Comparing inpatient versus outpatient bowel preparation in children and adolescents undergoing appendicovesicostomy

Accepted Manuscript Comparing inpatient versus outpatient bowel preparation in children and adolescents undergoing appendicovesicostomy David L. Weatherly, Konrad M. Szymanski, Benjamin M. Whittam, William E. Bennett, Jr., Shelly King, Rosalia Misseri, Martin Kaefer, Richard C. Rink, Mark P. Cain PII:

S1477-5131(17)30304-2

DOI:

10.1016/j.jpurol.2017.07.013

Reference:

JPUROL 2620

To appear in:

Journal of Pediatric Urology

Received Date: 23 May 2017 Revised Date:

1477-5131 1477-5131

Accepted Date: 8 July 2017

Please cite this article as: Weatherly DL, Szymanski KM, Whittam BM, Bennett Jr. WE, King S, Misseri R, Kaefer M, Rink RC, Cain MP, Comparing inpatient versus outpatient bowel preparation in children and adolescents undergoing appendicovesicostomy, Journal of Pediatric Urology (2017), doi: 10.1016/ j.jpurol.2017.07.013. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Comparing inpatient versus outpatient bowel preparation in children and adolescents undergoing appendicovesicostomy

RI PT

David L. Weatherly, Konrad M. Szymanski *, Benjamin M. Whittam, William E. Bennett Jr., Shelly King, Rosalia Misseri, Martin Kaefer, Richard C. Rink, Mark P. Cain

SC

Division of Pediatric Urology, Riley Hospital for Children at Indiana University Health,

M AN U

Indianapolis, IN, USA

* Corresponding author. Division of Pediatric Urology, Riley Hospital for Children at IU Health, Indiana University School of Medicine, 705 Riley Hospital Dr. #4230, Indianapolis, IN 46205, USA. Tel.: +1 317 944 7469; fax: +1 317 944 7481.

TE D

E-mail address: [email protected] (K.M. Szymanski).

Summary

EP

Purpose

The need for mechanical inpatient bowel preparation (IBP) in reconstructive pediatric urology

AC C

has come under scrutiny, secondary to literature demonstrating little benefit regarding outcomes. Starting in 2013, a majority of patients undergoing reconstructive procedures at our institution no longer underwent IBP. We hypothesized that outpatient bowel preparation (OBP) would reduce length of stay (LOS) without increasing postoperative complications after appendicovesicostomy surgery. Materials and methods

1

ACCEPTED MANUSCRIPT

An institutional database of patients undergoing lower urinary tract reconstruction between May 2010 and December 2014 was reviewed. Starting in 2013, a departmental decision was made to replace IBP with OBP. Patients undergoing an augmentation cystoplasty or continent

RI PT

ileovesicostomy were excluded because of insufficient numbers undergoing OBP. Patients undergoing IBP were admitted 1 day prior to surgery and received polyethylene

glycol/electrolyte solution. A personalized preoperative OBP was introduced in 2013. Cost data

SC

were obtained from the Pediatric Health Information System. Results

M AN U

Sixty-seven patients met the inclusion criteria, with 30 (44.8%) undergoing IBP. There were no differences with respect to gender, age, presence of ventriculoperitoneal shunt, body mass index, glomerular filtration rate, preoperative diagnosis, operative time, and prior or simultaneous associated surgeries (p ≥ 0.07). Patients undergoing an IBP had a longer median LOS (7 vs. 5

TE D

days, p = 0.0002) and a higher median cost (US$4,288, p = 0.01). Postoperative complications in both groups were uncommon and were classified as Clavien–Dindo grade 1–2, with no statistical difference (IBP 20.0% vs. OBP 5.4%, p = 0.13). No serious postoperative complication occurred,

Discussion

EP

such as a dehiscence, bowel obstruction, or shunt infection.

