Bladder outlet obstruction (BOO) in men with castration-resistant prostate cancer

Authors


Abstract

Objective

  • To evaluate the frequency of bladder outlet obstruction (BOO) and detrusor overactivity (DO) in patients with castration-resistant prostate cancer (CRPC) and lower urinary tract symptoms (LUTS).

Patients and Methods

  • Our prospective urodynamics database was queried.
  • Inclusion criteria were CRPC and an International Prostate Symptom Score (IPSS) ≥ 20.
  • Exclusion criteria were previous local therapy to the prostate gland, known urethral stricture disease, and a neurological component of LUTS.
  • Twenty-one patients were identified.
  • Urodynamic findings were analysed and compared with those of a matched cohort of 42 patients with benign prostatic enlargement (BPE).

Results

  • The median age of patients in the CRPC group was 74 years, and the median prostate-specific antigen (PSA) level at the time of the urodynamic study was 90 ng/mL.
  • According to the BOO index, three patients (14%) were obstructed, three were equivocally obstructed (14%) and 15 were unobstructed.
  • DO was seen in 12 patients (57%).
  • Compared with the BPE group, patients with CRPC had lower cystometric bladder capacities (P = 0.003), were less likely to have BOO (14 vs 43%, P = 0.009) and more likely to have DO (57 vs 29%, P = 0.028).

Conclusions

  • This study generates the hypothesis that only a minority of CRPC patients with LUTS have BOO, and that more than half of patients have DO.
  • LUTS in CRPC may therefore be seldom attributable to BOO, but are, at least in part, related to DO and reduced cystometric capacity.
  • A urodynamic investigation may be necessary before palliative transurethral resection of the prostate to select appropriate candidates.
  • Larger prospective studies are needed to confirm our findings.

Introduction

Castration-resistant prostate cancer (CRPC) causes significant morbidity to the lower urinary tract, including haematuria, UTI, urinary retention and LUTS [1, 2]. For treatment of LUTS and urinary retention, palliative TURP is often carried out. Palliative TURP is considered a safe procedure and improves LUTS and the average urinary flow rate in selected patients; however, the risks of postoperative urinary retention, re-operation, urinary incontinence and long-term catheterization are substantially higher for these patients than for patients treated with TURP for BPH [3-6]. In recent studies, up to 43% of patients who underwent palliative TURP failed the initial voiding trial, 11–21% required chronic drainage, and up to 29% underwent re-TURP [3-5]. Overall, this treatment does not convey palliative benefit for ∼40% of patients [4]. The reasons for these poorer outcomes of palliative TURP are largely unknown. We speculated that an appreciable proportion of patients with LUTS and CRPC do not have urodynamic BOO, which may explain the limited benefit of this procedure. This hypothesis is underscored by a recent study, which suggested that outcomes of palliative TURP are improved by patient selection with preoperative urodynamics [7].

In the present study, we report on a consecutive group of patients with CRPC and LUTS, and compared the urodynamic findings with a matched cohort of patients with LUTS and benign prostatic enlargement (BPE).

Patients and Methods

Patient Population and Study Design

Since 2007, data on patients referred for an urodynamic investigation have been collected prospectively in a database. After approval by the review board of the Medical University of Vienna (protocol registration number 1346/2013), we studied this database retrospectively. Patients with CRPC and an IPSS ≥20 were eligible. CRPC was defined according to criteria set by the Prostate Cancer Working Group [8]. Patients with previous local therapies to the prostate gland (e.g. radical prostatectomy, radiotherapy, brachytherapy, TURP, thermotherapy), known urethral stricture disease, or a neurological component of LUTS (as defined by an abnormal neurological examination) were excluded. A total of 21 patients were identified.

The goal of the present study was to compare the urodynamic findings in patients with LUTS and CRPC with those with LUTS and BPE (prostatic volume >30 mL). To accomplish this, each patient with CRPC was matched with two patients referred for an urodynamic investigation of LUTS in conjunction with BPE. Similarly to the CRPC cohort, patients with previous local therapies to the prostate gland, known urethral stricture disease or a neurological component of LUTS, were excluded. All patients had an IPSS ≥ 20, a PSA level less than the age-specific reference range [9] and a negative DRE. Patients were matched according to age and IPSS using matching based on the propensity score.

Multichannel Urodynamic Investigation

It is our standard practice to perform urodynamic studies in every patient who has CRPC with severe LUTS, if they are candidates for palliative transurethral therapy. The urodynamic studies were undertaken by two of the authors (M.R and T.K.) according to the ‘good urodynamic practices’ of the ICS [10]. An 8-F transurethral double-lumen catheter and a rectal balloon catheter were used to measure the intravesical and intra-abdominal pressure, respectively. Normal saline (0.9%) at 37°C was filled at 30–50 mL/min. The maximum cystometric capacity, presence of detrusor overactivity (DO), maximum urinary flow rate (Qmax), and detrusor pressure at maximum urinary flow rate (pdetQmax) were recorded. The degree of obstruction was measured using the BOO index (BOOI = pdetQmax – 2Qmax). The degree of contractility was measured using the bladder contractility index (= pdetQmax + 5 Qmax) [11]. Urodynamic data were reviewed independently by two of the authors (M.R and T.K.).

