination. The first male cat that presented to the HUVTH after each obstructed cat, and did not suffer from UO, was selected as a control cat. Data extracted from the medical record of all cats included the signalment, date and time of presentation, geographic location (eg, urban/rural), housing conditions (eg, indoors/outdoors), nutrition data (eg, dry/wet/combination), and the presence of other animals in the house. Data were further collected from the medical records of cats in the study group and included clinicopathological data (ie, complete blood count (CBC), serum chemistry, urinalysis), urine culture and sensitivity results, outcome and recurrence data (see below).

In the assessment of physical examination finding by the attending clinicians, cats presenting tachycardia, weak peripheral pulse, pale mucous membranes, hypothermia and cold extremities were clinically considered in a state of shock.

Idiopathic obstruction was defined as an obstruction with no identifiable cause, based on the diagnostic work-up performed during hospitalisation. Diagnostic work-up included CBC (66 cats), partial serum chemistry (77 cats), urinalysis (urine dipstick, 77 cats; sediment evaluation 49 cats), abdominal radiographs (24 cats) and abdominal ultrasonography (29 cats). In five cats both abdominal ultrasound and radiographs were performed.

Laboratory data

Blood samples for CBCs were collected in potassium–ethylenediaminetetraacetic acid (EDTA) tubes upon admission, and analyses were performed using an automatic impedance cell analysers (Abacus or Arcus, Diatron, Austria), calibrated for feline blood. Blood for biochemical analysis was collected in plain tubes or heparin tubes, serum or plasma were separated by centrifugation, and analysis was performed using a wet chemistry auto-analyser (Cobas Mira, Roche, Rottreutz, Switzerland, MaxMat PL, Kone) or a dry chemistry analyser (Reflovet, Roche, Germany), respectively. Urinalysis was performed on urine collected by cystocentesis or by a urinary catheter, and included dipstick analysis (Urilux; Roche, Germany) and sediment evaluation. The urine specific gravity (USG) was evaluated using a standard refractometer. Urine cultures were performed by plating the urine on blood, MacConkey and nutrient agar, and sensitivity analysis was performed using the disc diffusion technique. Proteinuria and haematuria were defined as any positive reaction of the protein pad and the occult blood pad of the urine dipstick, respectively. When evaluating the urine sediment, haematuria was defined as presence of ≥8 red blood cells/high power field (HPF), and pyuria was defined as presence of ≥8 white blood cells/HPF.

Climate information

The risk for UO was also evaluated based on the season and the climate. The year was partitioned into four seasons as follows: winter (December–February), spring (March–May), summer (June–August) and fall (September–November). Temperature and rain precipitation data were collected from the Israeli Meteorological Service. A rainy month was defined as a month with at least 7 rainy days or a month with rain precipitations higher than 60 cm. A warm month was defined as a month with at least 20 days in which the temperature was >25°C, while a cold month was defined as a month with at least 20 days in which the temperature was <15°C. Similarly, a warm week was defined as a week with at least 3 days in which the temperature was >25°C, while a cold week was defined as a week with at least 3 days in which the temperature was <15°C.

Survival and recurrence

Survivors were defined as cats that were discharged from the hospital, and were free of clinical signs during the week following discharge. Non-survivors were cats that either died during the episode, or those that were discharged, but died from complications associated with the current obstructive episode. Long-term survival and recurrence of UO were obtained from the owners by a telephone interview. The follow-up period ranged from 2–4 years.

Statistical analysis

Normality of distribution was assessed using the Shapiro–Wilk test. The association between categorical variables was assessed using the χ² or the Fischer exact test. Continuous parameters were compared between the study and the control group, and between survivors and non-survivors within the study group, using the Student's t-test or the Mann–Whitney-U test, according to data distribution. One-way analysis of variance (ANOVA) or the Kruskal–Wallis test was used to compare continues parameters among three groups or more, according to data distribution. The Spearman rank correlation test was used to assess the correlation between different continuous variables. The Kaplan–Meier product limit method was used to estimate long-term survival curves for different potential risk factors within the study group. For each categorical risk factor the log rank test was used to compare curves. P<0.05 was considered statistically significant. All calculations were performed using SPSS 17.0 for Windows.

