The American Journal of Surgery (2014) 208, 591-596
Association of Women Surgeons
Factors influencing disease recurrence after ileocolic resection in adult and pediatric onset Crohn’s disease Iuliana D. Bobanga, M.D.*, Shiyu Bai, B.S., Marco A. Swanson, B.A., Bradley J. Champagne, M.D., Harry J. Reynolds, M.D., Conor P. Delaney, M.D. Ph.D., Edward M. Barksdale, Jr, M.D., Sharon L. Stein, M.D. Division of Colon and Rectal Surgery, Department of Surgery, Case Western/University Hospital Case Medical Center, 11100 Euclid Ave, LKS 5047, Cleveland, OH 44106, USA KEYWORDS: Ileocecectomy; Crohn’s disease; Pediatric; Recurrence; Immunoprophylaxis
Abstract BACKGROUND: Factors influencing recurrence of ileocecal Crohn’s disease (CD) after surgical resection may differ between adolescents and adults. METHODS: CD patients who underwent ileocecectomy were retrospectively divided into pediatric onset (age at diagnosis % 16 years, n 5 34) and adult onset (.16, n 5 108) patients to evaluate differences in risks of endoscopic and clinical recurrence. RESULTS: In 142 patients, rates of any recurrence, endoscopic recurrence, and clinical recurrence at 5 years were 78%, 88%, and 65%, respectively. Risks of recurrence were similar between groups. Younger patients were more likely to be on immunologics preoperatively and more likely to be started on immunoprophylaxis postoperatively. Immediate postoperative prophylaxis was predictive of delayed clinical recurrence only in the older group. CONCLUSIONS: Despite increased preoperative and postoperative immunoprophylaxis in younger patients, recurrence rates of CD after ileocecectomy do not differ between these groups. Immediate postoperative prophylaxis was predictive of delayed clinical recurrence only in patients with adult onset CD. Ó 2014 Elsevier Inc. All rights reserved.
Crohn’s disease (CD) is a chronic inflammatory gastrointestinal disorder with an unpredictable clinical course and a high rate of recurrence. Despite increased use of
The authors declare no conflicts of interest. * Corresponding author. Tel.: 11-216-844-3027; fax: 11-216-8448201. E-mail address: [email protected] Manuscript received February 2, 2014; revised manuscript May 2, 2014 0002-9610/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjsurg.2014.06.008
immunomodulators and biologic therapy, surgical resection is still required in approximately 70% to 80% of patients with CD, and up to 70% of patients will require a second operation.1,2 Subclinical endoscopic recurrence precedes the development of symptoms. One year after surgical resection, more than 70% of patients have evidence of anastomotic endoscopic recurrence.3,4 Thirty percent of patients progress to symptomatic recurrence at 3 years, and 60% have endoscopic recurrence by 5 years.2,3 The pathophysiology of recurrence is poorly understood, and importantly,
592 there is little consensus on the best algorithm to prevent recurrence. While several small randomized clinical trials have identified potential treatment options, details of patient selection and timing of prophylaxis remain elusive.5 Several factors have been evaluated as potential risk factors for postoperative recurrence. Smoking, prior resection, genetic markers, perforating disease type, ileocolonic location, and short disease duration before first surgery may increase risk of postoperative recurrence.1,6,7 Some studies indicate that a younger age of disease onset is associated with an increased incidence of surgical recurrence.8–10 Additional studies demonstrate a more aggressive phenotype of CD in young patients.11,12 Studies evaluating differences in risk factors associated with postsurgical recurrence in pediatric compared with adult onset CD are sparse.13,14 In this study, we review the records of patients with pediatric onset CD (age at diagnosis % 16 years) compared with older patients with CD (age at diagnosis . 16 years) who underwent ileocecal resection. The purpose of this study was to investigate whether predictors of postoperative disease recurrence in patients undergoing ileocecectomy for CD differ with age. We hypothesized that there are distinct differences between early-onset and adult-onset CD and that different risk factors are associated with recurrence.