Statistical Analyses

Continuous data are presented as median values and interquartile ranges (IQRs). Categorical data are presented as numbers and proportions. Differences in factors that were not matched were evaluated with the Kruskal–Wallis and chi-squared tests or Fisher's exact tests. Linear regression models were fitted to identify clinical predictors of continuously coded BOOI in patients with CRPC. All statistical testing was two-sided, and a P-value <0.05 was considered to indicate statistical significance. The statistical package stata 12 (College Station, TX, USA) was used for all analyses.

Results

Characteristics

The median age of the 21 patients was 74 (68–82) years, and the median (IQR) PSA at the time of the urodynamic study was 90 (36–362) ng/mL. Bone metastases were present in 15 patients (71%). All patients were on LHRH analogues, and 12 (57%) had received docetaxel-based chemotherapy. None of the patients in either group was receiving anticholinergic medication or 5-α reductase inhibitors. Five patients (24%) in the CRPC group were receiving α-blockers, but this was not significantly different from the control group (P = 0.63). Seven patients had a history of urinary retention in the past 6 months, but had undergone a successful trial without catheter. At the time of the urodynamic study, none of the patients had gross haematuria or a UTI.

Urodynamic Findings

According to the BOOI, three patients (14%) were obstructed, three were equivocally obstructed (14%) and 15 (71%) were unobstructed. DO was seen in 12 patients (57%). Both obstruction and DO was noted in two patients with BPE, but in none of those with CRPC. In univariable linear regression analyses, age (P = 0.22), PSA level (P = 0.50), chemotherapy (P = 0.56) and presence of bone metastases (P = 0.72) were not associated with the BOOI.

Compared with patients who had BPE (Table 1), those with CRPC were less likely to have BOO (14 vs 43%, P = 0.009) and more likely to have DO (57 vs 29%, P = 0.028). The median cystometric bladder capacity (252 vs 348 mL, P = 0.003) and the median bladder compliance (25 vs 29 mL/cmH2O, P = 0.049) were significantly lower.

Table 1. Clinical and urodynamic variables in patients with CRPC and LUTS (n = 21) and those with BPE (n = 42). Patients with CRPC were less likely to have BOO (P = 0.023) and more likely to have DO (P = 0.028)
VariableCRPCBPEP
  1. *Obstructed vs not obstructed.
Median (IQR) age, years74 (68–82)73 (65–79)0.38
Lower urinary tract morbidity, past 6 months, N (%)   
Gross haematuria4 (19)2 (5)0.09
UTI1 (5)3 (7)1.00
Urinary retention3 (14)4 (10)0.68
Stones0 (0)0 (0)1.00
Hydronephrosis1 (5)0 (0)0.33
Diabetes mellitus, N (%)4 (19)9 (21)1.00
Median (IQR) PSA, ng/mL90 (36–362)3.0 (1.6–4.4)<0.001
Median (IQR) BOOI8 (0–29)34 (10–50)0.013
BOO classification, N (%)  0.009
Obstructed3 (14)18 (43)0.023*
Equivocally obstructed3 (14)11 (26)
Unobstructed15 (71)13 (31)
DO, N (%)12 (57)12 (29)0.028
Detrusor hypocontractility (bladder contractility index <100), N (%)19 (90)33 (79)0.24
Median (IQR) Qmax, mL/s3 (1–5)2 (0–5)0.39
Median (IQR) pdetQmax, cmH2O10 (0–40)40 (20–61)0.026
Median (IQR) bladder capacity, mL252 (190–347)348 (293–490)0.003
Median (IQR) post-void residual urine volume, mL150 (95–282)181 (80–280)0.52
Median (IQR) compliance, mL/cmH2O25 (23–30)29 (25–34)0.049

On staging CT, extensive bladder infiltration (T4), pelvic lymphadenopathy and hydronephrosis were noted in six (21%), four (19%) and one patient (5%), respectively. None of these patients had BOO. DO was identified in three (50%), two (50%), and one patient, respectively.

Discussion

In the present study in patients with CRPC and LUTS, <15% of subjects had BOO and >50% had DO; these data suggest that LUTS in CRPC are seldom attributable to BOO, but may, at least in part, be related to DO and reduced cystometric capacity.