Results

Signalment, history, and housing conditions of cats

Eighty-two cats with UO fulfilled the inclusion criteria and were enrolled into the study group. Twelve cats with UO were excluded due to insufficient data in their medical records. All cats included in the study group were males, of which 68 (83%) were castrated. Sixty-eight cats (83%) of the study group were domestic shorthairs; seven Persian, two Siamese, two Himalayans, and one of the following breeds: Turkish Angora, Canadian Sphinx and British Shorthair. There was no statistically significant difference between the study and the control group in regard to breed distribution or castration status.

The majority of the obstructed cats (82%) were between 1–7 years of age. The mean age of cats in the study group was significantly lower compared to the control group (51.7±37.7 vs 75.7±61.3 months, respectively, P<0.001). The mean body weight of cats in the study group was significantly higher compared to the control group (5.6±1.4 vs 5.1±1.5 kg, respectively, P=0.031).

There was no statistically significant difference between the study and the control groups in regard to the proportion of cats living in urban vs rural areas, proportion of vaccinated or dewormed cats, and presence of additional animals in the house (Table 1). The proportion of indoors–outdoors cats was significantly lower in the study group compared to the control group, while the proportion of strict outdoor cats was higher in study compared to the control group, but the difference did not reach statistical significance (P=0.053) (Table 1).

Table 1.

Signalment, history and housing conditions of cats with UO compared to controls.

Parameter	Study group, n (%)	Control group, n (%)	P value	OR † (CI95) ‡
Consuming dry food only	68 (83)	44 (54)	<0.001	4.20 (2.04–8.63)
Consuming both dry and wet food	14 (17)	35 (42)	<0.001	3.62 (1.76–7.45)
Consuming wet food only	0 (0)	3 (4)	0.08	NA
Domestic shorthair cats	67 (82)	66 (81)	0.84	1.08 (0.46–2.37)
Castrated	68 (83)	66 (81)	0.56	2.27 (0.57–2.84)
Vaccinated	55 (68)	61 (74)	0.36	0.73 (0.37–1.44)
Dewormed	39 (48)	47 (55)	0.39	0.76 (0.41–1.41)
Urban areas	66 (81)	71 (87)	0.29	0.65 (0.30–1.54)
Additional animals in the household	41 (53)	40 (49)	0.63	0.86 (0.28–1.48)
Indoors *	46 (56)	37 (45)	0.16	1.56 (0.84–2.87)
Indoors–outdoors *	30 (37)	44 (54)	0.03	0.50 (0.27–0.93)
Outdoors *	6 (7)	1 (1)	0.053	6.72 (0.75–55.56)

NA=not applicable.

^*Each group was compared to the other two groups as a post-hoc analysis.

^†Odds ratio.

^‡95% confidence interval.

The proportion of cats consuming only dry food was significantly higher in the study compared to the control group, while the proportion of cats consuming both dry and wet food was significantly lower in the study compared to the control group (Table 1).

Cats with UO

Season and weather

The median number of cats presented during the winter, spring, summer and fall was, four (range, 1–9), four (range, 1–5), three (range, 1–6) and two (range, 1–5), respectively, with no statistically significant differences in the median number of presentation between the seasons. There was also no statistically significant difference in the median number of cats presenting during warm vs cold weeks (in both median was 1; range 0–3, P=0.55), warm vs cold months (median 4, range (1–9) vs 2.5 (range 1–6), respectively, P=0.46), or rainy vs dry months (median 3, range (1–9) vs 3 (range 1–6), respectively, P=0.71).

