Introduction

A chronic urinary retention (CUR) in non-neurogenic male patients can result from bladder outlet obstruction (BOO) or detrusor underactivity [1]. Symptoms related to CUR are the feeling of incomplete bladder emptying, voiding and storage problems, recurrent urinary tract infections (UTIs), and overflow incontinence [2]. Possible complications are urosepsis, renal impairment, and perineal and skin ulceration [1, 3]. There is much debate in the literature about when treatment is indicated for CUR. Historically, the approach of Abrams et al. [4] has been adopted, who chose a post-void residual (PVR) of > 300 mL based on when the bladder is palpable above the pubic bone. Moreover, detrusor muscle biopsies have shown a degenerative pattern of impaired contractility in men with BOO and a PVR of > 150 mL, suggesting that a lower volume than 300 mL might be clinically relevant [5]. Either way, the relation between symptoms and the PVR volume is unclear, leading Stoffel et al. to suggest categorization according to symptomology (symptomatic and asymptomatic) and complication risk (high and low) [1].

Although treatment may be indicated in symptomatic and/or high-risk patients, no guidelines exist for management of CUR. Initial treatment therefore involves bladder drainage by clean intermittent catheterization (CIC), urethral catheterization (UC), or suprapubic catheterization (SPC), before considering curative treatment of the underlying cause, such as benign prostatic obstruction [6] or bladder underactivity [7]. In asymptomatic, low-risk men, medication or watchful waiting may be appropriate but can be associated with poor outcomes and an increased risk of progression [2]. If patients do not respond to curative treatment attempts, catheterization often remains necessary in the long-term and may lead to complications, decreased quality of life (QoL) [8], and increased diagnostic and treatment costs. To date, research has mainly studied the consequences of catheterization in patients with neurogenic bladder [9, 10] or acute urinary retention [11, 12], and not in men with a CUR.

In this study, we aimed to provide real world data for men with CUR. Since CUR typically is a condition in elderly patients, we expect numbers to increase. Therefore, we find it important to analyze the problems these patients are facing to improve the management of CUR and the quality of life of patients with this chronic condition. Our analysis focused on three elements: (1) the treatments offered, (2) the complications of those treatments, and (3) the consequences of those complications.

Materials

Patient selection

This retrospective cohort study included male patients aged ≥ 18 years with a PVR of > 150 mL and an indication to start treatment because of bothersome symptoms and/or a high risk of complications. Participants who started treatment between January 1st and December 31st, 2014 were identified in a Dutch large non-academic teaching hospital. Case identification was by diagnosis treatment combination code (DBC code) ‘033’ or ‘a- or hypo-contractile bladder’ and ‘041’ or ‘benign prostatic hyperplasia’ in the CTcue® health records search engine (CTcue B.V., Amsterdam, the Netherlands). In the Netherlands, a DBC code is applied to the diagnosis given to each patient [13]. Patients were excluded if they had a PVR for < 6 months to ensure a chronic condition [1], a PVR was diagnosed before inclusion or it was a first episode of urinary retention which did not resolve within 6 months. PVR was measured twice with at least 6 months apart. Other exclusion criteria were a PVR of neurogenic origin, no clinical data, or a percutaneous nephrostomy catheter. Follow-up began on the date of their first treatment and ended on September 1st, 2020.

Data collection

Electronic patient records were analyzed for baseline characteristics and follow-up events, which included treatments (e.g., CIC, SPC, UC, prostate surgery, sacral neuromodulation, or no catheterization), treatment complications (classified by origin), and consequences (e.g., type of hospital contact, additional diagnostics, and additional treatments). Patients could have multiple events during the follow-up period. Patients only received ‘no catheterization’ if they strongly preferred it above active treatment and/or if the treating urologist deemed it safe for the patient. Infectious complications included UTI, urosepsis, epididymitis, and prostatitis, all diagnosed if patients had symptoms and a positive urine culture. Post-renal problems included hydronephrosis on ultrasonography and acute (< 3 months) or chronic (≥ 3 months) kidney failure based on the estimated glomerular filtration rate (eGFR) of < 60 mL/min [14]. Other complications included macroscopic hematuria, bladder calculi, catheter problems, urethral stricture, and catheter-related pain. Surgical complications were defined as those appearing within 30 days after surgery.

