Clinical benefits of retrograde bladder filling method prior to catheter removal after TURP for BPH: A prospective randomized trial

Transurethral resection of the prostate (TURP) is a gold standard for surgical intervention in men with benign prostatic hyperplasia (BPH) [1], with various interchangeable techniques including transurethral enucleation with bipolar (TUEB) producing comparable outcomes [2]. However, some patients after those surgeries are prone to fail initial voiding trials without catheters (TWOCs), with literature suggesting up to 12% of men undergoing TURP to experience acute urinary retention, as well as risk of complications including severe hematuria, urethral stricture, and infection [1, 3]. As such, accurate assessment of voiding function at initial void after catheter removal is critical for decision for discharge, especially in patients with a previous history of chronic retention [3].

Awaiting initial TWOC for evaluation for adequacy of discharge can be a time and resource consuming process in both inpatient and outpatient settings, and retrograde bladder filling with normal saline or sterile water has been attempted to expedite the process [4, 5, 6]. However, previous randomized clinical trials (RCTs) have produced statistically insignificant results, with limited literature on men with urological transurethral surgery. Moreover, research on patient’s satisfaction regarding the process is lacking and more so on the healthcare provider’s perspectives. No consensus has been established for the benefits and risks in different urinary catheter removal techniques. As such, this study intends to compare and evaluate clinical outcomes in patients undergoing removal of intra-vesical catheter by either retrograde fill (RF) or spontaneous voiding (SV) after transurethral prostatectomy (TURP) for BPH. Procedure-related questionnaires were used to assess patient and medical staff satisfaction.

This was a single institution, randomized prospective trial initiated after approval by the Institutional Review Board (no. B-1806-474-004) and registered with ClinicalTrials.gov (NCT04309032). Written informed consent for each participant was obtained after approval. The study is described adhering to the most recent CONSORT statement guidelines (Supplementary Table 1) [7]. Eligible patients included males between ages 20 and 80 undergoing endoscopic prostate resection for BPH and/or lower urinary tract symptoms from November, 2018 to October, 2019. Otherwise healthy patients who required intra-vesical catheter indwelling after surgery and removal prior to discharge were included. Those with history of intermittent catheterization due to neurogenic bladder and non-removal of catheter prior to discharge were excluded. Out of 83 eligible patients within the 1-year study period, 56 patients were enrolled. Randomization of patient allocation was established via online generator (http://www.randomization.com/), with 28 patients equally distributed (1:1) to the SV (non-filling) and RF group. All 28 patients in the RF group were included for final analysis; 2 patients in the SV group were excluded due to patient request for discontinuation. No subjects were excluded in the final analysis due to other non-urologic clinical needs. A flow diagram based on the CONSORT diagram is available (Supplementary Fig. 1). Sample size was calculated based on a type 1 error (α) 0.05 and power (β) 0.8 to detect a 26.9-minute difference in discharge based on the results from a previous study [8], with total 48 patients required with minimal 24 participants in each group [9].

Patients underwent TURP by 3 experienced urologists with either bipolar TURP or TUEB, depending on the surgeon’s preference and discretion. Surgeons were blind to patient allocation, with groups identified at the day of intra-vesical catheter removal. Intra-vesical catheters were removed at post-operative day 2 to 8 (mean 3.1, median 3 days), with baseline at 3 days and varied depending on patient’s medical needs (e.g., gross hematuria, lab abnormality, and patient’s general condition). Immediately prior to catheter removal, patients in the RF group were filled with normal saline up to 250 mL or until patients’ filling of moderate urge to void. RF groups were instructed to void within 1 hour after catheter removal and if not, to check the intra-vesical volume by ultrasound to prevent severe urinary retension. SV subjects were removed without retrograde filling, and instructed to void within 3 hours or when sufficient desire-to-void was reached. Patients in both groups were measured for voided volume (VV) and residual urine (RU) via bladder scan, based on which the attending physician determined whether the patient was fit for discharge. Patients were instructed to attempt secondary voiding trial if VV was inadequate for accurate assessment of urinary retention (VV<150 mL) or if RU>150 mL was identified. Subjects unable to void after two trials were re-catheterized and instructed to return to the outpatient clinic for catheter removal and voiding trials within 1 week. Time-to-void and time-to-discharge were measured, as well as satisfaction surveys for patients and medical personnel involved in the overall process, including physicians, nurses, and physician assistants (PAs) before discharge. Time-to-discharge was set at the point where the attending clinician determined whether the patient was fit for discharge without further intervention. Patient satisfaction consisted of 8 questions on a 5-point Likert scale, asking if the overall removal process was conducted smoothly, causing a delay, was painful, was helpful to treatment, assisted safe voiding, or accelerated discharge, as well as patient needs to utilize the same procedure in the future. Medical personnel were asked similar questions regarding the smoothness of the RF process and opinions on potential delay and pain during the procedure, as well as professional opinions on whether the RF method can help early detection of urinary retention. Overall satisfaction of the RF method was scored in a 10-point scale in both patients and staff questionnaires.

