Effect of prazosin on feline recurrent urethral obstruction

Abstract

Objectives

The aim of this study was to determine if male cats treated with 7 days of prazosin following relief of urethral obstruction (UO) experienced decreased rates of recurrent urethral obstruction (rUO) within 30 days vs those treated with 7 days of placebo.

Methods

All castrated male cats presenting for the first time with UO from May 2014 to August 2017 were eligible for enrollment. Exclusion criteria included the administration of medications or passage of a urinary catheter prior to referral, the presence of heart disease or hypertension requiring medication, prior treatment with glucocorticoids, non-steroidal anti-inflammatory medications, prazosin or phenoxybenzamine, or radiographic identification of cystoliths. Cats were treated with standardized anesthetic and analgesic protocols, standardized indwelling urinary catheter management, and were hospitalized for care. A random numbers table was generated prior to study initiation and cats were randomized to receive either prazosin (0.5 mg PO q12h for 7 days) or placebo in a blinded fashion. A 30-day follow-up with owners via telephone was performed to identify the rate of rUO. Cats that did not receive the full course of study medication were removed from the analysis. The study was unblinded at the end of data collection.

Results

Eighty cats were enrolled and 65 cats completed the study; 12 were excluded because they did not receive the study medication. Sixteen of 65 cats experienced rUO (25%). Of the 16 cats experiencing rUO, five received placebo (n = 5/28 [18%]) and 11 received prazosin (n = 11/37 [30%]). Ten of the cats that experienced rUO reblocked while still hospitalized. There was no significant difference in frequency of rUO in cats treated with prazosin vs placebo (P = 0.27).

Conclusions and relevance

Prazosin administered at 0.5 mg PO q12h did not decrease the rate of rUO in this population of obstructed male cats vs placebo. These results further support evidence suggesting that prazosin may not be beneficial in prevention of feline rUO.

Introduction

Urethral obstruction (UO) is common, accounting for 4.6% of all feline cases and 8.1% of all male cat cases presenting on emergency. ¹ Causes include urethral plugs, urolithiasis, crystalluria, urinary tract infection and idiopathic disease.^2–4 Although life-threatening, UO is treatable, with a 93.6% rate of survival to hospital discharge reported. ⁵ Hospitalization is typically required, with treatment consisting of analgesia, anesthesia, retrograde flushing to relieve the obstruction, placement of an indwelling catheter connected to a closed urine collection system for 24–48 h, intravenous (IV) fluid therapy and correction of electrolyte disturbances.^5–7

A high percentage of recurrent UO (rUO) has been reported. In a study by Gerber et al, ⁸ 51% of cats with UO displayed recurrent lower urinary tract signs, with 36% experiencing rUO within 17 days. Other reports suggest a rUO rate as low as 15% and as high as 74% within 7 days.^8–10 Euthanasia rates of up to 23% have been reported in cats after rUO.^1,8

Studies have evaluated risks associated with rUO, including urinary catheter size, duration of catheterization and diet.^{1,6,8,11–14} In addition, urethral spasm has been hypothesized as a potential contributor to the development of rUO.^12,15–17 The proximal 28–37% of the feline urethra is composed primarily of smooth muscle and contains alpha (α)₁-adrenergic receptors that contract in response to norepinephrine (noradrenaline) release from the hypogastric nerve.^18,19 A number of drugs targeting these receptors have been used to reduce urethral muscular spasm; however, controlled clinical studies are lacking and the use of α-adrenergic receptor antagonists or other urethral relaxants is based largely on clinician preference. Prazosin, an α₁-adrenergic receptor antagonist, has been experimentally shown to reduce urethral pressure.^16,18 In the cat, blockade of the α-receptors prevents sympathetically mediated smooth muscle contraction throughout the preprostatic and prostatic urethral segments, leading to reduction in urethral wall tension. ¹⁸ Because of these effects, prazosin may be useful in treating urethral dyssynergia or trauma associated with urethral catheterization. ¹⁸