Statistical analysis

Data are presented as means ± standard deviations if normally distributed, as medians and ranges if skewed, and as numbers and percentages if categorical. To visualize the treatment sequence, we created a Sankey plot in Display R (www.displayr.com). Logistic regression analysis was used to identify associated factors. All variables showing a univariate association (p < 0.3) were entered in the multivariable regression, and variables with the highest p-value were subsequently excluded until all independent variables had a p-value of < 0.05. All complications and consequences, except those related to surgery, were expressed as incidence rates in patient-years. Each treatment step was managed as an individual element. The observed number of complications was divided by the total number of follow-up years from the treatment option. Fisher’s exact mid P test was used to calculate the incidence rate, the incidence rate ratio (IRR), and the associated 95% confidence intervals (CIs). We used IBM SPSS version 25.0 (IBM Corp., Armonk, NY, USA) for all analyses and considered 2‑tailed p-values < 0.05 statistically significant.

Ethics

The study was submitted to a legally constituted ethics committee, the Medical Ethical Research Committee of Isala Clinics Zwolle, which waived the need to obtain informed consent. The study received no funding. Declarations of interest: none.

Results

Descriptive statistics

We identified 900 patients in 2014 with the applicable DBC codes in CTcue®. Of these, 177 patients were eligible and the remainder were excluded because their PVR was < 150 mL (n = 422), had been present < 6 months (n = 258), or was neurogenic in origin (n = 27). A few patients were excluded because of a lack of clinical data (n = 12) or because they received a nephrostomy catheter for a reason other than PVR (n = 4).

The baseline characteristics are shown in Tab. 1. The patients had a median age of 77 years (range, 44–94), a mean International Prostate Symptom Score (IPSS) of 17 ± 8, and an IPSS-QoL score of 3 ± 1.5 (n = 36). The median PVR was 332 mL (range, 150–1200) with a maximum flow rate of 6.3 mL/s (range, 1.2–30.3), and a voided volume of 140 mL (range, 9–787). Other patients had an acute urinary retention at the time of inclusion and were not able to void (n = 30), uroflowmetry was not requested because the urologist did not find it contributory, or patients refused it (n = 26), or patients first received treatment and uroflowmetry was performed later on in the follow-up period (n = 68). Only 35 patients had a urodynamic study sometime during the follow-up. Prior to any treatment 24 patients had some type of post-renal problems e.g., hydronephrosis or acute kidney failure. During the median follow-up of 68 months (range, 1–319), patients had a median of 8 events (range, 1–51) and 75 (42%) died, though no deaths were related to a urological problem. Follow-up data were not available for 6 patients (4%) who received treatment elsewhere.

Table 1 Baseline patient characteristics
Full size table

Treatments

Most patients first underwent UC (74%). The final treatments were more diverse with SPC, CIC, and UC in 41, 35, and 24%, respectively. However, most patients (87%) were still dependent on a form of catheterization at the end of the follow-up period. In the Sankey diagram (Fig. 1) each column represents one treatment step, and the colored flows between the columns represent the patients’ transition from one treatment step to the next. The diagram shows that during follow-up, patients switched between the different catheterization types, periods with no catheterization, and prostate surgery. Patients typically required 3 treatment steps (range 1–18) before their final treatment. After initial treatment with CIC, one patient underwent sacral neuromodulation. Forty-nine patients (28%) underwent de-obstructive prostate surgery of the prostate (32 had one surgery, 13 had two, and 4 had three). Surgery involved transurethral resection of the prostate (TURP) (n = 51), GreenLight™ laser therapy (n = 15), diode laser vaporization (n = 1), or bladder neck incision (n = 3). Patients who received de-obstructive surgery were more likely stop catheterization (33%) than patients who did not have surgery (6%) (p < 0.001; odds ratio, 4.18; 95% CI, 2.38–7.35). Multivariable regression revealed that no baseline characteristic significantly affected the final treatment option.

