Postoperative Bladder Filling After Outpatient Laparoscopic Hysterectomy and Time to Discharge: A Randomized Controlled Trial

Abstract

Objective:

To determine if backfilling the bladder postoperatively will reduce time to discharge in patients undergoing outpatient laparoscopic hysterectomy.

Methods:

In a single-blind, randomized controlled trial among women undergoing outpatient laparoscopic hysterectomy, patients were randomly assigned to a backfill-assisted void trial or a trial of spontaneous voiding. The primary outcome was time to discharge assessed by length of time spent in the post-anesthesia care unit. Secondary outcomes included time to first spontaneous void, urinary retention rates, and postoperative complications within 8 weeks. We estimated that 152 patients (76 per group) would provide greater than 80% power to identify a 30-minute difference in the primary outcome with a SD of 56 minutes and a two-sided α of 0.05.

Results:

Between June 2017 to May 2018, 202 women were screened, 162 women were randomized, and results analyzed for 153 women. 75 patients (group A) who had a backfill-assisted voiding trial and 78 patients (group B) who had a spontaneous voiding trial were included in the analysis. The mean time to discharge was 273.4 minutes for group A vs. 283.2 minutes for group B, which was not found to be significant (p=0.45). The mean time to first spontaneous void was 181.1 minutes in group A vs. 206.0 minutes in group B. There was a statistically significant reduction of 24.9 minutes in time to first spontaneous void with patients randomized to the backfill group (p=0.04). Five out of 75 patients (6.7%) in group A and 10/78 patients (12.8%) in group B developed urinary retention postoperatively and required re-catheterization prior to discharge, which was also not significant (p=0.20).

Conclusion:

Bladder filling prior to removing the Foley catheter is a safe and simple procedure shown to reduce time to first spontaneous void, but not time to discharge in patients undergoing outpatient laparoscopic hysterectomy.

PRÉCIS

Postoperative bladder filling reduces time to first spontaneous void but not time to discharge from the hospital in patients undergoing outpatient laparoscopic hysterectomy.

INTRODUCTION

Same-day discharge following laparoscopic hysterectomy is being practiced by an increasing number of providers [2-5]. Not only is same-day discharge safe and well tolerated, it is also associated with decreased costs [6]. The “enhanced recovery after surgery (ERAS)” concept emerged as a multimodal approach directed at standardizing perioperative care, improving surgical outcomes, and enhancing the success of early discharge programs [7]. Minimally invasive surgery is an integral component of ERAS and associated with improved post-operative outcomes [8]. ERAS pathways call for early removal of urinary catheters following gynecologic surgery to optimize early discharge [7]. Earlier removal of urinary catheters resulted in a shorter time to first void, larger volume of urine at first void, less frequent catheterization for urinary retention, and a shorter length of stay [9].

Urogynecologic trials have shown that a backfill-assisted voiding trial is superior to a spontaneous voiding trial after transvaginal surgery [10, 11]. Studies have yet to examine the impact of a day-of-surgery voiding trial after laparoscopic hysterectomy. Successful spontaneous void is often a discharge criterion after same-day laparoscopic hysterectomy. Oftentimes, waiting for this spontaneous void can delay discharge.

We designed a single-blind, randomized controlled trial to evaluate whether backfilling the bladder postoperatively would reduce time to discharge in patients undergoing outpatient laparoscopic hysterectomy. We hypothesized that backfilling the bladder postoperatively will reduce time to spontaneous void and subsequent discharge from the post-anesthesia care unit.

METHODS

Approval was obtained from the Institutional Review Board of the University of Pittsburgh (PRO17010292) and the trial was registered with ClinicalTrials.gov (NCT03126162). Women 18-75 years of age undergoing outpatient laparoscopic hysterectomy through the ERAS protocol at Magee-Womens Hospital of the University of Pittsburgh Medical Center, a tertiary care institution, were approached to participate in the study. For women undergoing outpatient laparoscopic hysterectomy, the standard of care is to remove the Foley catheter at the end of the procedure. Hysterectomies included either a total laparoscopic hysterectomy or laparoscopic supracervical hysterectomy with or without bilateral salpingectomy or salpingo-oophorectomy performed by five minimally invasive gynecologic surgeons. Concurrent procedures including excision of endometriosis, ureterolysis, lysis of adhesions, enterolysis, mini-laparotomy (for specimen removal in cases with large fibroid uteri), and cystoscopy with insertion and removal of bilateral ureteral stents (used in difficult cases where retroperitoneal fibrosis is anticipated to aid in identification of ureters) were also recorded. The ERAS protocol developed at our institution for laparoscopic hysterectomies included premedication with acetaminophen, celecoxib, and perphenazine and the utilization of total intravenous anesthesia (TIVA) intraoperatively with the goal to avoid intraoperative narcotic administration. Exclusion criteria included the inability to provide informed consent, history of prior or concurrent urogynecologic procedures (mid-urethral sling, urethral bulking, anterior and posterior colporrhaphy, perineorrhaphy, burch urethropexy, sacrospinous ligament fixation, uterosacral ligament suspension, sacrocolpopexy), history of multiple sclerosis, known malignancy of the bladder, intraoperative bladder or ureteral injury, anticipated prolonged bladder catheterization, women being treated for an active urinary tract infection at the time of surgery, or planned postoperative overnight admissions. Written informed consent was obtained. Each participant was randomly assigned to a backfill-assisted void trial (group A) or a trial of spontaneous voiding with no backfilling of the bladder (group B).

