Abstract
Introduction and Hypothesis
The objective was to determine if a bowel prep prior to minimally invasive sacrocolpopexy (MIS) influences post-operative constipation symptoms. We hypothesized that women who underwent a bowel prep would have an improvement in post-operative defecatory function.
Methods
In this randomized-controlled trial, women undergoing MIS received a pre-operative bowel prep or no bowel prep. Our primary outcome was post-operative constipation measured by the Patient Assessment of Constipation Symptoms (PAC-SYM) 2 weeks post-operatively. Secondary outcomes included surgeons’ perception of case difficulty. Both intention to treat (ITT) and per protocol analyses (PPA) were performed. Analyses were done using t-test, Fisher’s exact test, Wilcoxon’s test and χ2.
Results
Of 105 enrolled women, 95 completed follow up (43 prep and 52 no prep). Baseline characteristics and rates of complications were similar. No differences were noted on ITT. The post-operative abdominal PAC-SYM subscale was closer to baseline for women who received a bowel prep on PPA (change in score 0.74 vs 1.08, p=.045). Women who underwent a prep were less likely to report strain (6.0% vs 26.7%, p=.009) or type 1 Bristol stool on their first post-operative bowel movement (4.3% vs 17.5%, p=.047). Surgeons were more likely to rate the complexity of case as “more difficult than average” (54.4% vs 40.1%, p=.027) in those without a bowel prep.
Conclusions
While there was no difference in ITT analysis, women who underwent a bowel prep prior to MIS demonstrated benefit for post-operative defecatory function with a corresponding improvement in surgeon’s perception of case complexity.
Keywords: sacrocolpopexy, pelvic organ prolapse, bowel prep, constipation, PAC-SYM
Brief summary
The use of a bowel prep prior to minimally-invasive sacrocolpopexy improved postoperative defecatory function and surgeon’s perception of case complexity.
Introduction
Pre-operative mechanical bowel preparations have been used by gynecologic and colorectal surgeons with the goal of improving visualization and bowel manipulation; however, studies have failed to show a patient benefit from mechanical bowel prep [1, 2]. Literature from colorectal surgery suggests an increase in surgical site infections while gynecologic studies have revealed no improvement in surgical field visualization during laparoscopy or vaginal prolapse surgery [3–5]. Additionally, there was some pre-operative patient discomfort that was attributed to the bowel prep [3, 5]. Abdominal sacrocolpopexy is a commonly performed surgery used to treat pelvic organ prolapse that requires access to the sacral promontory and cul de sac. There may be a role for mechanical bowel prep in minimally invasive sacrocolpopexies that has not been appreciated with other gynecologic surgeries.
There is a high prevalence of constipation and bowel symptomatology in women undergoing prolapse repair surgeries. One study found the prevalence of constipation to be 36% using ROME II criteria in a urogynecology population [6]. Another study reported 83% of patients presenting to a urogynecology office endorsed at least one positive finding on the Colorectal-Anal Distress Inventory subscale of the Pelvic Floor Distress Inventory [7]. Surgery, anesthesia and post-operative narcotics can all contribute to post-operative constipation. In a multi-center study of women undergoing abdominal sacrocolpopexy, approximately 12% reported post-operative nausea or bloating [8]. Moreover, it is theorized that straining and defecatory dysfunction associated with constipation and other functional disorders can increase a woman’s risk of pelvic organ prolapse or recurrence [9]. Improving post-operative constipation and defecatory dysfunction with the use of a pre-operative mechanical bowel prep may improve overall patient satisfaction with their post-operative experience.
To our knowledge, there has been no study specifically aimed to assess the role of mechanical bowel prep during minimally invasive sacrocolpopexy. Additionally, prior literature has not focused on post-operative constipation in relation to performance of a bowel preparation [4, 5]. For this reason, we performed a randomized-controlled trial assessing the impact of a magnesium citrate mechanical bowel prep versus no prep prior to minimally invasive sacrocolpopexy. We sought to determine if a pre-operative mechanical bowel prep improves patient’s post-operative constipation. We also assessed how well a pre-operative mechanical bowel prep is tolerated by the patient and the surgeon’s perception of case difficulty.
Materials and Methods
This was a single-blind, randomized trial of mechanical bowel prep versus no prep in women undergoing minimally invasive sacrocolpopexy. Surgeries were completed by 8 board certified FPMRS physicians at a large, academic university between 4/2013 and 4/2017. Eligible, non-pregnant women who were planning to undergo laparoscopic or robotic-assisted sacrocolpopexy were recruited. Exclusion criteria included a history of ulcerative colitis, Crohn’s disease or gastroparesis, renal insufficiency, prior bowel resection, prior pelvic radiation or abdominopelvic malignancy. Women were also excluded if concomitant bowel surgery, anal sphincteroplasty or rectovaginal fistula repair was planned. This study obtained approval from the University of Pittsburgh Institutional Review Board and all patients provided written informed consent.
