Original Research

Mechanical Bowel Preparation Before Laparoscopic Hysterectomy A Randomized Controlled Trial Matthew T. Siedhoff, MD, MSCR, Leslie H. Clark, MD, Kumari A. Hobbs, MD, Austin D. Findley, MD, MSCR, Janelle K. Moulder, MD, and Joanne M. Garrett,

PhD

OBJECTIVE: To examine the influence of mechanical bowel preparation on surgical field visualization during laparoscopic hysterectomy.

CONCLUSION: Mechanical bowel preparation is welltolerated but does not influence surgical field visualization for laparoscopic hysterectomy.

METHODS: The study’s primary outcome was the percentage of operations rated “good” or “excellent” in terms of surgical field visualization at the outset of the case by the primary surgeon. Additional measures included assessment of visualization during the case and patient perioperative comfort. The study was powered to detect a 20% absolute difference in the proportion of cases rated as “good” or “excellent.”

CLINICAL TRIAL REGISTRATION: ClinialTrials.gov, www.clinicaltrials.gov, NCT01576965.

RESULTS: Seventy-three patients were assigned to mechanical bowel preparation and 73 to no mechanical bowel preparation. The groups were comparable regarding patient and surgery characteristics. No differences were found for this rating between groups (mechanical bowel preparation, 64 of 73 patients [87.7%], compared with no mechanical bowel preparation, 60 of 73 patients [82.2%], P5.36). Surgeons guessed patient assignment correctly 59% of the time (42 of 71 patients) with mechanical bowel preparation and 55% of the time (41 of 75 patients) with no mechanical bowel preparation.

From the University of North Carolina at Chapel Hill, Chapel Hill, North Carolina. Funding was provided by a departmental research fund. Preliminary data were presented as an abstract at the 42nd Annual AAGL Global Congress on Minimally Invasive Gynecology, November 12, 2013, Washington, DC. Corresponding author: Matthew T. Siedhoff, MD, MSCR, University of North Carolina at Chapel Hill, Department of Obstetrics & Gynecology, Division of Advanced Laparoscopy & Pelvic Pain, Campus Box 7570, Chapel Hill, NC 27599-7570; e-mail: [email protected]. Financial Disclosure The authors did not report any potential conflicts of interest. © 2014 by The American College of Obstetricians and Gynecologists. Published by Lippincott Williams & Wilkins. ISSN: 0029-7844/14

562

VOL. 123, NO. 3, MARCH 2014

(Obstet Gynecol 2014;123:562–7) DOI: 10.1097/AOG.0000000000000121

LEVEL OF EVIDENCE: I

M

echanical bowel preparation is thought to offer several advantages to the surgeon in abdominopelvic surgery, including enhanced visualization of the surgical field, improved intraoperative bowel handling, and reduced fecal contamination in the setting of bowel injury. This routine practice, however, is largely untested and must be weighed against patient risks and discomfort associated with mechanical bowel preparation. Historically, mechanical bowel preparation use in gynecology was largely extrapolated from colorectal surgery practice.1–4 Newer literature in that discipline, however, suggests mechanical bowel preparation does not improve surgical outcomes2,3 and may even increase the risk of adverse events such as anastomotic leaks after bowel resection.3 Nearly all of these operations, however, were laparotomies in the supine position, a situation with markedly different visualization needs from gynecologic laparoscopy. A recent review5 cites a paucity of literature on the topic, and indeed there is no national standard of care or even consistent practice among gynecologic surgeons at our institution. In this trial, we sought to build on previous work investigating the value of mechanical bowel preparation in advanced gynecologic laparoscopy, focusing specifically on hysterectomy to standardize the procedure and because hysterectomy involves operating deep in the pelvis where the putative benefits of mechanical bowel

OBSTETRICS & GYNECOLOGY

preparation may be most relevant. Overweight patients or those with prior surgeries were not excluded to maximize the applicability of our findings.

