A, Global recurrence based on family history, smoking, and disease location. Recurrence events were significantly negatively associated with cessation of smoking (P < .05) and with distal ileum-colon localization of the disease (P < .01). B, Surgical recurrence in side-to-side isoperistaltic strictureplasty (SSIS) based on family history of inflammatory bowel disease and disease location. Surgical recurrence in SSIS alone was associated with significant reduction in patients without a family history and with a disease localization in the distal part of the bowel.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Fazi M, Giudici F, Luceri C, Pronestì M, Tonelli F. Long-term Results and Recurrence-Related Risk Factors for Crohn Disease in Patients Undergoing Side-to-Side Isoperistaltic Strictureplasty. JAMA Surg. 2016;151(5):452–460. doi:10.1001/jamasurg.2015.4552
Side-to-side isoperistaltic strictureplasty (SSIS) is useful in patients undergoing surgery for Crohn disease (CD) to avoid wide small-bowel resections. To our knowledge, there are no definitive data regarding its recurrence risk factors.
To evaluate the results obtained in a monocentric population of patients with CD who have undergone SSIS.
Design, Setting, and Participants
From August 1996 to March 2010, 91 patients with CD underwent SSIS in our center. In this prospective observational study, side-to-side isoperistaltic strictureplasty was according the Michelassi technique in 69 patients and the Tonelli technique in 22 patients. Factors relating to the patient and the CD, surgery, and pharmacological therapy during the preoperative and perioperative periods were evaluated in association with medical or surgical recurrence.
Side-to-side isoperistaltic strictureplasty.
Main Outcomes and Measures
The recurrence-free curve was estimated using Kaplan-Meier analysis. Patients were stratified into cohorts in relation to the considered categorical variables and data were compared by using the Mantel-Cox log-rank test. Cox proportional hazard regression analysis was used to set up a predictive model simultaneously exploring the effects of all independent variables on a dichotomous outcome recurrence in relation to time.
Among the 91 patients, the mean (SD) age was 39.5 (11.2) years and preoperative disease duration was 97.9 (85.8) months; 83 patients (91.2%) were followed up, of whom 37 (44.58%) experienced a recurrence at a mean (SD) of 55.46 (36.79) months after surgery (range, 9-140 months). The recurrence in the SSIS site at a mean (SD) of 48.25 (29.94) months after surgery affected 24 of 83 patients (28.9%), 9 being medical and 15 being surgical recurrence. Recurrence in the SSIS was statistically significantly associated with the time elapsed between diagnosis and surgery (P = .03). A borderline association between family history of CD and surgical recurrence (P = .054) was also found. Multivariate analysis identified the age at diagnosis (χ2 = 5.56; P = .02) and at surgery (χ2 = 7.77; P = .005), family history (χ2 = 6.26; P = .01), and smoking habit (χ2 = 10.06; P = .007) as independent risk factors for recurrence.
Conclusions and Relevance
In the short-term, SSIS leads to a resolution of symptoms in more than 90% of cases and the recurrence rate in the SSIS area is acceptable, even after long-term follow-up.
The surgical outcome of Crohn disease (CD) is characterized by recurrence even after “radical” operations, which remove both microscopically and macroscopically affected bowel. Although the surgical approach in the past has been characterized by radical resections of the affected areas, the present therapeutic approach is to perform conservative operations and reduce the risk for short-bowel syndrome. Included in this approach are techniques such as minimal resection and strictureplasty (SXPL). Lee and Papaioannou were the first to use SXPL in patients with CD in 1982.1 Over the last 30 years, the SXPL method has further expanded and been performed even in long strictures.
In 1996, Michelassi2 described a new type of SXPL known as side-to-side isoperistaltic strictureplasty (SSIS). In 2000, a variation in the original technique in the section of the neoterminal ileum and colic cul-de-sac at the level of a previous side-to-side or end-to-side ileo-colic anastomosis performed for CD was described by one of the authors (F.T.).3
In 2004, our center evaluated patients who had undergone SSIS, but at the time, the cases were still limited (n = 21) and the follow-up relatively short.4 In 2007, a retrospective observational, multicenter international study was conducted with the purpose of assessing the results achieved with SSIS in its first 10 years of use in treating CD. The evidence obtained proved how it could be considered a safe technique, with very low morbidity and mortality rates and an acceptable risk for recurrence.5
Because the evaluation of risk factors implicated in the recurrence of CD in the strictureplasted bowel is still extremely limited, the purpose of this study was to evaluate the results obtained in a large population of patients with CD who have undergone SSIS, after long-term follow-up, and especially to identify possible recurrence-related factors at the SSIS site.