AC C

This is the first analysis of hospitalization costs and IBP, showing a higher median cost of US$4,288 compared with OBP. The LOS was shorter with an OBP (figure), similar to a previous report. Similar complication rates between the groups add to the growing body of literature that avoidance of IBP is safe in pediatric lower urinary tract reconstruction. Being a retrospective review of a practice change, differences in care that influenced cost and LOS may be missing. Also, as the surgeons do not know if a usable appendix is initially present, our data may not

2

ACCEPTED MANUSCRIPT

extrapolate to all patients. Despite these potential limitations, our data support the safety of utilizing OBP in patients with a high likelihood of a usable appendix, including those undergoing a synchronous Malone antegrade continence enema via a split-appendix technique.

RI PT

Conclusion

In patients undergoing an appendicovesicostomy, preoperative IBP led to longer LOS and higher costs of hospitalization. OBP was not associated with increased risks of postoperative

SC

complications.

M AN U

KEYWORDS

Appendicovesicostomy; Mechanical bowel preparation; Cost; Length of stay; Neurogenic bladder

TE D

Introduction

The use of a preoperative mechanical inpatient bowel preparation (IBP) for surgical procedures involving the bowel was popularized in the 1970s [1]. It subsequently evolved into the standard

EP

of practice for many reconstructive surgical procedures involving the gastrointestinal (GI) tract. Advocates cite that mechanical bowel preparation (MBP) lowers the bacterial load inside the GI

AC C

tract, lowers intraluminal pressures, and improves bowel mobility [2]. These theoretical benefits of IBP were heralded to reduce surgical site infection and other perioperative complications (anastomotic dehiscence, pulmonary complications, ileus, etc.). However, this has been challenged by numerous studies reporting similar or lower rates of wound infection and anastomotic complications after eliminating MBP [3–5]. Most of these studies involve adult patients with normal bowel function, while pediatric urologic reconstructive procedures are

3

ACCEPTED MANUSCRIPT

frequently performed in patients with neuropathic bowel managed with a daily bowel regimen. The safety of eliminating an IBP in this patient population has not been extensively studied [4,6]. Given the current evidence demonstrating no clear benefit of MBP, we elected to stop

RI PT

performing IBP for planned lower urinary tract reconstruction using bowel

(appendicovesicostomy [APV], ileovesicostomy, bladder augmentation) in 2013. The initial purpose of this study was to compare clinical outcomes of children undergoing reconstruction at

SC

our institution with IBP and outpatient bowel preparation (OBP). Unfortunately, because of the continued high rate of IBP in patients undergoing ileovesicostomy or bladder augmentation, the

M AN U

analysis was focused on children undergoing APV. We hypothesized that eliminating an IBP in patients undergoing APV would not impact clinical outcomes nor alter length of stay (LOS).

TE D

Methods

Institutional Review Board approval was obtained (IRB 0908-70). A retrospective review was performed of all patients under 21 years of age undergoing

EP

lower urinary tract reconstruction with bowel at our institution between May 2010 and December 2014, performed by six pediatric urologists. From 2010 to 2012, patients were admitted the day

AC C

before surgery for IBP with polyethylene glycol solution with electrolytes (25 mL/kg by mouth or via nasogastric tube). In 2013, a preoperative protocol was adopted for all patients undergoing lower urinary tract reconstruction to eliminate a preoperative admission with IBP and forgo postoperative nasogastric drainage. Patients were routinely assessed in a preoperative clinic coordinated by a nurse practitioner (S.K.) 2–3 weeks prior to scheduled surgery. Bowel status was evaluated based on history and an abdominal radiograph. A majority of these patients were already on a daily bowel 4

ACCEPTED MANUSCRIPT

regimen. Based on the level of bowel dysfunction and stool burden, a personalized OBP was prescribed to resolve constipation or fecal impaction, and improve postoperative comfort. Patients with a satisfactory history and imaging were maintained on their current program.

