Laparoscopy is a valuable approach for selected patients undergoing ileocecal resection for ileocolonic Crohn’s disease (CD) [1–3]. The benefits of laparoscopy over laparotomy for these patients, who may be candidates for several abdominal operations during their lives, are not only improved cosmetic results and reduction of pain [4–7], but also a lower postoperative morbidity rate, a shorter hospital stay, and an earlier return to full activity [1–3, 8]. Furthermore, findings have shown laparoscopy to be associated with a recurrence rate after primary resection similar to that for the open approach [9–11]. However, the severity of the disease (i.e., the presence of an enteric fistula, an intraabdominal abscesses or a mass, and a history of previous intestinal resection) can be associated with a higher conversion rate and still remains a relative contraindication to the laparoscopic approach for CD [12, 13].

Recently, several reports from experienced centers for both CD surgery and laparoscopy have shown that the laparoscopic approach could be feasible and safe for selected patients with a complicated form of CD including abscess and fistulizing disease [14–16]. Only few studies focused on reoperation for CD recurrence after a previous ileocolonic resection. Although few studies of open surgery have suggested that reoperation for CD recurrence is more challenging and complex than a primary resection [17], to our knowledge, only one study, including 16 patients, has been devoted to the role of laparoscopy in iterative surgery for CD recurrence [18]. Furthermore, two other studies including 10 and 16 patients respectively reported that laparoscopy could be feasible for this indication [15, 19]. However, only one of these studies was comparative with a selected group of primary resections through laparoscopy, but it included no comparison with iterative surgery through the open approach [18]. The question whether laparoscopy can be performed safely for patients with CD recurrences after a primary resection, especially after a primary open procedure, still remains undetermined, and the risk factors for failure of the laparoscopic approach for these patients are unknown.

Thus, this comparative study aimed to evaluate the feasibility of the laparoscopic versus the open approach for patients with recurrence of CD after a primary ileocolonic resection and to determine the risk factors and consequences for a postoperative course involving a conversion for laparoscopic patients.

Methods

Patients

The data in this study were gathered from our prospective CD database. Between 1998 and 2008, among 231 patients who underwent surgery for ileocolonic CD in our institution, we selected all the patients undergoing an iterative ileocolonic resection for CD recurrence.

Preoperative assessment and indications for resection

All the patient candidates for elective surgery had a clinical, biologic, endoscopic, and radiographic assessment of their disease before surgery. An upper endoscopy and a complete colonoscopy were performed before surgery for all the patients except those in the context of an emergency surgery. Patients with a preoperative diagnosis of intraabdominal abscess were initially drained percutaneously whenever feasible. For these patients, a helical abdominal computed tomography (CT) scan was performed before surgery to evaluate persistence or collapse of collection. Beginning in 2006, all patients had a preoperative magnetic resonance imaging (MRI) enterography to determine better the number, length, and location of small bowel lesions.

The indications for surgery were discussed during multidisciplinary meetings with gastroenterologists, radiologists, and surgeons. Except for emergency cases, an iterative resection was proposed for patients with disease resistant to several lines of medical treatment (steroids, azathioprine, methotrexate, or infliximab) who presented with an intraabdominal abscess or an internal fistula.

Surgical procedure

Before 2002, patients who underwent surgery for CD recurrence were treated using a conventional approach with a midline laparotomy. A thorough exploration of the peritoneal cavity with a map of lesions on the bowel was conducted. Beginning in 2002, our policy called for a systematic attempt to perform a laparoscopic procedure if it was feasible in safe conditions, even for patients who already had undergone a previous laparotomy for intestinal resection to manage CD.

The criteria for the selection of patients specified no more than three previous abdominal operations and no history of diffuse peritonitis. A previous laparotomy was not considered a contraindication to laparoscopy. Indeed, we hypothesized that for these patients, eligible for numerous abdominal operations, the benefit of laparoscopy to preserve the integrity of abdominal wall was crucial.

Laparoscopic surgical procedure did not differ from the procedure used for primary resection that we have reported previously [6]. Briefly, mobilization of the bowel was performed intracorporeally. The bowel was then exteriorized by a 6-cm skin incision, in most cases in the right iliac fossa. The mesenteric division and bowel resection usually were performed extracorporeally. To complete exploration of the bowel and to ensure that no strictures were missed, the entire length of the small bowel was walked with atraumatic forceps.