Fig. 1
figure 1

Sankey diagram of management option step 1–7: each column represents one treatment step, and the colored flows between the columns represent the patients’ transition from one treatment step to the next. CIC clean intermittend catheterization, No cath no catheterization, SNM sacral neuromodulation, SPC Suprapubic catheter, UC urethral catheter. Note: treatment step 7–18 not included for better visualization. Percentages at step 7 are almost equal to the last treatment step, 92% of patients stay at the same treatment from step 7 forward. Note: Percentages are rounded, therefore, not all treatment steps equal 100%

Full size image

Complications

The patient who received sacral neuromodulation had no recorded complications. Regarding the other surgery-patients, 36 complications were recorded in the 30 days after prostate surgery. The most common were UTI (n = 17) and hematuria (n = 11). Other complications included frequency, urgency or urge incontinence (n = 7), and a bladder perforation (n = 1).

The incidence of complications for catheterization and no catheterization are presented in Tab. 2. Catheterization (i.e., SPC, UC, and CIC) resulted in significantly higher incidence rates for UTI (IRR, 3.68; 95% CI, 2.92–4.69; p < 0.001) and hematuria (IRR, 5.35; 95% CI, 2.29–15.12; p < 0.001) compared with no catheterization. However, no catheterization was associated with a much higher incidence of post-renal problems compared with catheterization (IRR, 25.36, 95% CI: 8.73–103.70; p < 0.001). Post-renal problems were present in 65% of cases before the start of the first treatment.

Table 2 Incidence of complications by management approach for men with chronic post-void residuals
Full size table

Comparing the different forms of catheterization revealed that CIC had a significantly lower incidence of all complications compared with UC and SPC (p < 0.01), except for hematuria. In addition, SPC had a significantly lower incidence of all complications compared with UC (p < 0.05), except for catheter pain.

Consequences of complications

Surgical cases required additional diagnostics, such as urine culture (n = 17), blood culture (n = 2), cystoscopy (n = 2), and computed tomography (n = 1), as well as additional treatments, such as blood transfusion (n = 1) and surgery to remove blood clots from the bladder (n = 6). Most complications were managed in outpatient (n = 15) or emergency (n = 3) departments, and some (n = 10) needed admission.

The consequences of complications that resulted from catheterization or no catheterization are displayed as incidence rates per patient-year in Tab. 3. Medical therapy included antibiotics (n = 685) for infection and antimuscarinics (n = 37) for catheter-related pain caused by bladder cramps. Some patients received blood transfusions (n = 10) for macroscopic hematuria. Surgical treatments included blood clot removal from the bladder, cystolithotripsy, treatment of urethral strictures, and intravesical injection of botulinum toxin A. Nephrologists were consulted in 8 cases that developed chronic kidney failure. Elective outpatient visits were the most common type of hospital contact (n = 726), with the remaining patients needing hospitalization (n = 157), emergency department visits (n = 52), or intensive care unit admissions (n = 2).

Table 3 Consequences of complications by Management Approach for Men with Chronic Post-void Residuals
Full size table

Conclusion

Implications

CUR is a condition in patients with increasing numbers due to the aging population in many countries. There is debate on treatment indications and patient selection to therapy. Management can be invasive e.g. de-obstructive prostate surgery, sacral neuromodulation, or more conservative e.g. a form of catheterization. To draw attention to the chronic condition of patients with CUR, we have analyzed the burden of treatments and complications of these treatments. Any form of catheterization places a considerable burden on patients with a symptomatic and/or high-risk CUR. In general, the resulting complications require additional diagnostics and treatments that increase costs. Our findings supplement existing research in different cohorts [9, 10] by providing data for a large cohort of men with CUR of non-neurogenic origin.