The American Journal of Surgery, Vol 208, No 4, October 2014 short- and long-term outcomes were recorded. Patients who underwent surgery on admission or during an index admission were considered ‘‘emergent’’ or ‘‘urgent’’ surgery, respectively. Study data were collected and managed using REDCap electronic data capture tools hosted at University Hospital Case Medical Center.19
Study outcomes The primary outcome of interest was time to any recurrence of CD at the neoterminal ileum, which was defined as endoscopic, clinical, or surgical recurrence. The secondary endpoints were endoscopic, clinical, and surgical recurrence at the neoterminal ileum. Endoscopic recurrence was demonstrated by Rutgeerts score of i2 or greater on postoperative colonoscopy.3 Patients requiring initiation or augmentation of medical therapy, hospitalization, or other interventions to control symptoms of CD were categorized as having a clinical recurrence at the earliest date of interventions. Patients who required reoperation at the neoterminal ileum were considered to have a surgical recurrence. Time to recurrence was measured from the date of surgery to the earliest date of recurrence, whether clinical, endoscopic, or surgical.
Statistical analysis
Methods Study population After obtaining institutional review board approval, patients who underwent ileocecectomy or right hemicolectomy for CD between January 2004 and October 2013 at Case Western University Hospitals Case Medical Center and Rainbow Babies and Children’s Hospital in Cleveland, Ohio, were identified from a prospectively maintained database of colon and rectal surgery patients and from the hospital medical records department. A retrospective chart review was performed on all patients. Patients were excluded if pathology report was negative for CD. Patients with concomitant procedures (such as additional bowel resection, stricturoplasty, treatment of perianal disease) were included provided that the main focus of the surgery was active terminal ileum or ileocecal CD. Our retrospective chart review included surgical and gastroenterology office notes, surgical and endoscopic operative reports, hospital admission and discharge notes, pathology reports, and radiology studies. Patients were divided into 2 groups based on age at diagnosis of CD: pediatric onset group (age at diagnosis % 16 years) and adult onset group (.16 years). This division was based on prior studies that have shown distinct physiologic characteristics between pediatric and adult onset CD11,15,16 and use the age of 16 as a cutoff to differentiate early and adult onset inflammatory bowel disease.17,18 Demographics, operative and perioperative data,
Statistical analysis was performed using SPSS version 21.0 (SPSS, Inc, Chicago, IL). Descriptive statistics were used to generate frequencies and percentages for categorical variables and mean with standard deviations for continuous variables. Categorical variables were compared using the chi-square test and continuous variables were compared using independent-sample t test. To evaluate factors associated with recurrence, we developed a Cox proportional hazards model of both endoscopic and clinical recurrence and performed univariate and multivariate analyses separately for younger and older adults. Crude and adjusted hazards ratios (HRs) with 95% confidence interval and P values are reported. All variables with a P value of less than .150 on univariate analysis were included in the multivariate model. Patients who were lost to follow-up were censored at the time of last follow-up. Time to recurrence between the 2 groups was compared using the Kaplan–Meier method and the log-rank test. A P value of less than or equal to .05 was considered statistically significant throughout the analysis.
Results One hundred and sixty-three patients underwent ileocecectomy or right hemicolectomy for CD between June 2004 and October 2013. Three patients were excluded because of cecal cancer, 4 were excluded because the pathology report was not consistent with CD, 6 were excluded because the date of diagnosis could not be obtained, and 8 were
I.D. Bobanga et al.