MATERIALS AND METHODS Research ethics approval was provided by the institutional review board at the University of North Carolina at Chapel Hill. The trial was a single-masked, randomized controlled trial carried out according to the guidelines of the most recent CONsolidated Standards of Reporting Trials (CONSORT) statement6 and was registered with www.clinicaltrials.gov (Identifier: NCT01576965). Beginning in April 2012, all nonpregnant women ages 18 years and older undergoing elective laparoscopic hysterectomy for benign indications and able to provide informed consent in English were invited to participate. Exclusion criteria included inability to perform enema mechanical bowel preparation, known or suspected malignancy, and severe endometriosis of the posterior cul de sac where bowel resection might be needed at the time of surgery. In a prior study comparing two types of mechanical bowel preparation,7 approximately 90% of the operations were rated as “excellent” or “good” (as opposed to “fair” or “poor”) in terms of surgical field visualization. Using a superiority design, we calculated 72 women would be needed in each arm to demonstrate a 20% absolute difference in the proportion of women whose surgeries were rated in these categories at 80% power with a two-sided a of .05. To account for dropout and nonadherence, we aimed to recruit 160 women. Patients were randomized (one to one) to mechanical bowel preparation compared with no mechanical bowel preparation using a six-subject computer-generated blocked randomization scheme. The study was briefly introduced to patients by written handout after a decision was made to proceed with laparoscopic hysterectomy, and full informed consent was obtained by telephone. A research assistant not involved in outcome assessment or data entry mailed patients preoperative instructions and a small monetary reward ($5) in an opaque sealed envelope. Mailings also included mechanical bowel preparation supplies for those allocated to that group. All patients were instructed to restrict their diet to clear liquids the day before surgery and to fast after midnight save a small sip of water for medications in the morning. Those assigned to the mechanical bowel preparation group were asked to administer a single sodium phosphate enema before going to bed the night before surgery. If stool was not clear in the

VOL. 123, NO. 3, MARCH 2014

Siedhoff et al

morning, mechanical bowel preparation patients were to administer one additional enema the morning of surgery. The primary outcome was measured using a selfadministered questionnaire completed immediately after the procedure by the attending surgeon. Four surgeons were involved over the course of the trial. Using a 4-point Likert scale—“excellent,” “good,” “fair,” or “poor”—surgeons rated the visual quality of the surgical field at the outset of the case. Other outcomes, including visualization at the conclusion of the case, ease of bowel handling, degree of small and large bowel preparation, visualization of the uterus, adnexae, posterior cul de sac, and overall difficulty of the operation, were rated on the same scale. Patients completed questionnaires regarding their relative comfort leading up to the surgery in the preoperative holding area, rating symptoms on a 100-mm visual analog scale. Those assigned to the mechanical bowel preparation group were asked if they were able to complete the regimen and if they would be willing to perform the same preparation before a future surgery. Patient characteristics such as body mass index (BMI, calculated as weight [kg]/[height (m)]2), age, and gastrointestinal comorbid conditions as well as operation characteristics (eg, estimated blood loss, uterine weight, intraoperative complications) were recorded immediately after the surgery. Patients were called 3–5 days after surgery to inquire about any problems with constipation. Postoperative complications that occurred within 6 weeks after the surgery date were reviewed in the electronic medical record. Although patients could not be masked to their allocation, they were asked not to reveal whether they performed mechanical bowel preparation to their surgeon and were reminded of this request when they were given the study survey. As a safety measure, patients could discuss their assignment with the anesthesiology team in the preoperative holding area so intravenous fluids could be managed appropriately and masking could be broken if necessary during the operation. Data analysis was carried out according to both an intent-to-treat and per-protocol manner. We examined baseline and operative characteristics as differences in means, medians, and proportions. The statistical software package STATA 11 was used for all data analyses. The primary outcome was represented as the proportion of women whose operations were rated as “good” or “excellent” by the primary surgeon in terms of visual quality at the outset of the operation, compared using Pearson’s x2 test. Ten percent of the operations were randomly selected for