Eligible women were randomized using simple randomization with a 1:1 ratio of mechanical bowel prep to no bowel prep. Randomization was done by research staff at the time of surgery scheduling. Physicians and fellows were blinded to group assignment. Patients randomized to the prep group were given written and verbal instructions to be on a clear liquid diet the afternoon prior to surgery and to consume one bottle of magnesium citrate the day before surgery. Those randomized to the no prep group could maintain a regular diet the day prior to surgery. Both groups were instructed not to eat or drink anything after midnight on the day of surgery as per routine practice at that time. Standard post-operative instructions were given along with a prescription for oxycodone, docusate and ibuprofen. Physicians were masked to the group assignment. Baseline characteristics, physical examination findings and complications were abstracted from the medical record.
Our primary outcome was post-operative constipation as measured by the change in pre-operative to post-operative global PAC-SYM score. Secondary outcomes included surgeon perceptions of case difficulty and patient tolerance of the bowel prep.
Participants were asked to complete two written questionnaires prior to surgery. The first was a pre-operative 7-day bowel diary which recorded characteristics about bowel movements. This item assessed narcotic and stool softener use, Bristol Stool Form Scale, the presence of straining or splinting, fecal leakage and the sensation of incomplete emptying. The Bristol Stool Form Scale is a 7-point scale for stool measurement with types 1 and 2 being hard stool consistent with constipation and types 6 and 7 loose stool consistent with diarrhea [10]. The second questionnaire was the Patient Assessment of Constipation Symptoms (PAC-SYM) which was used to assess baseline constipation and bowel symptomatology two weeks prior to surgery. This is a validated 12-item self-reported measure where patients are asked to rate their symptoms over the previous two weeks. Items are scored on a five-point Likert scale with a score of 0 corresponding to absent symptoms and 4 corresponding to very severe symptoms [11]. The PAC-SYM can be used as a global score or by subscales: abdominal (discomfort, pain, bloating, cramps), stool (hardness, size, straining, inability to pass stool), and rectal (pain, burning, bleeding, incomplete emptying). The global PAC-SYM score and three subscales have been validated for constipation [11]. Participants also completed a second bowel diary during the week following surgery and a repeat PAC-SYM questionnaire two weeks post-operatively. The timing of the second PAC-SYM was chosen based on its validation criteria of symptoms during a two-week period.
To assess tolerance of the bowel prep, patients completed a visual analog scale in the pre-operative area regarding general and bowel symptomology in the 24 hours preceding surgery. Patients marked their symptom score on a 10 cm line with point 0 indicating no symptoms and point 10 corresponding to maximal symptoms. Surgeons completed an immediate post-operative questionnaire assessing their perception of the quality of surgical field, the ease of bowel manipulation and the overall surgical case complexity. The surgeon questionnaire was developed for this study and not previously validated.
In addition to the intention to treat analysis, we performed a per protocol analysis. The intention to treat analysis was performed using the assigned randomization groups. For the per protocol analysis, patients were considered to have completed a bowel prep if they answered “Yes” to the following question: “If you took a bowel preparation, were you able to complete it?” Those who answered “No” or did not provide an answer were considered to have not completed a bowel prep. We do not have information regarding the type of bowel prep used by participants who were allocated to the no prep group.
The sample size for the primary outcome used a reported PAC-SYM score for post-operative constipation following abdominal pelvic organ prolapse surgery of 0.96 (± 0.72) with a minimal important difference of 0.45 [11, 12]. We calculated that we would need 41 patients in each arm assuming 80% power with a significance of 0.05. We accounted for a 10% drop-out rate and a 20% rate of surgery cancellation to give a final number of 110 participants with 55 in each group. This was projected to allow us to detect a difference in PAC-SYM score of 0.45 between groups. The primary outcome was compared using a two-sample t-test. All other between group comparisons were conducted by Wilcoxon’s test for continuous variables and Chi-square or Fisher’s exact test for categorical variables. All analyses were performed using SAS/STAT version 9.4 (SAS Institute Inc., Cary, NC).
