Objective
The objective of the study was to compare time to first bowel movement (BM) after surgery in subjects randomized to placebo or senna with docusate.
Study Design
Ninety-six subjects completed a baseline 7-day bowel diary before and after surgery. After pelvic reconstructive surgery, the subjects were randomized to either placebo (n = 45) or senna (8.6 mg) with docusate (50 mg) (n = 48). Time to first BM and postoperative use of magnesium citrate were compared.
Results
There was a significant difference in the time to first BM in those receiving senna with docusate vs placebo (3.00 ± 1.50 vs 4.05 ± 1.50 days; P < .002). More subjects in the placebo group needed to use magnesium citrate to initiate a bowel movement (43.6% vs 7.0%; P < .001).
Conclusion
The use of senna with docusate decreases time to first BM in those undergoing pelvic reconstructive surgery compared with placebo. Subjects using senna with docusate are also significantly less likely to use magnesium citrate.
Postoperative bowel function is a significant concern for women undergoing pelvic floor reconstructive surgery, as well as for their physicians. There are studies looking at treating constipation using fiber and wheat supplementation in pregnancy, general surgery, colorectal surgery, orthopedic surgery, and hospitalized patients. Collectively, these studies show some benefit to using additional fiber and/or wheat to help relieve or prevent constipation.
Prior research has found a significant discordance between surgeon and patient impression of postoperative constipation, highlighting the need for better patient education. Typically, the first bowel movement (BM) may not occur until at least the second postoperative day, and subsequent delays can predispose patients to impaction and discomfort. The use of preoperative bowel preparations, lasting effects of anesthetic medications, narcotic pain medications, surgical manipulation, and general postoperative immobility predispose this patient population to constipation.
Senna has been shown to decrease gut transit time. A study using a regimen of senna and docusate following different anorectal surgeries found that 80% of patients were able to have a BM on postoperative day 1 after only 2 tablets of senna with docusate. No patients in this study required an enema, and the only side effect reported was 1 case of diarrhea in a patient with “a history of loose stools in the past.”
There is no standardized medication regimen that has been tested for relief of constipation in the benign gynecology or urogynecology postoperative patients. The aim of this study was to compare the use of senna with docusate with placebo after pelvic reconstructive surgery to determine whether senna with docusate reduces time to first BM.
Materials and Methods
The institutional review board of Hartford Hospital approved this study. All subjects enrolled signed informed consent. All women, aged 18 years and older, undergoing pelvic reconstructive surgery requiring overnight admission, were eligible for enrollment. Patients with a history of ulcerative colitis, Crohn’s disease, irritable bowel disease, and gastroparesis; those undergoing concurrent bowel resection anal sphincteroplasty or repair of rectovaginal fistula; and those with intraoperative bowel or rectal injury or extensive lysis of adhesions were excluded. Those who were dependent on bowel stimulants or laxatives were also excluded. Subjects were allowed to continue the use of fiber supplements if they had been on these before surgery. Any patients who were unable to provide informed consent or follow the instructions to complete study-related materials were also excluded.
After enrollment, subjects completed a 7-day bowel diary to provide a baseline assessment of bowel frequency, type of BM using the validated Bristol stool scale, degree of strain and pain (measured on an 11-point visual analog scale from 0–10), and medications taken. The Bristol stool scale provides a reliable and valid measure of gut transit time based on the shape, consistency, and appearance of BMs. The subjects rate the type of BM on a scale from 1–7. Types 1 and 2 are consistent with reduced gut transit time; types 3 and 4 represent normal transit time; and types 5, 6, and 7 indicate increased gut transit time.
After surgery, if there was no unintentional bowel or rectal injury or significant adhesions, the subject was randomized to either senna with docusate (SennaS; Purdue Pharmaceuticals, Stamford, CT) or matching placebo. The placebo pills were generated to match the senna with docusate pills by Pioneer Health compounding pharmacy (Vernon, CT).
The subjects were provided with identical instructions on how to take the medications after surgery. The subjects started with 2 tablets at night on postoperative day 1 (either at home or in the hospital); followed by 2 tablets the following morning if no BM; followed by 3 tablets in the evening if no BM; and then, if no BM the following morning, 3 additional tablets. The subjects used 3 tablets twice per day until their first BM or took magnesium citrate if they did not have a BM on the fourth day after surgery.