Presentation, clinical signs and clinicopathologic data

Most of the cats in the study group were presented to the HUVTH with UO for the first time (62, 76%), but in 14 (17%), four (5%) and two (2%), this was the second, third and fifth episode of obstruction. The median temperature was 37.8°C (range 32.2–40.8°C), median heart rate was 160 bpm (range 60–300 bpm), and median respiratory rate was 40 breaths/min (range 16–100 breaths/min). Abnormalities of vital signs included tachypnoea (>40 breaths/min, 38, 57%), hypothermia (<37°C, 29, 51%) and bradycardia (<120 bpm, 17, 23%). The most common clinical signs at presentation were anorexia (48, 58%), vomiting (19, 23%) and circulatory shock (12, 15%).

CBC was performed in 66 cats. Median white blood cell count was 15.3×10³ cells/μl (range 3.7–70.2×10³ cells/μl, reference interval (RI) 5–17×10³ cells/μl), and median haematocrit was 40.8% (range 21.6–67.5%, RI 25–45%). The most common CBC abnormalities at presentation were leukocytosis (27 cats, 41%) and increased haematocrit (23 cats, 35%). Median serum creatinine was 4.43 mg/dl (range 0.5–29.1 mg/dl, RI 1.1–1.24mg/dl), median ionised calcium was 1.04 mmol/l (range 0.63–1.31, RI 1.1–1.24 mmol/l), median potassium concentration was 5.3 mmol/l (range 2.7–11.6 mmol/l, RI 3.5–5.5 mmol/l), and median sodium concentration was 144.3 mmol/l (range 119.0–178.1 mmol/l, RI 146–154 mmol/l). The most common abnormalities in the serum chemistry were azotaemia (64 cats, 85%), ionised hypocalcaemia (32 cats, 56%), hyperkalaemia (37 cats, 48%), and hyponatraemia (36 cats, 55%). Positive correlations were found between serum creatinine concentration and the following: serum potassium concentration (r=0.81, P<0.001) and white blood cell count (r=0.41, P<0.001); while negative correlations were found between serum creatinine concentration and the following: ionised calcium concentration (r=−0.68, P<0.001), USG (r=−0.49, P<0.001) and chloride concentration (r=−0.40, P=0.001). There was no statistically significant correlation between potassium concentration and heart rate.

Median USG was 1.025 (range 1.011–1.050), and was below 1.035 in 47 (67%) cats. Urine dipstick analysis revealed median urine pH of 7.0 (range 5.0–9.0). Common abnormalities included haematuria (72, 100%), varying degrees of proteinuria (66, 88%), glucosuria (25, 34%), alkaluria (pH >7.5, 19 cats, 26%), and bilirubinuria (nine, 12%). Forty-nine cats had their urine sediment evaluated and the most common findings were haematuria (47, 96%), pyuria (32, 65%), bacteruria (21, 43%), struvite crystalluria (19, 39%), presence of epithelial cells (eight, 16%) and granular casts (two, 4%). Urine culture and sensitivity were submitted in 23 cats, of which 13 (56%) were negative and 10 (44%) positive, however in six of the latter a urinary catheter was introduced by a referring veterinarian prior to presentation at the HUVTH, and in an additional three cats the urine culture was obtained from a urinary catheter. Isolates included Escherichia coli (eight cats), Pseudomonas aeruginosa, and Proteus mirabilis (one of each).

Hospitalisation and treatment

The cause for the UO was not identified in any of the cats. UO was relieved under general anaesthesia in all cats. The obstruction was relieved using a urinary catheter in 80 cats (97.6%) or by a gentle massage of the tip of the penis in two cats. A urine catheter was left in place after relieving the obstruction in 75 cats (91.5%) for a mean duration of 42±27.6 h. The most common treatments included intravenous crystalloid fluids (lactated Ringer's solution or saline 0.9%), (80 cats, 97.6%), phenoxybenzamine (Dibenyline, Goldshield pharmaceutical), (69 cats 84.1%), antibiotics (67 cats, 81.7%), analgesia (butorphanol: Turbogesic, Fort Dodge), (30 cats, 36.6%) and diazepam (Assival, Teva), (six cats, 7.3%). Seventy-three (89%) of the cats were hospitalised to a median hospitalisation time of 3.5 days (range 1–13).