Crohn’s recurrence after ileocecectomy
excluded because they had a large operation for multifocal CD and ileocecectomy was a minor part of the operation. Our final cohort of 142 patients (48% male, mean age at diagnosis 28 years, range 5 to 73) included 34 patients 16 years of age or younger at diagnosis (pediatric onset group) and 108 patients greater than 17 years of age at diagnosis (adult onset group). Fifty-four patients had a prior bowel resection for CD, 46 of which had a previous ileocecectomy (8 in the pediatric onset group). The 2 cohorts received similar preoperative medications before surgery with the exception of immunomodulators which were more common in the pediatric onset group (65% vs 21%, P , .001). Treatments were similar in terms of surgical approach, incidence of diverting, or permanent stoma. Groups were also similar with respect to preoperative disease duration, prior resection, multifocal disease, pathologically positive margins, length of stay, incidence of postoperative complication, and readmission within 30 days (Table 1). Length of follow-up was 2.4 years overall and did not differ significantly between the 2 groups (Table 2). Patients in the pediatric onset group were placed on postoperative immunoprophylaxis earlier than adults (.9 vs 5.1 months, P 5 .002), and often within 4 weeks of surgery (53% vs 22%, P 5 .002). Forty-seven percent of the total cohort underwent a postoperative endoscopy, and of those, 72% had endoscopic recurrence at the neoterminal ileum, while 32% Table 1
593 had clinical recurrence (Table 2). Adjusted for length of follow-up, the overall recurrence rate (including endoscopic, clinical, or surgical) was 37% at 1 year, 65% at 3 years, and 78% at 5 years. Time to any ileocolic recurrence did not differ significantly between the 2 groups. The censored endoscopic recurrence rate was 45% at 1 year, 73% at 3 years, and 88% at 5 years. Clinical recurrence rates were 28% at 1 year, 52% at 3 years, and 65% at 5 years. The log-rank test comparing time with endoscopic recurrence based on age did not find a statistically significant difference between the younger and older adult groups (P 5 .918). Similarly, there was no difference in time to clinical recurrence (P 5 .914) or any recurrence (P 5 .972) between the 2 age groups (Fig. 1). On univariate and multivariate analysis of time to endoscopic or clinical recurrence (Tables 3 and 4), there were no significant predictors of decreased time to endoscopic recurrence or clinical recurrence in the pediatric onset group. In the adult onset group, positive margins on pathology was predictive of a shorter time to endoscopic recurrence (HR 3.3, P 5 .036), while initiation of postoperative prophylaxis within 4 weeks of surgery was correlated with longer time to clinical recurrence in multivariate analysis (HR .3, P 5 .043). Preoperative disease duration, presence of multifocal disease, or multiple areas of resection were not associated with disease recurrence in either group.
Baseline demographic, clinical, and perioperative characteristics
Male Mean age at diagnosis (range) Mean age at surgery (range) Preoperative disease duration (years) Prior resection Preoperative medications Steroids Aminosalicylates Immunomodulators Biologics Antibiotics Urgent surgery Emergent surgery Laparoscopic procedure Diverting stoma Multifocal disease Positive margins (n 5 84) Length of TI resected (cm) Length of right colon resected (cm) Total bowel resected (cm) Length of stay (days) Readmission within 30 days Postoperative complications
All patients (n 5 142)
Age at diagnosis % 16 years (n 5 34)
Age at diagnosis . 16 years (n 5 108)
P value
68 (48) 28 (5–73) 39 (13–86) 12 6 10 54 (38) 119 (84) 85 (60) 43 (30) 44 (31) 28 (20) 37 (26) 35 (25) 11 (8) 81 (57) 24 (17) 46 (32) 29 (35) 23 6 17 10 6 7 36 6 20 5.9 6 3.3 15 (11) 51 (36)
19 (56) 13 (5–17) 23 (13–60) 9.7 6 8.6 11 (32) 31 (91) 23 (68) 14 (41) 22 (65) 8 (24) 8 (24) 5 (15) 4 (12) 21 (62) 5 (15) 12 (35) 7 (24) 15 6 8 12 6 9 28 6 12 6.2 6 3.5 2 (6) 11 (32)
49 (45) 32 (17–73) 44 (17–86) 12.2 6 10.5 43 (40) 88 (81) 62 (58) 29 (27) 22 (21) 20 (19) 29 (27) 30 (28) 7 (7) 60 (56) 19 (18) 34 (32) 22 (40) 26 619 967 38 6 22 5.8 6 3.3 13 (12) 40 (37)
.328 ,.001 ,.001 .193 .544 .285 .322 .135 ,.001 .621 .824 .171 .461 .557 .797 .680 .571 ,.001 .093 ,.001 .609 .522 .685
Values are given as mean 6 standard deviation or number (percentage) unless otherwise specified. TI 5 terminal ileum.
