Postoperative intestinal care after major gynecologic surgery has changed considerably. The purpose of this review was to describe these changes. Our findings are that (1) preoperative mechanical bowel preparation does not lower the risk of anastomotic leakage and infection, (2) elective postoperative nasogastric tube decompression increases postoperative pneumonia and does not decrease the incidence of other postoperative complications, (3) early feeding after major gynecologic surgery reduces hospital stay and does not increase (and may decrease) pneumonia and other postoperative complications, and (4) early feeding, gum chewing, bowel stimulation, alvimopan, and ketorolac may decrease the incidence of postoperative ileus.
Postoperative intestinal care after major gynecologic surgery has changed considerably. The purpose of this review was to describe changes in preoperative mechanical bowel preparation, elective postoperative nasogastric tube decompression, early postoperative feeding, and a mechanism to help reduce postoperative ileus.
Preoperative bowel preparation
For more than a century, the presence of unprepped bowel during surgery was believed to increase anastomotic leakage and infection rates. Logically, preoperative mechanical bowel preparation could reduce fecal flora and thus lower the risk of anastomotic leakage and infection. As such, this dogma has been passed down through the years on the basis of expert opinion and not scientific studies. In a review by Guenaga et al of 13 prospective randomized trials that included 4777 patients who underwent large bowel resection, anastomotic leakage occurred in 4% of the patients with a preoperative mechanical bowel preparation, compared with 3% of the patients with no preparation ( P = .15). Wound infection occurred in 10% of the patients with preparation, compared with 8% of the patients with no preparation ( P = .09). The authors concluded that the notion that mechanical bowel preparation is necessary before elective colorectal surgery should be reconsidered. The Canadian Society of Colon and Rectal Surgeons recommends that preoperative mechanical bowel preparation should be omitted.
Similarly, preoperative oral antibiotic bowel preparation is associated with increased nausea, vomiting, and abdominal pain and does not decrease postoperative infections and therefore is not recommended.
Although preoperative mechanical bowel preparation should be omitted for elective abdominal cases, there is controversy as to whether its use before advanced gynecologic laparoscopic procedures improves the visibility of the surgical field. In a prospective randomized trial of 162 patients who underwent gynecologic laparoscopic procedures, oral sodium phosphate did not improve the quality of the surgical field. Unfortunately, randomized data on this topic are limited.
The traditional postoperative feeding protocol
Postoperative intestinal care after major gynecologic abdominal surgery has changed considerably over the last 2 decades. The traditional postoperative feeding protocol after major abdominal surgery has included nasogastric decompression. Nasogastric tubes are removed when bowel sounds, flatus, and bowel movement are present and when nausea, vomiting, and abdominal distention are absent. If nausea and vomiting do not develop after removal of the nasogastric tube, a clear liquid diet is begun, and diet is advanced slowly to solids over several days. The rationale for the traditional postoperative feeding protocol is to prevent vomiting and ileus. It was believed that vomiting increased the incidence of aspiration pneumonia and that ileus increased the incidence of wound dehiscence and intestinal leaks. In the early days of surgery, vomiting was common after major abdominal surgery, in part because of the anesthetic agents that were used and the lack of centrally acting antiemetics (such as ondansetron). The traditional postoperative feeding protocol is still used. In a survey of the Society of Gynecologic Oncologists in 1998, the traditional postoperative feeding protocol that included nasogastric decompression was used by 67% of gynecologic oncologists after ovarian cytoreduction, by 34% after lymphadenectomy, by 29% after radical hysterectomy, and by 15% after routine abdominal hysterectomy. Two major trends in the late 1980s and 1990s have led to challenging the traditional postoperative feeding protocol. Originally, laparoscopy was restricted to diagnostic procedures and tubal sterilization. After laparoscopy, patients were fed immediately rather than according to the traditional postoperative feeding protocol. In the late 1980s and 1990s, major gynecologic surgical procedures were performed laparoscopically. Many physicians abandoned the traditional postoperative feeding protocol when these surgical procedures were performed laparoscopically rather than through a large abdominal incision. The second major impetus was health care economics that insisted on early postoperative discharge, which challenged physicians to reevaluate the traditional postoperative feeding protocol.
