Background
Rates of postoperative incomplete bladder emptying vary significantly after pelvic reconstructive surgery. With enhanced recovery protocols the paradigm is shifting towards same-day discharge and the rates of incomplete bladder emptying are expected to increase. The optimal length of time for postoperative catheter drainage has not been clearly established. There are no current studies that assess the optimal timing of a repeat voiding trial in women who have unsuccessful same day voiding trials.
Objective
This study aimed to compare the outcomes of a second voiding trial performed 2–4 days (earlier group) vs 7 days (later group) postoperatively in women with incomplete bladder emptying after vaginal prolapse surgery. Secondary aims included postoperative urinary tract infection rates, total days with a catheter, and patient-reported catheter bother between groups.
Study Design
Across 2 sites, women undergoing multicompartment vaginal repair were enrolled. Within 6 hours postoperatively, subjects had an active retrograde voiding trial. Those who passed this voiding trial exited the study; those who had persistent incomplete bladder emptying (postvoid residual >100 mL) had a transurethral indwelling catheter placed and were randomized to return for an earlier (postoperative day 2–4) vs later (postoperative day 7) follow-up office voiding trial. Subjects were followed for 6 weeks after surgery. The primary outcome was the rate of unsuccessful repeat office voiding trial. Secondary outcomes included rates of urinary tract infection, total days with a catheter, and subjective catheter bother. A power calculation based on a projected 31% difference, a power of 0.8, and an alpha of 0.05 revealed that 30 subjects were needed in each group.
Results
A total of 102 subjects were enrolled; 38 exited on postoperative day 0, leaving 64 subjects for randomization (4 of whom withdrew after randomization). A comparison of data revealed that randomization was effective, with no differences between the earlier and later groups in terms of demographic data or surgical procedures. Using an intention-to-treat analysis, women in the earlier group were more likely to be unsuccessful in their follow-up office voiding trial (23.3%) than the later group (3.3%), with a risk difference of 20% (95% confidence interval, 3.56–36.44) and a relative risk of 7.00 (95% confidence interval, 0.92–53.47; P =.02). A number-needed-to-treat calculation found that for every 5 patients using a catheter for 7 days postoperatively, 1 case of persistent postoperative incomplete bladder emptying was prevented. Rates of catheter bother did not differ between groups at the time of the follow-up office voiding trial or at 6 weeks ( P =.09 and P =.20, respectively). Urinary tract infection rates were higher in the earlier group but did not reach statistical significance (23% vs 7%, P =.07). Regression analysis revealed that subjects who required additional pain medication refills were 9.6 times (95% confidence interval, 1.24–73.77) more likely to have persistent incomplete bladder emptying after the follow-up office voiding trial.
Conclusion
Women with incomplete bladder emptying after multicompartment prolapse repair had a 7-fold higher risk of an unsuccessful repeat office voiding trial if performed within 4 days of surgery than when performed within 7 days of surgery. In addition, requiring additional prescriptions for analgesia increased the risk of an unsuccessful follow-up office voiding trial.
Active voiding trials (VTs) are routinely performed after pelvic floor reconstructive surgery to guide the decision as to which patients may be safely discharged without a catheter because of the increased risk of transient incomplete bladder emptying. Rates of temporary postoperative catheter use range from 12% to 83% in the urogynecology literature. This wide range is due to the lack of consensus of VT methods, timing and different definitions of a successful VT. Although short-term catheterization after pelvic floor reconstructive surgery may be common, less than 5% of patients require prolonged catheterization up to 6 weeks. It is critical to identify patients with incomplete bladder emptying to prevent pain, myogenic and neurogenic damage, ureteral reflux, and bladder overdistension that may further impair voiding function. , , However, extending bladder drainage beyond the necessary recovery period may be associated with higher rates of urinary tract infection (UTI) and patient bother. ,
Why was this study conducted?
The optimal length of time for catheter drainage after prolapse reconstruction and a same-day unsuccessful initial voiding trial has not been previously studied.
Key findings
Repeat voiding trials at least 7 days postoperatively have significantly lower rates of persistent incomplete bladder emptying as compared with voiding trials occurring fewer than 4 days postoperatively.
What does this add to what is known?
This is the first study to evaluate the timing of a second voiding trial after an unsuccessful initial same-day postoperative voiding trial.