Routine physical evaluation, history taking, and lab tests were conducted prior to surgery, as well as assessment of uroflowmetry (UFM). Prostate volume (PV) was measured with transrectal ultrasonography. All patients were returned to the outpatient clinic within 2 weeks after discharge, and assessed for voiding function via UFM and prostate specific antigen (PSA) change. Routine follow-up was conducted based on physician preference and histopathology, usually at 3 to 6 months intervals. Student t-test and Pearson chi tests were used for statistical analysis of continuous and categorical variables, respectively. Logistic regression was performed to analysis complication risk and urinary retention. All statistical analyses were conducted using IBM SPSS software package version 22.0 (IBM Corp., Armonk, NY, USA), with p-value <0.05 considered significant.

Baseline and perioperative characteristics are described in Table 1. No significant differences were observed in clinical parameters, including age, body mass index, diabetes mellitus/hypertension, 5-alpha reductase inhibitors (5-ARI) usage, pyuria, UTI, PV, and PSA. Bladder filling groups were filled with mean 227.0±37.5 mL urine prior to removal. While no significant differences were observed between the RF and SV groups in VV (252.3±82.0 vs. 228.5±148.3, respectively) and RU (57.8±68.0 vs. 94.8±102.2, respectively), time-to-void and time to discharge were significantly shorter in RF patients, with SV patients taking 77 minutes more to void (p<0.001) and 50 minutes more to be discharged from clinic (p=0.046; Table 2). Operation specifics including operation time, concurrent TRUS-bx in the operating room, concurrent urologic surgery (cystolitholapaxy or bladder neck resection), and pathologic findings were similar between the two groups.

Table 1
Baseline characteristics between two groups

• Click for larger image
• Click for full table
• Download as Excel file

Table 2
Post-operative outcomes between two groups

• Click for larger image
• Click for full table
• Download as Excel file

Rate of immediate postoperative complication and reintervention within 3 months of surgery were assessed. Rate of overall complications were not statistically different, with 29.6% (n=16) of overall patients suffering from any complications, with 10 and 6 patients in the RF and SV group, respectively (p=0.150; Table 3). Out of 5 RF patients who had grade ≥3 complications, 3 underwent coagulation for bleeding control. Rate of reintervention within a follow-up window of 3 months showed similar outcome, with 6 and 4 patients undergoing additional operation such as EIU or serial dilation for stricture or urinary retention (p=0.568). More patients in the SV group (5 out of 26) were discharged with catheter reinserted due to voiding failure, compared to only 1 patient in the RF group, although the difference was not statistically significant (p=0.067). On regression analysis to predict inadequate initial VV (<150 mL), only RF was significantly associated with decreased risk of void trial failure (odds ratio [OR], 0.059; 95% confidence interval [CI], 0.007–0.507; p=0.010) as opposed to insignificant associations with age, PV, and preoperative Qmax (p=0.695, 0.791, and 0.250, respectively). The postoperative catheterization duration did not influence postoperative Qmax >15 mL/s (OR, 0.689; p=0.128) or Qmax improvement (OR, 1.521; p=0.226) when weighted by age and PV. Only age was an independent predictor of Qmax >15 mL/s after surgery (OR, 0.812; 95% CI, 0.700–0.943; p=0.006).