Mechanical Bowel Prep Before Laparoscopic Hysterectomy 563

evaluation by the first assistant surgeon using the same Likert scales, and agreement was measured using a kappa statistic. Secondary outcomes were reported as mean visual analog scale scores for preoperative symptoms in both groups, compared using Student’s t test. Subanalyses were performed comparing these outcomes among factors known to influence laparoscopic hysterectomy,8 including normal-weight compared with obese patients, those with deeply infiltrating endometriosis (defined as lesions greater than 5 mm in depth or affecting other organs or ligaments9) or dense adhesions requiring lysis to complete the operation, and those with small (less than 500 g) compared with large (greater than 500 g) uteri. We also examined outcomes according to surgeon ratings of the difficulty of the operation and the presence of comorbid gastrointestinal conditions.

RESULTS Between April 2012 and August 2013, a total of 377 patients were evaluated for participation in the trial (Fig. 1). Of these, 31 did not meet eligibility requirements: 21 did not speak English, three were prescribed bowel preparation by their surgeon, two had suspected or known deeply infiltrating endometriosis, one had suspected malignancy, one had an end ileostomy, Assessed for eligibility N=377

Randomized: n=160

Excluded: n=217 • Not meeting inclusion criteria: 31 • Declined to participate or not able to be reached: 186

Allocated to mechanical bowel preparation: n=79 • Received allocated intervention: 73 • Did not receive allocated intervention; cancelled surgery: 6

Allocated to no mechanical bowel preparation: n=81 • Received allocated intervention: 73 • Did not receive allocated intervention; cancelled surgery: 8

Discontinued intervention: n=2 Unable to continue: 1 Did not understand instructions: 1

Discontinued intervention: 0

Analyzed: n=73 • Intent-to-treat: n=73 • Per protocol: n=71

Analyzed: n=73 • Intent-to-treat: n=73 • Per protocol: n=75

Fig. 1. Study flow diagram. Clark. Mechanical Bowel Prep Before Laparoscopic Hysterectomy. Obstet Gynecol 2014.

564

Siedhoff et al

one was quadriplegic, one had mental disability, and one was incarcerated. Among the remaining 346 patients, 54% (n5186) either could not be reached by phone or declined participation. Thus, 160 women consented and were randomized. Fourteen patients canceled surgery after allocation, leaving 146 who completed the study. Of those, 73 were randomized to mechanical bowel preparation and 73 to no mechanical bowel preparation. Among the women who canceled, six (7.6%) had been randomized to mechanical bowel preparation and eight (9.9%) to no mechanical bowel preparation (P5.61, Pearson’s x2 test). Two patients assigned to mechanical bowel preparation did not complete the preparation: one was unable and the other did not understand the instructions. The groups were comparable in terms of patient characteristics (Table 1). On average, women in the study were 42 years old (range 22–67 years old) and obese with BMIs of 33.4 kg/m2 (range 18–62 kg/m2). All patients elected total, rather than supracervical, hysterectomy and the vast majority (141 patients [97%]) was performed with conventional, as opposed to robotic, laparoscopy. The most common indications for surgery included abnormal bleeding (80 patients [55%]), leiomyomas (78 patients [53%]), and pelvic pain (65 patients [45%]). Other indications included adnexal mass (eight patients), gender reassignment (four patients), pelvic organ prolapse (two patients), and cervical dysplasia (one patient). Characteristics of surgery also did not differ between groups (Table 2). Mean surgery time was 139 minutes (range 50–350 minutes), estimated blood loss 61 mL (range 10–500 mL), and uterine weight 385 g (range 34–3,300 g). Fifty-one surgeries (34.9%) were rated as “difficult” or “moderately difficult” by the primary surgeon as opposed to “fair” or “easy.” Seven patients experienced a complication, including one who sustained two (both urinary retention and urinary tract infection). The most serious complication was a ureteral injury, which presented as a ureterovaginal fistula requiring ureteral reimplantation. Other complications included an unavoidable small bowel enterotomy during adhesiolysis, which was repaired laparoscopically, an umbilical incision separation, which healed by secondary intention, ureteral kinking (diagnosed by lack of efflux on cystoscopy and corrected by resuturing the vaginal cuff), vestibuloplasty incision separation revised in the operating room, and a cuff separation repaired vaginally in the operating room. None of the complications were felt to be specifically the result of poor visualization. At least one additional procedure was performed at the time of hysterectomy for 127 patients (87%) (excluding