The optimal length of time for postoperative catheter drainage after vaginal prolapse surgery has not been clearly established, and the recent trend of same-day hospital discharge has further challenged surgeons as to the ideal time to schedule a repeat VT in those with incomplete bladder emptying. Same-day discharge after pelvic reconstructive surgery is both feasible and safe with the help of enhanced recovery after surgery pathways but has been associated with higher rates of an unsuccessful VT than when performed on postoperative day (POD) 1. , Pelvic floor reconstructive surgeries are expected to increase by almost 50% within the next 40 years, and same-day discharges will likely become more prevalent as enhanced recovery pathways become more ubiquitous. At this time, there are no published studies that assess the optimal timing of a repeat VT in women who have an unsuccessful same-day postoperative VT. One cross-sectional study of providers noted that follow-up VTs ranged from 1 day to 2 weeks postoperatively. This study aims to investigate the optimal timing of a repeat office VT in women with incomplete bladder emptying who are discharged with an indwelling transurethral catheter after vaginal multicompartment pelvic floor reconstructive surgery.
Materials and Methods
This was a prospective randomized controlled trial that was conducted at 2 sites between February 2017 and November 2019. Institutional review board approval was obtained at both clinical sites (Wake Forest Baptist Health, Winston-Salem, NC and Cleveland Clinic Florida, Weston, FL), and the study was registered on www.clinicaltrials.gov ( NCT03048682 ). Subjects were eligible for the study if they were at least 18 years old, were able to give informed consent, had at least stage II pelvic organ prolapse (POP), underwent vaginal reconstructive surgery in at least 2 of 3 vaginal compartments (anterior, posterior, and apical), and had persistent incomplete bladder emptying after the same-day VT (postvoid residual [PVR] >100 mL). Participants were excluded if they had a prior surgery for urinary incontinence, had neurologic conditions that affect bladder function (ie, multiple sclerosis, stroke, spinal cord injuries), were using preoperative narcotic medications for chronic pain, or had incomplete bladder emptying preoperatively (defined as PVR >200 mL). Subjects who required prolonged catheterization after surgery, either because of a genitourinary abnormality or injury, prolonged vaginal packing, or more intense postoperative monitoring (as determined by the attending surgeon at the conclusion of the surgery) were deemed ineligible after enrollment. Subjects who received a concomitant anti-incontinence procedure were permitted to participate in the study.
All eligible subjects were recruited preoperatively, and all participants signed a consent form before surgery. All participants received a single prophylactic dose of intravenous antibiotics at the time of the surgery per standard surgical protocols. All surgeries were performed in teaching institutions with fellowship-trained surgeons and clinical fellows. The specific surgical plan was made preoperatively by the surgeons on the basis of the personalized treatment plan in conjunction with the patients’ desires. The use of vaginal packing at the conclusion of the case was per surgeon discretion. A standardized postoperative VT within 6 hours of surgery was performed in all subjects after vaginal packing was removed. The subject was helped into a standing position to ensure that the bladder was drained completely via the transurethral catheter. The bladder was then backfilled with 300 mL of normal saline solution through the transurethral catheter. The catheter was removed, and the subjects were given 1 hour to void spontaneously. Next, PVR was checked by using either a catheter or an ultrasound. A successful VT was defined as a PVR of <100 mL. If the subject was successful in their VT, they exited the study. If a subject had an unsuccessful VT, a transurethral catheter was reinserted and the subject was then randomized to either an earlier (2–4 days postsurgery) follow-up office VT (EVT) or a later (7 days postsurgery) follow-up office VT (LVT) in a 1:1 allocation ratio before discharge. Treatment allocation was made by a computer-generated program that used randomly permuted blocks of varying sizes (2, 4, 6, and 8). Group allocation was revealed by opening sequentially numbered opaque envelopes. Given the nature of the study, neither the surgeons nor the subjects were blinded to the group allocation. However, the nurses performing the office VT were not aware of the subjects’ group allocation.
Although all subjects underwent a same-day VT, not all subjects were discharged on POD #0. The decision for same-day discharge was based on the standard routines at each site, so all subjects at one institution were discharged on POD #0 (n=46), whereas all those at the second site were discharged on POD #1 (n=18). Subjects discharged on POD #1 received similar supportive care that could have been achieved at home (same medications and diet orders). Both groups were discharged from the hospital with the indwelling transurethral catheter in place and prescriptions for narcotics and antibiotics for UTI prophylaxis (nitrofurantoin 100 mg orally until the office VT unless allergic, then trimethoprim-sulfamethoxazole was used). The EVT group was scheduled to return to the office for their follow-up VT between postoperative days 2–4 on the basis of weekday availability, whereas the LVT group were all scheduled to return to the office 7 days postsurgery. The office VT was performed in the same standardized fashion as the same-day VT. If the subject had an unsuccessful office VT, defined as PVR >100 mL, they were given the option of continuing with a transurethral catheter or performing intermittent self-catheterization, and follow-up was directed by their surgeon. Subjects were monitored for the first 6 weeks after surgery. Subjects reported the number of narcotic pills taken within 24 hours and catheter bother on a Likert scale (1=very bothersome, 5=not bothersome at all) on the date of their follow-up office VT and at the 6-week follow-up. Intraoperative and recovery room morphine dose equivalents were calculated, as were the doses taken within 24 hours of the office VT. Urine cultures were performed if patients had symptoms of a UTI, such as dysuria or urgency, during the follow-up period, and a UTI was defined as a positive culture with >10,000 colony-forming units of a single organism. A sample size calculation was performed, and based on a projected 31% difference (40% elevated PVR in the early group vs 9% in the prolonged group), a power of 0.8, and a 2-sided alpha of 0.05, 60 subjects (30 per group) were required.