Postoperative ileus
Significance and incidence
A common complication of abdominal surgery is postoperative ileus (colonic stasis) that leads to patient discomfort, increased hospital stay, and increased cost. In a review of the Premier Perspective Database, which includes about one-sixth of all US hospitalizations, postoperative ileus was found to increase hospital stay by approximately 30% and to result in an increase in cost in excess of $5000-10,000 per ileus and $1.5 billion per year in the United States. The incidence of postoperative ileus after gynecologic surgery ranges from 5-25%; the highest rates occur after laparotomy. Thus, postoperative ileus is common, painful, and expensive.
Definition, cause, and site
Although ileus is defined as the inhibition of bowel motility, there is no standard definition of postoperative ileus. The symptoms of postoperative ileus include nausea and vomiting, abdominal distention and tenderness, and delayed passage of flatus and stool. When postoperative ileus in clinical trials is reported on, commonly used surrogate definitions include prolonged absence of flatus, persistent nausea and vomiting, the need for insertion of nasogastric tube for decompression, prolonged hospitalization, and readmission for vomiting. Although there is no agreement on the definition, it is usually accepted that postoperative ileus has resolved with the passage of flatus/stool and tolerance of adequate oral intake.
The exact cause of postoperative ileus is unknown, but the 3 most likely mechanisms that are involved in its development are gastrointestinal inflammatory response and neurogenic and pharmacologic mechanisms. The acute phase of postoperative ileus is believed to be primarily neurogenic that results from peritoneal irritation that causes stimulation of afferent reflexes. The prolonged phase of postoperative ileus is believed to be a stress response to tissue trauma that triggers a gastrointestinal inflammatory response. The pharmacologic mechanism mainly involves the use of perioperative opioids. The final pathway of the neurogenic, gastrointestinal inflammatory, and pharmacologic mechanisms is believed to be activation of the gastrointestinal mu-opioid receptors, which inhibit the release of acetylcholine from the mesenteric plexus which in turn disrupts peristalsis that leads to uncoordinated nonpropulsive contractions.
Although the stomach and small intestines recover quickly after surgery (return to normal activity within 8 hours), it is the colon, specifically the rectosigmoid, which can take up to 3-5 days before the return of normal peristalsis. Woods et al placed electrodes in the stomach, small bowel, ascending colon, and descending colon in monkeys. Myoelectric activity of the intestine was observed at a baseline, after anesthesia, and after anesthesia and a “sham” abdominal surgery. Although anesthesia alone had little effect on intestinal motility, anesthesia with abdominal surgery caused a significant change in motility. The stomach and small intestines returned to normal activity within 8 hours. However, the right colon took 48 hours and the sigmoid colon took 72 hours to return to normal contractility in this study. Wilson performed a similar study in humans. The night before surgery, 39 patients swallowed radiotelemetry capsules. Abdominal x-ray films the morning of surgery localized the transmitters in the right ascending colon. After abdominal surgery, normal right colonic activity was delayed for 48 hours, independent of the actual surgical procedure that had been performed. These studies confirmed that the length of the surgical procedure and the amount of intestinal manipulation had little effect on colonic stasis with 1 exception: retroperitoneal dissection was associated significantly with colonic stasis compared with other abdominal procedures. In a report of 46 patients who underwent radical paraaortic lymphadenectomy for gynecologic cancer, 50% of them experienced postoperative ileus.