594
The American Journal of Surgery, Vol 208, No 4, October 2014
Table 2
Postoperative outcomes and recurrence characteristics
Length of follow-up (years) Postoperative time to medication prophylaxis (months) Medication restarted within 4 weeks of surgery Patients with postoperative endoscopy Time to postoperative endoscopy (months) Postoperative time to any recurrence (years) Any recurrent TI disease Endoscopic recurrence (n 5 67) Clinical recurrence Surgical recurrence
All patients (n 5 142)
Age at diagnosis % 16 years (n 5 34)
Age at diagnosis . 16 years (n 5 108)
P value
2.4 6 2.5 3.9 6 8.6
3.1 6 2.7 .9 6 1.5
2.2 6 2.4 5.1 6 9.8
.063 .002
42 (30) 67 (47) 11.5 6 7.8 1.3 6 1.6 61 (43) 48 (72) 45 (32) 7 (5)
18 (53) 15 (50) 10.7 6 6.3 1.6 62 17 (50) 13 (87) 13 (38) 2 (6)
24 (22) 50 (45) 11.2 6 7.2 1.2 6 1.5 44 (41) 35 (70) 32 (30) 5 (5)
.002 .574 .793 .212 .427 .760 .400 .673
Values are given as mean 6 standard deviation or number (percentage). TI 5 terminal ileum.
Comments Predicting postoperative recurrence of CD after ileocecal resection remains a major challenge. Because of the variable course of CD, it has proven difficult to establish strict postoperative management guidelines to prevent recurrence. In fact, there are 3 different algorithms described by De Cruz et al1 as ‘‘valid approaches’’ in the management of postoperative CD: (1) postoperative drug treatment stratified according to the risk of recurrence; (2) aggressive approach of placing all patients on biologics postoperatively; and (3) a tailored prophylactic approach based on findings at
Figure 1 Time to recurrence based on age at diagnosis %16 or .16 years.
postoperative endoscopy.1 Prospective studies to validate which approach should be used are lacking. The association between early postoperative immunoprophylaxis and a decreased association with recurrence is an important finding of our study in the era of unclear postoperative management algorithms to prevent recurrence. In the only other study to evaluate the timing of prophylaxis in preventing recurrence after surgery for ileocecal CD, Bordeianou et al5 reported no difference between immediate (within 1 month of surgery) and tailored prophylaxis (after first endoscopy). Our results may differ because our cohort included patients who had previous resections for CD, while the study by Bordeianou et al only included patients undergoing their first operation for CD. Further studies to investigate this important clinical question should be carried out, preferably as a large prospective randomized trial. Studies in adults have established multiple predictors of disease recurrence after surgical resection, which include smoking, history of prior resection, genetic markers, perforating disease type, ileocolonic location, and short preoperative disease duration.1,6,7 Younger age at disease onset has also been associated with higher risk of recurrence, but results have been less conclusive because of the potential confounding effect of longer follow-up among younger patients.1,8–10 A fundamental question of inflammatory bowel disease remains unanswered: whether pediatric onset CD has a different pathogenesis leading to a different phenotype or whether it is the same disease process but merely occurring at an earlier time point.11,16 Some experts believe that genetic contribution has a greater role than environmental factors in pediatric and adolescent CD.16 Several consistent differences between pediatric onset CD and its adult counterpart, including disease type, disease location, disease behavior, sex preponderance, and genetically attributable risks, support the theory that pediatric onset CD represents a distinct disease phenotype.11,16,20 Thus, one would expect
I.D. Bobanga et al.
Crohn’s recurrence after ileocecectomy
595
Table 3 Cox proportional hazards regression of possible predictors of endoscopic recurrence in two age groups with univariate and multivariate analysis Age at diagnosis % 16 years (n 5 15)
Age at diagnosis . 16 years (n 5 50)
Univariate analysis Predictor
Crude HR
95% CI
P value
Crude HR
95% CI
P value
Age at diagnosis Age at surgery Preoperative disease duration Emergent procedure Prophylaxis within 4 weeks Positive margins
.9 1.1 1.1 1.4 .6 .6 Multivariate analysis
.73–1.21 .96–1.17 .97–1.25 .42–4.77 .01–.59 .10–3.39
.626 .277 .133 .573 .016 .558
1.0 1.0 1.0 2.2 .4 3.9
.99–1.03 .99–1.03 .99–1.06 .83–5.67 .16–1.13 1.46–10.37
.992 .496 .183 .116 .086 .007
Adjusted HR
95% CI
P value
1.1 .5 3.3
.27–4.20 .13–1.82 1.08–10.14
.937 .288 .036
95% CI
P value
.4
.74–1.14
.439
.9
.01–5.303
.917
Adjusted HR Preoperative disease duration Emergent procedure Prophylaxis within 4 weeks Positive margins
CI 5 confidence interval; HR 5 hazards ratio.
