Duron J, Hay JM, Msika S, Gaschard D, Domergue J, Gainant A, Fingerhut A, for the French Association for Surgical Research. Prevalence and Mechanisms of Small Intestinal Obstruction Following Laparoscopic Abdominal SurgeryA Retrospective Multicenter Study. Arch Surg. 2000;135(2):208-212. doi:10.1001/archsurg.135.2.208
The prevalence and mechanisms of intestinal obstruction following laparoscopic abdominal surgery have not been studied extensively.
Retrospective review of cases of intestinal obstruction after laparoscopic surgery.
Sixteen surgical units performing laparoscopy in France.
Twenty-four patients with intestinal obstruction.
Main Outcome Measures
Prevalence values and descriptive data.
The 3 most frequent primary procedures responsible for intestinal obstruction were cholecystectomy (10 cases), transperitoneal hernia repair (5 cases), and appendectomy (4 cases). Prevalences of early postoperative intestinal obstruction after these procedures were 0.11%, 2.5%, and 0.16%, respectively. Intestinal obstruction was due to adhesions or fibrotic bands in 12 cases and to intestinal incarceration in 11 cases. Obstruction was located at the trocar site in 13 cases (9 incarcerations and 4 adhesions), mainly at the umbilicus, and in the operative field in 10 cases (2 incarcerations in a wall defect after transperitoneal inguinal hernia repair, 4 adhesions, and 4 fibrotic bands). The small intestine was involved in 23 of 24 cases; the other was due to cecal volvulus following unrecognized intestinal malrotation. Intestinal obstruction was treated by laparoscopic adhesiolysis in 6 patients and by laparotomy in 18 patients, 6 of whom required small intestine resection. Three postoperative complications but no deaths occurred.
Intestinal obstruction following laparoscopic abdominal surgery can occur irrespective of the type of operation; the prevalence is as high as (cholecystectomy and appendectomy) or even higher than (transperitoneal hernia repair) that seen in open procedures.
ABDOMINAL laparoscopic surgery began in 1985,1 increased in usage,2,3 and became widespread by the 1990s. Smaller incisions, minute dissection, and experimental data4,5 have led one to expect that laparoscopic abdominal surgery would be associated with a low or nonexistent intestinal obstruction rate.6,7 However, intestinal obstruction complicating diagnostic gynecological laparoscopy has been recognized and reported as early as 1968.8 Short series of 1 to 4 cases of small-bowel obstruction after various kinds of laparoscopic abdominal surgery9- 11 have been reported since 1992, but do not provide any hard data on the mechanisms and prevalence of postlaparoscopic instestinal obstruction. The goal of this study was to analyze data on 24 patients operated on for mechanical intestinal obstruction secondary to laparoscopic abdominal surgery.
From March 1, 1990, to March 1, 1995, 24 patients (15 women and 9 men; mean ± SD age, 56 ± 19 years; age range, 24-82 years) were retrospectively selected from 16 surgical units (8 university hospitals, 7 teaching hospitals, and 1 private hospital). The median number of cases per surgical unit was 1 (range, 1-3). All adult patients previously operated on laparoscopically for gastrointestinal disease or transperitoneal hernia repair and then reoperated on for mechanical intestinal obstruction were eligible.
All patients who had had a previous laparotomy at any time, those undergoing gynecological laparoscopic procedures, those who underwent a conversion to an open procedure or who had only a laparoscopically assisted procedure, those operated on for intestinal obstruction associated with intraperitoneal infection, and those who improved through nonoperative management were not included.
The prevalence was calculated only for the 22 patients who underwent reoperations in the same institution. The 2 other patients were initially operated on in another center from which data necessary to calculate the prevalence were not available. Initial procedures12- 14 and the prevalence14 of obstruction in our series were compared with those from patients who had open surgery (Table 1 and Table 2).
Other data collected through a preestablished questionnaire included (1) mechanism of obstruction (adhesion, band, or incarceration); (2) location of obstruction (at the operative field or at the trocar site), location of the trocar (umbilical or lateral),15 size of the obstruction,9 and whether the fascia was closed or not11- 15; (3) interval time between the initial procedure and the reoperation; and (4) procedure used to deal with obstruction. As regards the mechanism, incarceration was defined as trapping of the intestines in the abdominal wall; adhesions as union of intestines to a surface (or conglutination); bands as fibrotic cordlike processes or anatomical structures that bind the intestines to other parts, whether encircling another structure or not; and incisional hernia as protrusion of intestines through a trocar incision. Early obstruction was defined as obstruction occurring during the first 6 postoperative weeks.14
Statistical comparisons for nonparametric variables were made with the Mann-Whitney test. The χ2 test was used to compare proportions.
Compared with 3 series of open surgery12- 14 (Table 1), in our series there were more obstructions observed after supracolonic surgery mainly because there were more cholecystectomies performed and no cases of gastric surgery were culled. The initial infracolonic operations were not significantly different except for inguinal hernia, which was more frequent in open surgery.