RI PT

Patients not on a daily bowel regimen are started on polyethylene glycol powder (typically 1 or 2 caps daily) for 7–10 days preoperatively. Patients already on a daily bowel regimen and with significant stool burden on imaging had their dosage increased (typically doubled) for the same

SC

period. Patients with rectal stool on the radiograph received enemas for one or two nights before surgery. Any patients on a daily enema program were instructed to continue this practice. All

M AN U

patients were given instructions for a clear liquid diet the day before surgery, with standard perioperative NPO instructions. Patients were included regardless of bowel dysfunction and bowel regimen.

Despite a plan to eliminate IBP for all lower urinary tract reconstruction involving bowel,

TE D

only 18.6% (11/59) patients undergoing bladder augmentation and 13.3% (6/45) undergoing an ileovesicostomy underwent the OBP protocol at the time of analysis. This made any statistical comparison involving these procedures underpowered. In an attempt to compare similar

EP

operations with sufficient numbers, we elected to exclude all patients who had a bowel anastomosis (bladder augmentation and/or ileovesicostomy).

AC C

Abstracted data included age at surgery, gender, baseline glomerular filtration rate

(GFR), body mass index (BMI), ventriculoperitoneal shunt status (VPS), preoperative diagnosis, associated surgeries, length of surgery, length of nasogastric tube drainage, time to clear liquid diet, time to regular diet, and LOS. GFR was calculated using the Schwartz formula. Diagnoses were divided into three categories: lipo/myelomeningocele, other spinal abnormalities (e.g., spinal cord injuries, tethered cord), and other (exstrophy, non-neurogenic neurogenic bladder).

5

ACCEPTED MANUSCRIPT

Three primary outcomes were assessed: (1) LOS was obtained from the medical record; (2) complications within 30 days of surgery were obtained by medical record review and classified using the Clavien–Dindo system [7]. To ensure no long-term VPS complications were

RI PT

missed, charts were reviewed for 12 months after surgery; (3) hospitalization costs were obtained from the Pediatric Health Information System database, which contains inpatient, emergency department, ambulatory surgery, and observation data from 43 not-for-profit, tertiary care

SC

pediatric hospitals in the United States.

Non-parametric tests were used for analysis: the Mann–Whitney U test for continuous

M AN U

data and the Fisher exact test for discrete variables. An exploratory analysis of the postoperative LOS and total cost using multivariate logistic regression was performed. Outcomes were binary, based on the median postoperative LOS (≤ 5 vs. > 5 days) and cost (≤ US$16,000 vs. > US$16,000). Outcomes were dichotomized due to concerns for extreme values of LOS and

TE D

costs biasing the analysis. Variables included in the final multivariate model were those with a p < 0.3 on univariate analysis. Screened variables were selected a priori based on clinical plausibility (age, gender, VPS status, surgery other than APV, baseline GFR, and whether IBP

EP

was performed). Total LOS was also screened for inclusion in the exploratory analysis of costs.

AC C

A critical p = 0.05 was used (Stata 10.1; StataCorp, College Station, TX, USA).

Results

Population characteristics

Of the 67 patients meeting inclusion criteria, 30 were admitted 1 day preoperatively for an IBP and 37 patients underwent OBP. There were no significant differences between the two groups with respect to gender, age at surgery, VPS status, BMI, GFR, preoperative diagnosis, length of surgery, and simultaneous associated surgeries (p ≥ 0.07) (Table 1). 6

ACCEPTED MANUSCRIPT

Overall, 26 (70.3%) of the OBP group were on a home bowel regimen prior to surgery. Of these, two patients (7.7%) remained on an unchanged home regimen as OBP. The remaining 24 patients (92.3%) had their bowel regimen changed at the preoperative visit, as described in

RI PT

the Methods section. Briefly, eight had daily polyethylene glycol powder doubled and the rest were started on a daily dose. Four patients continued on their daily enema, eight started on them

SC

1–2 days prior to surgery, and the remainder did not have a preoperative enema.