to also find differences in factors predicting disease recurrence in pediatric and adult onset CD. Few studies have evaluated risk factors for postoperative recurrence in younger patients. Baldassano et al13 found high Pediatric Crohn’s Disease Activity Index, colonic or ileocolonic disease, and preoperative use of 6-mercaptopurine to be associated with high rates of postoperative recurrence in children and Griffiths et al21 also showed similar results. We were unable to find any studies in the literature that directly compared factors that decrease the time to recurrence in children compared with adults. The results of this study show that while the time to any recurrence, endoscopic recurrence, or clinical recurrence is not statistically different between the 2 age groups, there are different predictive factors that increase the risk of
recurrence in pediatric and adult onset CD. After subgroup multivariate analysis, we found that positive margins at ileocecectomy were predictive of endoscopic recurrence only in the adult group. Furthermore, restarting or initiating postoperative prophylactic medications within 4 weeks of surgery was predictive of delayed time to clinical recurrence only in the adult onset group. There were several findings that differentiated the patients with pediatric onset disease from the adult onset patients in our study. The younger patients at diagnosis were more frequently treated with immunomodulators preoperatively, and were restarted on early postoperative immunotherapy more often than older adults. Despite this more aggressive medical management, the incidence and time to postoperative recurrence was not only equal, but
Table 4 Cox proportional hazards regression of possible predictors of clinical recurrence in 2 age groups with univariate and multivariate analysis Age at diagnosis % 16 years (n 5 34)
Age at diagnosis . 16 (n 5 108)
Univariate analysis P value
Crude HR
95% CI
P value
1.1 .87–1.31 1.0 .91–1.04 .9 .86–1.02 .5 .06–3.9 .5 .12–1.81 .6 .12–2.60 Multivariate analysis
.537 .430 .224 .505 .267 .455
1.0 1.0 1.0 3.1 .4 3.4
.96–1.01 .97–1.02 .95–1.01 1.05–8.91 .15–.92 1.42–8.09
.266 .215 .521 .041 .032 .006
Adjusted HR
P value
Adjusted HR
95% CI
P value
Predictor
Crude HR
Age at diagnosis Age at surgery Preoperative disease duration Emergent procedure Prophylaxis within 4 weeks Positive margins
Age at surgery Prophylaxis within 4 weeks Positive margins CI 5 confidence interval; HR 5 hazards ratio.
95% CI
95% CI
1.4 .3 2.0
.35–5.50 .08–.96 .75–5.35
.646 .043 .169
596 also trended toward increased recurrence in the younger group (50% vs 41%). These associations between medical treatment and recurrence suggest a more aggressive disease phenotype in pediatric onset CD. Our results have important implications for the clinician who may choose to advise patients with early onset CD of the more aggressive disease phenotype and increased risk of recurrence after surgical intervention. Further studies should focus on whether a more aggressive surgical approach is warranted in these younger patients. Ideally, the timing and type of pre- and postoperative medication prophylaxis to prevent recurrence in pediatric onset CD should be evaluated in a prospective multi-institutional setting. There are several limitations to our study. Retrospective studies are limited in their ability to show associations and not causative factors. Our study size limited our ability to explore subgroups and trends that may provide important differentiations between unique groups with CD. Another limitation in CD studies is the difficulty in accurately discerning the time of clinical recurrence because of the variety of presentations. While endoscopic and surgical recurrence can be standardized based on objective criteria, the definition of clinical recurrence varies with provider assessment of patient symptoms and is not always accompanied by an objective test. Finally, our study is limited to a single tertiary care multi-hospital institution and may be difficult to generalize the findings to other hospital settings. In conclusion, our retrospective analysis comparing risk factors for recurrence between patients with pediatric and adult onset CD who undergo ileocecal resection found that the rate of recurrence is not different between the 2 age groups. Despite more aggressive management with immunologics preoperatively and earlier resumption of postoperative immunoprophylaxis, younger patients are at equal risk of recurrence. Immediate postoperative immunoprophylaxis was associated with decreased time to clinical recurrence only in patients with adult onset CD. Further studies designed to analyze recurrence of CD in younger patients is warranted to help identify best practice for these patients.