A randomization scheme using a block size of four was generated using a website (http://www.randomization.com). Sealed randomization envelopes were opened for each subject after anesthesia was initiated in the operating suite. This study was single-blinded to the participant. After group allocation, the surgeon was informed of whether or not the bladder will be backfilled at the completion of the surgery prior to Foley catheter removal. If the patient was assigned to group A, 200 mL of room temperature, sterile normal saline was instilled retrograde into the bladder at the completion of the surgery prior to Foley catheter removal and the Foley was subsequently removed intraoperatively. If the patient was assigned to group B, the Foley catheter was removed intraoperatively at completion of the procedure. Our standard protocol is to use a 16F Foley catheter for gynecologic laparoscopy cases and patients from both groups received the same size catheter. Two hundred mL of sterile saline was chosen as the amount to backfill to prevent postoperative overdistension as the patient will continue to naturally produce urine while in the recovery room awaking from anesthesia. A normal adult bladder capacity can range from 400-550 mL, with a normal desire to void at 300-400 mL [12].

Routinely following a laparoscopic hysterectomy, patients at our institution were expected to have spontaneously voided by 6 hours. After arrival in the post-anesthesia care unit, the nursing staff received standardized written orders to follow a specific voiding protocol for our study participants (Figure 1). To evaluate for overdistension of the bladder, assessment of spontaneous void occurred at 5 hours for all participants rather than the routine 6 hours. A successful voiding trial was defined as a voided volume of at least 200 mL in order to be discharged home. If unable to void at least 200 mL by 5 hours after surgery, a bladder scan was performed and the volume recorded. If 200 mL or greater of urine was noted on the scan, a Foley catheter was reinserted and the patient was discharged home with the plan to return to the outpatient clinic for removal. If less than 200 mL of urine was noted on the scan, patients were given one additional hour to hydrate either orally or intravenously. If unable to spontaneously void 200 mL by 6 hours after surgery, a Foley catheter was reinserted and the patient was discharged home with the plan to return to the outpatient clinic 3 or 4 days for a retrograde voiding trial. The amount of urine voided, time of spontaneous void, and reinsertion of the Foley catheter were recorded by the nursing staff. Criteria for discharge included adequate pain control (pain score < 5), tolerating oral intake without nausea or emesis, independent ambulation, and completion of either a spontaneous void trial or home with a Foley catheter for urinary retention. Any surgical or medical complication for which the surgeon or anesthesiologist wanted to observe the patient overnight was also a criterion for overnight admission.

Figure 1.

Postanesthesia care unit voiding protocol.

Preoperative data collection included demographics, medical and surgical history, medications, and indication for hysterectomy. Intraoperative data including surgical procedure, estimated blood loss, total intravenous fluids received intraoperatively and in the post-anesthesia care unit, urine output, total operative time, narcotic medications, and surgical complications were all recorded for each participant. Data pertaining to the postoperative course (e.g., time of arrival in the post-anesthesia care unit, time of discharge, time and volume of first spontaneous void, and voiding trial results) were documented by the nursing staff. Voiding trial results were subsequently collected by a co-investigator and confirmed through chart review. Data on urinary retention rates, narcotics received in the post-anesthesia care unit, occurrence of postoperative urinary tract infection, occurrence of postoperative voiding dysfunction, and postoperative surgical complications were collected through medical chart review by a co-investigator for up to 8 weeks postoperatively.