From August 1996 to March 2010, 91 patients with CD (48 men and 43 women; median age, 39 years; range, 20-71 years) consecutively underwent SSIS in our center; the operations were performed by the same surgeon. The study was approved by the local ethics committee of Azienda Ospedaliera Careggi. Oral informed consent was obtained from the study participants. At the time of first surgery, 34 patients were smokers, 23 were former smokers, and 34 had never smoked. The main indication for surgery was bowel obstruction in 68 cases, malnutrition in 11 patients, and fistula and/or abscess in 12 patients. Of those patients with fistulizing CD, 7 had entero-enteric fistulae, 1 had entero-colic fistulae, 2 had entero-sigmoideal fistulae, and 3 had multiple fistulae (between the ileum and the bladder, vagina, skin, colon, and sigma). Seven of these patients also had an abscess: mesenteric in 5, pelvic in 1, and near the bladder in 1.
Fifty-two patients had already undergone 1 (n = 30), 2 (n = 12), 3 (n = 4), 4 (n = 5), or 5 (n = 1) operations for CD. Appendectomy had been previously performed in 17 patients. A total of 94 SSISs had been performed, as 2 SSISs were carried out simultaneously during the same operation in 3 patients. The clinical characteristics of the patients are shown in Table 1.
The mean extent of diseased bowel was 69.97 cm (range, 20-150 cm). Strictures were classified as short (≤6 cm in length) or long (>6 cm in length). The strictures involved in the SSIS were long and continuous in 18 patients, short and multiple in 54, and both short and long in 19. The median length of the long continuous strictures was 15 cm (range, 7-80 cm) and 2.5 cm for the short ones (range, 0.5-6 cm). The median thickness of the intestinal wall at the section where the SSIS was performed was 8.5 mm (range, 5-16 mm). The mean length of residual small bowel was 263.53 cm (range, 120-450 cm).
Side-to-side isoperistaltic strictureplasty was performed with the technique described by Michelassi in 1996 in 69 patients, whereas we used the previously described personal technique in the remaining 22 patients who had an ileal relapse at a previous ileo-colic anastomosis.2,3
Side-to-side isoperistaltic strictureplasty was performed at the level of the duodenum-jejunum in 5 patients, at the jejuno-ileum in 40 patients, and on the distal ileum-colon in 46 patients.
The median length of the diseased bowel where the SSIS was performed was 55 cm (range, 10-140 cm), and the median length of the SSIS was 28.41 cm (range, 8-72 cm). A synchronous intestinal resection was necessary in 41 patients; in 28 of these, the diseased bowel resected was located in the middle of the tract where SSIS had to be performed with a mean length of 28.5 cm (range, 3-92 cm). Resection was performed at different disease locations cranially or distally from the SSIS site in the remaining 12 patients.
A total of 41 patients had undergone additional SXPLs (16 singles and 25 multiples, up to a maximum of 11). The fistulizing diseased bowel was resected in 6, whereas the fistula was sutured without resection in 5 patients. Abdominal abscess was drained intraoperatively in 4 of 7 patients, whereas the abscesses were included in the resection in the remaining 3. Side-to-side isoperistaltic strictureplasty was performed together with appendectomy in 12 cases, cholecystectomy in 2 cases, hemicolectomy in 2 cases, and a subtotal colectomy in 1 case. Every patient was biopsied to exclude the presence of neoplastic tissue at the SSIS site.
At discharge, 5-amino-salicylic acid (5-ASA) by mouth was prescribed, while corticosteroid therapy administered preoperatively was gradually decreased. No prophylactic immunosuppressive or biological therapy were scheduled in these patients.
Patients were scheduled for outpatient clinical assessment 1, 3, 6, and 12 months and then every 6 months following surgery to evaluate the symptoms, bowel habit, body weight, and results of laboratory examinations and imaging studies performed.
We considered CD recurrence to be the relapse of disease symptoms in the presence of radiologically and/or endoscopically confirmed stricturing and/or inflammatory lesions requiring medical treatment or surgery.
Statistical analysis was carried out using the STATGRAPHICS Centurion XV statistical package (StatPoint Inc) and the use of GRAPHPAD Prism 5 software. Differences in the distribution of each variable among patients in the recurrence and nonrecurrence categories were evaluated using the χ2 test or t test, as appropriate.