Results
We enrolled 105 patients with 95 (90.5%) providing outcome data (43 assigned to prep group and 52 assigned to the no prep group). Figure 1 shows the study participant flow diagram. Baseline demographic and clinical characteristics were similar between groups (Table 1). The average age of our study population was 60.1 (± 6.4) years, 78 (82.1%) had stage III prolapse, 31 (32.6%) had a prior hysterectomy and 24 (25.3%) self-reported a history of baseline constipation. Overall, 61 (64.2%) underwent a supracervical hysterectomy with 85 (89.5%) having a laparoscopic sacrocolpopexy and 10 (10.5%) a robotic-assisted sacrocolpopexy. The mean length of stay was 1.1 (± 0.7) days and the mean operative time was 208.7 (± 56.2) minutes. Surgical complications and readmissions were similar between groups (all p>.05). No complication was directly attributable to undergoing a bowel prep or lack thereof. Compliance with the assigned treatment was poor with only 23 (53%) women in the prep group and 27 (51.9%) women in the no prep group adhering to their assigned group. Our per protocol analysis had 50 women in the prep group and 45 in the no prep group.
Figure 1.
Flow of participants
ITT = Intention to treat which includes all participants allocated to each arm
PPA = Per protocol analysis which includes participants who completed a bowel prep or did not complete a bowel prep, regardless of randomization arm
Table 1.
Participant demographic, clinical and perioperative characteristics
| Characteristic | Intention to Treat |
Per Protocol Analysis |
||
|---|---|---|---|---|
| Bowel Prep n=43 (%) | No Prep n=52 (%) | Bowel Prep n=50 (%) | No Prep n=45 (%) | |
| Agea (y) | 59.5±6.0 | 60.7±6.7 | 60.0±6.1 | 60.3±6.7 |
| BMIa (kg/m2) | 27.4±4.6 | 26.6±4.8 | 27.1±3.9 | 26.9±5.5 |
| Gravidityb | 2 (2, 4) | 3 (2, 4) | 3 (2, 4) | 3 (2, 4) |
| Parityb | 2 (2, 3) | 2.5 (2, 3) | 2 (2, 3) | 2 (2, 3) |
| Current smoker | 4 (9.3%) | 3 (5.8%) | 4 (8.0%) | 3 (6.7%) |
| Post-menopause | 41 (95.3%) | 49 (94.2%) | 47 (94.0%) | 43 (95.6%) |
| Prior hysterectomy | 14 (32.6%) | 17 (32.7%) | 20 (40.0%) | 34 (75.6%) |
| Stage of prolapse | ||||
| II | 2 (4.7%) | 9 (17.3%) | 7 (14.0%) | 4 (8.9%) |
| III | 39 (90.7%) | 39 (75.0%) | 40 (80.0%) | 38 (84.4%) |
| IV | 2 (4.7%) | 4 (7.7%) | 3 (6.0%) | 3 (6.7%) |
| Self-reported baseline constipation* | 8 (18.6%) | 16 (30.8%) | 14 (28.0%) | 10 (22.2%) |
| Laparoscopic sacrocolpopexy | 37 (86.1%) | 47 (90.4%) | 47 (94.0%) | 37 (82.2%) |
| Robotic-assisted sacrocolpopexy | 5 (11.6%) | 5 (9.6%) | 3 (6.0%) | 7 (15.6%) |
| Supracervical hysterectomy | 28 (65.1%) | 33 (63.5%) | 29 (58.0%) | 32 (71.1%) |
| Midurethral sling | 2 (4.7%) | 4 (7.7%) | 3 (6.0%) | 3 (6.7%) |
| Length of staya (days) | 1.1±1.0 | 1.0±0.4 | 1.1±0.9 | 1.0±0.4 |
| Operative timea (min) | 214.3±60.2 | 204.1±52.9 | 206.7±53.1 | 210.0±60.0 |
| Estimated blood lossa,c (ml) | 50.7±48.5 | 57.2±63.2 | 43.9±36.2 | 65.8±72.0 |
| Intra-operative complications | 4 (9.3%) | 3 (5.8%) | 3 (6.0%) | 4 (8.9%) |
| Post-operative complications | 7 (16.3%) | 6 (11.5%) | 5 (10.0%) | 8 (17.8%) |
| Readmission | 0 (0.0%) | 2 (3.9%) | 1 (2.0%) | 1 (2.2%) |
With the exception of EBL, all p-values >0.05
Data are presented as mean ± SD
Data are presented as median (IQR)
p=.036 for per protocol analysis
Self-reported baseline constipation was obtained from pre-operative 7-day bowel diary
There were no group differences in our primary outcome, post-operative constipation, as measured by the change from baseline in the global PAC-SYM score in the intention to treat analysis (Table 2). Additionally, there were no group differences in our secondary outcomes of pre-operative bowel symptoms and surgeon perception of case difficulty in our intention to treat analysis (Tables 3 & 4).
Table 2.