Outcomes

M AN U

Postoperative nasogastric tube drainage was used in 11 patients in the IBP group, with all of them removed on postoperative day 1 or 2. No nasogastric tubes were left in place postoperatively in the OBP group. The use of nasogastric drainage was not associated with longer operative times (p = 0.32). Eight of 10 patients with a gastrostomy tube left the operating

TE D

room with it capped. Two patients, both in the IBP group, had their gastrostomy tube open to drainage post operatively and capped later (5 and 7 days). Time to starting clear liquids and a regular diet was shorter in the OBP group (2 vs. 3 postoperative days, and 3 vs. 4 postoperative

EP

days, respectively, p≤0.02) (Table 2).

The median perioperative LOS was longer in patients undergoing IBP than OBP (7 vs. 5

AC C

days, p = 0.002) (see Summary figure). The 2 day longer LOS in the IBP group was due to an additional 1 day of preoperative admission (p = 0.0001), and an additional day of postoperative admission (5 vs. 6 days), although the postoperative LOS difference did not reach statistical significance (p = 0.053).

Eight complications were recorded in the first 30 days after surgery, all grades 1 or 2 on the Clavien–Dindo classification. The IBP group had six complications: three episodes of ileus (2 requiring total parenteral nutrition), two febrile urinary tract infections (UTIs), and one 7

ACCEPTED MANUSCRIPT

Clostridium difficile colitis. The OBP group had two complications: a wound infection, and an episode of ileus (requiring a nasogastric tube on postoperative day 1). No VPS complications were identified in the 18 patients with a shunt over 12 months of follow-up. There was no

RI PT

statistical difference in overall postoperative complications between the IBP and the OBP groups (p = 0.13). In other words, the risk of urinary or wound infections was low in both groups (0–3% wound infections, 0–7% UTIs). In addition, the risk of either of these two complications was

SC

similar between bowel preparation groups, regardless of whether a concurrent Malone antegrade continence enema (MACE) procedure was performed (OBP, 0% with MACE and 4.6% without;

M AN U

IBP, 7.7% with MACE and 5.9% without, p = 0.88).

Hospitalization costs for patients undergoing an IBP were US$18,610, compared with US$14,322 for the OBP group. The difference of US$4,288 (29.9%) was statistically significant

Exploratory analysis

TE D

(p = 0.01).

On exploratory multivariate analysis of postoperative LOS, IBP versus OBP was not a

EP

statistically significant predictor of prolonged hospitalization (OR 2.2, p = 0.13). VPS status and undergoing additional surgery other than APV were not associated with increased LOS either

AC C

(p ≥ 0.23).

On exploratory multivariate analysis of total costs, higher costs were primarily driven by

prolonged LOS (p < 0.001). IBP and undergoing surgery other than a urinary channel were not independent predictor of a costs >US$16,000 (p ≥ 0.73).

Discussion

8

ACCEPTED MANUSCRIPT

We present a retrospective study comparing postoperative clinical outcomes, LOS, and hospitalization costs for 67 pediatric and adolescent patients undergoing an APV procedure with IBP or OBP. We found a significant 2-day decrease in the length of stay and 30% lower costs for

RI PT

patients undergoing an OBP, with no difference in adverse postoperative outcomes. Infection rates were low in both groups: 0–3% wound infections, 0–7% UTIs, which are similar to

reported rates [3–6]. On multivariate analysis, the presence of a VPS and undergoing associated

SC

surgeries did not affect postoperative LOS. On the other hand, prolonged LOS was the main driver of higher hospitalization costs. This study adds to the growing body of evidence that IBP

M AN U

for some pediatric urology procedures can be safely avoided [4,6].

Multiple studies report high rates of constipation in patients with spina bifida, other spinal pathology, and developmental delay [8–11]. A daily outpatient bowel regimen is a standard of care in managing these patients [12]. This typically involves a combination of diet,

TE D

fiber, osmotic agents, motility agents, and enemas. In this study, 70% of patients in the OBP group were on a daily outpatient bowel regimen. This limits any study attempting to evaluate complete avoidance of MBP. We feel it is important to eliminate fecal impaction or severe

EP

constipation prior to reconstructive procedures, as postoperative narcotic use and immobility will potentially worsen pre-existing constipation [13–16]. Preoperative evaluation allows for

AC C

identification of patients who may require additional preparation prior to abdominal surgery in an attempt to improve postoperative recovery. Few studies of pediatric urologic reconstructions without MBP have been published.