Mechanical Bowel Prep Before Laparoscopic Hysterectomy

OBSTETRICS & GYNECOLOGY

Table 1. Baseline Characteristics of Study Participants Patient Characteristics

Total (n5146)

Mechanical Bowel Preparation (n573)

No Mechanical Bowel Preparation (n573)

41.668 33.568 16 (11.0) 7 (4.8) 2 (1.4) 6 (4.1) 1 (0–4) 1 (0–6)

40.968 32.367 9 (12.3) 5 (6.9) 1 (1.4) 3 (4.1) 1 (0–4) 0 (0–6)

42.669 34.669 7 (9.6) 2 (2.7) 1 (1.4) 3 (4.1) 1 (0–4) 1 (0–5)

Age (y) BMI (kg/m2) Reflux disease Irritable bowel syndrome Inflammatory bowel disease Chronic constipation Prior laparoscopies Prior laparotomies

BMI, body mass index. Data are mean6standard deviation, n (%), or median (range). Continuous variables (age and BMI) are listed as means. Ordinal variables (number of prior laparoscopies and prior laparotomies) are listed as medians. Categorical variables (reflux disease, irritable bowel syndrome, inflammatory bowel disease, and those with chronic constipation) are listed as proportions.

standard cystoscopy), the most common being removal of fallopian tubes, ovaries, or both (82.2%) followed by adhesiolysis (20.1%) and appendectomy (17.1%). There was no clinically meaningful or statistically significant difference between the groups for the study’s primary outcome, the proportion of surgeries rated as “good” or “excellent” by the primary surgeon in terms of surgical field visualization after trocar placement and Trendelenburg positioning (Table 3). In the mechanical bowel preparation group, 64 patients (87.7%) received this rating compared with 60 patients (82.2%) in the no mechanical bowel preparation (P5.36, Pearson’s x2 test). No differences were noted among any of the other surgeon ratings or patient preoperative assessments apart from a small improvement in insomnia among the mechanical bowel preparation group (Table 4). There was moderate agreement among primary and assistant surgeon ratings (kappa5.56). Three surgeries required an

extra technique to improve visualization (eg, resection or suturing of anterior abdominal wall fat, suturing large bowel epiploica), one woman in the mechanical bowel preparation group and two in the no mechanical bowel preparation group (P5.50, Fisher’s exact test). None of the results differed when analyzed by subgroup: obese (BMI greater than 30 kg/m2) compared with not obese, deeply infiltrating endometriosis, dense adhesions requiring lysis, large (uterine weight greater than 500 g) compared with small (uterine weight less than 500 g) uteri, comorbid irritable bowel syndrome, inflammatory bowel disease (ie, Crohn’s disease or ulcerative colitis), gastroesophageal reflux disease, or a history of chronic constipation. Including the two patients assigned to mechanical bowel preparation, but who did not complete the preparation, in the no mechanical bowel preparation group (ie, per protocol rather than intent to treat) also did not change any statistical result.

Table 2. Characteristics of Surgery Characteristics of Surgery Surgery time (min) Estimated blood loss (mL) Uterine weight (g) Length of stay (d) Dense adhesions requiring lysis Deeply infiltrating endometriosis Rated moderately difficult or difficult Experienced intraoperative or postoperative complication Had additional procedure(s)

Total (n5146)

Mechanical Bowel Preparation (n573)

No Mechanical Bowel Preparation (n573)

139655 61668 3856528 1 (0–5) 30 (20.6) 6 (4.1) 51 (34.9) 7 (4.8)

139657 58664 3146493 1 (0–5) 17 (23.3) 2 (2.7) 25 (34.3) 2 (2.7)

138653 64672 4576555 1 (0–2) 13 (17.8) 4 (5.5) 26 (35.6) 5 (6.9)

127 (87.0)

64 (87.7)

63 (86.3)

Data are mean6standard deviation, median (range), or n (%). Continuous variables (surgery time, estimated blood loss, uterine weight) are listed as means. Length of stay is listed as medians. Categorical variables (had additional procedure[s], presence of dense adhesions, presence of deeply infiltrating endometriosis, rated as moderately difficult or difficult by the surgeon, experienced complication[s]) are listed as proportions.