Outcome and risk factors for mortality

Eight cats (10%) underwent a perineal urethrostomy (PU) surgery after medical management had failed to resolve the obstruction. In four cats the surgery was performed in the current episode, which was the first episode in three cats and the second episode in one cat. In the remaining four cats, surgery was performed within the following 2 months of presentation due to recurrence episodes. All of the cats that underwent PU survived.

Outcome data were available in 71 cats, of which six cats (8.5%) did not survive the episode. There was no statistically significant difference in the age of survivors compared to non-survivors (median, 48 months (range 4–132) vs 57 months (range 24–144), respectively, P=0.35). Similarly breed, castration, season, housing conditions, presence of additional animals in the house, vaccination, deworming, and number of obstruction episodes were not found to be associated with mortality. Prior treatment at a referring veterinarian was statistically more common among non-survivors compared to survivors (six (100%) vs 19 (30%), respectively, P<0.001). None of the clinical signs at presentation was associated with increased mortality.

Serum creatinine and potassium concentrations were higher in non-survivors, but the difference did not reach statistical significance. Likewise, there was a trend (P=0.08) for higher proportion of hyperkalaemia in non-survivors compared to survivors (Table 2). Median concentration of ionised calcium was significantly higher in survivors compared to non-survivors (1.08 mmol/l, range 0.65–1.28 vs 0.88 mmol/l, range 0.66–1.11, respectively, P=0.037), and the proportion of hypocalcemia was lower (51% vs 100%, respectively, P=0.024) (Table 2).

Table 2.

Comparison of clinicopathological data of cats with UO between survivors and non-survivors.

Parameter	Survivors	Non-survivors	P value	n (%) below RR		n (%) above RR
	Median (range)	Median (range)	Survivors	Non-survivors	Survivors	Non-survivors
White blood cells (103/μl)	15.6 (5.4–70.2)	17.7 (10.9–27.4)	0.65	0 (0)	0 (0)	22 (43)	3 (50)
Haematocrit (%)	42.0 (23.3–67.5)	41.3 (27.0–54.3)	0.94	1 (2)	0 (0)	19 (38)	2 (33)
Creatinine (mg/dl)	4.2 (0.8–17.0)	6.0 (0.9-10.0)	0.69	0 (0)	0 (0)	53 (88)	5 (83)
Potassium (meq/l)	5.1 (2.7–11.6)	7.7 (3.0–9.7)	0.22	5 (9)	1 (17)	27 (46)	5 (83)
Sodium (meq/l)	146 (119–162)	142 (134–148)	0.13	24 (49)	5 (83)	9 (18)	0 (0)
Ionised calcium (mmol/l)	1.08 * (0.65–1.28)	0.88 (0.66–1.11)	0.037	22 (51) *	6 (100)	0 (0)	0 (0)
USG	1.025 (1.011–1.050)	1.030 (1.016–1.030)	0.86		NA
Urine pH	7 (6–9)	6 (6–7)	0.13	0 (0)	0 (0)	14 (24)	0 (0)

NA = Not applicable, RR = Reference range.

^*Significant difference between survivors and non-survivors.

Median serum creatinine concentration of the survivors before discharge was available in 68 cats and was 1.2 mg/dl (range 0.5–1.9 mg/dl). Only seven cats (10.3%) were discharged with serum creatinine higher the upper limit of the reference interval (1.6 mg/dl).

Recurrence

Fifty cats that survived the episode were available for long-term outcome, of which 11 (22%) had experienced at least one recurrence episode within the 6 months after discharge. Of the 11 cats, two died, four had a PU surgery within this period and all survived, one was lost for follow-up, and four survived the obstruction episode with medical care. One cat had an additional recurrence episode within 1 year of discharge, while the rest did not experience any recurrence episodes until the last follow-up period. No complications such as UTI or stricture were reported in any of the cats that underwent PU surgery in the follow-up period.