Gundeti et al. [4] compared outcomes of children undergoing ileal enterocystoplasty in 24 patients with MBP and 22 patients without MBP. They reported no difference in LOS,

9

ACCEPTED MANUSCRIPT

postoperative UTI, and wound infections. It is unclear how many patients were on a daily bowel regimen preoperatively. Some believe a concomitant MACE procedure may increase the risk of postoperative

RI PT

infection because of exposure of the operative field to colonic flora [4]. Our findings indicate that a concurrent MACE procedure in the setting of an APV did not increase the risk of postoperative infections. Victor et al. [6] reviewed 158 pediatric patients undergoing

SC

enterocystoplasty, 85% augmented with sigmoid colon. No patients were on a preoperative

bowel regimen and 68% had a diagnosis of a neuropathic bladder. Postoperative complications

M AN U

occurred in 10%, the most common being a urinary fistula (2%), with a wound infection rate of 2%, and 3% requiring a surgical exploration [6]. Their report supports the safety of omitting a MBP for complex lower urinary tract reconstructions even with planned usage of colon. Despite the current level of evidence, the use of IBP in reconstructive pediatric urology

TE D

remains prevalent with experts citing distal fecal impaction causing high pressures at the anastomotic site, high stool volume increasing technically difficult and lengthening the operative time as reasons to continue IBP [4].

EP

The current study represents the first analysis of the relationship between hospitalization costs and IBP in pediatric patients. Gundeti et al. [4] have previously reported that patients

AC C

without MBP before enterocystoplasty were discharged 1 day earlier. We found a 2-day reduction in the LOS in patients undergoing APV with OBP: 1 day before and after surgery. In our population, this longer LOS was the single independent predictor of increased hospitalization costs, corresponding to a 30% increase in the costs of the hospitalization for the IBP group without a significant increase in postoperative complications.

10

ACCEPTED MANUSCRIPT

VPS infection is a severe complication in this population, often requiring externalization of the shunt. We did not identify any postoperative VPS complications in the 20 patients with a VPS over a period of 12 months. Previous reports of VP shunt infections after reconstructive

RI PT

urologic procedures have shown rates of 0–20% [17–20]. Typical care for the VPS during

procedures at our facility include early identification of the shunt tubing, wrapping with an

antibiotic soaked sponge and tucking it away from the operative field for the duration of the

SC

procedure.

This study is a retrospective review of a select population, and thus has several

M AN U

limitations. As the OBP and IBP groups underwent surgery in different years, variables relating to changes in postoperative care may not have been captured. This is not a randomized study, instead comparing results based on a practice change at our institution in 2013 to eliminate IBP. This change included attempting to avoid nasogastric tube drainage postoperatively. However,

TE D

the results are still likely to be valid, as patients and the surgeries performed were similar. While complications managed at other institutions may not have been captured, we believe that patients would be equally likely to be seen elsewhere for these problems regardless of whether they had

EP

OBP and IBP.

The true cost of IBP may be higher than presented in this paper, as we relied only on

AC C

direct in-hospital costs from an administrative database. Societal costs, including the family’s financial burden of work lost due to an additional 2 days of hospital admission in the IBP group, are not accounted. Additionally, the study was underpowered to adequately identify clinically significant differences in postoperative complications. For a sufficiently powered study (80% power) to be performed to detect a 50% risk reduction of 30-day complications (20% vs. 10%)

11

ACCEPTED MANUSCRIPT

between IBP and OBP groups, 219 children would need to be enrolled in each group. This would likely need to be a multicenter study. Lastly, and perhaps most importantly, it needs to be noted that before surgery, a surgeon

