By James E. Barone, MD
“Never let the sun to rise or set on a small bowel obstruction.”
Although these words were written more than 100 years ago, some surgeons still believe them. But recent studies show that about 75% of patients with a small-bowel obstruction will get better without surgery. So how many sunsets should you wait to operate?
A paper presented at the 2012 annual meeting of the American Association for the Surgery of Trauma suggests that surgeons probably should wait no longer than three days.
In an email to General Surgery News, senior author Frederick H. Millham, MD, chair, Department of Surgery, Newton–Wellesley Hospital, in Newton, Mass., and associate professor of surgery at Harvard Medical School, in Boston, said, “[Small-bowel obstruction] is a complex problem. Some patients need surgery right away; some need it eventually; and some don’t need it at all. The first group tells us they need surgery. It’s challenging to try to distinguish the second group from the third in some cases. We need to figure out where to draw the line.”
Dr. Millham and his group reviewed the Nationwide Inpatient Sample for 2009. It is a large database sponsored by the Agency for Healthcare Research and Quality and contains information from 44 states. Patients undergoing gynecologic procedures and those in the category of early postoperative small-bowel obstruction were excluded. There were more than 27,000 patients with adhesive small-bowel obstruction, with some 4,800 patients (18%) who underwent surgery for this condition.
The main outcomes studied were complications, number of resections, deaths and postoperative length of stay (LOS) longer than seven days. The ages of the patients were similar in the nonoperative and operative groups, but the average LOS was three days longer for those who had surgery (P<0.001).
About 60% of the patients managed nonoperatively were discharged by the end of day 3, with 91% discharged by the end of day 7.
Women comprised 60% of those admitted with a small-bowel obstruction and 66% of those who had surgery (P<0.001 for both). The authors speculated that more women required surgery because of pelvic adhesions that probably do not resolve as easily as generalized abdominal adhesions.
Patients who underwent surgery usually had their operations within the first few days, with 80% having had surgery by the end of the third day of admission.
For the patients having surgery, 19% experienced complications; 25% required resection of intestine; 3% died; and 32% stayed longer than seven days in the hospital postoperatively. A delay of four or more days until surgery resulted in a complication rate of 25%, a resection rate of 28% and a mortality rate of 5%, with 43% of patients hospitalized longer than seven days after surgery. All of these differences were statistically significant compared with earlier surgery. But after multivariate analysis, only the risk for death (increased by 60%) and postoperative LOS longer than seven days remained significantly related to delaying surgery.
Invited discussant Rao R. Ivatury, MD, chair, Division of Trauma, Critical Care and Emergency Surgery at Virginia Commonwealth University, in Richmond, asked Dr. Millham if he could explain why the delayed group had a higher mortality rate considering that the rate of resection was similar.
Dr. Millham said that it was possible that there was a bias toward higher physiologic acuity in the early surgery group, due to the large percentage of that group who underwent surgery on the day of admission.
“When the early and delayed [surgery] groups are compared, the patients in the early group may mask the patients in the delayed group who physiologically deteriorated,” he said.
Dr. Millham acknowledged a few limitations of the study. For one, physiologic data and the indications for surgery were not available in the National Inpatient Sample, which is composed of administrative data only. “It is possible that the lack of effect of delay on resection results from the high resection rate in patients operated on at the time of admission,” he said.
Also, a possible increase in the percentage of resections in the delayed group might not be readily apparent in the data. Dr. Millham saw this study as hypothesis-generating and thought a prospective study that included indications for surgery would sort this out.
“We can definitely say that patients stay longer and are more likely to die with delay, but are left looking for additional data,” Dr. Millham said. “I hope this work will stimulate a multicenter trial to get a better answer.”
Dr. Ivatury echoed this and said, “I agree that these answers cannot come from a large database but from a multicenter, prospective study.”