Indoors–outdoors cats had a trend for lower recurrence rate compared to other cats (9.5% vs 31.0%, P=0.07, odds ratio (OR) 23, 95% confidence interval (CI₉₅) 0.44–1.22). Consumption of diets designed for cats with lower urinary tract diseases was not associated with a significant reduction in the recurrence rate. Cats with recurrence had a trend for higher mean USG compared to cats without recurrence (1.034±0.01 vs 1.028±0.01, P=0.09), and a significantly lower urine pH at presentation (6.6±0.5 vs 7.2±1.5, P=0.05). No other risk factors for recurrence were identified in this study.

Discussion

Several risk factors for UO in cats were identified in this study, as well as several risk factors for mortality and recurrence. Indoors–outdoors cats were found to have decreased risk for UO, which is in agreement with previous studies of FLUTD. ^6,12 It has been hypothesised that indoors cats, which are less active and presumably urinate less frequently, are more likely to develop urine crystals that may become incorporated into urethral plugs. ²

Increased body weight was also found in this study to be a risk factor for UO. Increased body weight has been previously associated with a higher incidence of FLUTD, ^2,6 as well as increased risk for urolithiasis in human patients. ¹³ Obesity is associated with variety of diseases, ¹⁴ but it is currently unknown whether it has a direct or indirect effect on FLUTD. Alternatively, obesity may merely be a marker of decreased activity level, and thus indirectly increases the risk for FLUTD as discussed above.

Cats with UO were younger than the control cats. It has been shown that the frequency of different lower urinary tract diseases in cats changes over time, with a gradual decrease in the frequency of feline idiopathic cystitis and UO, and a gradual increase in frequency UTI. ¹ The age range of UO parallels the reported age range of feline idiopathic cystitis, ⁹ therefore, it is possible that UO is a complication of feline idiopathic cystitis due to the aforementioned reasons (ie, presence of inflammation, crystals, oedema), and thus UO is more common at a relatively young age.

The proportion of cats with UO consuming dry food was higher compared to the control group, as has been found in other studies evaluating the effect of diets on FLUTD. ^6,10,15–18 It is possible that the main protective effect of the wet food is the water content rather than nutrient content. Consumption of a food containing 75% moisture compared to the same food containing 10% moisture was associated with increased urine volume and decreased USG. ¹⁹ High urine volume and decreased USG are associated with higher frequency of urination and decreased activity product of urinary crystals.

The effect of the season and weather was previously evaluated as factors that may influence the incidence of non-obstructive FLUTD, with conflicting results. A study from Canada has reported a higher frequency of non-obstructive FLUTD in the winter, ²⁰ while a study from New Zealand demonstrated no statistically significant increased risk for FLUTD during the winter time, but risk did increase after a rainy month. ⁶ It has been hypothesised that cold weather is associated with decreased activity and urine retention, particularly in cats that urinate outdoors, but it cannot be excluded that the cold weather can be a direct stress factor that increases the risk for feline idiopathic cystitis. In the current study a difference in the median number of UO among different seasons was not found, nor was there a difference in this number in cold/rainy weeks or months compared to warm/dry weeks or months. It is likely that if weather does have an effect on the incidence of UO it will be more prominent in countries in which the winter is cold and rainy. The criteria selected in this study to define a cold/rainy period might not have been cold or rainy enough to influence the cats’ behaviour, and thus to demonstrate a significant difference in the presentation rate. However, selecting different criteria would have resulted in a reduction of the number of cold and rainy weeks/months that would have fulfilled these criteria, thus rendering any comparison extremely underpowered.