Subjects were to use the minimum number of tablets of the medication to maintain soft regular BMs and start to cut back as they resumed their usual diet and activities. They were instructed to stop the medication if they had loose stools or diarrhea. If they had to use magnesium citrate on the fourth day after surgery and then became constipated, they could use either polyethylene glycol or milk of magnesia instead of the study medication.
Randomization was performed using computer-generated blocks by a neutral individual. Investigators, study nurse, and subjects were blinded to treatment allocation and were unblinded only if necessary for subject safety.
The study medication was dispensed to the subject on the postoperative floor prior to discharge by a neutral individual. The study medication was started on postoperative day 1. All subjects were given another diary to complete using the Bristol stool scale, pain, strain, and list of medications being taken. They recorded information on the diary until the first BM. If no BM occurred by the fourth postoperative day, subjects were instructed to use 5 ounces of magnesium citrate to initiate a BM. If they did not have a BM with the magnesium citrate, they were to call the office for further instructions to use an enema or come to the office for evaluation.
All subjects received a phone call within the first postoperative week about any pain, cramps, flatulence, or discomfort associated with the study medication. After the first BM, the subjects could follow the same regimen with the study medication, stop the study medication, or use a different medication to maintain regular BMs. All subjects were instructed to stop the study medication if they had loose stools or diarrhea.
Subjects were enrolled in a 1:1 ratio. We estimated that the patients taking senna with docusate would have a BM 2 days earlier than those receiving placebo, extrapolating from the study by Corman, in which a majority of the patients receiving senna with docusate had a bowel movement on the first postoperative day after only 2 doses of the medication.
Because we did not know whether these patients had a preoperative bowel preparation, and preoperative bowel preparations can increase time of return of bowel movements after surgery, we estimated that 1 additional day would be necessary to initiate a BM in the senna with docusate group and those receiving placebo would not have a bowel movement until postoperative day 4 or beyond. Thus, a sample size of 49 subjects in each group (a total sample size of 98) was calculated to afford 90% power to detect a difference in mean time to BM of 2.0 days, assuming that the common SD was 3.0 days, using a 2-group Student t test with a 0.05 2-sided significance level.
This number of participants also would afford sufficient power to detect a statistically significant difference in the expected proportion of patients requiring magnesium citrate. We estimated a 5% dropout or exclusion rate of those who consented to participate. Therefore, we planned to enroll 104 patients, 52 subjects in each group, into the study.
The primary outcome, time to first BM in each group, was compared with a 2-group Student t test. Time to first BM was calculated as the difference between the date and time of the subject’s surgery and the date and time of first BM after surgery, as listed on the postoperative bowel diaries.
The secondary outcome, use of magnesium citrate in each group, was compared using Fisher’s exact test. Other comparisons were type of BM based on Bristol stool scale, frequency of bowel movements, straining at stool, and pain with bowel movements. A constipation score for subjects was calculated based on preoperative stool frequency, postoperative days to first BM, stool consistency, percentage of straining, and sensation of incomplete evacuation.
Constipation scores were calculated as follows: days to first BM (2 points if >4 days, 0 points if 1–3 days); stool consistency (1 point if types 1 or 2 on Bristol stool scale, 0 points for types 3–7 on the Bristol stool scale); strain rating (0.5 points if for ≥5 on ≥25% of the BMs; 0 points for ≤4 on 0–25% of the BMs; 0.5 point for rating of mild, moderate, or severe for sense of incomplete evacuation to the question: “Have you experienced the following symptom: incomplete bowel movement, like you did not finish?” with 0 point for rating of none).
The scoring system was developed by a study in cardiac patients using lactulose after surgery for treatment of constipation. A score of 2 or more was consistent with constipation based on the Rome III criteria. All analyses were conducted with SPSS 14.0 (SPSS, Inc, Chicago, IL) at the 0.05 significance level such that all comparisons yielding P < .05 were deemed statistically significant.
Results
A total of 82 subjects had complete data for analysis ( Figure ). There was a Food and Drug Administration (FDA) recall on senna with docusate before completion of the study because of 2 missing ingredients (lactose and tartaric acid) on the label of the medication, so enrollment was stopped early. In addition, analysis of results found that there was a significantly higher need for subjects in the placebo arm to use magnesium citrate, so for ethical reasons we elected not to resume enrollment once the medications were available. Three subjects were enrolled but not randomized because of the recall on the senna with docusate medications prior to surgery and were not included in this analysis. Two subjects were not randomized because of an intraoperative rectal injury that was recognized and repaired.