There are few studies evaluating α-adrenergic receptor antagonists for management of feline UO. A retrospective study comparing rUO rates in cats treated with prazosin (0.5–1mg PO q8–12h) with those treated with phenoxybenzamine, a non-selective α-adrenergic receptor antagonist, identified a significant decrease in short-term rUO in those receiving prazosin. ¹ Although the results of that study suggest that prazosin may be beneficial in the management of rUO, it was limited by its retrospective design, lack of a control group and confounding variables such as catheter size, duration of catheterization, and differing anesthetic and analgesic protocols. Another prospective study evaluating various risk factors for rUO determined that administration of either prazosin or phenoxybenzamine was not associated with a decreased risk of rUO; however, this study was observational, not randomized and did not include a control group. ⁶ A more recent double-blinded prospective interventional study specifically evaluated the effect of prazosin on outcome in feline UO and failed to find a significant difference in the incidence of rUO in cats receiving prazosin (0.25 mg/cat PO q12h) and those receiving placebo for 30 days following relief of obstruction. ¹⁰ This study was relatively small, evaluating only 45 cats with several uncontrolled variables, including the duration of catheterization and urinary catheter size. ¹⁰

Our prospective, randomized, placebo-controlled, blinded study aimed to evaluate statistical differences in the rate of 30-day rUO in male cats receiving prazosin vs those receiving placebo.

Materials and methods

Sample size determination

An improvement from an expected frequency of 36% rUO over 30 days in the control group to 4% in the treated group using a log rank test with 80% power and 5% confidence assuming a loss to follow-up of 5% or less in each group was used to determine a necessary sample size of 80 (40 cats per group).

Case selection

Castrated male cats presenting with UO for the first time to the Emergency Department at a private multispecialty practice were eligible for inclusion in the study. Written informed consent was obtained from each owner prior to enrollment of their cat for treatment and participation in the study. The study protocol was reviewed and approved by the hospital Animal Care and Use Committee. A UO was defined as the presence of a non-expressible bladder or non-productive stranguria documented by a veterinarian. All cats were hospitalized for treatment of UO. Cats were excluded prior to enrollment if they had heart disease or hypertension requiring medical therapy, and if they had treatment within the previous 2 weeks with glucocorticoids, non-steroidal anti-inflammatory medications, prazosin, phenoxybenzamine or other medications for UO. Cats were excluded if an indwelling urinary catheter was placed prior to enrollment. Cats whose temperament prevented administration of oral medications were excluded, as well as those with radiopaque cystoliths identified on a lateral abdominal radiograph performed after relief of the obstruction.

Therapy

Cats had an IV catheter placed and were initially administered methadone (0.2 mg/kg IV or IM; Mylan Institutional) and IV crystalloid therapy, and any electrolyte disturbance was corrected. Propofol (Zoetis) was administered IV to effect for sedation to relieve the obstruction. Oxygen was provided via a face mask, and electrocardiogram, blood pressure and pulse oximetry were monitored throughout sedation. The prepuce and perineal region were shaved of hair and aseptically prepared. A 3.5 F polypropylene catheter (Covidien) or an olive-tipped urethral catheter (Jorgensen Laboratories) was used for relief of the obstruction. An indwelling 3.5 F red rubber catheter (Covidien) was placed, permitting collection of a urine sample for urinalysis and culture if desired, complete drainage of the bladder and flushing with sterile 0.9% saline. The red rubber catheter was secured in place and connected to a closed urinary collection system. Following catheter placement, a lateral abdominal radiograph was performed to evaluate for radiopaque cystoliths. Urinalysis and urine culture and susceptibility were performed based upon clinician discretion with antimicrobial therapy initiated if indicated. Buprenorphine (0.015–0.02 mg/kg TM or IV q8h; Par Pharmaceutical) or long-acting buprenorphine (0.24mg/kg SC q24h [Simbadol; Zoetis]) were administered for post-procedural pain relief. If cats received alternative medications that did not lead to study exclusion, these medications were recorded. An indirect arterial blood pressure reading was obtained via Doppler within 2 h of initiating unblocking treatment, and those with a reading of <80 mmHg were withdrawn from the study. An indirect arterial blood pressure was obtained 1 h after initiating treatment with the study drug. Cats were randomly assigned to receive either prazosin (0.5 mg/cat PO q12h for 7 days; placed in a gel capsule) or a placebo in identical capsules (1 capsule PO q12h for 7 days) within 8 h of urinary catheter placement. The prazosin and placebo were prepared by a licensed pharmacist who identified the prescriptions by a letter. The study investigators and cat owners were blinded to the contents of the capsules. A random number table was created prior to study initiation to randomly distribute patients between groups and was followed with each enrollment.