Our primary aim was to determine if backfilling the bladder in the operating room immediately prior to removal of the Foley catheter would reduce time to discharge in patients undergoing outpatient laparoscopic hysterectomy. We hypothesized that patients randomized to group A, who had their bladders backfilled with 200 mL of normal saline would have a shorter time to discharge compared to patients randomized to group B who just had their Foley catheters removed at the completion of surgery. We hypothesized that patients randomized to group A would have an overall shorter time to discharge. The outcome variable of interest was time spent in the post-anesthesia care unit, measured in minutes from arrival time to the post-anesthesia care unit to time of discharge from the hospital. Our secondary aim was to determine if backfilling the bladder postoperatively prior to removal of the Foley catheter reduced time to spontaneous void. Our outcome variable of interest was time to first spontaneous void, measured in minutes from the end of surgery. In addition, we aimed to determine if covariates, such as narcotic usage and total intravenous fluids received, influenced time to spontaneous voiding. Other important secondary outcomes included group comparisons for rates of urinary retention, postoperative complications, readmissions, emergency department visits, and urinary tract infections.

An a priori power analysis was undertaken to determine an adequate sample size for our primary outcome, time to discharge. We estimated that 152 patients (76 per group) would provide greater than 80% power to identify a 30-minute difference in the primary outcome (time to discharge assessed by length of time spent in post-anesthesia care unit) from baseline length of time spent in the post-anesthesia care unit of 227 minutes (based on prior study data) with a standard deviation of 56 minutes and a two-sided significance level of 0.05 [10].

All statistical analyses were performed using SAS (Version 9.0; SAS Institute, Cary, NC, USA). Data are presented as means (±standard deviation) for continuous variables or medians (interquartile range) for data not normally distributed. Categorical data are presented as frequency (percentage). Group means that follow normal distributions were compared using Student’s t tests. Non-normal distributions were compared using Wilcoxon rank-sum tests. Chi-squared analyses or Fisher’s exact tests were used to compare proportional data. Multivariable modeling was done using logistic regression. All outcomes were evaluated at a 0.05 level of significance. Individual participant data (including data dictionaries) will not be available or shared.

RESULTS

Recruitment occurred from June 2017 to May 2018 at Magee-Womens Hospital of the University of Pittsburgh Medical Center. Of the 202 women screened for eligibility, 172 consented for enrollment, and 162 patients were randomized. 81 patients were allocated to group A, the backfill-assisted voiding group and 81 to group B, the spontaneous voiding group. Six patients from group A and three patients from group B were admitted overnight and therefore excluded in our analysis, resulting in 75 patients in group A and 78 patients in group B (Figure 2).

Figure 2:

Consolidated Standards of Reporting Trials (CONSORT) flow diagram.

ERAS, enhanced recovery after surgery.

Baseline characteristics were similar between the two groups (Table 1). There was no difference in baseline demographics, prior surgical history, or indication for hysterectomy between the two groups. The most common indications for hysterectomy were symptomatic uterine fibroids (38.7%, 37.2%) and pelvic pain (37.3%, 41%) for groups A and B, respectively. The most common type of hysterectomy performed was a total laparoscopic hysterectomy (98.7%) (Table 1). There were also no differences in concomitant procedures performed at the time of laparoscopic hysterectomy between the two groups as well as intraoperative factors including length of surgery (time of incision to skin closure), total intravenous fluids given intraoperatively, urine output, or estimated blood loss. Postoperatively in the post-anesthesia care unit, there was no difference between the two groups with the length of time spent in the post-anesthesia care unit or amount of intravenous fluids given. Both groups received a similar amount of narcotics preoperatively, intraoperatively, and postoperatively (Table 2).

Table 1.

Baseline Characteristics

Demographic	Group A Backfill- assisted void trial (n=75)	Group B Spontaneous void trial (n=78)
Age (y)	42.7±6.4	42.2±6.8
BMI (kg/m2)	30.3±7.5	31.0±8.3
Race
Caucasian	51(68)	59 (75.6)
African American	22 (29.3)	18 (23.1)
Asian	2 (2.7)	1 (1.3)
Previous abdominal surgery	20 (26.7)	25(32.1)
Previous pelvic surgery	58 (77.3)	60 (76.9)
Laparoscopy for endometriosis	14 (18.7)	14 (18)
Endometrial ablation	10 (13.3)	13 (16.7)
Tubal ligation	22 (29.3)	17 (21.8)
Cesarean section	25 (33.3)	23 (29.5)
Tube and/or ovarian surgery	10 (13.3)	9 (11.5)
Myomectomy	2 (2.7)	3 (3.9)
Indication for hysterectomy
Symptomatic fibroids	29 (38.7)	29 (37.2)
Pelvic pain	28 (37.3)	32 (41)
Abnormal uterine bleeding	12 (16)	15 (19.2)
Endometrial hyperplasia	1(1.3)	0
Cervical dysplasia	2 (2.7)	1 (1.3)
Endometriosis	2 (2.7)	0
Genetic predisposition	1 (1.3)	1 (1.3)
Type of laparoscopic hysterectomy
Total	74 (98.7)	77 (98.7)
Supracervical	1 (1.3)	1 (1.3)
Concomitant Procedures
Salpingectomy*	60 (80)	62 (79.5)
Oophorectomy*	21 (28)	22 (28.2)
Ureterolysis*	10 (13.3)	13 (16.7)
Excision of endometriosis	23 (30.7)	27 (34.6)
Ureteral stents†	20 (26.7)	24 (30.8)
Cystoscopy	29 (38.7)	32 (41)
Mini-laparotomy	13 (17.3)	13 (16.7)
Lysis of adhesions	2 (2.7)	2 (2.6)
Enterolysis	3 (4)	4 (5.1)