Clinical signs of UO at presentation can be categorised to local lower urinary tract signs, resulting from the obstruction, and systemic signs associated with the degree of UO and the resulting accumulation of uraemic toxins and electrolyte and acid–base disorders. The most common clinical signs of cats with UO in this study were firm and distended urinary bladder (which were part of the inclusion criteria), dysuria and stanguria, anorexia and tachypnoea, while bradycardia, vomiting and shock were reported less frequently. Most of the systemic clinical signs were consistent with the metabolic derangement of cats with UO, as reflected by the common clinicopathological findings of this study that included azotaemia, hyperkalaemia, hyponatraemia, and hypocalcaemia. Accumulation of uraemic toxins, acid–base and electrolyte disorders result in UO from a sharp reduction in the GFR. The latter is a consequence of the increasing hydrostatic pressure within the urinary system, up to the level of Baumann's space. When the pressure is high enough it may oppose the capillary hydrostatic pressure, which is the main force for filtration, and thus may stop glomerular filtration, and consequently lead to the accumulation of nitrogen waste products (85% of the cats in this study were azotaemic).

The relatively high frequency of tachypnoea observed in this study might have resulted from pain, excitement, metabolic acidosis (or a combination of these factors), and is less likely to result from a direct effect of the uraemic toxins on the respiratory system (ie, uraemic pneumonitis). As no cat presented with dyspnoea it is more likely that tachypnoea was an extra respiratory manifestation that resulted from the combination of the aforementioned reasons.

Bradycardia in UO is attributed to hyperkalaemia (documented in 48% of the cats in this study), which results mainly from GFR reduction, and is exacerbated by the metabolic acidosis. A recent study has found that severe hyperkalaemia (>8 mmol/l) is associated with decreased heart rate, ²¹ however, in the current study the correlation between heart rate and serum potassium concentration was not significant, suggesting that other factors probably play a role. Stress and pain at presentation may activate the sympathetic system and thus mask an underlying bradycardia. Bradycardia might have predisposed cats in this study to circulatory shock, as seen in 15% of the cats. However, in cats, bradycardia can also be the result of circulatory shock; ²² thus a cause and effect relationship is difficult to establish in a retrospective study, and further studies are warranted to resolve this issue.

Hypothermia (51% of the cats) was also a common finding in this present study, and most likely resulted from circulatory shock and accumulation of uraemic toxins. When renal function decreases, an accumulation of dozens of uraemic toxins ensues, some of which decrease heat production and may promote hypothermia. ^23,24

Urinalysis results included haematuria, pyuria, proteinuria, glucosuria and bilirubinuria. One should keep in mind that the interpretation of the urine dipstick may be misleading as the urine colour of most cats with UO is abnormal, and thus may interfere with the colour reading on the urine dipstick. The high prevalence of haematuria is most likely a result of urinary bladder bleeding due to inflammation, high pressure within the bladder, and less likely due to previous cystocentesis and catheterisation attempts. Inflammation is also supported by the high proportion of pyuria. Likewise the origin of the proteinuria in cats with UO is most likely post-renal. In this study proteinuria was assessed only using the urine dipstick test, which in a recent study has been shown to be have questionable reliability. ²⁵ The cause for the bilirubinuria, documented in 12% of the cats, is unclear. As pre-hepatic (ie, haemolyisis), hepatic, or post-hepatic causes were not identified in any of the cats in this study it probably represents an artifact due to urine colour. Glucosuria detected in 34% of the cats may be associated with stress-induced hyperglycaemia or when present in the face of normoglyacemia it may indicate tubular damage and acute kidney injury. ²³ As blood glucose concentrations were not reported commonly in this study it is impossible to differentiate between the two, but as only two cats had cylinduria, and 89.7% of the cats had normal creatinine concentrations at discharge, it is more likely that hyperglycaemia was associated with stress or represented an artifact.

The proportion of bacteruria and positive urine cultures found in this study is higher than the reported incidence in the literature, ³ however, in the present study most cats with positive urine cultures had a urinary catheter introduced by a referring veterinarian prior to the arrival at the HUVTH, or the urine was obtained at the HUVTH via a urinary catheter. A recent study from Norway documented a 25% incidence of UTI in cats with FLUTD. ²⁶ In light of these findings urine cultures should be considered in cats with UO, especially in those that had previous urinary catheter placement. ²⁷ Despite the above, the use of antibiotics to prevent UTI when urine catheter is placed is not considered good practice.