The indwelling urinary catheter remained in place for at least 24 h and no more than 48 h. Each cat received IV fluids to meet replacement, maintenance and ongoing fluid losses. Following urinary catheter removal, each cat was monitored for dysuria.

Each cat was discharged with the study medication to complete a 7-day course of therapy along with transmucosal buprenorphine for pain control. Standardized discharge instructions were provided, including recommendations for litter tray management, strategies for increasing water consumption, and environmental modification and enrichment. Transition to a prescription urinary tract diet was recommended on a case-by-case basis depending upon the presence of crystalluria in the initial urine sample.

Data analysis

Patient follow-up was performed via standardized telephone interview by the primary investigator within 1 week of hospital discharge and again after 30 days. Owner compliance with administration of the study medication and occurrence of rUO was recorded. If owners noted any adverse effects, these were also recorded. If owner communication was not possible, referring veterinarian records were evaluated for follow-up information.

Medical records for each cat were reviewed and the age, weight, presenting vital signs and results of available laboratory analysis, and duration of catheterization and hospitalization were recorded. Normality of distribution was assessed using the Shapiro–Wilk test. The Mann–Whitney U-test was used for comparison of presenting variables between cats treated with prazosin and placebo as well as cats that experienced rUO and those that did not. Additionally, a two-sample χ² test for equality of proportions was used to compare the difference in rUO between the prazosin and placebo groups. P values <0.05 were considered to be statistically significant. To assess the impact of study variables on the outcome measure (ie, odds of developing rUO), a multivariate logistic regression model was built using a stepwise forward inclusion (P <0.25) backward elimination approach (P <0.05). Multicollinearity was assessed for using variance inflation factor <2.5. Finally, a Cox proportional-hazard model was used to analyze the risk of rUO development chronologically between the two study groups. All analyses were performed in R 3.5.1.

Results

From May 2014 to August 2017, 80 cats meeting the criteria were enrolled in the study. Forty cats were randomized into the prazosin group and 40 cats in the placebo group. Following randomization, one cat from the placebo group was withdrawn from the study by its owner. Two cats from the placebo group were lost to follow-up. Of the 77 cats for which there was follow up information, 65 received the study medication, three received only one dose and nine cats did not have any of medication administered following discharge. No specific adverse effects were reported by owners. The 12 cats that did not receive the complete course of study medication were excluded from the data analysis. Sixty-five cats (37 treated with prazosin and 28 treated with placebo) were therefore included in the final evaluation.

Of the 65 cats that began the study, there were 47 domestic shorthairs, 11 domestic longhairs, four domestic mediumhairs, two Maine Coons and one Russian Blue cat. The median weight of the cats was 6 kg (range 3.7–10.8) and median age was 5 years (range 1–19). Characteristics of cats treated with prazosin and placebo and those experiencing rUO vs those that did not are displayed in Table 1. Based on the results from the Mann–Whitney U-test, at inclusion, there were no significant differences between cats treated with prazosin vs placebo in any of the covariates evaluated except for duration of catheterization, with prazosin cats having a significantly shorter duration of urinary catheterization than placebo cats (Table 1). No significant differences were identified in any of the covariates evaluated between cats that experienced rUO and those that did not, with the exception of a significantly lower presenting packed cell volume (PCV) in cats that experienced rUO (Table 1). No cat experienced hypotension and there was no significant difference in post-treatment Doppler blood pressure in cats treated with prazosin and placebo, and those that experienced rUO and those that did not.

Table 1

Selected characteristics in 65 cats treated with prazosin and placebo and those that did and did not experience recurrent urethral obstruction (rUO): results from the Mann–Whitney U test