BMI, body mass index

Data are n (%), mean ± standard deviation, or median (interquartile range)

^*Unilateral and bilateral

^†Bilateral

Table 2.

Intraoperative and Postoperative Factors

	Group A Backfill-assisted void trial (n=75)	Group B Spontaneous void trial (n=78)	P
Intraoperative Factors
Length of surgery (minutes	124 (58)	118 (57)	0.45
IVF (mL)	1500 (600)	1400 (600)	0.37
EBL (mL)	100 (50)	100 (100)	0.43
Urine output (mL)	300 (200)	300 (300)	0.33
Postoperative Factors
Length spent in PACU (minutes)	273.4±78	283.2±80.5	0.45
PACU IVF (mL)	500 (600)	500 (500)	0.44
Total IVF (mL)	1900 (700)	1950 (900)	0.99
Total narcotics (MME/day)	37.5 (11.5)	38 (11.5)	0.47

IVF, intravenous fluids; EBL, estimated blood loss; PACU, post-anesthesia care unit; MME, morphine milligram equivalents

Data are n (%), mean ± standard deviation, or median (interquartile range)

For the primary outcome, time to discharge was assessed by the length of time spent in the post-anesthesia care unit. The mean time to discharge was 273.4 minutes for group A vs. 283.2 minutes for group B, a difference of 9.8 minutes which was not found to be significant (p=0.45) (Table 2). The mean time to first spontaneous void was 181.1 minutes in group A vs. 206.0 minutes in group B (Table 3). There was a significant reduction of 24.9 minutes in time to first spontaneous void with patients randomized to the backfill group (p=0.04). The total amount of intravenous fluids received was not associated with time to spontaneous void (p=0.26). The total amount of narcotics received also did not influence time to first spontaneous void (p=0.31). However, when the total amount of narcotics received was calculated prior to spontaneous void, more narcotics taken pre-void slowed down time to void, which was found to be significant (p=0.03). Bladder filling prior to removing the Foley catheter was not associated with increased rates of postoperative urinary retention. Five patients (6.7%) in group A and 10 patients (12.8%) in group B developed urinary retention postoperatively and required re-catheterization prior to discharge, which was also not significant (p=0.20).

Table 3.

Primary and Secondary Outcomes

Outcomes	Group A Backfill-assisted void trial (n=75)	Group B Spontaneous void trial (n=78)	P
Time to discharge (minutes)	273.4±78	283.2±80.5	0.45
Time to spontaneous void (minutes)	181.1±66	206±77.6	0.04
Postoperative urinary retention	5(6.7)	10(12.8)	0.20
ED visits*	4(5.3)	5(6.4)	0.78
Postoperative UTI†	1(1.3)	3(3.8)	0.33
Postoperative readmissions*,‡	2(2.7)	2(2.6)	0.98

ED, emergency department; UTI, urinary tract infection;

Data are n (%), mean ± standard deviation, or median (interquartile range)

^*Includes events within 8 weeks postoperatively

^†Defined as a positive urine culture or symptoms and antibiotic treatment

^‡Postoperative readmission reasons include: fever, incisional cellulitis, pelvic abscess, and pelvic hematoma

For additional secondary outcomes, the number of emergency department visits, rates of urinary tract infections, and readmissions were similar between the two groups (Table 2). Emergency department visits were assessed up until 8 weeks postoperatively and were similar between groups A and B (5.3% vs. 6.4%, p=0.78). Similarly, rates of urinary tract infections were low and not different between groups A and B (1.3% vs. 3.8%, p=0.33). The number of readmissions was also similar (Table 2). In group A, one patient was admitted on postoperative day 3 for workup of a fever and another was admitted for inpatient management of incisional cellulitis on postoperative day 7. In group B, there was one patient admitted on postoperative day 9 for management of a pelvic abscess and another patient who was admitted on postoperative day 1 with a large pelvic hematoma due to a mechanical fall unrelated to surgery. None of the postoperative complications were directly related to the study intervention. Although our findings revealed similarities between the two groups in rates of emergency department visits, urinary tract infections, and readmissions, these counts of adverse outcomes are low and therefore underpowered to generalize that there is no true difference.