VOL. 123, NO. 3, MARCH 2014

Siedhoff et al

Mechanical Bowel Prep Before Laparoscopic Hysterectomy 565

Table 3. Surgeon Rating of Visual Field Surgeon Rating*

Total (n5146)

Visual field at start Visual field at end Quality of bowel handling Degree of small bowel preparation Degree of large bowel preparation Visualization of uterus Visualization of adnexae Visualization of posterior cul de sac

124 138 132 138 110 126 129 114

Mechanical Bowel Preparation (n573)

(84.9) (94.5) (90.4) (94.5) (75.3) (86.3) (88.4) (78.1)

64 69 65 69 58 62 67 57

(87.7) (94.5) (89.4) (94.5) (79.5) (84.9) (91.8) (78.1)

No Mechanical Bowel Preparation (n573) 60 69 67 69 52 64 62 57

(82.2) (94.5) (91.8) (94.5) (71.2) (87.7) (84.9) (78.1)

P .36 .99 .57 .99 .25 .23 .20 .99

Data are n (%) unless otherwise specified. * Number (%) rated by the primary surgeon as “good” or “excellent” rather than “fair” or “poor” on a 4-point Likert scale.

Nineteen (13.0%) of patients struggled with postoperative constipation. This did not differ between mechanical bowel preparation (11 patients [15.1%]) and no mechanical bowel preparation (eight patients [11.0%]) groups (P5.46, Pearson’s x2 test). All 19 patients with postoperative constipation did not have a history of preoperative chronic constipation, and none of the six patients with preoperative chronic constipation struggled with the problem postoperatively. In the 75 operations in which no mechanical bowel preparation was done, surgeons guessed the correct allocation only 55% (41 patients) of the time. Among the 71 in whom patients completed mechanical bowel preparation, surgeons performed only slightly better at 59% (42 patients). Neither of these were significantly different from a 50:50 guess (P5.56 for the no preparation group, P5.27 for the preparation group, Pearson’s x2 test).

DISCUSSION This trial demonstrates no significant value for mechanical bowel preparation before laparoscopic

hysterectomy. Although patients generally did not rate negative symptoms strongly and the vast majority would be willing to repeat mechanical bowel preparation in a subsequent operation, the procedure did not offer any advantage nor even prevent postoperative constipation. A compelling additional finding was that surgeons were only able to correctly guess patient allocation a little more then half of the time in this study. Only four trials regarding mechanical bowel preparation in gynecologic laparoscopy have been performed.6,10–12 One did not include a control group, two involved relatively simple operations, and none specifically target hysterectomy. Strengths of this study include the standardization of the operation, the randomized, single-masked design, and the inclusion of a population with complex pathology. Some might argue an enema mechanical bowel preparation is an insufficient preparation of the bowel. Prior research, however, has demonstrated equivalence between oral and enema mechanical bowel preparation,7 and more often it is the colon that interferes with

Table 4. Patient Rating of Symptoms Patient Rating*

Total (n5146)

Mechanical Bowel Preparation (n573)

No Mechanical Bowel Preparation (n573)

P

Cramps Hunger Bloating Embarrassment Insomnia Weakness Dizziness Thirst Nausea Incontinence

17624 42633 17625 10623 21629 13623 9618 39634 8619 6616

18623 43634 17625 12623 16624 11619 9619 35632 6614 7615

15625 41632 16625 9623 26634 15626 8618 43636 9623 6618

.46 .60 .78 .47 .04 .30 .90 .16 .31 .55

Data are mean6standard deviation unless otherwise specified. * Mean (standard deviation) patient rating for symptoms experienced in the 24 hours before surgery rated on a 100-mm visual analog scale.