Variable and applicable RI	Cats treated with prazosin (n = 37)	Cats treated with placebo (n = 28)	P value	Cats with rUO	Cats without rUO	P value
Age (years)	5 (1–17)	4.5 (1–19)	0.63	5 (2–9)	4 (1–19)	0.94
Weight (kg)	5.8 (3.9–9.9)	6 (3.7–10.8)	0.52	5.6 (3.3–9.0)	6 (4.0–10.8)	0.44
Temperature (°F)	100.9 (97.8–104)	101.7 (99–104.4)	0.07	100.9 (97.8–102.8)	101.3 (98.7–104.4)	0.48
HR (bpm)	210 (160–280)	210 (180–280)	0.78	205 (160–280)	220 (160–280)	0.18
RR (breaths/min)	60 (30–156)	47 (24–124)	0.19	47 (32–120)	60 (24–156)	0.38
SBP post-prazosin/placebo (mmHg)	123 (92–190)	118 (86–170)	0.75	121 (100–168)	120 (86–190)	0.81
Creatinine (mg/dl; RI 0.7–2.1)	1.8 (1.05–12.9)	2.05 (0.91–18.1)	0.21	2.45 (1.35–18.1)	1.86 (0.91–12.3)	0.21
PCV (%; RI 35–55)	44 (31–55)	45 (31–56)	0.42	43 (32–52)	45 (31–56)	0.04
TS (g/dl; RI 5.5–7.5)	8 (6.4–10)	7.6 (6.2–8.6)	0.21	7.8 (6.2–8.6)	8 (6.6–10)	0.33
Potassium (mmol/l; RI 3.6–5.6)	3.8 (3.2–10)	4 (3–9.3)	0.78	3.8 (3.3–10)	3.9 (3–8.3)	0.74
Duration of catheterization (h)	24 (17–44)	28 (22–48)	0.03	29 (24–42)	26 (17–48)	0.16
Duration of hospitalization (h)	40 (24–72)	41 (26–72)	0.81	46 (26–56)	40 (24–72)	0.24

P <0.05 was considered as statistically significant. Data are presented as median (range)

RI = reference interval; HR = heart rate; bpm = beats/min; RR = respiratory rate; SBP = systolic blood pressure; PCV = packed cell volume; TS = total solids

Sixteen cats experienced recurrent urethral obstruction (n = 16/65 [25%]). Of the 16 cats experiencing rUO, five were in the placebo group (n = 5/28 [18%]) and 11 were in the prazosin group (n = 11/37 [30%]).

Importantly, results from the χ² proportional test showed no significant difference in the incidence of rUO in cats treated with prazosin vs placebo (P = 0.27). Of the 16 cats with rUO, 10 (62%: two placebo cats and eight prazosin cats) experienced rUO after urinary catheter removal while still in hospital (median 6 h; range 6–16) and six (38%; three placebo cats and three prazosin cats) experienced rUO within the 30-day follow-up period (median 4.5 days; range 1–26). When looking at the results from the multivariate logistic regression analysis, only creatinine levels (P = 0.03) and total solids (TS; P = 0.05) were found to significantly affect the odds of developing rUO (Table 2). Multicollinearity of these variables was assessed using the variance inflation factor (VIF; creatinine VIF 1.201, TS VIF 1.201). Consistent with the results from the χ² proportional test, the variable ‘treatment’ was not found to be statistically significant (P = 0.27). Finally, results from the Cox-proportional hazard model showed no significant difference in time to rUO between cats treated with placebo and cats treated with prazosin (P = 0.24) (Table 3 and Figure 1).

Table 2

Results of the multivariate logistic regression model with risk factors found to significantly affect the probability of developing recurrent urethral obstruction

Variable	Regression coefficient	Odds ratio	95% confidence interval	P value
Creatinine	0.195	1.215	1.018–1.475	0.03
Total solids	−1.134	0.322	0.093–0.936	0.05

Table 3

Results from the Cox-proportional hazard model showing no significant difference in time to recurrent urethral obstruction between cats treated with placebo and cats treated with prazosin

Variable	Coefficient	Hazard ratio	95% confidence interval	P value
Treatment group (prazosin vs placebo)	0.626	1.870	0.650–5.385	0.246

Figure 1

Cumulative survival distribution function for time to recurrent urethral obstruction (days) in 65 cats with lower urinary tract obstruction treated with prazosin (n = 37) or placebo (n = 28), including the number of cats at risk in each treatment group

After study completion, a posterior sample size computation using the study’s incidence rate of 25% (α: 0.05; 1-beta [β]: 0.8) found that 199 cats would be required to truly determine statistically significant differences between study groups.

Discussion

In the current study, the rUO rate of all cats (25%) was similar to rates previously reported.^8–10 Between treatment groups, the rUO rate was not significantly higher in cats that received placebo (18%) compared with those that received prazosin (30%). These results compare to the results of a recently published prospective study ¹⁰ evaluating the effect of prazosin on outcome in feline UO. In that study there was no significant difference in rUO rates in cats treated with 0.25 mg q12h prazosin vs placebo. The higher prazosin dose used in this study was the same as the dose used in a retrospective study ¹ wherein cats receiving prazosin at 0.5 mg/cat PO q12h had a lower rate of rUO than cats treated with phenoxybenzamine.