References 1. De Cruz P, Kamm MA, Prideaux L, et al. Postoperative recurrent luminal Crohn’s disease: a systematic review. Inflamm Bowel Dis 2012;18:758–77.
The American Journal of Surgery, Vol 208, No 4, October 2014 2. Sachar DB. The problem of postoperative recurrence of Crohn’s disease. Med Clin North Am 1990;74:183–8. 3. Rutgeerts P, Geboes K, Vantrappen G, et al. Predictability of the postoperative course of Crohn’s disease. Gastroenterology 1990;99: 956–63. 4. Olaison G, Smedh K, Sjo¨dahl R. Natural course of Crohn’s disease after ileocolic resection: endoscopically visualised ileal ulcers preceding symptoms. Gut 1992;33:331–5. 5. Bordeianou L, Stein SL, Ho VP, et al. Immediate versus tailored prophylaxis to prevent symptomatic recurrences after surgery for ileocecal Crohn’s disease? Surgery 2011;149:72–8. 6. Yamamoto T. Factors affecting recurrence after surgery for Crohn’s disease. World J Gastroenterol 2005;11:3971–9. 7. Borley NR, Mortensen NJ, Jewell DP. Preventing postoperative recurrence of Crohn’s disease. Br J Surg 1997;84:1493–502. 8. Scarpa M, Angriman I, Barollo M, et al. Risk factors for recurrence of stenosis in Crohn’s disease. Acta Biomed 2003;74(Suppl 2):80–3. 9. Ryan WR, Allan RN, Yamamoto T, et al. Crohn’s disease patients who quit smoking have a reduced risk of reoperation for recurrence. Am J Surg 2004;187:219–25. 10. Post S, Herfarth C, Bo¨hm E, et al. The impact of disease pattern, surgical management, and individual surgeons on the risk for relaparotomy for recurrent Crohn’s disease. Ann Surg 1996;223:253–60. 11. Sauer CG, Kugathasan S. Pediatric inflammatory bowel disease: highlighting pediatric differences in IBD. Gastroenterol Clin North Am 2009;38:611–28. 12. Beaugerie L, Seksik P, Nion-Larmurier I, et al. Predictors of Crohn’s disease. Gastroenterology 2006;130:650–6. 13. Baldassano RN, Han PD, Jeshion WC, et al. Pediatric Crohn’s disease: risk factors for postoperative recurrence. Am J Gastroenterol 2001;96: 2169–76. 14. Hojsak I, Pavic AM, Misak Z, et al. Risk factors for relapse and surgery rate in children with Crohn’s disease. Eur J Pediatr 2014;173: 617–21. 15. Lacher M, Helmbrecht J, Schroepf S, et al. NOD2 mutations predict the risk for surgery in pediatric-onset Crohn’s disease. J Pediatr Surg 2010;45:1591–7. 16. Ruel J, Ruane D, Mehandru S, et al. IBD across the age spectrum-is it the same disease? Nat Rev Gastroenterol Hepatol 2014;11:88–98. 17. Law ST, Li KK. Age-related differences in the clinical course of Crohns disease in an Asian population: a retrospective cohort review. Indian Pediatr 2013;50:1148–52. 18. Boualit M, Salleron J, Turck D, et al. Long-term outcome after first intestinal resection in pediatric-onset Crohn’s disease: a populationbased study. Inflamm Bowel Dis 2013;19:7–14. 19. Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)da metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009;42:377–81. 20. Van Limbergen J, Russell RK, Drummond HE, et al. Definition of phenotypic characteristics of childhood-onset inflammatory bowel disease. Gastroenterology 2008;135:1114–22. 21. Griffiths AM, Wesson DE, Shandling B, et al. Factors influencing postoperative recurrence of Crohn’s disease in childhood. Gut 1991;32: 491–5.