Elimination of postoperative nasogastric decompression
In 1995, a metaanalysis combined 26 prospective randomized trials that compared postoperative nasogastric decompression with no postoperative nasogastric decompression after elective major abdominal surgery in 3964 patients. One of the reasons for the use of postoperative nasogastric decompression is to prevent vomiting and aspiration pneumonia. However, withholding nasogastric decompression had no effect on emesis (10% [no nasogastric decompression] vs 8% [nasogastric decompression]; P = .11), and only 5% of the patients without nasogastric decompression required postoperative nasogastric tube insertion. Postoperative pneumonia was increased in the nasogastric decompression group (6% [nasogastric decompression] vs 2% [no nasogastric decompression]; P = .0001), which refutes this theory. Another reason commonly used in support of postoperative nasogastric decompression is to prevent ileus and decrease the incidence of wound dehiscence and intestinal leaks. However, nasogastric decompression did not significantly affect the incidence of postoperative dehiscence (2% [nasogastric decompression] vs 1% [no nasogastric decompression]; P = .06) and intestinal leaks (1% [nasogastric decompression] vs 1% [no nasogastric decompression]; P = .93), although the trend favored the omission of nasogastric tube use. Similar findings were noted in a recent review of 28 prospective randomized trials in 4194 patients in which patients who did not have a nasogastric tube placed routinely had earlier return of bowel function ( P < .00001) and fewer pulmonary complications ( P = .01) with a nonsignificant trend towards fewer wound infections ( P = .22) and ventral hernias ( P = .09).
Therefore, class I data consistently demonstrate that routine nasogastric tube decompression after surgery does not accomplish any of the aforementioned goals and is associated with an increased incidence of pneumonia. Hence, elective postoperative nasogastric tube decompression is not indicated.
It should be noted that patients with preoperative bowel obstruction were excluded from these trials; thus, the safety of withholding postoperative nasogastric decompression in patients who undergo surgery for bowel obstruction is unknown.
Early feeding after major gynecologic surgery
Recently, several prospective randomized trials reported on the safety of early postoperative feeding after major gynecologic surgery, mostly major gynecologic oncology surgery ( Table 1 ). As opposed to the traditional postoperative feeding protocol, early or immediate feeding consists of a diet of the patient’s preference once she has recovered from anesthesia. Schilder et al randomly assigned 96 patients to the traditional postoperative feeding protocol vs immediate feeding and reported no increase in aspiration pneumonia, wound dehiscence, or intestinal leakage in the early feeding group. Length of hospital stay was reduced from 4-3 days ( P = .008). Pearl et al performed a similar trial of the traditional postoperative feeding protocol vs immediate feeding on 200 patients and also reported that no increase occurred in aspiration pneumonia, wound dehiscence, or intestinal leakage and that the hospital stay was reduced from 6-5 days ( P = .001) in the immediate feeding group. Steed et randomly assigned 96 patients and reported no increased complications and a hospital stay that was reduced 2 days in the immediate feeding group (4 vs 6 days; P = .0001). Cutillo et al randomly assigned 122 patients and reported significantly faster time to toleration of diet (3 vs 5 days; P < .01), passage of first stool (3 vs 4 days; P < .01), and shorter hospital stay (5 vs 6 days; P < .05) in the immediate feeding group, with no difference in the complication rate. Similarly, Minig et al randomly assigned 143 patients to early feeding vs the traditional postoperative feeding protocol and not only confirmed a shorter hospital stay (5 vs 6 days; P = .006) but also reported fewer complications in the early feeding group (17% vs 39%; P = .003). Postoperative infections were lower (3% vs 14%; P = .02) and patient satisfaction scores were higher (83 vs 72 mm; P = .001) in the early feeding group. Finally, a recent review did not show any difference in complications between patients that were fed early or had delayed feeding, except for death, which was decreased significantly in the early feeding group; however, this difference may not be attributable to early feeding directly. Although individual complications in that review failed to reach statistical significance, the trend clearly favored early feeding that demonstrated fewer postsurgical complications in the early feeding group.
| Study | n | Complications | P value | Length of hospital stay | P value |
|---|---|---|---|---|---|
| Schilder et al | 96 | No significant difference | Less by 1 d | .008 | |
| Pearl et al | 200 | No significant difference | Less by 1 d | .001 | |
| Steed et al | 96 | No significant difference | Less by 2 d | .0001 | |
| Cutillo et al | 122 | No significant difference | Less by 1 d | .05 | |
| Minig et al | 143 | Fewer overall complications | .003 | Less by 1 d | .006 |
| Fewer infections | .02 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