RI PT

does not know with certainty whether the appendix will be usable for an APV and that an

ileovesicostomy with a bowel anastomosis will be not be required. Patients who had a bowel resection were excluded from this study in order to compare only patients without a bowel

SC

anastomosis. Data on patients with a bowel anastomosis continue to be collected. As a result, short-term results reported in this study may not extrapolate to patients without a usable

M AN U

appendix who undergo an ileovesicostomy instead, or undergoing a bladder augmentation. On the other hand, our findings do support avoiding IBP in patients with a high likelihood of a usable appendix, including patients where a synchronous MACE channel was made split-

TE D

appendix technique.

Conclusions

IBP can be safely omitted in patients undergoing an APV. The use of an IBP is associated with

EP

increased LOS and costs. No difference in postoperative complications was observed when

AC C

patients were managed with OBP.

Conflicts of interest None.

Funding None.

12

ACCEPTED MANUSCRIPT

References [1]

Nichols RL, Smith JW, Garcia RY, et al. Current practices of preoperative bowel

preparation among north American colorectal surgeons. Clin Infect Dis 1997;24:609–19. Platell C, Hall J. What is the role of mechanical bowel preparation in patients undergoing

RI PT

[2]

colorectal surgery. Dis Colon Rectum 1998;41:875–83. [3]

Leys CM, Austin MT, Pietsch JB, et al. Elective intestinal operations in infants and

Gundeti M, Godbole P, Wilcox D. Is bowel preparation required before cystoplasty in

children. J Urol 2006;176:1574–7. [5]

M AN U

[4]

SC

children without mechanical bowel preparation: A pilot study. J Pediatr Surg 2005;40:978–82.

Fa Si Oen PR, Verwaest C, Buitenweg J, et al. Effect of mechanical bowel preparation

with polyethyleneglycol on bacterial contamination and wound infection in patients undergoing

[6]

TE D

elective open colon surgery. Clin Microbiol Infect 2005;11:158–60.

Victor D BC, Corbetta J.P Sentagne A, et al. Augmentation cystoplasty in children

without preoperative mechanical bowel preparation. J Pediatr Urol 2012;8:201–4. Dindo D, Demartines N, Clavien PA. Classification of surgical complications. Ann Surg

2004;240:205–13.

Veugelers R, Benning M, et al. Prevalence and clinical presentation of constipation in

AC C

[8]

EP

[7]

children with severe generalized cerebral palsy. Dev Med Child Neurol 2010:e216–e21. [9]

Krogh K, Lie HR, Bilenberg N, et al. Bowel function in Danish children with

myelomeningocele. APMIS Suppl 2003;109:81–5. [10]

Lie HR, Lagergren J, et al. Bowel and bladder control of children with myelomeningocele:

A Nordic study. Dev Med Child Neurol 1991;33:1053–61.

13

ACCEPTED MANUSCRIPT

[11]

Pigeon N, Leroi AM, Devroede G, et al. Colonic transit time in patients with

myelomeningocele. Neurogastroenterol Motil 1997;9:63-70. Bischoff A, Levitt M. Bowel management for the treatment of pediatric fecal

incontinence. Pediatr Surg Int 2009;25:1027–42. [13]

Thorpe DM. Management of opioid-induced constipation. Curr Pain Headache Rep

SC

2001;5:237–40. [14]

RI PT

[12]

Cameron JC. Constipation related to narcotic therapy. A protocol for nurses and

M AN U

patients. Cancer Nurs 1992;15:372–7. [15]

Canty SL. Constipation as a side effect of opioids. Oncol Nurs Forum 1994;21:739–45.

[16]

Lovino P, Chiarioni G, Bilancio G, et al. New onset of constipation during long-term

physical inactivity: A proof-of-concept study on the immobility-induced bowel changes. PLoS

[17]

TE D

One 2013;8:e72608.