Association of Women Surgeons
Factors influencing disease recurrence after ileocolic resection in adult and pediatric onset Crohn’s disease Iuliana D. Bobanga, M.D.*, Shiyu Bai, B.S., Marco A. Swanson, B.A., Bradley J. Champagne, M.D., Harry J. Reynolds, M.D., Conor P. Delaney, M.D. Ph.D., Edward M. Barksdale, Jr, M.D., Sharon L. Stein, M.D. Division of Colon and Rectal Surgery, Department of Surgery, Case Western/University Hospital Case Medical Center, 11100 Euclid Ave, LKS 5047, Cleveland, OH 44106, USA KEYWORDS: Ileocecectomy; Crohn’s disease; Pediatric; Recurrence; Immunoprophylaxis
Abstract BACKGROUND: Factors influencing recurrence of ileocecal Crohn’s disease (CD) after surgical resection may differ between adolescents and adults. METHODS: CD patients who underwent ileocecectomy were retrospectively divided into pediatric onset (age at diagnosis % 16 years, n 5 34) and adult onset (.16, n 5 108) patients to evaluate differences in risks of endoscopic and clinical recurrence. RESULTS: In 142 patients, rates of any recurrence, endoscopic recurrence, and clinical recurrence at 5 years were 78%, 88%, and 65%, respectively. Risks of recurrence were similar between groups. Younger patients were more likely to be on immunologics preoperatively and more likely to be started on immunoprophylaxis postoperatively. Immediate postoperative prophylaxis was predictive of delayed clinical recurrence only in the older group. CONCLUSIONS: Despite increased preoperative and postoperative immunoprophylaxis in younger patients, recurrence rates of CD after ileocecectomy do not differ between these groups. Immediate postoperative prophylaxis was predictive of delayed clinical recurrence only in patients with adult onset CD. Ó 2014 Elsevier Inc. All rights reserved.
Crohn’s disease (CD) is a chronic inflammatory gastrointestinal disorder with an unpredictable clinical course and a high rate of recurrence. Despite increased use of
The authors declare no conflicts of interest. * Corresponding author. Tel.: 11-216-844-3027; fax: 11-216-8448201. E-mail address: [email protected] Manuscript received February 2, 2014; revised manuscript May 2, 2014 0002-9610/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjsurg.2014.06.008
immunomodulators and biologic therapy, surgical resection is still required in approximately 70% to 80% of patients with CD, and up to 70% of patients will require a second operation.1,2 Subclinical endoscopic recurrence precedes the development of symptoms. One year after surgical resection, more than 70% of patients have evidence of anastomotic endoscopic recurrence.3,4 Thirty percent of patients progress to symptomatic recurrence at 3 years, and 60% have endoscopic recurrence by 5 years.2,3 The pathophysiology of recurrence is poorly understood, and importantly,
592 there is little consensus on the best algorithm to prevent recurrence. While several small randomized clinical trials have identified potential treatment options, details of patient selection and timing of prophylaxis remain elusive.5 Several factors have been evaluated as potential risk factors for postoperative recurrence. Smoking, prior resection, genetic markers, perforating disease type, ileocolonic location, and short disease duration before first surgery may increase risk of postoperative recurrence.1,6,7 Some studies indicate that a younger age of disease onset is associated with an increased incidence of surgical recurrence.8–10 Additional studies demonstrate a more aggressive phenotype of CD in young patients.11,12 Studies evaluating differences in risk factors associated with postsurgical recurrence in pediatric compared with adult onset CD are sparse.13,14 In this study, we review the records of patients with pediatric onset CD (age at diagnosis % 16 years) compared with older patients with CD (age at diagnosis . 16 years) who underwent ileocecal resection. The purpose of this study was to investigate whether predictors of postoperative disease recurrence in patients undergoing ileocecectomy for CD differ with age. We hypothesized that there are distinct differences between early-onset and adult-onset CD and that different risk factors are associated with recurrence.