566

Siedhoff et al

Mechanical Bowel Prep Before Laparoscopic Hysterectomy

OBSTETRICS & GYNECOLOGY

visualization in the pelvis, an area likely well addressed with enema mechanical bowel preparation. Although there were a few patients with unanticipated advanced endometriosis, this trial specifically excluded patients suspected of needing a bowel resection for that disease, a debatable exclusion criterion and a limitation of the study. Safety or prevention of complications was not an intended outcome of this trial because it has already been demonstrated bowel resection can be safely performed in unprepped patients.4 However, many surgeons would prefer not having hard stool directly in the line of an intended anastomosis, so simple measures such as placement of a bisacodyl suppository the night before anticipated bowel resection might be reasonable. Although the study did not have adequate power to test differences among all subgroups, none of the additional analyses suggested a clinically meaningful difference between mechanical bowel preparation and no mechanical bowel preparation, suggesting simply enrolling more patients would not have produced a different result. REFERENCES 1. Hares MM, Alexander-Williams J. The effect of bowel preparation on colonic surgery. World J Surg 1982;6:175–81. 2. Platell C, Hall J. What is the role of mechanical bowel preparation in patients undergoing colorectal surgery? Dis Colon Rectum 1998;41:875–82. 3. Mahajna A, Krausz M, Rosin D, Shabtai M, Hershko D, Ayalon A, et al. Bowel preparation is associated with spillage

VOL. 123, NO. 3, MARCH 2014

Siedhoff et al

of bowel contents in colorectal surgery. Dis Colon Rectum 2005;48:1626–31. 4. Guenaga KF, Matos D, Castro AA, Atallah AN, WilleJørgensen P. Mechanical bowel preparation for elective colorectal surgery. The Cochrane Database of Systematic Reviews 2011, Issue 2. Art. No.: CD001544. DOI: 10.1002/14651858. CD001544. 5. Cohen SL, Einarsson JI. The role of mechanical bowel preparation in gynecologic laparoscopy. Rev Obstet Gynecol 2011;4: 28–31. 6. Schulz KF, Altman DG, Moher D; CONSORT Group. CONSORT 2010 statement: updated guidelines for reporting parallel group randomized trials. Obstet Gynecol 2010;115:1063–70. 7. Yang LC, Arden D, Lee TT, Mansuria SM, Broach AN, D’Ambrosio L, et al. Mechanical bowel preparation for gynecologic laparoscopy: a prospective randomized trial of oral sodium phosphate solution vs single sodium phosphate enema. J Minim Invasive Gynecol 2011;18:149–56. 8. Siedhoff MT, Carey ET, Findley AD, Riggins LE, Garrett JM, Steege JF. Effect of extreme obesity on outcomes in laparoscopic hysterectomy. J Minim Invasive Gynecol 2012;19:701–7. 9. Kennedy S, Bergqvist A, Chapron C, D’Hooghe T, Dunselman G, Greb R, et al. ESHRE guideline for the diagnosis and treatment of endometriosis. Hum Reprod 2005;20:2698–704. 10. 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; 85:689–93. 11. Lijoi D, Ferrero S, Mistrangelo E, Casa ID, Crosa M, Remorgida V, et al. Bowel preparation before laparoscopic gynaecological surgery in benign conditions using a 1-week low fibre diet: a surgeon blind, randomized and controlled trial. Arch Gynecol Obstet 2009;280:713–8. 12. Won H, Maley P, Salim S, Rao A, Campbell NT, Abbott JA. Surgical and patient outcomes using mechanical bowel preparation before laparoscopic gynecologic surgery: a randomized controlled trial. Obstet Gynecol 2013;121:538–46.

Mechanical Bowel Prep Before Laparoscopic Hysterectomy 567

Mechanical bowel preparation before laparoscopic hysterectomy: a randomized controlled trial.

To examine the influence of mechanical bowel preparation on surgical field visualization during laparoscopic hysterectomy...
154KB Sizes 1 Downloads 0 Views