The recurrence-free curve was estimated using Kaplan-Meier analysis of different outcomes. Patients were stratified into cohorts in relation to the considered categorical variables and data were compared using the Mantel-Cox log-rank test. Cox proportional hazard regression analysis was used to set up a predictive model simultaneously exploring the effects of all independent variables on a dichotomous outcome recurrence in relation to time (dependent variable values: censored/complete). The quantitative variables were transformed into dichotomous categorical ones, and hazard ratios (HRs) and 95% CIs were calculated using standard methods.
The univariate analysis included the following independent variables: patient factors (sex, age at diagnosis and at surgery, smoking habits, number of cigarettes per day, years of smoking, and family history of inflammatory bowel disease [IBD]); disease factors (time between diagnosis and first surgery, number of previous surgical procedures, presence of fistulae and/or abscesses, and presence of associated perianal diseases); surgery factors (technique, number of strictures in SSIS, number of SXPLs, final length of the SSIS, disease location, and associated appendectomy); pharmacological therapy factors (presurgical and postsurgical therapy); and perioperative factors (blood transfusions and total parenteral nutrition). Other preoperative factors included the presence of anemia, increase in inflammatory indexes, and malnutrition.
All the independent variables of the univariate analysis were forced in the Cox proportional hazard model with backward elimination applied to select the most promising subset of predictors by multiple regression.
None of the 91 patients died during surgery but 24 experienced postoperative complications. Four patients had an enterorrhagia between the third and eighth postoperative days treated with medical therapy and 1 developed a hematoma in the rectal abdominal muscles that was drained percutaneously. One patient had anastomotic dehiscence of the SSIS; another patient had dehiscence of associated restrictureplasty (Heineke-Mikulicz [HM]) SXPL. Both patients were reoperated on the second and seventh postoperative days with suture of the dehiscence and peritoneal toilette, respectively. One patient had dehiscence of 2 associated short SXPLs with peritonitis requiring ileal resection and temporary terminal ileostomy. One patient had a pelvic abscess treated with computed tomography–guided drainage.
Ten patients had a bloodstream infection from the central venous catheter, 3 patients developed suppuration of the surgical wound, and 2 cases had prolonged postoperative paralytic ileum.
Patients’ intestinal activity restarted between the third and 13th postoperative days (mean, day 5), and nutrition was resumed gradually from postoperative day 2.
The median postoperative stay was 12 days (range, 7-87 days); 13 patients needed 1 or more blood transfusions due to serious anemia, 6 of whom during the preoperative period, 6 during the postoperative period, and 1 during surgery.
After surgery, all patients had resolution of obstructive symptoms. During the third month after surgery, 85% of patients (n = 77) had increased body weight, and 68% (n = 62) experienced normalization of inflammation indexes. Obstructive symptoms relapsed within 2 weeks after discharge in only 1 patient who was rehospitalized and the symptoms were conservatively treated with medical therapy.
Seven of 91 patients (7.7%) died owing to causes not related to CD during the follow-up.
Eight patients were lost to follow-up. The remaining 83 patients were regularly followed up with laboratory and instrumental tests. The mean duration of follow-up was 85.97 months (range, 6-180 months).
During long-term follow-up, most patients reported 1 to 2 semisolid evacuations per day. Five patients (without recurrence) had 3 to 4 evacuations per day with mild abdominal pain.
Eleven patients were treated with 5-ASA, 23 with 5-ASA associated with corticosteroids, 19 with immunosuppressants, 20 with biological drugs (anti–tumor necrosis factor α), and 10 patients with a combination of all the previous drugs. A total of 47 patients periodically took vitamins, folic acid, and probiotics. Four patients developed iron-deficiency anemia.
Among the 83 patients regularly followed up, 37 (44.6%) had a clinical recurrence at a mean (SD) of 55.46 (36.79) months after surgery (range, 9-140 months). Global recurrence was controlled with medical therapy in 15 patients (40.5%) and surgically (reoperation) in 22 (59.5%).
The recurrent disease was outside of the SSIS site in 13 patients (15.7%), being medical in 6 cases and surgical in 7 patients.
In the other 24 patients (28.9%), recurrence occurred after a mean of 48.25 months (range, 9-140 months) inside the SSIS site, being medical in 9 and surgical in 15 patients. The surgical recurrence affected the SSIS body in 8 patients (in 2 with long stricture and in 6 with short ones), the inlet in 4 patients, and the outlet in 3 cases; it was treated with SSIS removal in 2 cases (both affected by more than 2 long strictures at SSIS body) and strictureplasty (HM) of the stricture in 13 patients.