Patient constipation and bowel symptomatology
| Intention to Treat |
p | Per Protocol |
p | |||
|---|---|---|---|---|---|---|
| Bowel Prep n=43 (%) | No Prep n=52 (%) | Bowel Prep n=50 (%) | No Prep n=45 (%) | |||
| Pre-op PAC-SYMa | ||||||
| Global | 0.50±0.52 | 0.62±0.52 | .26 | 0.48±0.46 | 0.66±0.57 | .08 |
| Abdominal subscale | 0.51±0.63 | 0.65±0.71 | .31 | 0.54±0.60 | 0.64±0.75 | .45 |
| Rectal subscale | 0.32±0.46 | 0.44±0.44 | .19 | 0.32±0.39 | 0.47±0.51 | .10 |
| Stool subscale | 0.66±0.76 | 0.76±0.70 | .52 | 0.58±0.67 | 0.87±0.76 | .046 |
| Post-op PAC-SYMa | ||||||
| Global | 0.97±0.63 | 1.04±0.61 | .59 | 0.82±0.45 | 1.22±0.71 | .002 |
| Abdominal subscale | 1.46±0.89 | 1.51±0.77 | .75 | 1.28±0.64 | 1.72±0.94 | .010 |
| Rectal subscale | 0.59±0.53 | 0.66±0.61 | .55 | 0.51±0.43 | 0.76±0.68 | .033 |
| Stool subscale | 0.87±0.72 | 0.95±0.88 | .63 | 0.67±0.62 | 1.18±0.91 | .002 |
| Difference in PAC-SYMa | ||||||
| Global | 0.47±0.51 | 0.42±0.57 | .65 | 0.34±0.44 | 0.56±0.63 | .06 |
| Abdominal subscale | 0.95±0.75 | 0.86±0.86 | .60 | 0.74±0.65 | 1.08±0.93 | .045 |
| Rectal subscale | 0.27±0.54 | 0.22±0.51 | .64 | 0.19±0.40 | 0.29±0.62 | .24 |
| Stool subscale | 0.20±0.64 | 0.19±0.91 | .92 | 0.10±0.68 | 0.31±0.90 | .20 |
| In past 24 hours, how many bowel movements?a | 5.9±6.2 | 5.2±5.0 | .52 | 8.6±5.9 | 2.0±1.4 | <.0001 |
| Time until first post-operative BMa (days) | 3.1±1.4 | 3.3±1.2 | .19 | 3.3±1.5 | 3.1±1.1 | .31 |
| First BM straining | 8 (19.6%) | 7 (13.5%) | .77 | 3 (6.0%) | 12 (26.7%) | .009 |
| First BM sensation of empty | 26 (60.5%) | 28 (53.9%) | .78 | 32 (64.0%) | 22 (48.9%) | .33 |
| First BM splinting | 4 (9.3%) | 3 (5.8%) | .84 | 3 (6.0%) | 4 (8.9%) | .59 |
| First BM leaking | 5 (11.6%) | 5 (9.6%) | .86 | 6 (12.0%) | 4 (8.9%) | .64 |
| Post-operative stool softener usea (days) | 6.6±1.0 | 6.3±1.5 | .69 | 6.4±1.4 | 6.4±1.2 | .92 |
| Post-operative narcotic usea (days) | 2.9±2.3 | 3.2±2.0 | .37 | 2.9±2.2 | 3.2±2.0 | .41 |
| First BM Bristol | .58 | .046 | ||||
| Type 1 | 5 (12.5%) | 4 (8.5%) | 2 (4.3%) | 7 (17.5%) | ||
| Type 2, 3, 4 or 5 | 20 (50.0%) | 20 (42.6%) | 20 (42.6%) | 20 (50.0%) | ||
| Type 6 or 7 | 15 (37.5%) | 23 (48.9%) | 25 (53.2%) | 13 (22.5%) | ||
| First BM Bristol | .57 | .11 | ||||
| Type 1 or 2 | 7 (17.5%) | 7 (14.9%) | 5 (10.6%) | 9 (22.5%) | ||
| Type 3, 4 or 5 | 18 (45.0%) | 17 (36.2%) | 17 (36.2%) | 18 (45.0%) | ||
| Type 6 or 7 | 15 (37.5%) | 23 (48.9%) | 25 (53.2%) | 13 (32.5%) | ||
Data are presented as mean ± SD
Table 3.
