26, 2007 — Mechanical bowel preparation for elective colorectal surgery may not be necessary, according to the results of a multicenter, randomized, noninferiority trial reported in the December 22/29 issue of The Lancet. Although this trial showed no benefits associated with mechanical bowel preparation other than lower rate of abscess, the editorialist in an accompanying commentary disagrees with the study authors' conclusion that it is no longer necessary.
"Symptomatic anastomotic leakage is the most important surgical complication after colorectal surgery and can cause morbidity and mortality," write Caroline M.E. Contant, MD, from the Ikazia Hospital in Rotterdam, the Netherlands, and colleagues. "Mechanical bowel preparation has been regarded as an efficient strategy to prevent anastomotic leakage and septic complications. . . . We aimed to compare the rate of anastomotic leakage after elective colorectal resections and primary anastomoses between patients who did or did not have mechanical bowel preparation."
At 13 hospitals, a total of 1431 patients scheduled for elective colorectal surgery were randomized to receive or not to receive mechanical bowel preparation. Those who did not have mechanical bowel preparation had a normal meal on the day before surgery, whereas those who did received a fluid diet and mechanical bowel preparation with either polyethylene glycol or sodium phosphate. The main outcome measure was anastomotic leakage.
As a noninferiority trial, the study was designed to test the hypothesis that patients who receive preoperative mechanical bowel preparation do not have a lower risk for anastomotic leakage than those who do not. Median follow-up was 24 days (interquartile range, 17 - 34 days), and analysis was as per protocol.
Of 77 patients who were excluded, 46 did not undergo bowel resection; 21 were missing outcome data; and 10 withdrew, cancelled, or were excluded for other reasons. The rate of anastomotic leakage was statistically similar in both groups: 32 (4.8%) of 670 patients who had mechanical bowel preparation and 37 (5.4%) of 684 who did not (difference, 0.6%; 95% confidence interval, −1.7% to 2.9%; P = .69).
Compared with patients who did not have mechanical bowel preparation, those who did had fewer abscesses after anastomotic leakage (0.3% [2/670] vs 2.5% [17/684]; P = .001). Other septic complications, fascial dehiscence, length of stay, and mortality were similar in both groups.
"The conclusion that elective colorectal surgery can be safely done without mechanical bowel preparation is justified," the study authors write. "In view of possible disadvantages of this practice, patient discomfort, and the absence of clinical value, we advise that mechanical bowel preparation before elective colorectal surgery should be abandoned."
Limitations of this trial include lack of blinding, not all eligible patients who could potentially have been enrolled were registered, use of 2 different oral regimens for mechanical bowel preparation, failure to record the exact height of anastomosis below the pelvic verge, and analysis of only 1354 patients.
In an accompanying commentary, Cameron Platell and John Hall, from the St. John of God Hospital in West Perth, and the University of Western Australia in Perth, note several concerns with methodologic issues and analyses in this study. They point out that although the absolute difference in leak rates was less than 1%, it amounted to a 20% relative difference between the groups.
"Some doubts still need to be resolved, especially for patients undergoing low rectal anastomoses," Drs. Platell and Hall write. "Although evidence from trials favours not having mechanical bowel preparation, we should consider each case carefully, otherwise the chance of making an inappropriate decision exists, with great consequences for patients."
Drs. Platell and Hall have disclosed no relevant financial relationships.
Lancet. 2007;370:2073-2075, 2112-2117.
Clinical Context
Observational data suggest that mechanical bowel preparation before colorectal surgery reduces fecal mass and bacterial count in the lumen, but the practice has been questioned because the bowel preparation liquefies feces, which could increase the risk for intraoperative spillage, and may be associated with bacterial translocation and electrolyte disturbance. Mechanical bowel preparation is still commonly practiced without the benefit of evidence from randomized trials, and 2 of 3 meta-analyses suggest a higher rate of anastomotic leakage with mechanical bowel preparation.
This is a randomized clinical trial conducted at 13 hospitals in 1 country in patients undergoing elective colorectal surgery to compare the outcomes of mechanical bowel preparation with no preparation on anastomotic leakage, infection, hospital stay, and mortality in patients undergoing elective colorectal surgery.
Study Highlights
Included were patients 18 years or older scheduled for elective colorectal surgery with primary anastomosis.
Excluded were those who required acute laparotomy, had laparoscopic surgery, or had contraindication to bowel preparation or a previous deviating ileal stoma.
Mechanical bowel preparation consisted of 2 to 4 liters of polyethylene glycol bowel lavage solution with bisacodyl or sodium phosphate solution, with a fluid diet the day before the operation.
The fluid diet consisted of beverages, yogurt, and soup.
Patients assigned to no bowel preparation had normal meals before the operation.
All patients were given prophylactic intravenous antibiotics to prevent infection at the surgical site, and all procedures were done by laparotomy.
Anastomoses were done according to the surgeons' judgment.
The primary endpoint was anastomotic leakage diagnosed by clinical signs followed by contrast radiography.
Secondary endpoints were septic complications, fascial dehiscence, duration of hospitalization, and mortality.
Intra-abdominal abscess was diagnosed by a combination of clinical presentation and computed tomography or ultrasonography.
Of 1431 patients enrolled, 707 were randomly assigned to bowel preparation and 724 to no preparation.
Assignment was not blinded.
670 and 684 patients, respectively, were assessed for endpoints.
77 patients were excluded from analysis because of receiving no surgery or having missing data.
Mean age was 67 years, 50% were women, American Society of Anesthesiologists (ASA) classification was I or II in 88%, 10% had diabetes, 25% were smokers, and colorectal cancer was the indication for surgery in three fourths of patients.
Median follow-up was 24 days.
The rate of anastomotic leakage was approximately 5% whether or not patients had bowel preparation (difference, 0.6%; P = .69).
The treatment effect did not differ among the 13 centers.
30 of 69 anastomotic leakages were verified by radiologic examination, and 57 of 69 needed relaparotomy.
The rate of relaparotomy after anastomotic leakage was 4% in both groups.
The rate of intra-abdominal abscess after anastomotic leakage was lower in those who underwent bowel preparation (0.3% vs 2.5%; P = .001).
However, even accounting for abdominal abscess, outcomes were similar for reintervention, length of hospital stay, and mortality.
Other septic complications, fascial dehiscence, and mortality did not differ between the groups.
Fecal contamination, days until resumption of normal diet, and duration of hospital stay were similar between the 2 groups.
Risk factors identified by univariate analysis as significant predictors of anastomotic leakage were type of anastomosis (ileocolic, colocolic, and colorectal), ASA classification, and blood loss during the operation.
The requirement for a stoma during the operation did not affect the leakage rate.
The authors concluded that elective colorectal surgery could be safely performed without mechanical bowel preparation.
Pearls for Practice
Patients who undergo mechanical bowel preparation before elective colorectal surgery vs those who do not undergo preparation have similar outcomes of anastomotic leakage, wound infection, mortality, length of hospital stay, and fecal contamination but a lower rate of abscess after anastomotic leakage.
Predictors of anastomotic leakage after elective colorectal surgery are type of anastomosis, ASA classification, and blood loss during the operation
|