Twelve patients (50%) of 24 experienced recurrence after more than 5 years, during which they experienced well-being and good condition.
The postoperative global recurrence and the surgical or medical recurrence at the SSIS site were analyzed (Table 2).
The frequency of global recurrence was lower in former smokers compared with active smokers or those who had never smoked (P < .01). Those with recurrence also had a significantly shorter time between diagnosis and surgery (mean [SD], 55.46 [36.79] vs 122.30 [17.20] months; P = .02).
Patients with recurrence received postoperative therapy that was significantly different; in particular, none of them had received mesalazine alone while most of them had therapies with corticosteroids, immunosuppressant drugs, or biological therapies.
Moreover, recurrent patients had a family history of CD or IBD at higher frequency compared with the no recurrence cases (32.4% vs 13%) but this difference did not reach statistical significance (P = .06).
Instead, patients with surgical recurrence in SSIS had a significantly shorter time between diagnosis and surgery (P = .002).
The estimation of the actuarial curves according to the Kaplan-Meier method was calculated with the log-rank test to evaluate the relationship between the examined variables and postoperative recurrence. The global risk for recurrence was moderately lower in patients without family history of CD (HR, 0.45; P = .06) and higher for those with diseases located in the jejunum-ileum compared with the distal ileum-colon localizations (HR, 2.45; 95% CI, 1.15-5.22; P = .002) (Figure, A; Table 3). Moreover, former smokers had a significantly lower risk for recurrence compared with smokers and those patients who had never smoked (Figure, A). When analyzing the risk factors associated with surgical recurrence in SSIS alone, we found a reduced risk in patients without a family history of CD (HR, 0.13; P = .006) and with a disease localization in the distal part of the small bowel (HR, 0.23; P = .002) (Figure, B).
Multivariate analysis identified age at diagnosis (χ2 = 5.56; P = .02) and at surgery (χ2 = 7.77; P = .005), family history (χ2 = 6.26; P = .01), and smoking habit (χ2 = 10.06; P = .007) as independent risk factors for postoperative recurrence. These risk factors were more relevant in men because in women, the only statistically significant risk factor was age at surgery (Table 4).
Our study indicated that the following factors were related to global recurrence of CD in our patients who had undergone SSIS: smoking habit, a short time between diagnosis and surgery, and postoperative therapy. Moreover, a family history of IBD was a borderline risk factor.
It is of interest that former smokers (patients who had stopped smoking after surgery) had a significantly lower recurrence rate. Cigarette smoking is considered both an etiological risk factor for the disease as well as its recurrence, with studies showing that smokers not only have a higher risk for developing a postoperative recurrence compared with nonsmokers (around 2.5 times more), but also that this risk increases in relation to the number of cigarettes smoked (especially for heavy smokers, >10 cigarettes a day).6-10 Our data suggest that smoking probably plays a decisive role in the development of CD but when these patients quit, they may have significantly reduced the risk for recurrence, as the “aggressiveness” of the disease in their case may have been strongly influenced by smoking. Literature evidence is rare on this point, and the most recent contributions do not seem to show a significant difference between former smokers and nonsmokers.11
Sex, age at diagnosis or surgery, number of previous operations, duration of the disease, localization of the disease and of the SSIS, number of strictures and associated small-bowel resection within the SSIS, other resections, perianal CD, fistulizing disease, anemia, inflammatory index, blood transfusion, and malnutrition or total parenteral nutrition did not seem related to recurrence.
Univariate analysis of patients with medical or surgical recurrence at the SSIS site identified similar risk factors: the time between diagnosis and surgery and the postoperative therapy. Specifically, patients with a surgical recurrence in the SSIS site showed a significantly shorter time between the diagnosis and surgery (mean [SD], 45.64 [13.01] compared with 122.3 [17.20] months for patients without recurrence). This finding was not unexpected because patients undergoing surgery with a short time between diagnosis and surgery usually have a more aggressive form of the disease.
We observed that the risk for relapse was significantly lower in patients without a family history of IBD; 35.7% of the SSIS patients who had a surgical recurrence also had a family history of IBD compared with 13% of the no recurrence cases.