Preoperative bowel symptoms as reported on visual analog scale
| Intention to Treat |
p | Per Protocol |
P | |||
|---|---|---|---|---|---|---|
| Bowel Prep n=43 | No Prep n=52 | Bowel Prep n=50 | No Prep n=45 | |||
| Cramping or Pain | 1.37±2.04 | 1.23±1.89 | .96 | 1.57±1.93 | 0.98±1.94 | .09 |
| Bloating or swelling | 1.30±1.67 | 1.24±1.70 | .91 | 1.77±1.94 | 0.70±1.08 | .009 |
| Hunger | 3.11±2.72 | 3.58±2.60 | .45 | 4.21±2.69 | 2.42±2.29 | .0008 |
| Thirst | 3.41±2.95 | 3.79±2.95 | .54 | 3.56±2.89 | 3.69±3.03 | .90 |
| Nausea | 0.96±1.97 | 1.27±2.26 | .43 | 1.68±2.50 | 0.50±1.40 | .033 |
| Vomiting | 0.24±0.95 | 0.35±1.20 | .86 | 0.44±1.47 | 0.15±0.26 | .40 |
| Dizziness | 0.58±1.36 | 0.44±0.80 | .86 | 0.72±1.35 | 0.25±0.57 | .17 |
| Weakness | 1.01±1.92 | 0.85±1.27 | .71 | 1.25±1.80 | 0.54±1.21 | .13 |
| Sleep disruption | 1.22±1.89 | 1.49±2.47 | .70 | 1.90±2.33 | 0.77±1.94 | .002 |
| Embarrassment | 0.42±0.97 | 0.64±1.63 | .68 | 0.73±1.45 | 0.34±1.27 | .13 |
| Fecal Incontinence | 0.62±1.32 | 1.22±2.30 | .75 | 1.28±2.20 | 0.57±1.53 | .038 |
| Overall Discomfort | 1.78±2.10 | 1.91±2.37 | .97 | 2.22±2.02 | 1.44±2.43 | .014 |
| How difficult was the bowel prep to complete?a | ||||||
| Easy | 12 (27.9%) | 7 (13.5%) | 19 (38.0%) | N/A | ||
| Tolerable | 7 (16.3%) | 11 (21.2%) | 18 (36.0%) | N/A | ||
| Slightly difficult | 1 (2.3%) | 3 (5.8%) | 4 (8.0%) | N/A | ||
| Moderately difficult | 2 (4.7%) | 1 (1.9%) | 3 (6.0%) | N/A | ||
| Very difficult | 1 (2.3%) | 2 (3.9%) | 3 (6.0%) | N/A | ||
Data are presented as mean ± SD
Missing data for 3 patients
Scores are on a Visual Analog Scale with 0 representing the absence of symptoms and 10 representing maximal symptoms
Table 4.
Surgeon perceptions
| Intention to Treat |
P | Per Protocol |
P | |||
|---|---|---|---|---|---|---|
| Bowel Prep n=43 (%) | No Prep n=52 (%) | Bowel Prep n=50 (%) | No Prep n=45 (%) | |||
| Quality of surgical field | .65 | .61 | ||||
| Poor | 7 (16.3) | 7 (13.5) | 6 (12.0) | 8 (17.8) | ||
| Fair | 8 (18.6) | 15 (28.9) | 10 (20.0) | 13 (28.9) | ||
| Good | 16 (37.2) | 16 (30.8) | 19 (38.0) | 13 (28.9) | ||
| Excellent | 11 (25.6) | 14 (26.9) | 14 (28.0) | 11 (24.4) | ||
| Overall ability to retract bowel | .73 | .49 | ||||
| Extremely poor | 4 (9.3) | 3 (5.8) | 3 (6.0) | 4 (2.9) | ||
| Below average | 9 (20.9) | 16 (30.8) | 10 (20.0) | 15 (33.3) | ||
| Average | 9 (20.9) | 7 (13.5) | 10 (20.0) | 6 (13.3) | ||
| Above average | 7 (16.3) | 9 (17.3) | 8 (16.0) | 8 (17.8) | ||
| Excellent | 14 (32.6) | 17 (32.7) | 19 (38.0) | 12 (26.7) | ||
| Overall ability to safely access surgical field | .80 | .91 | ||||
| Extremely poor | 4 (9.3) | 3 (5.8) | 4 (8.0) | 3 (6.8) | ||
| Below average | 5 (11.6) | 8 (15.4) | 6 (12.0) | 7 (15.6) | ||
| Average | 9 (20.9) | 12 (23.1) | 10 (20.0) | 11 (24.4) | ||
| Above average | 10 (23.3) | 8 (15.4) | 9 (18.0) | 9 (20.0) | ||
| Excellent | 15 (34.9) | 21 (40.4) | 21 (42.0) | 15 (33.3) | ||
| Overall complexity of the case | .68 | .027 | ||||
| Easier than average | 6 (14.0) | 11 (21.2) | 6 (12.0) | 11 (24.4) | ||
| Average | 16 (37.2) | 17 (32.7) | 23 (46.0) | 10 (22.2) | ||
| More difficult than average | 21 (48.8) | 24 (46.2) | 21 (42.0) | 24 (53.3) | ||
| Do you think surgery was compromised by adequacy of prep? | ||||||
| 6 (14.0) | 9 (17.3) | .78 | 3 (6.0) | 12 (26.7) | .010 | |
| Do you feel increased complexity was due to bowel prep? | ||||||
| 5 (23.8) | 6 (26.1) | .92 | 2 (9.5) | 9 (37.5) | .06 | |
| Do you think participant used bowel prep? | .91 | .013 | ||||
| No | 23 (53.5) | 26 (50.0) | 20 (40.0) | 29 (64.4) | ||
| Yes | 20 (46.5) | 25 (48.1) | 30 (60.0) | 15 (33.3) | ||
Our per protocol analysis found significant differences between groups. When compared to no prep, the post-operative PAC-SYM score for the prep group was significantly lower for the global score and all three subscales (abdominal, rectal and stool). This remained significant with the change in score on the abdominal subscale only (Table 2). Women who completed a bowel prep were significantly less likely to report strain on their first post-operative bowel movement (6.0% vs 26.7%, p=.009, Table 2). Additionally, women who received a bowel prep were less likely to report a Bristol type 1 stool as their first post-operative bowel movement (4.3% vs 17.5%, p=.047). There was no difference in days until first bowel movement, total days post-operative narcotic use, or total days stool softener use (all p>.05). Additionally, there was no difference in the sensation of emptying, splinting or leaking on the first post-operative bowel movement (all p>.05, Table 2).