All cats in the current study were defined as idiopathic, however, one has to take into consideration that a thorough diagnostic work-up was not performed for all cats, thus the true incidence of idiopathic obstruction in this study may be an overestimation of its true prevalence.

UO is a life-threatening condition; however, in the current study the survival rate was high. The results of the current study are in agreement with results of additional two studies that have reported survival rates of 91.1% and 94.2%. ^4,11 These data suggest that with proper medical management, UO is associated with a favourable outcome. Moreover, long-term survival of these cats was also high, suggesting a complete recovery of the cats from the episode. These data can be used to encourage owners and veterinarians to pursue treatment in cats with UO. Perineal urethrostomy is a salvage procedure and is associated with complication as stricture formation and UTI. ^28,29 Nevertheless, the long-term outcome of all cats that underwent this surgery was also excellent, as has been shown previously. ²⁸

Most of the parameters evaluated in this study (eg, age, weight, castration, housing conditions, nutrition, vaccination, deworming, clinical signs at presentation and clinicopathological data) were not associated with increased mortality. Nevertheless, none of these can be excluded as risk factors, as a limitation of the risk analysis in this study is the low mortality rate. Consequently, comparisons made between survivors and non-survivors suffered for a low statistical power, and were prone to type II error. Further studies evaluating the above parameters prospectively and using more extensive diagnostic work-up may shed light on further risk factors for mortality in cats with UO.

Decreased ionised calcium concentration was common in this study, as has been previous shown, ^4,30,31 and was found to be a risk factor for mortality. Ionised hypocalcaemia is likely a consequence of phosphorus accumulation and thus may serve as an indirect marker of the obstruction severity. This is also supported by the correlation between serum creatinine concentration and ionised calcium. Ionised hypocalcaemia may exacerbate the cardiotoxicity caused by the hyperkalaemia, and therefore, may be a direct cause of death. ³² Hyperkalaemia has been suggested as a risk factor for mortality. In this study 56% of the cats presented with hyperkalaemia, and there was a trend for increased mortality among cats with hyperkalaemia.

Another risk factor for mortality in this study was an initial treatment at a referring veterinarian. Due to the nature of this retrospective study it is difficult to assess this further, however, it is likely that cats that were presented to the HUVTH after initial treatment at a referring veterinarian represented a more severe and complicated cases.

The recurrence rate of UO in this study was relatively low (11/50, 22% within 6 months and 12/50, 24% within 2 years), compared to a higher reported recurrence rate. ^10,11,33 It is plausible that the recurrence rate varies among different geographical areas due to the potential influence of the climate, housing, environment, food consumption, and owner complaints. Urine pH was significantly lower in cats with recurrence. It is unknown what the exact cause for this finding is. It is possible that the high urine pH played a role in the pathophysiology of some cats with UO (ie, promoted struvite crystal formation); therefore, in these cats the underlying cause was more amenable to dietary modification, as many of the diets designed for cats with lower urinary tract disease aim for a urine pH that decreases the solubility of struvite crystals. In cats with low urine pH at presentation the dietary modification might not have played such important role, and thus these cats were more prone to recurrent episodes. This speculation needs further evaluation in other studies.

In summary, several risk factors for UO were found in this study including age (ie, young adult cats), increased body weight, and consumption of dry food. Indoors–outdoors cats were found at decreased risk. Despite a variety of clinical signs and severe clinicopathological abnormalities the vast majority of cats with UO survived following the treatment. Ionised hypocalcaemia and hyperkalaemia should alert clinician and promote aggressive management and stabilisation prior to de-obstructing cats under general anaesthesia. The recurrence rates are relatively low and the long-term outcome is good, thus therapy should always be perused.