Compared with the open surgery series of Stewart et al,14 the prevalence of intestinal obstruction in our series was significantly higher in infracolonic procedures, especially in transperitoneal hernia repair (Table 2).
As summarized in Table 3, 50% of obstructions were due to adhesions (33%) or bands (17%) while 46% were due to intestinal incarceration, accounting for 23 cases, all involving the small intestine. The last patient who initially underwent laparoscopic cholecystectomy was reoperated on 72 hours later for cecal volvulus due to unrecognized intestinal malrotation.
Intestinal obstruction (excepting the case of cecal volvulus) occurred at the trocar site in 54% of cases and in the operative field in 42% of cases (Table 3).
Ten (77%) of 13 cases involved 10-mm trocars while 3 (23%) of 13 involved 12-mm trocars. Nine trocar ports were at the umbilicus. As regards the 4 lateral trocar sites (three 10-mm trocars and one 12-mm trocar) (all with incarceration), 1 was in the left iliac fossa and 3 were in the right iliac fossa. Of the 13 patients with trocar site obstruction, 6 had undergone adequate fascia closure, 5 in the umbilicus and 1 lateral. Of the 9 incarcerations at the trocar site, 3 were closed while 6 were not. Slowly resorbable suture material was used in all cases.
All 4 obstructions due to bands and half of the adhesions originated from the operative field (Table 2). Only 2 of 11 incarcerations originated in the operative field and these were due to a peritoneal defect following transperitoneal mesh inguinal hernia repair.
The median interval to reoperation was significantly shorter (P<.01) for incarceration (8 days) than for adhesions (25 days) or bands (22.5 days). All intestinal obstructions occurred less than 2 years after the initial procedures (Table 3); 21 (88%) in the early postoperative course and 3 later (180, 420, and 650 days, respectively). All 3 late obstructions were due to adhesions or bands. The early postoperative course of the initial procedure was uneventful in 14 cases, whereas 10 patients underwent reoperation within 8 days.
Intestinal lysis was performed in 6 cases via laparoscopy (4 bands, 1 adhesion in the operative field, and 1 adhesion at the trocar site). In 17 patients lysis was performed through open laparotomy, including 6 small intestine resections—4 for incarcerated loops (2 at the umbilical trocar site and 2 behind the mesh in a hernia repair) and 2 for adhesions at the trocar site. One patient with cecal volvulus required open cecostomy.
The postoperative course was uneventful in 21 cases. Three patients had complications; 1 had a large mural intestinal hematoma following reoperation for intestinal incarceration at the umbilical trocar site after laparoscopic cholecystectomy, 1 had acute pancreatitis after reoperation for adhesions at the umbilical site following laparoscopic cholecystectomy, and 1 had acute postoperative duodenal ulcer bleeding following reoperation for intestinal incarceration behind the mesh in an inguinal hernia repair. None of these patients required another operation and there were no deaths.
Laparoscopic surgery does not eliminate postoperative obstruction, whether involving the operative field or the incision. The prevalence of obstruction after laparoscopy might even be greater (transperitoneal inguinal hernia repair) or as high (cholecystectomy and appendectomy) as that seen with open operation. The number of incarcerations was nearly the same as that of adhesions or bands combined (Table 3). Most obstructions occurred at the umbilical trocar site, 10 mm or more, even when the fascia was closed.
The initial procedures in laparoscopic surgery differ from those in open series (Table 1), simply because some operations are performed less often laparoscopically than traditionally. Examples are gastric and rectal operations, which are often responsible for postoperative mechanical obstruction in open surgery but which are presently not widely performed laparoscopically.
The relatively high prevalence of small-bowel obstruction in laparoscopic transperitoneal hernia repair (Table 2) is explained by the presence of the small intestines near the umbilical and lateral trocar sites, whereas the peritoneal insult in traditional hernia repair is minimal (compared with that necessary for insertion of a transperitoneal mesh) and is likely most often covered by the nearby colon. For cholecystectomy or appendectomy, the lengths of the multiple port incisions are probably equivalent to one long incision and each port incision in itself is a potential cause of obstruction. Because all patients undergoing the initial procedures were not followed up prospectively, some of these patients could have been operated on for mechanical intestinal obstruction in another institution without the knowledge of the initial center. The prevalence calculated herein was therefore a minimal value. The exact prevalence of intestinal obstruction after laparoscopic or open surgery is and will remain difficult to evaluate. Most diseases treated by laparoscopic surgery are common and generally benign conditions: long-term follow-up is therefore difficult to obtain.