The presence and severity of symptoms in the 24 hours preceding surgery differed between groups in the per protocol analysis (Table 3). Women who completed a bowel prep were more likely to report discomfort on their pre-operative questionnaire. Specifically, women who completed a bowel prep were more likely to report increased abdominal bloating or swelling, hunger, nausea, fecal incontinence, sleep disruption and overall discomfort in the 24 hours prior to surgery (all p<.05). Moreover, they had more bowel movements in the 24 hours prior to surgery with the prep group reporting a mean of 8.6 (± 5.9) bowel movements compared with 2.0 (± 1.4) in the no prep group (p<.0001, Table 2). The experience of completing a bowel prep was reported as easy or tolerable by 37 (74.0%) women and slightly, moderately or severely difficult by 10 (20.0%, Table 3).
The surgeon was more likely to feel that the surgery was compromised by the adequacy of the bowel prep in the no prep group on per protocol analysis (26.1% vs 6.1%, p=.010, Table 4). In the absence of a bowel prep, surgeons were more likely to report their perception of the case as more difficult than average (53.3% vs 42.0%, p=.024). Surgeons correctly identified whether a bowel prep was used in 60.0% of the prep group and in 64.4% of the no prep group (p=.013). There was no reported difference in the quality of the surgical field, overall ability to retract the bowel or overall ability to safely access the field in both sets of analyses (all p>.05).
Discussion
In this population of women undergoing minimally invasive sacrocolpopexy, there was no difference in post-operative constipation, patient tolerance of bowel prep or surgeon perception of case difficulty among women randomized to receive a bowel prep or no bowel prep. However, among women who completed a magnesium citrate mechanical bowel prep, there was a beneficial impact on post-operative defecatory function. Global PAC-SYM scores were not significantly different between groups but individual post-operative subscales and the difference in abdominal subscale showed a closer return to baseline among women who completed a bowel prep. Additionally, women who underwent a bowel prep were less likely to strain on their first post-operative bowel movement or report a type 1 Bristol stool form. Despite experiencing some discomfort from the pre-operative bowel prep, only 20% of women felt the bowel prep was difficult to complete indicating tolerability for most participants. Surgeons found the bowel prep advantageous in terms of complexity of case and adequacy of prep, though this did not impact complication rates or the ability to safely access the surgical field.
In contrast to the majority of bowel prep studies which focus on surgeon’s assessment of intra-operative characteristics, we chose to study the patient centered outcome of post-operative constipation and associated defecatory dysfunction [4, 5, 13]. Though a bowel prep reduced the severity of constipation symptoms reported on the PAC-SYM post-operatively, only the difference in the abdominal PAC-SYM subscale met the previously reported minimal important difference (MID) for chronic constipation of 0.45–0.60 [11, 14]. The MID represents the smallest change in symptoms that are considered meaningful for patients and thus provides a practical application of any statistically significant results. It is important to note that the MID for peri-operative constipation is not known and may be different than that for chronic constipation. We chose to assess symptoms 2-weeks post-operatively as the PAC-SYM is designed to assess bowel function over the preceding two weeks. It is possible that the impact of a bowel prep may be short-lived, and a clinically meaningful difference may exist in symptomology closer to the date of surgery.
Patients who completed a bowel prep were more likely to report bloating or swelling, hunger, nausea, sleep disruption, fecal incontinence and overall discomfort. Most scores were low (under 2 on a 10-point Visual Analog Scale) suggesting a mild to moderate burden. These findings are consistent with previous studies of pre-operative bowel preps [3, 5]. However, most women reported the prep was “easy” or “tolerable” with only 6% saying it was “very difficult” to complete. This suggests that while patients experience more pre-operative symptoms with a bowel prep, they are not severely burdened by the experience of a mechanical bowel prep. Some patients may know their tolerance to a bowel prep from experience and it may be useful to incorporate a shared decision-making model during counselling on peri-operative optimization of bowel function.