Casperson K, Fronczak C, Siparsky G, et al. Ventriculoperitoneal shunt infections after

bladder surgery: Is mechanical bowel preparation necessary. J Urol 2011;186:1571–5. Yerkes E, Rink R, Cain M. shunt infection and malfunction after augmentation

EP

[18]

cystoplasty. J Urol 2001;165:2262–4. Matthews GJ, Churchill BA, McLorie GA. Ventriculoperitoneal shunt infection after

AC C

[19]

augmentation cystoplasty. J Urol 1996;155:686–8. [20]

Kreder KJ, Webster GD, Oakes WJ. Augmentation cystoplasty complicated by

postoperative ventriculoperitoneal shunt infection. J Urol 1990;144.

14

ACCEPTED MANUSCRIPT

Table 1 Population characteristics.

Ventriculoperitoneal shunt

0.99

8.2 (4.5–18.0)

7.7 (4.0–13.7)

0.36

10 (27.0%)

Body mass index (kg/m2) (median, IQR) GFRa (median, IQR)

0.99

17.6 (15.4–24.7)

16.5 (14.4–20.0)

0.07

104.3 (46.7–159.5)

118.8 (39.0–167.7)

0.92

M AN U

12 32.4%)

Other spine pathology Other bladder pathology Associated surgeries

Bladder Otherb a

8 (26.7%)

5 (16.7%)

21 (56.8%)

17 (56.7%)

25 (67.6%)

22 (73.3%)

0.79

15 (40.5%)

13 (43.3%)

0.99

17 (46.0%)

13 (43.3%)

0.99

1 (2.7%)

1 (3.3%)

0.99

AC C

EP

Data missing for four patients in the outpatient group, seven in inpatient group. Other surgeries included ovarian cystectomy (benign) and urachal cyst excision (benign).

b

0.74

4 (10.8%)

TE D

MACE

p

8 (26.7%)

Diagnosis Lipo/myelomeningocele

RI PT

Age (median, IQR)

Inpatient bowel prep (n = 30) 19 (63.3 %)

SC

Male gender

Outpatient bowel prep (n = 37) 23 (62.2 %)

ACCEPTED MANUSCRIPT

Table 2 Patient outcomes. Mechanical bowel preparation group (n = 30) (%)

5 4–9 3–14

7 5–11 5–29

Clavien grade 1 Clavien grade 2 Cost a, US$ Median IQR Range a

0 0–0 0–1 5 4–9 3–14 2 (5.4)

1 1–1 1–2

SC

LOS days Median IQR Range Preoperative LOS days Median IQR Range Postoperative LOS (days Median IQR Range Complications at 30 days (any)

6 4–10 4–28 6 (20.0)

1

1

1

5

14,322 9,998–25,338 9,315–33,740

18,610 13,433–30,682 10,430–94,164

EP

TE D

Data missing for two patients in outpatient group, one in inpatient group.

AC C

Difference

p

+2.0

0.0002

RI PT

Outpatient bowel preparation group (n = 37)

M AN U

Outcome

+1.0

0.0001

+1.0

0.053

+15.6%

0.13

+ 4,288 (+29.9%)

0.01

00 15 Number 2 3 4 5 6 7 8 1 9 10 Length Outpatient Mechanical of bowel patients hospital prepration stay

(days)

RI PT SC TE D

M AN U

10

AC C

EP

5 0

Number of patients

15

ACCEPTED MANUSCRIPT

1

2

3

4

5

6

7

8

Length of hospital stay (days)

Outpatient bowel prepration

9

10

Mechanical bowel prepration

00 15 Number 2 3 4 5 6 7 8 1 9 10 Length Outpatient Mechanical of bowel patients hospital prepration stay

(days)

RI PT SC TE D

M AN U

10

AC C

EP

5 0

Number of patients

15

ACCEPTED MANUSCRIPT

1

2

3

4

5

6

7

8

Length of hospital stay (days)

Outpatient bowel prepration

9

10

Mechanical bowel prepration