Table 3
Immediate postoperative complication and 3-month reintervention rates

• Click for larger image
• Click for full table
• Download as Excel file

Comparison of post-removal satisfaction showed no difference in overall satisfaction (p=0.883) nor in questions regarding pain (p=0.488) or procedural delay (p=0.957) (Fig. 1), as well as subjective patient opinion on assisting voiding (p=0.520). However, patient response in the RF group showed significantly higher scores regarding assisting discharge (p=0.001), suggesting patients may favor RF for facilitating discharge despite no difference in discomfort. While general distributions in survey response were similar in healthcare personnel concerning RF (Supplementary Fig. 2), physicians were more prone to use RF for early detection of acute urinary retention (AUR) than nurses or PAs despite similar scores in overall satisfaction.

Fig. 1
Patients’ response on the questionnaire about the clinical usefulness of retrograde bladder filling method.
^*Statistically significant p-value <0.05.

• Click for larger image
• Download as PowerPoint slide

In this prospective randomized trial, patients undergoing TURP for BPH were evaluated for differences in voiding function and postoperative complications. The RF method resulted in lowered risk for inadequate volume at initial voiding trials and thus significantly facilitated time-to-void and time-to-discharge by at 77 minutes and 50 minutes, respectively, without increasing procedure related pain or discomfort. Postoperative complication and intervention rates after 3 months of follow-up were similar to SV, suggesting RF is non-inferior to SV in evaluating voiding function and patient outcome. Patient and clinician satisfaction for RF were similar to SV, with no overt preference for either method.

Urinary retention and voiding failure are common complications in patients undergoing surgery or catheter indwelling for genitourinary disorders such as BPH, organ prolapse, or incontinence [10, 11, 12]. Previous reports on BPH men after TURP documents rate of AUR at initial voiding trials from 1% to 12%, with preoperative chronic retention, high PV, and low resection rates to be risk factors [3, 12]. Patients suffering from AUR are more likely to experience treatment failure and require indwelling or intermittent catheterization, hence requiring accurate assessment of voiding function prior to discharge [13]. While RF is common in real-world practice, literature is limited with mixed results, especially in male patients. Wilson et al. (2000) [4] tried evaluate the patients’ readiness for discharge between the RF and SV groups in 75 patients undergoing TURP. However, they only compared the discharge rate between the two groups from postoperative first day to tenth day without unified discharge plan for entire patients and no comparisons were made with time measurement from catheter removal to discharge readiness in their study. Similarly, an outpatient RCT including 15 TURP and 7 radical prostatectomy patients found RF patients to have significantly shorter time to discharge and increased catheter-free state (OR, 1.56; 95% CI, 1.03–2.36; p=0.03) [5]. On the other hand, the most recent study by Du et al. (2013) [6] found no significant difference by either method in men undergoing TURP, bladder neck incision, or admission for retention. Even small in number, the present study clearly demonstrated that the RF method had significant decrease in the time to void and time to discharge, which can be clinically beneficial for both patients and clinicians, which makes our study more clinically meaningful.

On the other hand, studies in female subjects have produced similarly inconsistent results. A RCT comprised of 40 women undergoing mid-urethral sling procedures showed no difference in discharge rate (p=0.71) nor in time-to-void (p=0.41) and time-to-discharge (p=0.43) [14]. Only success of void defined as RU ≥150 mL was significant with increased risk in the SV group (OR, 4.5; 95% CI, 1.2–17.4; p=0.025), a result that was paralleled in our study with decreased risk of void failure in the RF group (OR, 0.059; 95% CI, 0.007–0.507; p=0.010). A similar trial by Foster et al. (2007) [8] had previously found RF to decrease time in the anesthesia unit in transvaginal surgery patients (n=60) from 226.6 minutes for the SV group to 199.5 minutes in the RF group, despite failing to achieve statistical significance (p=0.08). However, the authors similarly found that RF was an independent predictor of successful voiding defined as VV ≥200 mL. A relatively large RCT in 153 laparoscopic hysterectomy patients found RF to reduce time to first void by 25 minutes but not to discharge [15]. In another study, RF was more accurate for detecting postoperative urinary retention, with decreased rate of voiding failure in 50 females undergoing surgery for pelvic organ prolapse, incontinence, or both (62% vs. 84% for RF and SV groups, respectively) [16], similar to a former study by Pulvino et al. (2010) [17] where RF was better correlated with successful void.