The primary outcome of the study was the rate of unsuccessful office VT after vaginal reconstructive surgery for multicompartment prolapse. We hypothesized that there would be a significantly higher rate of unsuccessful office VT in the EVT group than in the LVT group. Secondary outcomes included the rates of postoperative UTI, total days with a catheter (indwelling transurethral catheter±self-catheterization), and patient-reported catheter bother.
Descriptive statistics were calculated for the overall sample. The data were assessed using an intention-to-treat analysis. The Shapiro-Wilk test was used to test for normal distributions. Continuous data with normal distribution were analyzed using Student t -test, whereas skewed data were analyzed using the Wilcoxon signed-rank test. Categorical data, such as unsuccessful VT rates (the primary outcome), were analyzed using Pearson chi-square test or Fisher exact test if there were ≤5 subjects in 20% of the cells. Patient bother as noted on a Likert scale was considered as ordinal data. Bivariate analysis and a forward stepwise multivariate regression analysis were performed. P <.05 was considered statistically significant. All analyses were performed using JMP version 14.0.0 (SAS Institute Inc, Cary, NC).
Results
Overall, 102 subjects agreed to participate in the study preoperatively. Of these, 29 (28%) had a successful same-day VT and exited the study before randomization. Five patients required vaginal packing for longer than 6 hours and were thus excluded. Of the 68 remaining subjects who had an unsuccessful same-day VT, 4 opted out of the study before randomization and 4 opted out after randomization (2 from the EVT group and 2 from the LVT group, all of whom expressed a desire to repeat the VT on POD #1). Thus, 60 subjects (30 in EVT group and 30 in LVT group) were included in the outcome analysis ( Figure ).
Randomization was effective, and there were no significant differences between the baseline demographics of the 64 randomized subjects in the 2 groups. Both groups were primarily postmenopausal white women who were nonsmokers and had stage II or III multicompartment POP. Approximately one-third of both groups had received a diagnosis of stress urinary incontinence on urodynamic testing ( Table 1 ).
| Characteristic | Earlier VT group (n=32) | Later VT group (n=32) | P value |
|---|---|---|---|
| Age | 64.06±12.04 | 65.31±11.45 | .6719 |
| Race | .1191 | ||
| White | 23 (71.88) | 26 (81.25) | |
| Hispanic | 3 (9.38) | 5 (15.63) | |
| African American | 6 (18.75) | 1 (3.13) | |
| Site | .5782 | ||
| CCF | 22 (68.75) | 24 (75.00) | |
| WF | 10 (31.25) | 8 (25.00) | |
| Body mass index | 28.24±6.77 | 26.34±5.23 | .2980 |
| Parity (median [interquartile range]) | 2 [2, 3] | 2 [2, 3] | .8314 |
| Smoker | .7192 | ||
| Current | 0 (0.00) | 0 (0.00) | |
| Former | 4 (12.50) | 5 (15.63) | |
| Never | 28 (87.50) | 27 (84.38) | |
| Prior hysterectomy | 8 (25.00) | 9 (28.13) | .7772 |
| Postmenopausal | 28 (87.50) | 27 (84.38) | .7192 |
| Diabetes mellitus | 4 (12.50) | 2 (6.25) | .3911 |
| Vaginal atrophy on examination | 22 (68.75) | 24 (75.00) | .5782 |
| POP-Q stage | .3114 | ||
| 2 | 15 (46.88) | 10 (31.25) | |
| 3 | 15 (46.88) | 21 (65.63) | |
| 4 | 2 (6.25) | 1 (3.13) | |
| Preoperative PVR | 38.84±30.11 | 38.94±39.62 | .7617 |
| UDS incontinence | .7096 | ||
| None | 14 (50.00) | 14 (48.28) | |
| SUI | 11 (39.29) | 9 (31.03) | |
| ISD | 2 (7.14) | 3 (10.34) | |
| DO | 1 (3.57) | 3 (10.34) | |
| UDS maximum urethral closure pressure | 68.72±32.96 | 85.39±50.70 | .3636 |
| UDS maximum flow | 18.01±8.60 | 18.91±11.17 | 1.0000 |
| UDS detrusor pressure at maximum flow | 25.19±14.20 | 28.43±19.65 | .7450 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