Another risk factor seems to be the disease location, with an increased recurrence in patients with the disease located in the upper part of the intestine (2.5-fold higher) than those with diseases located in the distal ileum and colon.12-14
A previous retrospective study by Greenstein et al15 in 2009 found that the number of strictures and SXPLs was associated with postoperative recurrence risk. However, the authors examined only HM and/or Finney SXPLs. Our data suggest that the situation is different for SSIS because a higher number of strictures found on the SSIS segment (11-20 stenosis) is not related to a higher rate of recurrence.
The association between age at diagnosis or at first surgery and recurrence risk is still controversial16,17; our data indicated that the age at diagnosis or at surgery are independent risk factors only in men. Our results showed a lower percentage of global recurrence and surgical recurrence in patients treated with 5-ASA alone after surgery compared with those treated with corticosteroids, immunosuppressants, or anti–tumor necrosis factor α drugs. This is likely because 5-ASA alone was administered only in patients with no symptoms or mild disease, while immunosuppressive or biological therapies were administered in the presence of clinical relapse. However, 10 patients treated with a combination of all the drugs did not show any recurrence.
Many prospective studies18-21 or meta-analysis22 evaluating global CD recurrence after ileo-cecal resection are present but clinical trials evaluating the efficacy of prophylactic medical therapies in reducing postoperative recurrence after SSIS are still lacking.
Only a randomized study by Ardizzone et al23 evaluated, after conservative surgery (SXPL or minimal resection), the postoperative prophylactic use of azathioprine, showing after 24 months of follow-up that there was no significant difference in terms of clinical and surgical recurrence compared with the control group treated with 5-ASA.
Only rarely is there described an association between cancer and SXPL24-26 and in the present study, we had no case of intestinal cancer in SSIS, even after a long-term follow-up. However, careful surveillance of these patients is desirable.
Side-to-side isoperistaltic strictureplasty can be considered a useful alternative to resection, especially in extended or multioperated on patients with CD who might have short-bowel syndrome.
Currently, to our knowledge, there are no standard indications for SSIS and, therefore, the experience of the surgeon plays a fundamental role. Studies carried out so far, aiming at identifying specific risk factors associated with recurrences in patients treated with SSIS, have not provided homogeneous results.5,6,17,26,27 Furthermore, these few studies, owing to the relatively recent introduction of SSIS, included small cohorts of patients. Our relatively long experience with this technique, performed for more than 15 years, provides us with the opportunity of evaluating possible recurrence risk factors over long-term follow-up.
Furthermore, our data are discordant regarding previous experiences28,29 about the site of the surgical recurrence in SSIS. In 8 of our patients (53.7%), the recurrence affected the body of the SSIS, which was successfully treated with HM strictureplasty in 75% of them.
Moreover, we followed up by endoscopy the patients with an SSIS on the neoterminal ileum and we observed normal or mild erythematous mucosa in most of these patients. With time only, 27.3% (6 of 22 patients who had undergone SSIS on the neoterminal ileum) of them developed mucosal edema with aphthae or ulcers and subsequently stricture that required surgical treatment only in 2 cases.
A possible limitation of the study is the lack of uniformity in the postoperative medical therapy.
Our work shows that SSIS, in the short-term, leads to a resolution of symptoms in more than 90% of cases; only 5 patients (6%) all with a strictureplasted small bowel wider than 70 cm required a longer follow-up for resolution of anemia. We have noticed that some factors influence the recurrence of the disease, including the habit of cigarette smoking and the age at diagnosis, especially in men. A family history of IBD is another positive correlation with recurrence risk, as is the location of the disease in the upper part of the bowel. The other examined factors did not show any significance, but in some cases, this could depend on the relatively small size of our sample.
Corresponding Author: Marilena Fazi, MD, Department of Surgery and Translational Medicine, University of Florence, Largo Brambilla 3, 50134 Florence, Italy (firstname.lastname@example.org).
Accepted for Publication: September 16, 2015.
Published Online: December 30, 2015. doi:10.1001/jamasurg.2015.4552.
Author Contributions: Dr Fazi had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Fazi, Giudici, Pronestì, Tonelli.
Acquisition, analysis, or interpretation of data: Fazi, Giudici, Luceri, Pronestì.
Drafting of the manuscript: Fazi, Giudici, Pronestì.
Critical revision of the manuscript for important intellectual content: Fazi, Giudici, Luceri, Tonelli.
Statistical analysis: Luceri, Pronestì.
Obtained funding: Fazi.
Administrative, technical, or material support: Fazi, Giudici.
Study supervision: Fazi, Giudici, Tonelli.
Conflict of Interest Disclosures: None reported.
Create a personal account or sign in to: