- Email: [email protected]
Long-term outcomes of laparoscopic versus open ileocolic resection for Crohn's disease: Follow-up of a prospective randomized trial
Long-term outcomes of laparoscopic versus open ileocolic resection for Crohn’s disease: Follow-up of a prospective randomized trial Luca Stocchi, MD,a Jeffrey W. Milsom, MD, FACS,b and Victor W. Fazio, MD, FACS,a Cleveland, Ohio, and New York, NY
Background. The purpose of this study was to analyze long-term recurrence rates and complications in patients previously enrolled in a prospective randomized trial comparing laparoscopic (LC) and open ileocolectomy (OC) for ileocolic Crohn’s disease (CD). Methods. Follow-up data were available on 56 of 60 patients. Demographic data, recurrence rates, need for additional surgery related to primary procedure, and medication use were recorded. Results. Mean follow-up for 56 patients (27 LC vs 29 OC) was 10.5 years and comparable between LC and OC (10.0 vs 11.0, respectively; P = .64). One patient died 8 years after OC of causes unrelated to CD. Eight patients for each group underwent initial reoperative (26% LC vs 28% OC; P = .89). One patient underwent incisional hernia repair after LC (4%) versus 4 patients (14%) after OC (P = .61). Two patients in the LC group underwent adhesiolysis versus none after OC (P = .23). Incidences of anorectal disease, anorectal surgery, endoscopic or radiologic recurrence, and medication use were also similar between LC and OC. OC patients requiring operation during follow-up were significantly more likely than LC to require multiple operations (P = .006). Conclusions. Long-term data from this prospective randomized trial confirm that LC is at least comparable to OC in the treatment of ileocolic CD. (Surgery 2008;144:622-8.) From the Department of Colorectal Surgery, Cleveland Clinic,a Cleveland, Ohio; and the Department of Surgery, New York Presbyterian Hospital, Weill Medical College of Cornell University,b New York, NY
LAPAROSCOPIC ILEOCOLIC RESECTION (LC) for ileocolic Crohn’s disease (CD) has demonstrated short-term benefits, including reduced duration of postoperative ileus,1-4 decreased narcotic requirement,1,5-7 reduced length of hospital stay,1,2,4-9 and decreased costs10 when compared with open ileocolic resection (OC). Postoperative morbidity after laparoscopic approach has also been either comparable or decreased when compared with OC.1,2,4-8,11-16 Only 2 prospective randomized trials have been published on short-term outcomes after LC versus OC for CD, including the study by Milsom et al in 200114 and a multicenter trial from Europe more recently reported.17 On the other hand, only a handful of studies have longer term follow-up, Accepted for publication June 19, 2008. Reprint requests: Luca Stocchi, MD, Department of Colorectal Surgery, 9500 Euclid Avenue, Desk A-30, Cleveland Clinic, Cleveland, OH 44195. E-mail: [email protected]. 0039-6060/$ - see front matter Ó 2008 Mosby, Inc. All rights reserved. doi:10.1016/j.surg.2008.06.016
622 SURGERY
ranging from a mean of 17 months to 81 months,4,5,7,11,14,15,18 none of which was designed as a prospective randomized trial. The purpose of this study was therefore to evaluate the long-term rates of small bowel obstruction, reoperation, and recurrence rates after LC versus OC for CD in the cohort of patient originally enrolled into the study by Milsom et al.14 MATERIALS AND METHODS All patients previously enrolled into a prospective randomized trial evaluating LC versus OC for CD14 were included. Patient accrual for the original trial was completed in March 1998. Collected information included demographic data, date of original surgery, total number of operations after index surgical procedure, operations for abdominal recurrent CD, CD-related anorectal conditions and operations, biologic behavior of the disease according to the Vienna classification,19 use of medications to prevent or treat recurrence (steroids, antiinflammatory agents, immunomodulators), and tobacco consumption.
Surgery Volume 144, Number 4
With regard to smoking history, we subdivided patients into those who stopped smoking after their index surgical procedure and those who continued smoking after their index operation. Long-term data recorded included incidence, location, and behavior of recurrent disease; need for hospital admission; incidence of endoscopic recurrence; radiologic recurrence or recurrence requiring operation; type of operation(s) performed; postoperative outcomes after the second operation; and use of postoperative medication to reduce the incidence of further recurrence. Data were collected from our institutional, Institutional Review Board-approved, electronic CD database and supplemented by chart review as necessary. Patients who were not already in the CD database were contacted by mail and phone if necessary and asked to participate in the database, therefore providing data on the treatment(s) received for their disease after the time of their operation. The incidence of recurrence was recorded on a per-patient basis. The definition of recurrence excluded symptomatic recurrence only and was limited to recurrence based on imaging and/or operation. Radiologic recurrence was based on the readings from small bowel follow-through or abdominal and pelvic computed tomography. Endoscopic recurrence was based on the reading of the colonoscopy reports and was not further classified. The incidence of recurrence was collected separately and defined as anastomotic recurrence when involving the small bowel or colon or both at the level of the anastomosis, and small bowel or colonic recurrence when the disease recurred away from the anastomosis in the small bowel or colon. There were no recurrences in the upper gastrointestinal tract. We considered as a recurrence requiring surgery a condition requiring an abdominal operation. Use of any medications was recorded as a single variable regardless whether the medications were used on a regular basis or for a treatment course. The medications administered were also subdivided into antiinflammatory agents, steroids, azathioprine/6-MP, anti-tumor necrosis factor (TNF) agents and others (antibiotics, methotrexate, antidiarrheals, probiotics). Antidiarrheals were used after the index operations in all cases except 2 and we decided to consider them combined. However, the remaining medication categories were divided according to the time of their initiation, whether after the index operation or after the second operation or later in those patients who had >2 CD-related operations. We collected the incidence of any operations in the follow-up whose indication could be related to
Stocchi, Milsom, and Fazio 623
the index surgery. In particular, we excluded indications for surgery that were unrelated to CD or complications of the index operation, such as adhesions and incisional hernias. We accepted the diagnosis of adhesive small bowel occlusions requiring surgery based on the operative reports to ensure that recurrent CD was not a concern. We also collected the incidence of incisional hernia repairs related to the incisions used for the operations to treat CD. In addition, we collected data on colonoscopic balloon dilatation of anastomotic strictures related to recurrent CD. The decision to perform and repeat this intervention instead of considering surgery or continuing management with medical treatment alone was left to the treating gastroenterologist. The data on admissions to the hospital for medical treatment were based on documentation or patient report, whereas details on surgical procedures were based only on direct review of operative reports from our institution of obtained from our institution or from outside. Statistical analysis was conducted by v2 or Fisher’s exact tests for group comparisons with respect to categorical variables. Group comparisons with respect to quantitative and ordinal variables were carried out using the Wilcoxon rank sum test. P < .05 was considered significant. RESULTS Four out of 60 patients had no follow-up (2 OC and 2 LC), 2 of whom declined to provide updates regarding their conditions. This left 56 patients for analysis, 24 of whom were males. Mean age at the time of index surgery was comparable (36 vs 35 years old for LC and OC, respectively; P = .68). Mean follow-up for the entire patient cohort was 10.5 (± 2.5) years and was also comparable between the 2 groups (Table I). One patient died 8 years after OC for causes unrelated to CD. One additional patient underwent open sigmoid resection for diverticulitis 2 years after laparoscopic ileocolic resection. He developed an incisional hernia after his sigmoidectomy, which was therefore not considered as a long-term complication of his operative treatment for CD. A history of smoking was present in 21 out of 56 patients (43%) and was not significantly different when comparing the 2 groups (12 patients, 52% vs 9 patients, 35% for LC and OC, respectively; P = .22). Two patients had stopped smoking after their index OC, whereas all the remaining individuals with history of tobacco use in either groups continued smoking for variable periods of time during their follow-up.
624 Stocchi, Milsom, and Fazio
Surgery October 2008
Table I. Recurrent disease on 27 LC versus 29 OC patients Mean follow-up, yrs (SD) Colonoscopy during follow-up Endoscopic recurrence Radiological recurrence First surgical recurrence In-hospital medical treatment for recurrent disease anytime Location first recurrence Anastomotic Small bowel Colon Behavior first recurrence Nonstricturing, nonpenetrating disease Stricturing Penetrating Operation for first recurrence Resection and reanastomosis Small bowel resection Colonic resection Strictureplasty alone Laparoscopic resection for recurrent abdominal disease
LC (patients, %)
OC (patients, %)
9.97 ± 3.17 18 (66.7%) 12 (48.0%) 13 (48.1%) 8 (25.9%) 3 (11.1%)
10.98 ± 1.38 22 (75.9%) 19 (65.5%) 15 (51.7%) 8 (27.6%) 2 (6.9%)
13 (48.1%) 1 (3.7%) 0
12 (41.4%) 1 (3.4%) 2 (6.9%)
5 (14.8%) 9 (33.3%) 0
3 (10.3%) 8 (27.6%) 4 (13.8%)
7 (88%) 0 0 1 (12.5%) 4 (50%)
4 (50%) 1 (12.5%) 3 (37.5%) 0 1 (12.5%)
P value .64 .45 .2 .89 .89 .83 .78
.29
.13
.31
LC, Laparoscopic ileocolectomy; OC, open ileocolectomy.
The overall recurrence rate was 52% (29/56 patients). A comparable number of patients in the 2 groups experienced endoscopic, radiologic, or surgical recurrence (Table I). All cases of anastomotic recurrence requiring surgery had evidence of recurrent disease at either radiology (computed tomography and/or small bowel follow-through) or colonoscopy or both. Admission to the hospital for medical treatment of CD was reported infrequently and was again comparable between the 2 groups. Most of the cases of recurrent disease were located at the anastomosis and had comparable biologic behavior between LC and OC. With regard to operative procedures to treat the first recurrence after the index operations, LC patients had an increased incidence of surgical recurrence at the ileocolic anastomosis compared with OC patients. When an operation became necessary for abdominal recurrence of CD, in half of the cases (4/8 patients) originally treated with LC it was possible to treat the patient with a repeat laparoscopic procedure. Although a laparoscopic approach was successful in only 1 case of abdominal recurrent disease after original OC, this difference did not reach statistical significance (Table I). There were 2 readmissions after abdominal operation for recurrent disease, one in each group. One patient who had originally undergone OC developed prolonged ileus after his open operation for recurrent disease. One additional patient who had originally been
treated with LC developed an intra-abdominal abscess after repeat LC for recurrent abdominal disease. Two patients required creation of a stoma after treatment of abdominal recurrent CD after open index operation versus none in the laparoscopic group. Both these patients had developed recurrent disease in the colon and rectum. With regard to the overall need for operations after the index operative procedure, the average number of operations per patient did not differ significantly in the 2 groups. However, patients initially treated with OC who required additional operations were significantly more likely to require >1 additional operation when compared with patients initially treated with LC (Table II). In respect to specific surgical procedures during follow-up, 2 patients underwent lysis of adhesions after LC versus none after OC (Table III). One of these 2 patients also developed an incisional hernia at the extraction site that was repaired; he subsequently underwent open ileocolic resection for anastomotic recurrence disease. During his subsequent follow-up, he developed adhesive mechanical small bowel obstruction requiring an operation. The remaining LC patient requiring lysis of adhesions had not had any operations since her index laparoscopic procedure. No other LC patients developed an incisional hernia. Conversely, 4 patients developed an incisional hernia after OC, 1 of whom required 3 sequential surgical repairs for recurrent incisional hernias. Last, 5
Stocchi, Milsom, and Fazio 625
Surgery Volume 144, Number 4
Table II. Surgical procedures at follow-up on 27 LC versus 29 OC patients
Table III. Follow-up operations and interventions on 27 LC versus 29 OC patients
LC OC P (patients, %) (patients, %) value Mean number of 0.59 ± 1.01 operations at follow-up (± SD) Number of operations at follow-up 1 12 (44.4%) 2 0 3 0 5 1 (3.7%) Number of abdominal operations 1 8 (29.6%) 3 0 (0%) 4 1 (3.7%) 5 0 Number of anorectal operations 1 5 (18.5%) 2 0 Number of anorectal conditions 1 5 (18.5%) 2 1 (3.7%)
0.76 ± 1.21
.92
.006 3 3 3 1
(10.3%) (10.3%) (10.3%) (3.4%) .89
LC OC P (patients, %) (patients,%) value Incisional hernia repair One hernia repair Three hernia repairs Lysis of adhesions Colonoscopic dilatations of anastomotic strictures One dilatation Two dilatations Three dilatations
1 (3.7%) 1 (3.7%) 0 2 (7.4%) 1 (3.7%)
4 (13.7%) 3 (10.3%) 1 (3.4%) 0 4 (13.8%)
1 (3.7%) 0 0
1 (3.4%) 2 (6.9%) 1 (3.4%)
.61
.23 .74
LC, Laparoscopic ileocolectomy; OC, open ileocolectomy.
8 (27.6%) 1 (3.4%) 0 1 (3.4%) .75 3 (10.3%) 2 (6.9%) .86 4 (13.8%) 1 (3.4%)
LC, Laparoscopic ileocolectomy; OC, open ileocolectomy.
patients underwent endoscopic dilatation of their ileocolic anastomosis for recurrent disease, 4 of whom had undergone OC. Surgery ultimately became necessary in 1 of these 4 patients after a single endoscopic dilatation because of recurrent symptomatic anastomotic stricture. The remaining 4 patients have not undergone operation for recurrent CD to date and received 3 dilatations in 1 case and 2 dilatations in the remaining 2 patients (Table III). Their follow-up since their latest endoscopic dilatations ranges from 5 to 21 months. Use of medications was comparable between the 2 groups both when considering overall use of any medications and use of specific agents either after the index operation or beginning after additional operations for CD (Table IV). DISCUSSION The data from our study based on long-term follow-up confirm that LC is an appropriate option for operative treatment of CD. There were no significant differences between LC and OC with regard to use of medication to treat CD, recurrence rates, and in particular need for abdominal operation for recurrent CD. It is also reassuring to note that the average number of overall operations
during the follow-up period per patient was not statistically different between LC and OC. It is even intriguing that more patients required multiple operations after index OC than after index LC, although a logical, straightforward explanation of this phenomenon remains elusive. The prospective randomized design of the original study should have helped to reduce a possible patient selection bias affecting long-term follow-up because the original groups were well-matched with regard to body mass index, age, gender, and American Society of Anesthesiology classification.14 Furthermore, the use of medications, which might affect recurrence rates, was similar between our 2 groups. We therefore analyzed the specific long-term complications potentially requiring operation that should be reduced after LC, namely adhesive small bowel obstructions and incisional hernias.20 With regard to adhesive small bowel obstruction, our study could not demonstrate any appreciable advantages of LC over OC. We relied on patient reports to identify recrudescence of CD requiring admission for medical treatment; it is, therefore, possible that nonoperative adhesive small bowel obstructions are underreported or erroneously considered as recurrent CD. However, the incidence of admission for presumed CD was low and comparable between the groups. Most important, no patient with OC required operation for adhesive small bowel obstruction. On the other hand, the crude percentages on incidence of incisional hernia do suggest benefits in favor of LC; <4% of the patients treated laparoscopically required incisional hernia repair compared with >13% of patients treated with OC. However, this difference was not statistically significant, probably because of the small numbers in our series. Because we specifically excluded from our analysis operations
626 Stocchi, Milsom, and Fazio
Surgery October 2008
Table IV. Medication use on 27 LC versus 29 OC patients LC (patients, %) Use of any medications after index operation Steroids Anti-inflammatory agents Anti-TNF Azathioprine/6-MP Other medications Use of any medications after second operation Steroids Anti-inflammatory agents Anti-TNF Azathioprine/6-MP Other medications Overall use of antidiarrheals
21 6 14 5 6 7 8 2 4 2 5 2 3
(78%) (22%) (52%) (19%) (22%) (26%) (30%) (7%) (15%) (7%) (19%) (7%) (11%)
OC (patients, %) 23 6 17 4 4 9 7 2 5 1 4 3 5
(79%) (21%) (59%) (14%) (14%) (31%) (24%) (7%) (17%) (3%) (14%) (10%) (17%)
P value .89 .89 .61 .72 .5 .67 .64 1 1 .6 .72 1 .71
LC, Laparoscopic ileocolectomy; OC, open ileocolectomy; TNF, tumor necrosis factor.
carried out for conditions unrelated to the index operations, recurrent disease with its various sequelae remains the main reason for repeated operations in our study. It could therefore be tempting to speculate that OC might predispose patients to multiple and more aggressive recurrent CD during long-term follow-up because of a still unclear biologic effect on immunity. However, our numbers are not sufficient to support this contention and larger sample sizes are needed to confirm these findings. An additional advantage of LC that might be underestimated because of our small sample size is the ability to repeat laparoscopic operation for recurrent abdominal CD when the index operation was also completed laparoscopically. This was possible in half of our LC patients, as also reported by Lowney et al.18 In our LC group, it is notable that the surgeons evaluating patients for abdominal CD recurrence were not necessarily experienced in laparoscopic techniques; it is, therefore, possible that some patients underwent OC for their second operation owing to surgeon preference rather than inability to carry out the operation laparoscopically. Although our study offers the advantages of a prospective randomized trial design with long-term follow-up, it presents the possible disadvantage of lacking evaluation of quality of life (QOL) after LC versus OC. Evaluations of QOL in prospective randomized trials comparing LC and OC for colon carcinoma21 and ileocolic resection for CD17 have indicated minimal, if any, clinically appreciable short-term benefits in favor of LC. In addition, a possible evaluation of QOL during long-term follow-up after LC versus OC would become even more difficult as the effects on QOL attributable
to a specific operative technique are complicated by the effects on QOL of a disease at inherently high risk of recurrence such as CD.
REFERENCES 1. Benoist S, Panis Y, Beaufour A, et al. Laparoscopic ileocecal resection in Crohn’s disease: a case-matched comparison with open resection. Surg Endosc 2003;17:814-8. 2. Huilgol RL, Wright CM, Solomon MJ. Laparoscopic versus open ileocolic resection for Crohn’s disease. J Laparoendosc Adv Surg Tech A 2004;14:61-5. 3. Kishi D, Nezu R, Ito T, et al. Laparoscopic-assisted surgery for Crohn’s disease: reduced surgical stress following ileocolectomy. Surg Today 2000;30:219-22. 4. Msika S, Iannelli A, Deroide G, et al. Can laparoscopy reduce hospital stay in the treatment of Crohn’s disease? Dis Colon Rectum 2001;44:1661-6. 5. Alabaz O, Iroatulam AJ, Nessim A, et al. Comparison of laparoscopically assisted and conventional ileocolic resection for Crohn’s disease. Eur J Surg 2000;166:213-7. 6. Diamond IR, Langer JC. Laparoscopic-assisted versus open ileocolic resection for adolescent Crohn disease. J Pediatr Gastroenterol Nutr 2001;33:543-7. 7. Tabet J, Hong D, Kim CW, et al. Laparoscopic versus open bowel resection for Crohn’s disease. Can J Gastroenterol 2001;15:237-42. 8. Bemelman WA, Slors JF, Dunker MS, et al. Laparoscopic-assisted vs. open ileocolic resection for Crohn’s disease. A comparative study. Surg Endosc 2000;14:721-5. 9. Fichera A, Peng SL, Elisseou NM, et al. Laparoscopy or conventional open surgery for patients with ileocolonic Crohn’s disease? A prospective study. Surgery 2007;142:566-71. 10. Young-Fadok TM, HallLong K, McConnell EJ, et al. Advantages of laparoscopic resection for ileocolic Crohn’s disease. Improved outcomes and reduced costs. Surg Endosc 2001; 15:450-4. 11. Bergamaschi R, Pessaux P, Arnaud JP. Comparison of conventional and laparoscopic ileocolic resection for Crohn’s disease. Dis Colon Rectum 2003;46:1129-33. 12. Duepree HJ, Senagore AJ, Delaney CP, et al. Advantages of laparoscopic resection for ileocecal Crohn’s disease. Dis Colon Rectum 2002;45:605-10.
Surgery Volume 144, Number 4
13. Luan X, Gross E. Laparoscopic assisted surgery for Crohn’s disease an initial experience and results. J Tongji Med Univ 2000;20:332-5. 14. Milsom JW, Hammerhofer KA, Bohm B, et al. Prospective, randomized trial comparing laparoscopic vs. conventional surgery for refractory ileocolic Crohn’s disease. Dis Colon Rectum 2001;44:1-8. 15. Shore G, Gonzalez QH, Bondora A, Vickers SM. Laparoscopic vs conventional ileocolectomy for primary Crohn disease. Arch Surg 2003;138:76-9. 16. von Allmen D, Markowitz JE, York A, et al. Laparoscopic-assisted bowel resection offers advantages over open surgery for treatment of segmental Crohn’s disease in children. J Pediatr Surg 2003;38:963-5. 17. Maartense S, Dunker MS, Slors JF, et al. Laparoscopic-assisted versus open ileocolic resection for Crohn’s disease: a randomized trial. Ann Surg 2006;243:143-9. 18. Lowney JK, Dietz DW, Birnbaum EH, et al. Is there any difference in recurrence rates in laparoscopic ileocolic resection for Crohn’s disease compared with conventional surgery? A long-term, follow-up study. Dis Colon Rectum 2006;49:58-63. 19. Gasche C, Scholmerich J, Brynskov J, et al. A simple classification of Crohn’s disease: report of the Working Party for the World Congresses of Gastroenterology, Vienna 1998. Inflamm Bowel Dis 2000;6:8-15. 20. Duepree HJ, Senagore AJ, Delaney CP, Fazio VW. Does means of access affect the incidence of small bowel obstruction and ventral hernia after bowel resection? Laparoscopy versus laparotomy. J Am Coll Surg 2003;197:177-81. 21. Weeks JC, Nelson H, Gelber S, et al. Short-term qualityof-life outcomes following laparoscopic-assisted colectomy vs open colectomy for colon cancer: a randomized trial. JAMA 2002;287:321-8.
DISCUSSION Dr Susan Galandiuk (Louisville, Ky): The study by Dr Stocchi and colleagues is important. With respect to treatment of CD, there have been many factors arguing both for and against laparoscopic treatment of disease. On the ‘‘con’’ side has been the fear of overlooking segments of diseased bowel by not being able to explore the abdomen as well as in open procedures and not having the ‘‘tactile’’ feedback available in open procedures. On the ‘‘pro’’ side are the smaller incisions in immunocompromised patients, with decreased wound healing capability, more rapid recovery, and decreased need for pain medication in patients who are often already on significant medication and have an increased tolerance. I have several questions for the authors: Do you think that the Crohn’s patient in the laparoscopic group who subsequently underwent open surgery for ‘‘diverticulitis’’ really had this? I would assume this was diverticulitis associated colitis and in reality Crohn’s. Was this performed at your institution or elsewhere? If done elsewhere, I would have the pathology re-reviewed. The onset of recurrent disease can be greatly affected by medical treatment. With the advent of more aggressive medical therapy, information regarding need for reoperation is incomplete without comparing how many patients in each group were maintained on azathioprine or 6-MP,
Stocchi, Milsom, and Fazio 627
and how many patients similarly maintained on biologic therapy with anti-TNF agents. Were there differences? I applaud the authors for their study, and hope that it is the first of many examining the long-term outcome of patients undergoing minimally invasive surgery for CD. Dr L. Stocchi (Cleveland, Ohio): With regard to the patient who underwent sigmoid colectomy for diverticulitis, we did review both the operative report and the pathology report. And this operation occurred several years after the index surgery, so we felt that although initially we were suspicious, as you were, that this was not really diverticulitis, the direct review of the report confirmed that it was sigmoid diverticulitis. With regard to the use of postoperative medication, some of these patients especially at that time were given Pentasa postoperatively immediately at the time of discharge. The studies with regard to the advantages of that practice have been conflicting, and so we have not been very accurate in collecting this specific information, which we could have. The use of postoperative medications was not studied within the confines of this trial. As our numbers are small, we would expect that there are not any significant differences. But it is certainly something that we could look into. Dr Fabrizio Michelassi (New York, NY): Dr Stocchi, first of all I would like to congratulate you on a good, clear presentation and on really trying to find all those patients who were initially part of that randomized study published in 2001. And I would like to remind everybody that this, to my knowledge, was the first and the only prospective randomized study in laparoscopic resection of ileocolic or CD to date. So it is an important study, and therefore, as a consequence, your follow-up study assumes even more importance. Since that 2001 study, there have been some 20 papers that have looked at laparoscopic colectomies and CD, and so far all of them pretty are much retrospective. And 1 of them suggested that the laparoscopic approach would prevent adhesions and to a certain extent would prevent long-term obstructions after the initial procedure. I was interested in seeing that indeed in your group there were 2 patients who developed small bowel obstruction second to adhesions and both of them were in the laparoscopic group. I would like to ask you this: What was the experience of the entire cohort of patients and what did you find in these 2 particular patients? The other thing is that I notice that of the 8 patients who underwent a laparoscopic operation, several underwent surgery again. Four of them were handled laparoscopically and 4 of them were handled in an open fashion. Therefore, I would like to ask you how it was decided that 4 patients who had initial laparoscopic surgery underwent operation for recurrent CD in an open fashion? What were the criteria? And if you do have, what was the subjective impression of the patients in comparing the laparoscopic approach at the first time and the open approach at the second time? You also make a cogent argument that the laparoscopic procedure may indeed prevent against the
628 Stocchi, Milsom, and Fazio
incisional hernia, although you do have 1 in the laparoscopic group and you have 3 in the open. And to a certain extent it is probably fair to remind everybody that even if you do a procedure laparoscopically, you do have to remove the specimen, so there is still an incision of about 5 or 6 centimeters, and sometimes these incisions are even bigger than that, because the inflammatory mass in the---is not going to come out intact---let’s put it this way, intact---and through a small incision. So it does not surprise me that indeed the laparoscopic approach may not confer an advantage against incisional hernias. And although in your conclusion you state that the laparoscopic treatment may provide an advantage, I would like to remind you that you have 25% of patients lost to follow-up so you do not really know what happened to those other patients who were not able to be followed up. So I think that the judgment on this is still far away. But I would like to congratulate you again on this because, as Dr Galandiuk said, we are in need of longterm outcomes in laparoscopic surgery for CD. Dr L. Stocchi (Cleveland, Ohio): Thank you for your insightful questions, Dr Michelassi. I will try to answer them in the order that they were asked. With regard to the ileocolic resection and the laparoscopic approach, it is true that the study published in 2001 by Dr Milsom remains 1 of the important studies. There was a recent study published from Europe also looking at the QOL and published in Annals of Surgery, and it confirms the data from Dr Milsom’s original study on the advantages of short-term results. With regard to the adhesions, 1 of our patients developed adhesive small bowel obstruction requiring an operation after he had undergone the initial laparoscopic procedure, and then a second procedure, which had been performed open for treatment of recurrent Crohn’s disease. So we rigorously cannot determine whether this is really an occurrence that is related to the second operation, which was performed open. But I think it is appropriate to maintain the patient in the laparoscopic group in an intent-to-treat analysis. Having said that, I think that you are right, that these data that I just presented do not definitely demonstrate that laparoscopic operation prevents adhesions in this patient population. With regard to the repeat operation performed laparoscopically versus open, this was not performed under the surveillance of a properly designed trial. And basically, if the patient who experienced recurrent disease was evaluated by a surgeon who performed laparoscopic operations, this would be an option, depending on the individual surgeon. Indeed, there were a number of these
Surgery October 2008
patients who had been originally treated laparoscopically who had not been offered a repeat laparoscopic operation, and that covers the 4 remaining patients, which to me, although again without statistical significance we cannot say really anything strongly, to me, it supports that repeat laparoscopic operation might have been possible for a greater number of these patients. With regard to the incidence of hernia, it is true that it is possible to have a hernia, after laparoscapic surgery, which in this particular study occurred after laparoscopic strictureplasty at the extraction site in the only patient in the laparoscopic group. It did not occur after the initial index operation, but occurred in this group after the repeat laparoscopic procedure, which was the stricturoplasty. Dr Mary Francis Otterson (Milwaukee, Wis): With respect to the difference in open and then laparoscopic surgical procedures, if the purpose is a cosmetically superior incision, I would suggest that sometimes the Pfannenstiel is appropriate, particularly for young women. Second, I think it is important to look at where you put your incision for the extraction of the specimen. Because the last thing you want to do in a patient who may well end up with a stoma is to eliminate a viable option for stoma placement in the future. So although I applaud the prospective nature of your laparoscopic study, I think that we have to be mindful of these considerations for patients. Dr L. Stocchi (Cleveland, Ohio): With regard to the use of Pfannenstiel incision, this is possible. And I had some colleagues who have used it. I personally feel a little reluctant to use it for 2 reasons. Sometimes in these patients it is possible to do an ileocolic resection that does not require such an extensive mobilization as you would do for a cancer operation when you perform laparoscopic right hemicolectomy. However, it might be difficult to exteriorize the entire bowel that you need without creating tension at the anastomosis. In addition, I would have concern because this patient population is subject to recurrent disease that, as you suggested, you do not create an additional incision when you can maintain the same incision or utilize a portion of your incision for reoperation. With regard to the comments regarding the extraction site, I personally prefer to perform a periumbilical small incision to extract my specimen in this patient population, although there are surgeons who perform laparoscopic operations in favor of left lower quadrant or right lower quadrant incision to add for benefits in the immediate postoperative course. I do agree with you that these would not be my favorite incisions for extraction site in this patient population.
Background. The purpose of this study was to analyze long-term recurrence rates and complications in patients previously enrolled in a prospective randomized trial comparing laparoscopic (LC) and open ileocolectomy (OC) for ileocolic Crohn’s disease (CD). Methods. Follow-up data were available on 56 of 60 patients. Demographic data, recurrence rates, need for additional surgery related to primary procedure, and medication use were recorded. Results. Mean follow-up for 56 patients (27 LC vs 29 OC) was 10.5 years and comparable between LC and OC (10.0 vs 11.0, respectively; P = .64). One patient died 8 years after OC of causes unrelated to CD. Eight patients for each group underwent initial reoperative (26% LC vs 28% OC; P = .89). One patient underwent incisional hernia repair after LC (4%) versus 4 patients (14%) after OC (P = .61). Two patients in the LC group underwent adhesiolysis versus none after OC (P = .23). Incidences of anorectal disease, anorectal surgery, endoscopic or radiologic recurrence, and medication use were also similar between LC and OC. OC patients requiring operation during follow-up were significantly more likely than LC to require multiple operations (P = .006). Conclusions. Long-term data from this prospective randomized trial confirm that LC is at least comparable to OC in the treatment of ileocolic CD. (Surgery 2008;144:622-8.) From the Department of Colorectal Surgery, Cleveland Clinic,a Cleveland, Ohio; and the Department of Surgery, New York Presbyterian Hospital, Weill Medical College of Cornell University,b New York, NY
LAPAROSCOPIC ILEOCOLIC RESECTION (LC) for ileocolic Crohn’s disease (CD) has demonstrated short-term benefits, including reduced duration of postoperative ileus,1-4 decreased narcotic requirement,1,5-7 reduced length of hospital stay,1,2,4-9 and decreased costs10 when compared with open ileocolic resection (OC). Postoperative morbidity after laparoscopic approach has also been either comparable or decreased when compared with OC.1,2,4-8,11-16 Only 2 prospective randomized trials have been published on short-term outcomes after LC versus OC for CD, including the study by Milsom et al in 200114 and a multicenter trial from Europe more recently reported.17 On the other hand, only a handful of studies have longer term follow-up, Accepted for publication June 19, 2008. Reprint requests: Luca Stocchi, MD, Department of Colorectal Surgery, 9500 Euclid Avenue, Desk A-30, Cleveland Clinic, Cleveland, OH 44195. E-mail: [email protected]. 0039-6060/$ - see front matter Ó 2008 Mosby, Inc. All rights reserved. doi:10.1016/j.surg.2008.06.016
622 SURGERY
ranging from a mean of 17 months to 81 months,4,5,7,11,14,15,18 none of which was designed as a prospective randomized trial. The purpose of this study was therefore to evaluate the long-term rates of small bowel obstruction, reoperation, and recurrence rates after LC versus OC for CD in the cohort of patient originally enrolled into the study by Milsom et al.14 MATERIALS AND METHODS All patients previously enrolled into a prospective randomized trial evaluating LC versus OC for CD14 were included. Patient accrual for the original trial was completed in March 1998. Collected information included demographic data, date of original surgery, total number of operations after index surgical procedure, operations for abdominal recurrent CD, CD-related anorectal conditions and operations, biologic behavior of the disease according to the Vienna classification,19 use of medications to prevent or treat recurrence (steroids, antiinflammatory agents, immunomodulators), and tobacco consumption.
Surgery Volume 144, Number 4
With regard to smoking history, we subdivided patients into those who stopped smoking after their index surgical procedure and those who continued smoking after their index operation. Long-term data recorded included incidence, location, and behavior of recurrent disease; need for hospital admission; incidence of endoscopic recurrence; radiologic recurrence or recurrence requiring operation; type of operation(s) performed; postoperative outcomes after the second operation; and use of postoperative medication to reduce the incidence of further recurrence. Data were collected from our institutional, Institutional Review Board-approved, electronic CD database and supplemented by chart review as necessary. Patients who were not already in the CD database were contacted by mail and phone if necessary and asked to participate in the database, therefore providing data on the treatment(s) received for their disease after the time of their operation. The incidence of recurrence was recorded on a per-patient basis. The definition of recurrence excluded symptomatic recurrence only and was limited to recurrence based on imaging and/or operation. Radiologic recurrence was based on the readings from small bowel follow-through or abdominal and pelvic computed tomography. Endoscopic recurrence was based on the reading of the colonoscopy reports and was not further classified. The incidence of recurrence was collected separately and defined as anastomotic recurrence when involving the small bowel or colon or both at the level of the anastomosis, and small bowel or colonic recurrence when the disease recurred away from the anastomosis in the small bowel or colon. There were no recurrences in the upper gastrointestinal tract. We considered as a recurrence requiring surgery a condition requiring an abdominal operation. Use of any medications was recorded as a single variable regardless whether the medications were used on a regular basis or for a treatment course. The medications administered were also subdivided into antiinflammatory agents, steroids, azathioprine/6-MP, anti-tumor necrosis factor (TNF) agents and others (antibiotics, methotrexate, antidiarrheals, probiotics). Antidiarrheals were used after the index operations in all cases except 2 and we decided to consider them combined. However, the remaining medication categories were divided according to the time of their initiation, whether after the index operation or after the second operation or later in those patients who had >2 CD-related operations. We collected the incidence of any operations in the follow-up whose indication could be related to
Stocchi, Milsom, and Fazio 623
the index surgery. In particular, we excluded indications for surgery that were unrelated to CD or complications of the index operation, such as adhesions and incisional hernias. We accepted the diagnosis of adhesive small bowel occlusions requiring surgery based on the operative reports to ensure that recurrent CD was not a concern. We also collected the incidence of incisional hernia repairs related to the incisions used for the operations to treat CD. In addition, we collected data on colonoscopic balloon dilatation of anastomotic strictures related to recurrent CD. The decision to perform and repeat this intervention instead of considering surgery or continuing management with medical treatment alone was left to the treating gastroenterologist. The data on admissions to the hospital for medical treatment were based on documentation or patient report, whereas details on surgical procedures were based only on direct review of operative reports from our institution of obtained from our institution or from outside. Statistical analysis was conducted by v2 or Fisher’s exact tests for group comparisons with respect to categorical variables. Group comparisons with respect to quantitative and ordinal variables were carried out using the Wilcoxon rank sum test. P < .05 was considered significant. RESULTS Four out of 60 patients had no follow-up (2 OC and 2 LC), 2 of whom declined to provide updates regarding their conditions. This left 56 patients for analysis, 24 of whom were males. Mean age at the time of index surgery was comparable (36 vs 35 years old for LC and OC, respectively; P = .68). Mean follow-up for the entire patient cohort was 10.5 (± 2.5) years and was also comparable between the 2 groups (Table I). One patient died 8 years after OC for causes unrelated to CD. One additional patient underwent open sigmoid resection for diverticulitis 2 years after laparoscopic ileocolic resection. He developed an incisional hernia after his sigmoidectomy, which was therefore not considered as a long-term complication of his operative treatment for CD. A history of smoking was present in 21 out of 56 patients (43%) and was not significantly different when comparing the 2 groups (12 patients, 52% vs 9 patients, 35% for LC and OC, respectively; P = .22). Two patients had stopped smoking after their index OC, whereas all the remaining individuals with history of tobacco use in either groups continued smoking for variable periods of time during their follow-up.
624 Stocchi, Milsom, and Fazio
Surgery October 2008
Table I. Recurrent disease on 27 LC versus 29 OC patients Mean follow-up, yrs (SD) Colonoscopy during follow-up Endoscopic recurrence Radiological recurrence First surgical recurrence In-hospital medical treatment for recurrent disease anytime Location first recurrence Anastomotic Small bowel Colon Behavior first recurrence Nonstricturing, nonpenetrating disease Stricturing Penetrating Operation for first recurrence Resection and reanastomosis Small bowel resection Colonic resection Strictureplasty alone Laparoscopic resection for recurrent abdominal disease
LC (patients, %)
OC (patients, %)
9.97 ± 3.17 18 (66.7%) 12 (48.0%) 13 (48.1%) 8 (25.9%) 3 (11.1%)
10.98 ± 1.38 22 (75.9%) 19 (65.5%) 15 (51.7%) 8 (27.6%) 2 (6.9%)
13 (48.1%) 1 (3.7%) 0
12 (41.4%) 1 (3.4%) 2 (6.9%)
5 (14.8%) 9 (33.3%) 0
3 (10.3%) 8 (27.6%) 4 (13.8%)
7 (88%) 0 0 1 (12.5%) 4 (50%)
4 (50%) 1 (12.5%) 3 (37.5%) 0 1 (12.5%)
P value .64 .45 .2 .89 .89 .83 .78
.29
.13
.31
LC, Laparoscopic ileocolectomy; OC, open ileocolectomy.
The overall recurrence rate was 52% (29/56 patients). A comparable number of patients in the 2 groups experienced endoscopic, radiologic, or surgical recurrence (Table I). All cases of anastomotic recurrence requiring surgery had evidence of recurrent disease at either radiology (computed tomography and/or small bowel follow-through) or colonoscopy or both. Admission to the hospital for medical treatment of CD was reported infrequently and was again comparable between the 2 groups. Most of the cases of recurrent disease were located at the anastomosis and had comparable biologic behavior between LC and OC. With regard to operative procedures to treat the first recurrence after the index operations, LC patients had an increased incidence of surgical recurrence at the ileocolic anastomosis compared with OC patients. When an operation became necessary for abdominal recurrence of CD, in half of the cases (4/8 patients) originally treated with LC it was possible to treat the patient with a repeat laparoscopic procedure. Although a laparoscopic approach was successful in only 1 case of abdominal recurrent disease after original OC, this difference did not reach statistical significance (Table I). There were 2 readmissions after abdominal operation for recurrent disease, one in each group. One patient who had originally undergone OC developed prolonged ileus after his open operation for recurrent disease. One additional patient who had originally been
treated with LC developed an intra-abdominal abscess after repeat LC for recurrent abdominal disease. Two patients required creation of a stoma after treatment of abdominal recurrent CD after open index operation versus none in the laparoscopic group. Both these patients had developed recurrent disease in the colon and rectum. With regard to the overall need for operations after the index operative procedure, the average number of operations per patient did not differ significantly in the 2 groups. However, patients initially treated with OC who required additional operations were significantly more likely to require >1 additional operation when compared with patients initially treated with LC (Table II). In respect to specific surgical procedures during follow-up, 2 patients underwent lysis of adhesions after LC versus none after OC (Table III). One of these 2 patients also developed an incisional hernia at the extraction site that was repaired; he subsequently underwent open ileocolic resection for anastomotic recurrence disease. During his subsequent follow-up, he developed adhesive mechanical small bowel obstruction requiring an operation. The remaining LC patient requiring lysis of adhesions had not had any operations since her index laparoscopic procedure. No other LC patients developed an incisional hernia. Conversely, 4 patients developed an incisional hernia after OC, 1 of whom required 3 sequential surgical repairs for recurrent incisional hernias. Last, 5
Stocchi, Milsom, and Fazio 625
Surgery Volume 144, Number 4
Table II. Surgical procedures at follow-up on 27 LC versus 29 OC patients
Table III. Follow-up operations and interventions on 27 LC versus 29 OC patients
LC OC P (patients, %) (patients, %) value Mean number of 0.59 ± 1.01 operations at follow-up (± SD) Number of operations at follow-up 1 12 (44.4%) 2 0 3 0 5 1 (3.7%) Number of abdominal operations 1 8 (29.6%) 3 0 (0%) 4 1 (3.7%) 5 0 Number of anorectal operations 1 5 (18.5%) 2 0 Number of anorectal conditions 1 5 (18.5%) 2 1 (3.7%)
0.76 ± 1.21
.92
.006 3 3 3 1
(10.3%) (10.3%) (10.3%) (3.4%) .89
LC OC P (patients, %) (patients,%) value Incisional hernia repair One hernia repair Three hernia repairs Lysis of adhesions Colonoscopic dilatations of anastomotic strictures One dilatation Two dilatations Three dilatations
1 (3.7%) 1 (3.7%) 0 2 (7.4%) 1 (3.7%)
4 (13.7%) 3 (10.3%) 1 (3.4%) 0 4 (13.8%)
1 (3.7%) 0 0
1 (3.4%) 2 (6.9%) 1 (3.4%)
.61
.23 .74
LC, Laparoscopic ileocolectomy; OC, open ileocolectomy.
8 (27.6%) 1 (3.4%) 0 1 (3.4%) .75 3 (10.3%) 2 (6.9%) .86 4 (13.8%) 1 (3.4%)
LC, Laparoscopic ileocolectomy; OC, open ileocolectomy.
patients underwent endoscopic dilatation of their ileocolic anastomosis for recurrent disease, 4 of whom had undergone OC. Surgery ultimately became necessary in 1 of these 4 patients after a single endoscopic dilatation because of recurrent symptomatic anastomotic stricture. The remaining 4 patients have not undergone operation for recurrent CD to date and received 3 dilatations in 1 case and 2 dilatations in the remaining 2 patients (Table III). Their follow-up since their latest endoscopic dilatations ranges from 5 to 21 months. Use of medications was comparable between the 2 groups both when considering overall use of any medications and use of specific agents either after the index operation or beginning after additional operations for CD (Table IV). DISCUSSION The data from our study based on long-term follow-up confirm that LC is an appropriate option for operative treatment of CD. There were no significant differences between LC and OC with regard to use of medication to treat CD, recurrence rates, and in particular need for abdominal operation for recurrent CD. It is also reassuring to note that the average number of overall operations
during the follow-up period per patient was not statistically different between LC and OC. It is even intriguing that more patients required multiple operations after index OC than after index LC, although a logical, straightforward explanation of this phenomenon remains elusive. The prospective randomized design of the original study should have helped to reduce a possible patient selection bias affecting long-term follow-up because the original groups were well-matched with regard to body mass index, age, gender, and American Society of Anesthesiology classification.14 Furthermore, the use of medications, which might affect recurrence rates, was similar between our 2 groups. We therefore analyzed the specific long-term complications potentially requiring operation that should be reduced after LC, namely adhesive small bowel obstructions and incisional hernias.20 With regard to adhesive small bowel obstruction, our study could not demonstrate any appreciable advantages of LC over OC. We relied on patient reports to identify recrudescence of CD requiring admission for medical treatment; it is, therefore, possible that nonoperative adhesive small bowel obstructions are underreported or erroneously considered as recurrent CD. However, the incidence of admission for presumed CD was low and comparable between the groups. Most important, no patient with OC required operation for adhesive small bowel obstruction. On the other hand, the crude percentages on incidence of incisional hernia do suggest benefits in favor of LC; <4% of the patients treated laparoscopically required incisional hernia repair compared with >13% of patients treated with OC. However, this difference was not statistically significant, probably because of the small numbers in our series. Because we specifically excluded from our analysis operations
626 Stocchi, Milsom, and Fazio
Surgery October 2008
Table IV. Medication use on 27 LC versus 29 OC patients LC (patients, %) Use of any medications after index operation Steroids Anti-inflammatory agents Anti-TNF Azathioprine/6-MP Other medications Use of any medications after second operation Steroids Anti-inflammatory agents Anti-TNF Azathioprine/6-MP Other medications Overall use of antidiarrheals
21 6 14 5 6 7 8 2 4 2 5 2 3
(78%) (22%) (52%) (19%) (22%) (26%) (30%) (7%) (15%) (7%) (19%) (7%) (11%)
OC (patients, %) 23 6 17 4 4 9 7 2 5 1 4 3 5
(79%) (21%) (59%) (14%) (14%) (31%) (24%) (7%) (17%) (3%) (14%) (10%) (17%)
P value .89 .89 .61 .72 .5 .67 .64 1 1 .6 .72 1 .71
LC, Laparoscopic ileocolectomy; OC, open ileocolectomy; TNF, tumor necrosis factor.