Our study population demonstrated poor compliance with their prep assignment which we addressed with a per protocol analysis. Reasons for this may include both patient and provider preference as well as legacy counseling. A quarter of our population had a prior hysterectomy and may have previously been instructed to complete a bowel prep for that surgery. Those patients may have completed a bowel prep for this surgery thinking it was needed despite thorough counseling regarding their participation in the study. Additionally, some providers may have discussed a bowel prep at the pre-operative visit prior to study enrollment and randomization. Again, those patients may have taken pre-operative laxatives regardless of randomization group. Conversely, some women may have had a negative experience with a bowel prep in the past and declined to complete a prep despite their agreement to participate in this study. Yang and colleagues reported women were less willing to try an oral bowel prep using sodium phosphate a second time when compared to those receiving enemas suggesting a past negative experience could have impacted our randomization compliance [13].
The poor compliance with randomization limits our ability to draw conclusions from the intention to treat analysis and is a major limitation to this study. Additionally, we do not have information regarding the type of bowel prep used by those allocated to the no prep group who did not follow randomization. This potential heterogeneity among the bowel prep group in the PPA is a significant limitation. There were no difference in demographics or baseline characteristics between those who followed group assignment and those who did not. Use of a placebo agent would have masked patients to their group assignment and may have improved compliance to the assigned prep. This phenomenon of poor compliance with assigned bowel prep warrants a closer evaluation in the future, possibly with the use of qualitative measures.
Surgeons were more likely to perceive the case as more difficult than average and compromised by the adequacy of the bowel prep in the absence of a prep. This suggests there may be an unmeasured benefit of a bowel prep in minimally invasive sacrocolpopexy. Previous literature did not support improved visualization with a mechanical bowel prep during benign laparoscopy but these studies did not include sacrocolpopexies [3, 4]. A bowel prep may play a more important role during minimally invasive sacrocolpopexy given the required access to the sacral promontory and cul de sac. This area may be obstructed by a redundant sigmoid colon or a full small bowel making identification and delineation of important structures more difficult. While there was no difference in the overall ability to safely access the surgical field or retract the bowel, there may be an unquantifiable benefit to a bowel prep during minimally invasive sacrocolpopexy that is not present in other gynecologic surgeries. Our surgeons were able to identify whether a bowel prep was performed slightly more than half the time. This is similar to findings from previous studies on bowel preps prior to gynecologic surgery [4].
A strength of this study is the patient population which is representative of many urogynecology populations. Additionally, most sacrocolpopexies are currently being performed using minimally invasive techniques, as supported by data from the National Surgical Quality Improvement Program which showed that 82% of all sacrocolpopexies performed at participating sites in 2016 were minimally invasive [15]. All cases in this study were performed at a single academic institution by a total of 8 high volume, board-certified FPMRS surgeons who performed all sacrocolpopexies by minimally invasive route during the period of study enrollment. It is possible that less experienced surgeons may value the effects of a bowel prep differently. Other strengths of this study include its randomized design and adequately powered sample size.
A limitation of our study is the lack of a true placebo arm which may have influenced our poor compliance. All patients were instructed to use daily docusate post-operatively. It is possible that a stronger post-operative bowel regimen along with detailed patient instructions regarding the prevention of constipation may have had a similar impact on patient symptoms without the use of a bowel prep. This was not examined in this study and is an avenue for future research. Other limitations of this study include our survey measures. While the PAC-SYM has been validated for use with chronic constipation it has not been validated for perioperative constipation [11]. Additionally, our pre-operative questionnaire, bowel diary and surgeon questionnaire were developed for this study and therefore not previously validated. Limitations related to our poor compliance were discussed previously. While our decision to perform a per protocol analysis can affect the external validity of our study, our results are still generalizable to most women undergoing minimally invasive sacrocolpopexy.
This study adds to the available literature regarding the role of a mechanical bowel prep in gynecologic surgery. Specifically, it suggests a bowel prep confers some benefit to women’s post-operative abdominal symptoms and surgeon’s perception of the case difficulty during minimally invasive sacrocolpopexies with an associated pre-operative symptom burden that appears well tolerated by patients. Further research is needed to assess the full extent that a bowel prep contributes to minimally invasive sacrocolpopexy as well as patient and surgeon preferences regarding the decision to comply with a mechanical bowel prep.
Acknowledgements
Our research was supported by the National Institutes of Health through Grant Number UL1TR001857
Footnotes
Conflict of interest: The authors report no conflict of interest.