When the laparoscopic procedure was not feasible because of intraabdominal peritoneal adherences due to previous operations or inflammation, or when dissection proved to be difficult, the procedure was converted. A conversion was defined as any unplanned incision or any planned incision longer than 6 cm that was necessary for simple exteriorization of the resected specimen and used to fashion the anastomosis.

Anastomosis most often was a side-to-side anastomosis, either hand sewn or stapled. A temporary stoma was preferred to anastomosis for high-risk patients who presented more than two of the following risk factors already reported for septic complications [20, 21]: a steroid treatment before surgery, an abscess or fistula discovered during surgery, or poor nutritional status. However, even for patients with no more than two risk factors, in the case of intraoperative difficulties (adherences, small bowel injury, associated procedures), a temporary stoma was sometimes performed according to the surgeon’s preference.

Study criteria

To assess the feasibility of the laparoscopic procedure for iterative ileocolonic resection used to manage CD recurrence after a primary resection, univariate analysis was used to compare pre- and intraoperative characteristics, the postoperative morbidity rate with special reference to a severe complication rate defined as a classified 3 or more in Dindo’s scoring system [22] (i.e., requiring surgical, endoscopic, or radiologic intervention), and the mortality rate for patients treated with laparoscopy versus that for patients treated with an open approach.

To determine the risk factors for conversion, the patients treated with a laparoscopic approach were divided into two groups based on whether conversion was needed or not. Univariate analysis was used to examine the relationship between the occurrence of a conversion and the following 20 variables : age, gender, American Society of Anesthesiology (ASA) score, body mass index (BMI), weight loss exceeding 10%, the duration of CD from diagnosis, the interval between primary and iterative surgeries for CD, the number of previous acute episodes for CD, the number of previous abdominal operations, the type of surgical approach for primary resection (open or laparoscopic), an eventual steroid therapy within 3 months before surgery, indications for resection (stenosis, abscess, or internal fistula), the type of surgery (elective or emergency), intraoperative findings (peritoneal adherences, abscess, or internal fistula), the use of a temporary defunctioning stoma or anastomosis, the operative time, and the occurrence of intraoperative intestinal injury.

Finally, to determine whether the occurrence of a conversion increased the morbidity rate or not, the patients with completed laparoscopy and those undergoing conversion to laparotomy were compared.

Statistical analysis

Quantitative data were expressed as means ± standard deviation (SD) and qualitative data as frequency and percentage. Comparisons between the groups were analyzed with the chi-square test, the Mann–Whitney U test, and the Student’s t-test or Fisher’s exact test for qualitative and quantitative variables, as appropriate. All p values less than 0.05 were considered statistically significant. Statistical analysis was performed using the StatView software (version 5.0; SAS Institute, Cary, NC, USA).

Results

Patient characteristics

Between 1998 and 2008, 57 patients underwent 62 iterative ileocolonic resections for CD recurrence. The preoperative characteristics of these 57 patients are summarized in Table 1. Among the 62 iterative ileocolonic resections for CD recurrence, 29 (for 29 patients) were performed using laparoscopy, and 33 (for 28 patients) were performed using an open approach. The laparoscopy group consisted of 10 men and 19 women, and the open group comprised 10 men and 18 women (nonsignificant difference [NS]).

Table 1 Preoperative characteristics of 62 iterative ileocolonic resections for Crohn’s disease recurrence in 57 patients
Full size table

Table 2 compares the preoperative characteristics of the two groups. There were significant differences between the laparoscopy and open groups, including significantly more frequent associated perianal disease (p = 0.02) and a lower preoperative hemoglobin level (p < 0.001) in the open group. Preoperative physical status (i.e., ASA classification, BMI, and preoperative weight loss) and indications for resection were similar in the two groups (Table 2).

Table 2 Preoperative characteristics of 29 laparoscopic vs. 33 open ileocolonic resections for Crohn’s disease recurrence
Full size table

Concerning the type of previous resection in the two groups, patients who underwent a laparoscopy for recurrent disease already had undergone an ileocolonic resection by open (n = 18) or laparoscopic (n = 10) surgery, an intestinal resection by the open approach (n = 2), or a subtotal colectomy by the open approach (n = 1). Of the 18 ileocolonic resection, 2 were iterative resections. In the laparotomy group, previous operations had been performed by the open approach.

Intraoperative characteristics and postoperative outcome

Comparison of the intraoperative characteristics and postoperative outcomes between the laparoscopy and open groups is summarized in Table 3. The lesions of CD discovered during surgical exploration did not differ significantly in terms of type between the laparoscopy and open groups. However, the need for associated procedures was more frequent in the open group than in the laparoscopy group (17 vs. 2; p = 0.003). The occurrence of intraoperative intestinal injuries was more frequent in the laparoscopy group than in the open group (5 vs. 0; p = 0.01).

Table 3 Intraoperative characteristics and postoperative outcome for 29 laparoscopic vs. 33 open ileocolonic resections to manage Crohn’s disease recurrence
Full size table

After the operation, the mortality rate was nil in both groups. There were no differences in the morbidity rate or the rate of severe complications. The length of the postoperative hospital stay was similar in the two groups (Table 3). Four patients (2 in each group) underwent reoperation. One patient underwent a laparotomy for an intraabdominal hemorrhage on postoperative day 2 after a converted laparoscopy. Afterward, the postoperative course was uneventful.

The three remaining patients experienced postoperative diffuse peritonitis due to an anastomotic leakage. Reoperations were performed for these three patients on postoperative days 5, 7, and 8, respectively, and the anastomosis was removed in all cases. The postoperative course after the reoperation was uneventful for two of these three patients, and they were discharged from hospital on postoperative days 20 and 44, respectively. The remaining patient, initially treated with laparoscopy, experienced several intraabdominal abscesses and required a second reoperation on postoperative day 33. She was discharged from hospital on postoperative day 63.

Risk factors for failure of the laparoscopic approach

The results from the univariate analysis of risk factors for failure of the laparoscopic approach are presented in Table 4. Overall, among 29 laparoscopic procedures, 9 (31%) were converted to an open procedure. The reasons for conversion reported by the surgeon in the surgical report were the presence of dense intraabdominal peritoneal adherences difficult to dissect (n = 5), the occurrence of intraoperative intestinal injury (n = 3), and a problem during making of the anastomosis (n = 1).

Table 4 Univariate analysis of the risk factors for conversion to laparotomy for the 29 patients who underwent laparoscopy for Crohn’s disease recurrence
Full size table

Univariate analysis showed two risk factors for conversion to an open procedure: the intraoperative finding of an internal fistula (in 4 of 9 converted vs. 0 of 20 nonconverted patients; p = 0.02) and the occurrence of an intraoperative intestinal injury (in 5 of 9 converted vs. 0 of 20 nonconvered patients; p = 0.0001). Furthermore, there was a trend toward an increased risk of conversion for patients with a longer interval between the previous intestinal resection and the surgical procedure for iterative resection (12 vs. 6.5 years; p = 0.06).

Intraoperative and postoperative consequences of conversion to laparotomy

Comparison of the intraoperative characteristics and the postoperative outcomes for 20 completed laparoscopic procedures and 9 laparoscopic procedures converted to open operations is summarized in Table 5. There were no statistically significant differences between the converted and laparoscopically completed patients in terms of intraoperative blood loss, requirement of a temporary defunctioning stoma, and postoperative morbidity rate. The mean operating time was significantly longer for the converted procedures than for the completed laparoscopic procedures (296 vs. 194 min; p = 0.001). Finally, there was a trend toward a longer postoperative hospital stay for the converted patients than for the completed patients (11 vs. 8 days, NS).

Table 5 Univariate analysis of the impact that conversion had on the postoperative outcome for the 29 patients who underwent laparoscopy for Crohn’s disease recurrence
Full size table

Discussion

This study showed that although challenging and complex, the laparoscopic approach for CD recurrence is feasible and safe. Despite a higher incidence of intraoperative intestinal injuries, laparoscopy for CD recurrence was associated with mortality and morbidity rates similar to those observed after the open approach. In this study, the risk factors for failure of the laparoscopic approach were the intraoperative discovery of an internal fistula and the occurrence of an intraoperative intestinal injury. Finally, even when converted, laparoscopy was not associated with a higher complication rate.

To our knowledge, this study reports the largest series of laparoscopic procedures for iterative ileocolonic resection used to manage CD recurrence [15, 18, 19]. Furthermore, the 29 laparoscopic reoperations for CD recurrence were compared with 33 open iterative ileocolonic resections. Previous studies have suggested that open reoperations for CD recurrence are associated with higher morbidity rates than primary operations [17]. For these reasons, many authors, expecting technical difficulties during the operation, prefer the open approach for CD recurrence. However, laparoscopy has clearly demonstrated its benefits for primary resection of CD, with a lower morbidity rate and a shorter hospital stay, as suggested by a recent metaanalysis [1]. The question as to whether these expected benefits also can be observed for CD recurrence or not remains unanswered.