The most frequently encountered mechanisms in our series were adhesions (33%) or bands (17%). Although the pathophysiology of these 2 causes is similar,14 bands were encountered in the operative field only whereas adhesions were seen both in the operative field as well as at the trocar sites. Adhesions have been reported previously in only 2 cases after abdominal laparoscopic surgery.16 This apparent difference with the literature may be explained by the fact that the exact mechanism of intestinal obstruction was not always mentioned in the literature but was requested specifically in our questionnaire and the fact that the exact mechanism was not always reported back to the first surgeon when late obstruction occurred. The second mechanism in our series was incarceration (46%) at the trocar site or in a peritoneal defect in the operative field after transperitoneal hernia repair. Incarcerations, on the other hand, are the most frequently reported mechanism in the literature,17- 20 undoubtedly because obstruction due to incarceration occurs early in the postoperative course. In our series, the median interval from the initial procedure to reoperation was significantly shorter (P<.01) when obstruction was caused by incarceration (8 days) vs adhesions (25 days) (Table 3). In the literature9,10,15,17- 20 incarcerations occurred 1 to 9 days after operation, whereas the 2 cases of adhesions16 occurred at 3 weeks and at 6 months, respectively. Recommendations in the literature21,22 to decrease the prevalence of postlaparoscopic incarceration include: (1) opening the trocar valve to ambient air before port removal, as a partial vacuum can be created when the port is withdrawn, thus drawing omentum or intestines into the fascial defect; (2) shaking the abdominal wall, which may help free any temporary visceral incarceration or adhesions; and (3) removing the trocar under direct visual control while the pneumoperitoneum is maintained by the surgeon's finger, temporarily sealing the trocar site.
In our series, as in the literature,9- 11,15,18,20,23 the most common location of intestinal obstruction was the trocar site, but the lateral sites were involved less often compared with the umbilicus.15,20,21 There are at least 2 explanations for this. First, the lateral wall is composed of 2 fascial planes and muscle, making it theoretically less prone to dehiscence; second, the small intestine is less often in contact with the lateral trocar sites.
All trocars causing intestinal obstruction in our series were 10 or 12 mm in diameter. Incarcerations, however, can arise even in 5-mm holes when they are not closed.11,16 Although it has been stated that fascia closure of trocar holes, which is sometimes difficult,22,23 may preclude or decrease the incidence of obstructions,10,11 our experience (6 cases) as well as that in the literature20,24 indicates that adhesions or incarcerations, whether median or lateral, can still occur after fascial closure. On the other hand, no cases of obstruction due to incisional hernia at the trocar site (Richter hernia), as described previously,24 were found in our series. This seems to occur more often in the lateral sites.24
The second most frequent location of intestinal obstruction encountered in our study was the operative field (Table 2). Although this has certainly already occurred, to the best of our knowledge, obstruction arising from the operative field has not yet been described after laparoscopy. Even though laparoscopy is reported to be less traumatic than open surgery,7 there is no reason to believe that the obstruction rate will decrease in the operative field. The prevalence of intestinal obstruction after procedures such as laparoscopic transperitoneal inguinal hernia repair could actually be higher, as the peritoneum is violated to a greater extent as compared with the open techniques.25,26 Thus, the extraperitoneal route may be preferable to the transperitoneal route. Moreover, the rate of intestinal obstruction could even be higher than reported here, as operations such as colectomy that use more extensive dissection27- 29 (2 cases in our series) and the use of gauze swabs30 or other foreign bodies, which are known to promote adhesions and fibrotic bands, are performed more often laparoscopically.31
As after laparotomy,12,32 the small intestines were involved in the majority of obstructions complicating laparoscopic abdominal surgery (23 of 24). However, colonic herniation, as well as incarceration or herniation of the greater omentum through trocar ports,11 has been reported after gynecologic laparoscopic procedures.24
In 1 of our patients, a cecal loop volvulated 72 hours after laparoscopic cholecystectomy. This complication has already been reported after laparoscopy,33 and may be due to excessive slackness of the malrotated cecum after withdrawal of the pneumoperitoneum.34 As complete inspection of the abdomen is thought to be less satisfactory under laparoscopy,35 the right colon was not seen to be abnormally mobile during the primary operation.
Postlaparoscopic obstruction was severe. Six resections (26%) were necessary, 4 because of incarcerations and 2 because of adhesions. This resection rate, however, is comparable to that seen after open procedures (14%-20%).12- 14
Techniques for avoiding contact between the intestines and the trocar sites include assessment of subxiphoid microceliotomy36 for cholecystectomy, the extraperitoneal approach16 for inguinal herniorrhaphy, and laparoscopic surgery with low pressure37 or gasless laparoscopy.38
Corresponding author: Jean-Jacques Duron, MD, Service de Chirugie Digestive, Groupe Hospitalier Pitié-Salpêtrière, 83 boulevard de l'Hôpital, 75013 Paris, France.
P. Baillet, MD, Eaubonne; F. Benhamida, MD, Sousse-Tunisie; P. Cubertafond, MD, A. Gainant, MD, B. Descottes, MD, D. Valleix, Limoges; J. Domergue, MD, B. Millat, MD, Montpellier; J.-J. Duron, MD, N. Elian, MD, H. Levard, MD, Paris; J. M. Hay, MD, Y. Flamant, MD, G. Zeitoun, MD, Colombes; Y. Laborde, MD, Pau; P. LePicard, MD, Charenton; P. Marre, MD, Marly; J. Marescaux, MD, S. Evrard, MD, Strasbourg; S. Msika, MD, Meulan; Y. Soulier, MD, Montmorency; B. Desrousseaux, MD, Lomme, France.