Clinical Trial Registration:
1. Date of registration: March 6th, 2013
2. Date of initial participant enrollment: March 18th, 2013
3. Clinical trial identification number: NCT01805310
4. URL of registration site: https://clinicaltrials.gov/ct2/show/NCT01805310?cond=sacral+colpopexy&cntry=US&state=US%3APA&city=pittsburgh&rank=1
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of a an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
References
- 1.Wells T, Plante M, McAlpine JN, Preoperative bowel preparation in gynecologic oncology: a review of practice patterns and an impetus to change. Int J Gynecol Cancer, 2011. 21(6): p. 1135–42. [DOI] [PubMed] [Google Scholar]
- 2.Slim K, Vicaut E, Launay-Savary MV, Contant C, Chipponi J, Updated systematic review and meta-analysis of randomized clinical trials on the role of mechanical bowel preparation before colorectal surgery. Annals Surg, 2009. 249(2): p. 203–9. [DOI] [PubMed] [Google Scholar]
- 3.Muzii L, Bellati F, Zullo MA, Manci N, Angioli R, Panici PB, Mechanical bowel preparation before gynecologic laparoscopy: a randomized, single-blind, controlled trial. Fertil Steril, 2006. 85(3): p. 689–93. [DOI] [PubMed] [Google Scholar]
- 4.Siedhoff M, Clark LH, Hobbs KA, Findley AD, Moulder JK, Garrett JM, Mechanical bowel preparation before laparoscopic hysterectomy: a randomized controlled trial. Obstet Gynecol, 2014. 123(3): p. 562–7. [DOI] [PubMed] [Google Scholar]
- 5.Ballard A, Parker-Autry CY, Markland AD, Varner RE, Huisingh C, Richter HE, Bowel preparation before vaginal prolapse surgery: a randomized controlled trial. Obstet Gynecol, 2014. 123: p. 232–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Jelovsek J, Barber MD, Paraiso MF, Walters MD, Functional bowel and anorectal disorders in patients with pelvic organ prolapse and incontinence. AJOG, 2005. 193(6): p. 2105–11. [DOI] [PubMed] [Google Scholar]
- 7.Raza-Khan F, Cunkelman J, Lowenstein L, Shott S, Kenton K, Prevalence of bowel symtpoms in women with pelvic floor disorders. Int Urogynecol J, 2010. 21(8): p. 933–8. [DOI] [PubMed] [Google Scholar]
- 8.Whitehead W, Bradley CS, Brown MB, Brubaker L, Gutman RE, Varner E, Visco AG, Weber AM, Zyczynski H, Gastrointestinal complications following abdominal sacrocolpopexy for advanced pelvic organ prolapse. AJOG, 2007. 197(1): p. 78.e1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Weir L, Nygaard IE, Wilken J, Brandt D, Janz KF, Postoperative activity restrictions: any evidence? Obstet Gynecol, 2006. 107(2): p. 305–9. [DOI] [PubMed] [Google Scholar]
- 10.Blake MR, Raker JM, and Whelan K, Validity and reliability of the Bristol Stool Form Scale in healthy adults and patients with diarrhoea-predominant irritable bowel syndrome. Aliment Pharmacol Ther, 2016. 44(7): p. 693–703. [DOI] [PubMed] [Google Scholar]
- 11.Frank L, Kleinman L, Farup C, Taylor L, Miner P, Pyschometric validation of a constipation symptom assessment questionnaire. Scan J Gastro, 1999. 34(9): p. 870–7. [DOI] [PubMed] [Google Scholar]
- 12.Patel M, O’Sullivan D, LaSala C, Constipation symptoms before and after vaginal and abdominal pelvic reconstructive surgery. Int Urogynecol J, 2011. 22: p. 1413–1419. [DOI] [PubMed] [Google Scholar]
- 13.Yang L, Arden D, Lee TT, Mansuria SM, Broach AN, D’Ambrosio L, Guido R, Mechanical bowel preparation for gynecologic laparoscopy: a prospective randomized trial of oral sodium phosphate solution vs single sodium phosphate enema. JMIG, 2011. 18(2): p. 149–56. [DOI] [PubMed] [Google Scholar]
- 14.Yiannakou Y, Tack J, Dubois B, Quigley EMM, Ke MY, DaSilva S, Joseph A, Kerstens R, The PAC-SYM questionnaire for chronic constipation: defining the minimal important difference. Alimentary Pharm Therap, 2017. 46(11–12). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Linder B, Occhino JA, Habermann EB, Glasgow AE, Bews KA, Gershman B, A national contemporary analysis of perioperative outcomes for open versus minimally-invasive sacrocolpopexy. J Urol, 2018. [DOI] [PubMed] [Google Scholar]