DISCUSSION

This study demonstrates that a backfill-assisted void trial with 200 mL of normal saline instilled into the bladder prior to Foley catheter removal decreases time to first spontaneous void but does not significantly impact time to discharge in patients undergoing outpatient laparoscopic hysterectomy. Backfilling the bladder prior to a voiding trial is safe and feasible compared with a standard spontaneous voiding trial [15]. This technique is commonly implemented to test for voiding dysfunction following urogynecologic and urologic surgery; however, it is not routinely performed following other gynecologic procedures [11]. We are not aware of any previous studies specifically assessing time to discharge after postoperative bladder testing in outpatient laparoscopic hysterectomy patients.

While ERAS pathways have been shown to reduce length of stay in abdominal gynecologic surgery with early removal of urinary catheters, studies have not assessed length of stay for outpatient gynecologic procedures [16]. As many hospital systems are moving towards expedited recovery to decrease costs and complications associated with hospitalization, optimizing postoperative care and length of stay may lead to reduced hospital costs and decreased burden to nursing staff and other providers [11, 17-19]. Previous voiding studies in urogynecologic and urologic literature have reported mixed results with regards to time of discharge. Foster et al. found that subjects randomized to a backfill group after outpatient transvaginal surgery (e.g. midurethral sling, rectocele and/or cystocele repair, perineorrhaphy) were more likely to adequately empty their bladder and had a shorter mean stay of 27.1 minutes in the post-anesthesia care unit [10]. Boccola et al. showed that bladder filling prior to a trial of void in patients undergoing prostate surgery reduced time to discharge by 112.5 minutes [20]. Du et al. showed that bladder filling prior to a trial of void for patients post-transurethral resection of prostate, bladder neck incision, or admitted with urinary retention showed a reduction in time to discharge of 39.5 minutes [15]. We used a reduction in time to discharge of 30 minutes based on prior study data to calculate our sample size at a power of 0.8 [10]. Bladder filling prior to Foley catheter removal did not reduce time to discharge in our study population due to several possibilities. Factors such as nausea, postoperative pain, or social issues (e.g. transportation) could have contributed to a delay in discharge. Another possibility could be due to nursing staff not adequately accustomed to same-day discharge after laparoscopic hysterectomy. Since patients have traditionally stayed overnight, nurses may still be in that mindset and not comfortable with the newly implemented compressed timeline. This may result in reluctance to discharging patients in a shorter amount of time, even if patients have met their discharge milestones. Thus, a reduction of 24.9 minutes in time to spontaneous void in the backfill group may not have resulted in a reduced time to discharge.

Other limitations of this study include the lack of racial diversity. This study had predominantly Caucasian or African American participants and an underrepresentation of other races besides Asian when compared with the national population. This may be representative of local demographics and the fact that the study was carried out at a single institution, limiting the generalizability of our results. Another limitation was not collecting outcome measures related to voiding dysfunction in the first 4 weeks after surgery. Voiding dysfunction is infrequently seen after routine gynecologic procedures compared to urogynecologic procedures, especially after a successful spontaneous voiding trial. However, short-term abnormalities in bladder sensation are influenced by post-surgical factors such as pain, edema, inflammation, and anesthesia; thus a formal assessment with a validated questionnaire assessing patient satisfaction could have contributed to the results of this study [21, 22].

Major strengths of this study include the study design allowing a direct comparison of two commonly used techniques for assessing postoperative voiding. Furthermore, our enrollment acceptance rate of 85% and drop-out incidence of 5.5% reduces the concern for selection bias. Lastly, the nursing staff who carried out the study protocol were blinded to the intervention assignment (backfill-assisted versus spontaneous voiding) during data collection for our primary and secondary outcomes, limiting observer bias.

We conclude that bladder filling prior to removing the Foley catheter is a safe and simple alternative to a spontaneous voiding trial that can reduce time to first spontaneous void in patients undergoing outpatient laparoscopic hysterectomy. Despite being feasible with no additional morbidity, it has not been shown to significantly reduce time to discharge.

Authors’ Data Sharing Statement.

Will individual participant data be available (including data dictionaries)? No.

What data in particular will be shared: Not available

What other documents will be available? Not available.

When will data be available (start and end dates)? Not applicable.

By what access criteria will data be shared (including with whom, for what types of analyses, and by what mechanism)? Not applicable.