There are various treatment options for patients with prostate cancer with LUTS: surgery, systemic therapy, catheterization and α-blockers. Palliative TURP is frequently used in those who are not candidates for radical prostatectomy. Studies have shown that palliative TURP is a safe procedure with acceptable peri-operative morbidity [3, 5, 6]. Some authors have suggested a negative impact on oncological outcomes [12-14], whereas others have reported no impact [15]. From a functional aspect, the risks of postoperative urinary retention, re-operation, urinary incontinence and long-term catheterization are markedly high [3-6], and are greater in men who have progressed to CRPC [4]. The present study focused on this particular group, and did not include patients with prostate cancer of other stages because LUTS may also be induced by the benign prostatic component (especially in those with low-volume prostate cancer). In such patients, TURP is similarly termed ‘palliative’, but outcomes may be more similar to those for patients with benign prostatic disease who have undergone surgery.

The functional outcomes of palliative TURP can be improved by appropriate selection of patients. Gnanapragasam and Leonard [5] carried out preoperative urodynamics in 41 patients, and selected 19 cases with BOO for subsequent palliative TURP. Compared with the existing literature, the functional outcomes were markedly better [5]. The authors observed BOO in 49% of patients, which is higher than the 21% rate reported in the present study. Different inclusion criteria are likely to be responsible for this result because Gnanapragasam and Leonard [5] included patients at all stages whereas we chose only those with CRPC. We hypothesize that, with disease progression and subsequent progressive invasion of the bladder, a true BOO becomes less common. This is supported by the fact that none of the patients with extensive bladder infiltration, lymphadenopathy and hydronephrosis showed BOO. The number of patients in these groups is low, and no meaningful statistical conclusions can be drawn. Prospective evaluation of these variables and their association with urodynamic variables could be a direction for further research.

The high prevalence of patients with urodynamic DO was comparable with that from the study by Gnanapragasam and Leonard [5]. Furthermore, we found that patients with CRPC had lower bladder capacities and compliances than those with BPE, which suggests that LUTS in CRPC are seldom attributable to BOO, but may be related to DO and reduced cystometric capacity. We conclude that a urodynamic investigation may be helpful before palliative TURP to select appropriate candidates. This may be particularly important in patients with CRPC because survival time is limited and improving quality of life is the main therapeutic goal.

We found a high prevalence of detrusor hypocontractilty in the CRPC group. The inability to demonstrate obstruction in those patients may be secondary to primary detrusor failure. It has been shown that patients with a compromised detrusor function may also benefit from deobstructive surgery [16-18]. If we suspect obstruction in the presence of hypocontractility, patients who would be considered to be candidates for TURP would undergo cystoscopy. This diagnostic approach, however, is a matter of debate. Among patients with DO, we have different treatment approaches. If micturition is possible with acceptable post-void residual urine volume, we start with anticholinergic medication. Other patients undergo a channel-TURP, if cystoscopy shows obstructive tissue, and others (hypocontractility and DO) perform clean intermittent self-catheterization and are started on anticholinergic medication.

The present study has several limitations. Because of its retrospective design, several variables were not recorded in a standardized way. There were relatively few patients, which precludes extensive subanalyses and multivariable analyses. We did not include outcomes of palliative TURP because these results are confounded additionally by the surgeon, the surgical method, and the amount of resected tissue. In addition, outcomes of TURP were not evaluated in a standardized way. A prospective study is ongoing. Rather, we focused on urodynamic variables because these are objective measures that predict TURP outcomes [19, 20]. At our institution, it is our practice to perform urodynamic studies in every patient with CRPC, if they have severe LUTS and are candidates for palliative TURP; however, this group is subject to selection bias, as it only comprises a small proportion of patients with CRPC. That is supported by the fact that few patients showed pelvic lymphadenopathy and hydronephrosis. Likewise, patients with BPE are unlikely to have urodynamic studies as part of their routine evaluation, thereby introducing selection bias; however, it is likely that, among classical patients with LUTS and BPE, the proportion of subjects with BOO would be even higher and the proportion with DO even lower. A prospective study providing the highest level of evidence would be extremely helpful to eliminate biases and to allow high grade recommendations. Despite these limitations, we think that the present study shows that BOO is seldom observed in men with CRPC, and that LUTS may instead be related to DO and the reduced bladder capacity.

In conclusion, this study generates the hypothesis that only a minority of patients who have CRPC with LUTS have BOO, and that more than half of patients have DO. The data suggest that LUTS in CRPC is seldom attributable to BOO, but is, at least in part, related to DO and the reduced cystometric capacity. A urodynamic investigation may therefore be necessary before palliative TURP to select appropriate candidates. Larger prospective studies are needed to confirm our findings.

Conflict of Interest

None declared.

Abbreviations
DO

detrusor overactivity

CRPC

castration-resistant prostate cancer

BPE

benign prostatic enlargement

Qmax

maximum urinary flow rate

pdetQmax

detrusor pressure at maximum urinary flow rate

BOOI

BOO index

IQR

interquartile range

Ancillary