carried out for conditions unrelated to the index operations, recurrent disease with its various sequelae remains the main reason for repeated operations in our study. It could therefore be tempting to speculate that OC might predispose patients to multiple and more aggressive recurrent CD during long-term follow-up because of a still unclear biologic effect on immunity. However, our numbers are not sufficient to support this contention and larger sample sizes are needed to confirm these findings. An additional advantage of LC that might be underestimated because of our small sample size is the ability to repeat laparoscopic operation for recurrent abdominal CD when the index operation was also completed laparoscopically. This was possible in half of our LC patients, as also reported by Lowney et al.18 In our LC group, it is notable that the surgeons evaluating patients for abdominal CD recurrence were not necessarily experienced in laparoscopic techniques; it is, therefore, possible that some patients underwent OC for their second operation owing to surgeon preference rather than inability to carry out the operation laparoscopically. Although our study offers the advantages of a prospective randomized trial design with long-term follow-up, it presents the possible disadvantage of lacking evaluation of quality of life (QOL) after LC versus OC. Evaluations of QOL in prospective randomized trials comparing LC and OC for colon carcinoma21 and ileocolic resection for CD17 have indicated minimal, if any, clinically appreciable short-term benefits in favor of LC. In addition, a possible evaluation of QOL during long-term follow-up after LC versus OC would become even more difficult as the effects on QOL attributable
to a specific operative technique are complicated by the effects on QOL of a disease at inherently high risk of recurrence such as CD.
REFERENCES 1. Benoist S, Panis Y, Beaufour A, et al. Laparoscopic ileocecal resection in Crohn’s disease: a case-matched comparison with open resection. Surg Endosc 2003;17:814-8. 2. Huilgol RL, Wright CM, Solomon MJ. Laparoscopic versus open ileocolic resection for Crohn’s disease. J Laparoendosc Adv Surg Tech A 2004;14:61-5. 3. Kishi D, Nezu R, Ito T, et al. Laparoscopic-assisted surgery for Crohn’s disease: reduced surgical stress following ileocolectomy. Surg Today 2000;30:219-22. 4. Msika S, Iannelli A, Deroide G, et al. Can laparoscopy reduce hospital stay in the treatment of Crohn’s disease? Dis Colon Rectum 2001;44:1661-6. 5. Alabaz O, Iroatulam AJ, Nessim A, et al. Comparison of laparoscopically assisted and conventional ileocolic resection for Crohn’s disease. Eur J Surg 2000;166:213-7. 6. Diamond IR, Langer JC. Laparoscopic-assisted versus open ileocolic resection for adolescent Crohn disease. J Pediatr Gastroenterol Nutr 2001;33:543-7. 7. Tabet J, Hong D, Kim CW, et al. Laparoscopic versus open bowel resection for Crohn’s disease. Can J Gastroenterol 2001;15:237-42. 8. Bemelman WA, Slors JF, Dunker MS, et al. Laparoscopic-assisted vs. open ileocolic resection for Crohn’s disease. A comparative study. Surg Endosc 2000;14:721-5. 9. Fichera A, Peng SL, Elisseou NM, et al. Laparoscopy or conventional open surgery for patients with ileocolonic Crohn’s disease? A prospective study. Surgery 2007;142:566-71. 10. Young-Fadok TM, HallLong K, McConnell EJ, et al. Advantages of laparoscopic resection for ileocolic Crohn’s disease. Improved outcomes and reduced costs. Surg Endosc 2001; 15:450-4. 11. Bergamaschi R, Pessaux P, Arnaud JP. Comparison of conventional and laparoscopic ileocolic resection for Crohn’s disease. Dis Colon Rectum 2003;46:1129-33. 12. Duepree HJ, Senagore AJ, Delaney CP, et al. Advantages of laparoscopic resection for ileocecal Crohn’s disease. Dis Colon Rectum 2002;45:605-10.
Surgery Volume 144, Number 4
13. Luan X, Gross E. Laparoscopic assisted surgery for Crohn’s disease an initial experience and results. J Tongji Med Univ 2000;20:332-5. 14. Milsom JW, Hammerhofer KA, Bohm B, et al. Prospective, randomized trial comparing laparoscopic vs. conventional surgery for refractory ileocolic Crohn’s disease. Dis Colon Rectum 2001;44:1-8. 15. Shore G, Gonzalez QH, Bondora A, Vickers SM. Laparoscopic vs conventional ileocolectomy for primary Crohn disease. Arch Surg 2003;138:76-9. 16. von Allmen D, Markowitz JE, York A, et al. Laparoscopic-assisted bowel resection offers advantages over open surgery for treatment of segmental Crohn’s disease in children. J Pediatr Surg 2003;38:963-5. 17. Maartense S, Dunker MS, Slors JF, et al. Laparoscopic-assisted versus open ileocolic resection for Crohn’s disease: a randomized trial. Ann Surg 2006;243:143-9. 18. Lowney JK, Dietz DW, Birnbaum EH, et al. Is there any difference in recurrence rates in laparoscopic ileocolic resection for Crohn’s disease compared with conventional surgery? A long-term, follow-up study. Dis Colon Rectum 2006;49:58-63. 19. Gasche C, Scholmerich J, Brynskov J, et al. A simple classification of Crohn’s disease: report of the Working Party for the World Congresses of Gastroenterology, Vienna 1998. Inflamm Bowel Dis 2000;6:8-15. 20. Duepree HJ, Senagore AJ, Delaney CP, Fazio VW. Does means of access affect the incidence of small bowel obstruction and ventral hernia after bowel resection? Laparoscopy versus laparotomy. J Am Coll Surg 2003;197:177-81. 21. Weeks JC, Nelson H, Gelber S, et al. Short-term qualityof-life outcomes following laparoscopic-assisted colectomy vs open colectomy for colon cancer: a randomized trial. JAMA 2002;287:321-8.
DISCUSSION Dr Susan Galandiuk (Louisville, Ky): The study by Dr Stocchi and colleagues is important. With respect to treatment of CD, there have been many factors arguing both for and against laparoscopic treatment of disease. On the ‘‘con’’ side has been the fear of overlooking segments of diseased bowel by not being able to explore the abdomen as well as in open procedures and not having the ‘‘tactile’’ feedback available in open procedures. On the ‘‘pro’’ side are the smaller incisions in immunocompromised patients, with decreased wound healing capability, more rapid recovery, and decreased need for pain medication in patients who are often already on significant medication and have an increased tolerance. I have several questions for the authors: Do you think that the Crohn’s patient in the laparoscopic group who subsequently underwent open surgery for ‘‘diverticulitis’’ really had this? I would assume this was diverticulitis associated colitis and in reality Crohn’s. Was this performed at your institution or elsewhere? If done elsewhere, I would have the pathology re-reviewed. The onset of recurrent disease can be greatly affected by medical treatment. With the advent of more aggressive medical therapy, information regarding need for reoperation is incomplete without comparing how many patients in each group were maintained on azathioprine or 6-MP,
Stocchi, Milsom, and Fazio 627
and how many patients similarly maintained on biologic therapy with anti-TNF agents. Were there differences? I applaud the authors for their study, and hope that it is the first of many examining the long-term outcome of patients undergoing minimally invasive surgery for CD. Dr L. Stocchi (Cleveland, Ohio): With regard to the patient who underwent sigmoid colectomy for diverticulitis, we did review both the operative report and the pathology report. And this operation occurred several years after the index surgery, so we felt that although initially we were suspicious, as you were, that this was not really diverticulitis, the direct review of the report confirmed that it was sigmoid diverticulitis. With regard to the use of postoperative medication, some of these patients especially at that time were given Pentasa postoperatively immediately at the time of discharge. The studies with regard to the advantages of that practice have been conflicting, and so we have not been very accurate in collecting this specific information, which we could have. The use of postoperative medications was not studied within the confines of this trial. As our numbers are small, we would expect that there are not any significant differences. But it is certainly something that we could look into. Dr Fabrizio Michelassi (New York, NY): Dr Stocchi, first of all I would like to congratulate you on a good, clear presentation and on really trying to find all those patients who were initially part of that randomized study published in 2001. And I would like to remind everybody that this, to my knowledge, was the first and the only prospective randomized study in laparoscopic resection of ileocolic or CD to date. So it is an important study, and therefore, as a consequence, your follow-up study assumes even more importance. Since that 2001 study, there have been some 20 papers that have looked at laparoscopic colectomies and CD, and so far all of them pretty are much retrospective. And 1 of them suggested that the laparoscopic approach would prevent adhesions and to a certain extent would prevent long-term obstructions after the initial procedure. I was interested in seeing that indeed in your group there were 2 patients who developed small bowel obstruction second to adhesions and both of them were in the laparoscopic group. I would like to ask you this: What was the experience of the entire cohort of patients and what did you find in these 2 particular patients? The other thing is that I notice that of the 8 patients who underwent a laparoscopic operation, several underwent surgery again. Four of them were handled laparoscopically and 4 of them were handled in an open fashion. Therefore, I would like to ask you how it was decided that 4 patients who had initial laparoscopic surgery underwent operation for recurrent CD in an open fashion? What were the criteria? And if you do have, what was the subjective impression of the patients in comparing the laparoscopic approach at the first time and the open approach at the second time? You also make a cogent argument that the laparoscopic procedure may indeed prevent against the
628 Stocchi, Milsom, and Fazio
incisional hernia, although you do have 1 in the laparoscopic group and you have 3 in the open. And to a certain extent it is probably fair to remind everybody that even if you do a procedure laparoscopically, you do have to remove the specimen, so there is still an incision of about 5 or 6 centimeters, and sometimes these incisions are even bigger than that, because the inflammatory mass in the---is not going to come out intact---let’s put it this way, intact---and through a small incision. So it does not surprise me that indeed the laparoscopic approach may not confer an advantage against incisional hernias. And although in your conclusion you state that the laparoscopic treatment may provide an advantage, I would like to remind you that you have 25% of patients lost to follow-up so you do not really know what happened to those other patients who were not able to be followed up. So I think that the judgment on this is still far away. But I would like to congratulate you again on this because, as Dr Galandiuk said, we are in need of longterm outcomes in laparoscopic surgery for CD. Dr L. Stocchi (Cleveland, Ohio): Thank you for your insightful questions, Dr Michelassi. I will try to answer them in the order that they were asked. With regard to the ileocolic resection and the laparoscopic approach, it is true that the study published in 2001 by Dr Milsom remains 1 of the important studies. There was a recent study published from Europe also looking at the QOL and published in Annals of Surgery, and it confirms the data from Dr Milsom’s original study on the advantages of short-term results. With regard to the adhesions, 1 of our patients developed adhesive small bowel obstruction requiring an operation after he had undergone the initial laparoscopic procedure, and then a second procedure, which had been performed open for treatment of recurrent Crohn’s disease. So we rigorously cannot determine whether this is really an occurrence that is related to the second operation, which was performed open. But I think it is appropriate to maintain the patient in the laparoscopic group in an intent-to-treat analysis. Having said that, I think that you are right, that these data that I just presented do not definitely demonstrate that laparoscopic operation prevents adhesions in this patient population. With regard to the repeat operation performed laparoscopically versus open, this was not performed under the surveillance of a properly designed trial. And basically, if the patient who experienced recurrent disease was evaluated by a surgeon who performed laparoscopic operations, this would be an option, depending on the individual surgeon. Indeed, there were a number of these
Surgery October 2008
patients who had been originally treated laparoscopically who had not been offered a repeat laparoscopic operation, and that covers the 4 remaining patients, which to me, although again without statistical significance we cannot say really anything strongly, to me, it supports that repeat laparoscopic operation might have been possible for a greater number of these patients. With regard to the incidence of hernia, it is true that it is possible to have a hernia, after laparoscapic surgery, which in this particular study occurred after laparoscopic strictureplasty at the extraction site in the only patient in the laparoscopic group. It did not occur after the initial index operation, but occurred in this group after the repeat laparoscopic procedure, which was the stricturoplasty. Dr Mary Francis Otterson (Milwaukee, Wis): With respect to the difference in open and then laparoscopic surgical procedures, if the purpose is a cosmetically superior incision, I would suggest that sometimes the Pfannenstiel is appropriate, particularly for young women. Second, I think it is important to look at where you put your incision for the extraction of the specimen. Because the last thing you want to do in a patient who may well end up with a stoma is to eliminate a viable option for stoma placement in the future. So although I applaud the prospective nature of your laparoscopic study, I think that we have to be mindful of these considerations for patients. Dr L. Stocchi (Cleveland, Ohio): With regard to the use of Pfannenstiel incision, this is possible. And I had some colleagues who have used it. I personally feel a little reluctant to use it for 2 reasons. Sometimes in these patients it is possible to do an ileocolic resection that does not require such an extensive mobilization as you would do for a cancer operation when you perform laparoscopic right hemicolectomy. However, it might be difficult to exteriorize the entire bowel that you need without creating tension at the anastomosis. In addition, I would have concern because this patient population is subject to recurrent disease that, as you suggested, you do not create an additional incision when you can maintain the same incision or utilize a portion of your incision for reoperation. With regard to the comments regarding the extraction site, I personally prefer to perform a periumbilical small incision to extract my specimen in this patient population, although there are surgeons who perform laparoscopic operations in favor of left lower quadrant or right lower quadrant incision to add for benefits in the immediate postoperative course. I do agree with you that these would not be my favorite incisions for extraction site in this patient population.