To date, only one study comparing laparoscopy for CD recurrence and laparoscopy for primary resection has been reported [18]. However, we consider it more pertinent to compare laparoscopy with the open approach for CD recurrence because in cases of CD recurrence, the surgeon must choose between a laparoscopic or an open approach.

In our study, the morbidity rate after laparoscopic iterative ileocolonic resection was similar to that observed after the open approach and close to that reported after laparoscopic primary ileocecal resection for CD [5, 8]. However, a failure of the laparoscopic procedure was observed in 31% of cases because of dense intraperitoneal adherences due to the disease or previous operations and because of intraoperative intestinal injury, highlighting the complexity of these procedures. Nevertheless, the conversion rate reported in this series remains close to those already reported after primary ileocecal resection [12, 13]. Despite a higher rate of intestinal injury, the morbidity rate was not increased for converted patients. However, it can be hypothesized that the occurrence of intestinal injuries during the procedure may increase the risk of septic complications, and the absence of increased morbidity in the current series could be due to the sample size.

This study is the first to analyze the risk factors for conversion of laparoscopic iterative ileocolonic resection used to manage CD recurrence. Univariate analysis showed that both the intraoperative discovery of an internal fistula and the occurrence of an intestinal injury during surgery were risk factors for conversion. In a previous study involving a large cohort of patients undergoing a laparoscopic primary ileocecal resection for CD, we and others [12, 13] had already reported that intraoperative findings of intraabdominal abscess or internal fistula were independent risk factors for conversion.

In the current study, the occurrence of intraoperative intestinal injury also was associated with a significantly higher conversion rate (5/5 vs. 0/24; p = 0.0002). These intestinal injuries were mainly due to dense adherences difficult to dissect, which also was the principal cause of conversion (5 of 9 converted procedures). This can be explained by the fact that in the current series, 21 (72%) of the 29 patients who underwent laparoscopy already had undergone a laparotomy for a previous intestinal resection used to manage CD, which is supposed to increase the incidence of intraperitoneal adherences [23]. However, in our study, univariate analysis did not show any relationship between the intraoperative discovery of intraperitoneal adherences and conversion. Indeed, the majority of postoperative adherences can be dissected safely using laparoscopy [24]. Thus, we consider that laparoscopy is not a contraindication for patients who have surgery for CD recurrence, even if a previous resection was performed by the open approach.

In this series, the expected benefits of laparoscopy, such as a lower morbidity rate and a shorter postoperative hospital stay, demonstrated after primary resection for CD [1–3] were not observed after iterative resection. This could be explained in part by the small sample size, as suggested by the severe complication rate, which although not significant, was 15% in open group versus only 10% in laparoscopic group. From a pessimistic point of view, it can be considered that laparoscopy will never be superior to laparotomy for this indication. Because the primary end points (i.e., mortality and morbidity rates and hospital stay) were similar in the laparoscopy and open groups, secondary end points such as the possible reduction of postoperative pain, a better cosmetic result, and the possible reduction of postoperative adherences and wound dehiscence became primary objectives for these young patients, who are candidates for several abdominal operations during their lives. Consequently, our policy is to discuss systematically a laparoscopic reoperation approach for CD recurrence.

The current study has some limitations because the patients who had surgery through the laparoscopic and open approaches were not strictly comparable. Indeed, our policy is to discuss a laparoscopic approach with all patients who have ileocolonic CD recurrence, but we still consider that a history of multiple laparotomies (≥3) and diffuse peritonitis are contraindications for a laparoscopic approach. In the current study, the patients in the laparoscopic group had undergone no more than two previous abdominal operations and had no history of diffuse peritonitis.

Although a randomized study would provide a meaningful answer to the concern about reoperative surgery for CD, it is unlikely that the benefit of laparoscopy over laparotomy can be tested in a controlled trial due to the heterogeneity of the patients with recurrent CD.

In conclusion, our study suggests that the laparoscopic approach to iterative ileocolonic resection for the recurrence of Crohn’s disease is challenging and complex with an increased risk of small bowel injury, that it has a morbidity rate similar to that of the open approach, and that small bowel injury could thus possibly induce postoperative septic complications. However, we believe that laparoscopy can be recommended for selected patients with CD recurrence, especially patients with nonfistulizing disease. It gives a postoperative outcome similar to that of the open approach, even for converted patients, but with the other expected benefits of laparoscopy.