In patients who are dependent on catheterization, CIC is widely accepted as being preferable to indwelling catheters. Studies show that it has a lower complication rate [8] and is more effective and practical [10]. Our data support this position, indicating that CIC is better than any type of indwelling catheter in terms of overall complications. Compared with UC, a SPC was also associated with significantly lower incidence rates of UTIs, hematuria, and catheter-related problems, consistent with studies in patients with acute urinary retention [11, 12]. Although the International Continence Society states that SPCs offer easier access and may be more comfortable, our findings conflict with their stance that complications do not differ between UC and SPC [8].

In some patients, catheterization may be avoided by treating the underlying cause of CUR. For a long time, TURP has been the cornerstone of surgical treatment for bladder outlet obstruction due to benign prostatic obstruction [2]. Because a diagnosis of bladder outlet obstruction requires adequate detrusor contractility to measure pressure flow, absent or reduced contractility in patients with detrusor underactivity can exclude them from de-obstructive prostate surgery. Some authors have advised avoiding TURP in patients with detrusor underactivity because of the risks of surgery and the possibly of disappointing outcomes [15, 16]. Others have confirmed that these patients perform worse after TURP than patients without detrusor underactivity, but that improvements are possible compared with no action [17, 18]. In our study, patients who underwent de-obstructive prostate surgery were significantly more likely to have catheterization stopped. The surgical complications for TURP in this study were comparable to those reported in a previous meta-analysis [19]. No other factors, such as BMI or age, were found to have influence on the likelihood to stop with catheterization in our cohort.

A study concluded that high-risk complications are uncommon when using watchful waiting for patients with non-neurogenic CUR who refuse intervention, and that this approach is safe for at least 5 years [20]. They reported post-renal problems in 21%, UTIs in 14.5%, and bladder calculi in 7%; however, their population was much smaller, was younger, and included PVRs > 300 mL. Our data show that patients not treated with catheterization have a lower chance of UTI and hematuria compared with catheterized patients, but that they have a higher chance of post-renal problems. Although this resulted in chronic kidney failure requiring long-term treatment by a nephrologist in some, it should be noted that most patients had post-renal problems before their first treatment (65%).

Limits and perspectives

To the best of our knowledge, no prior study has recorded the long-term consequences (e.g., type of hospital contact, additional diagnostics, and additional treatments) of complications in this group of patients. However, some important limitations should be considered. First, this was a retrospective single center study for which translating results to a larger population will be vulnerable to bias. Second, there may be information bias when complications are treated by a general practitioner, or when information is not systematically documented. Third, only a small fraction of the patients underwent a urodynamic assessment (n = 35), of which 7 patients underwent de-obstructive prostate surgery. The decision hereof was made by the urologist. Because of this small number, we did not include the findings of the urodynamic studies in the results. However, it would be interesting to take these findings into considering especially regarding prostate surgery. Lastly, the absence of a validated and clinically relevant PVR is an important limitation when studying CUR. We chose a PVR of > 150 mL because degenerative changes in the detrusor are shown at this volume [5] and patients were often symptomatic or at increased risk of complications at this volume within this cohort. The median PVR volume was 332 mL (range 150–1200), confirming that men in this cohort are comparable with earlier reports on CUR [4]. Nevertheless, further research is needed to validate the optimal PVR threshold in relation to symptoms, risks, and treatment response.

Conclusions

Various treatment options are available for men with symptomatic and/or high-risk CUR. Most men are usually treated with some form of catheterization, and among the available forms, CIC causes the fewest complications. Additional diagnostics and treatments are typically required when complications develop. Finally, de-obstructive prostate surgery increases the likelihood of patients being able to stop catheterization, indicating that there is an argument for patients to be offered this surgical option.