Findings from our study support the argument that RF is a safe procedure with no increased risk of complications in a 3-month period, able to promote faster void and discharge with statistical significance. One of the strengths of this study is that all patients were limited to a single type of surgery (TURP for BPH) with inclusion of a wide range of clinical parameters including preoperative urine analysis, PSA, PV, comorbidities, and usage of 5-ARIs. Also, a detailed questionnaire survey was acquired from patients as well as medical staff involved in RF, offering an assessment of both patient and provider satisfactions with the overall procedure, the latter of which is especially important as RF can be a cumbersome additional procedure for medical personnel. Aside from surgical and administrative procedures that can be undertaken to facilitate discharge, RF is the one method that physicians can elect to perform that can directly affect length of hospital stay in actual practice. However, the safety and feasibility of such method has not been adequately discussed, hence the need for this randomized trial. Our results show that RF is non-inferior to SV and does not sacrifice patient nor clinician comfort for faster discharge, with no downside to patient prognosis and clinical outcome. One interesting concern raised by nurse/PA survey response was the risk of inducing reflex micturition by RF, masking patients with risk of urinary retention development after discharge. However, no differences were observed in rate of grade ≥3 immediate postoperative complications nor reintervention within 3 months. On the other hand, 50% of SV patients required additional voiding trials due to poor initial VV or increased RU whereas only 36% patients in the RF group required double voiding, suggesting SV to cause delay in discharge and increasing unnecessary hospital stay, though not statistically significant in this study (p=0.289).

Facilitated discharge from hospital may be met with enthusiasm from both patient and healthcare providers in the COVID-19 era, especially as inpatient outbreaks are real and potentially fatal adverse events that may negatively affect patient outcome [18]. Of particular importance may be the fact that hospital-acquired infection of COVID-19 not only patients within the same ward, but also staff members whose infection may be more contagious and cause uncontrolled spread [19]. Considering that the critical exposure time of COVID-19 is less than 1 hour in a small space even with sufficient active ventilation, decreasing hospital stay by even 60 minutes may hold potential significance [20]. As such, methods for expediting clinical decisions may have positive effects by decreasing unnecessary hospital stay and exposure to nosocomial infection, as well as other obvious gains such as saving economic cost or staff resources, especially as discharge delay is often cited as one of the most unpleasurable experiences of hospitalization [21]. However, preference for early discharge by both patient and healthcare providers must be properly leveraged against risk of complications and reintervention, as well as possible patient discomfort.

Our study is not without limitations, one of which is the relatively small sample size included for analysis. Also, complete blinding of the subject and researcher was not possible due to the nature of the study and may have produced inaccurate results. The actual time when the patient leaves the hospital can also be affected by administrative procedures, including – and not limited to – hospital billing, discharge medication retrieval, and even scheduling for further visits. While these may have influenced patient experience, the time-to-discharge endpoint set for this particular study was not confounded by these factors. Lack of ethnic variability may result in misrepresentation and loss of generalizability. Also, the difference in time-to-discharge, while statistically significant, may be marginal to drastically influence patient experience. Despite limitations, the results from our study provide valuable evidence for supporting the routine use of RF in actual clinical fields as a safe and satisfactory method in catheter removal after urologic surgery.

RF for catheter removal in urologic inpatient settings is a safe and satisfactory method that can facilitate void and time-to-discharge, decreasing length of unnecessary hospital stay. When feasible, the procedure can be undertaken in routine clinical practice. Larger trials with long-term follow-up are required to further validate our results.

Supplementary materials can be found via https://doi.org/10.4111/icu.20220233

Supplementary Table 1

Study description based on CONSORT checklist

Supplementary Fig. 1

CONSORT flow diagram.

Supplementary Fig. 2

Clinicians' response on the questionnaire about the clinical usefulness of retrograde bladder filling method. AUR, acute urinary retention.