Both the current study and the prior recent prospective study ¹⁰ actually identified a higher rate of rUO in cats treated with prazosin (30% in the current study, 37% at 6 months in Reineke et al ¹⁰ ) vs placebo (18% in the current study, 31% at 6 months in Reineke et al ¹⁰ ). Neither study found this to be statistically significant and both studies were underpowered, and therefore this finding must be interpreted with caution, but it is interesting to note. Future higher-powered studies are necessary to determine if there is a statistically significant trend towards higher rUO in cats treated with prazosin and examine possible causes if this is the case.

The only presenting variable found to be significantly different between cats treated with prazosin vs placebo was the duration of catheterization, with placebo cats having an indwelling urinary catheter in place significantly longer than those treated with prazosin (28 h vs 24 h; P = 0.03). There were a few cats in each group whose urinary catheter was self-removed by the patient slightly early (17–22 h). This was a study limitation in that removal of the urinary catheter occurred during a time range based upon clinical evaluation of the patient rather than at a uniform specific time. However, duration of catheterization was not found to be associated with any significant difference in development of rUO vs not (P = 0.24). A previous study found that a shorter duration of urinary catheterization was significantly associated with a higher probability of rUO, although this study had multiple uncontrolled variables. ⁶ Future studies more specifically evaluating the role of duration of catheterization on development of rUO are warranted.

Prior studies have suggested that cats with UO may develop anemia secondary to bladder hemorrhage related to inflammation and high intraluminal bladder pressure.^13,20 In the present study, cats experiencing rUO had a significantly lower PCV than those that did not. However, as there was no significant difference identified in PCV between the treatment groups in general, it is difficult to place high value on this finding and it is unlikely that this contributed in any major way to development of rUO. Additionally, the clinical implications of the decreased PCV in cats experiencing rUO are likely to have been low, as the median PCV in this group was 43%, well within the RI (35–55%).

Based on the results of multivariate logistic regression analysis, creatinine levels (P = 0.03) and TS (P = 0.05) were found to significantly affect the probability of developing rUO. Increasing creatinine was associated with an increasing probability of rUO, and increasing TS was associated with a decreasing probability of rUO. Prior studies have correlated the degree of azotemia in cats with UO with the length of hospitalization. ²¹ The severity of azotemia is hypothesized to correlate with a higher level of inflammation and hemorrhage within the bladder and kidneys, contributing to more urinary debris, requiring extended hospitalization. ²¹ However, no prior studies have specifically correlated these variables with the risk of rUO and, based upon our limited sample size, these findings should be interpreted with caution.

Sample size was an important limiting factor in this study. Of the 80 cats initially enrolled, 15 had to be excluded owing to loss to follow-up and non-compliance with medication administration. A larger number of these cats were lost from the placebo group than the prazosin group, leading to an imbalance in the study population − 28 cats were in the placebo group and 37 in the prazosin group, as opposed to the even distribution as originally intended. During the study design, 80 cats were initially calculated to be necessary; however, based on a posterior sample size computation using the current study’s incidence rate of 25% (α: 0.05; 1- β: 0.8), 199 cats would need to be evaluated to truly determine statistically significant differences between study groups. This is a smaller population than suggested by previous studies and may therefore be more attainable for future evaluation. ¹⁰ Administration of prazosin does not decrease the frequency of rUO based upon our results, lending support to prior similar conclusions; ¹⁰ however, owing to the underpowered nature of our study, this finding must be interpreted cautiously.

Additional limiting factors included the inability to strictly control for all variables that could contribute to rUO. A standardized analgesic and anesthetic protocol was used for this study, as was standardized placement of a 3.5 F indwelling urinary catheter; however, duration of catheterization – as discussed above – was not strictly controlled. Environmental and dietary factors were also not controlled following discharge. More strict control of variables between groups should be considered in future studies.

Larger, more strongly powered prospective studies are necessary and will help provide further definitive information about the potential benefit of prazosin in management of feline urethral obstruction.

Conclusions

Administration of prazosin 0.5 mg/cat PO q12h for 7 days did not reduce the frequency of rUO within 30 days of treatment for UO in this small population of cats presenting with UO for the first time.