The American Journal of Surgery, Vol 208, No 4, October 2014 short- and long-term outcomes were recorded. Patients who underwent surgery on admission or during an index admission were considered ‘‘emergent’’ or ‘‘urgent’’ surgery, respectively. Study data were collected and managed using REDCap electronic data capture tools hosted at University Hospital Case Medical Center.19
Study outcomes The primary outcome of interest was time to any recurrence of CD at the neoterminal ileum, which was defined as endoscopic, clinical, or surgical recurrence. The secondary endpoints were endoscopic, clinical, and surgical recurrence at the neoterminal ileum. Endoscopic recurrence was demonstrated by Rutgeerts score of i2 or greater on postoperative colonoscopy.3 Patients requiring initiation or augmentation of medical therapy, hospitalization, or other interventions to control symptoms of CD were categorized as having a clinical recurrence at the earliest date of interventions. Patients who required reoperation at the neoterminal ileum were considered to have a surgical recurrence. Time to recurrence was measured from the date of surgery to the earliest date of recurrence, whether clinical, endoscopic, or surgical.
Statistical analysis
Methods Study population After obtaining institutional review board approval, patients who underwent ileocecectomy or right hemicolectomy for CD between January 2004 and October 2013 at Case Western University Hospitals Case Medical Center and Rainbow Babies and Children’s Hospital in Cleveland, Ohio, were identified from a prospectively maintained database of colon and rectal surgery patients and from the hospital medical records department. A retrospective chart review was performed on all patients. Patients were excluded if pathology report was negative for CD. Patients with concomitant procedures (such as additional bowel resection, stricturoplasty, treatment of perianal disease) were included provided that the main focus of the surgery was active terminal ileum or ileocecal CD. Our retrospective chart review included surgical and gastroenterology office notes, surgical and endoscopic operative reports, hospital admission and discharge notes, pathology reports, and radiology studies. Patients were divided into 2 groups based on age at diagnosis of CD: pediatric onset group (age at diagnosis % 16 years) and adult onset group (.16 years). This division was based on prior studies that have shown distinct physiologic characteristics between pediatric and adult onset CD11,15,16 and use the age of 16 as a cutoff to differentiate early and adult onset inflammatory bowel disease.17,18 Demographics, operative and perioperative data,
Statistical analysis was performed using SPSS version 21.0 (SPSS, Inc, Chicago, IL). Descriptive statistics were used to generate frequencies and percentages for categorical variables and mean with standard deviations for continuous variables. Categorical variables were compared using the chi-square test and continuous variables were compared using independent-sample t test. To evaluate factors associated with recurrence, we developed a Cox proportional hazards model of both endoscopic and clinical recurrence and performed univariate and multivariate analyses separately for younger and older adults. Crude and adjusted hazards ratios (HRs) with 95% confidence interval and P values are reported. All variables with a P value of less than .150 on univariate analysis were included in the multivariate model. Patients who were lost to follow-up were censored at the time of last follow-up. Time to recurrence between the 2 groups was compared using the Kaplan–Meier method and the log-rank test. A P value of less than or equal to .05 was considered statistically significant throughout the analysis.
Results One hundred and sixty-three patients underwent ileocecectomy or right hemicolectomy for CD between June 2004 and October 2013. Three patients were excluded because of cecal cancer, 4 were excluded because the pathology report was not consistent with CD, 6 were excluded because the date of diagnosis could not be obtained, and 8 were
I.D. Bobanga et al.
Crohn’s recurrence after ileocecectomy
excluded because they had a large operation for multifocal CD and ileocecectomy was a minor part of the operation. Our final cohort of 142 patients (48% male, mean age at diagnosis 28 years, range 5 to 73) included 34 patients 16 years of age or younger at diagnosis (pediatric onset group) and 108 patients greater than 17 years of age at diagnosis (adult onset group). Fifty-four patients had a prior bowel resection for CD, 46 of which had a previous ileocecectomy (8 in the pediatric onset group). The 2 cohorts received similar preoperative medications before surgery with the exception of immunomodulators which were more common in the pediatric onset group (65% vs 21%, P , .001). Treatments were similar in terms of surgical approach, incidence of diverting, or permanent stoma. Groups were also similar with respect to preoperative disease duration, prior resection, multifocal disease, pathologically positive margins, length of stay, incidence of postoperative complication, and readmission within 30 days (Table 1). Length of follow-up was 2.4 years overall and did not differ significantly between the 2 groups (Table 2). Patients in the pediatric onset group were placed on postoperative immunoprophylaxis earlier than adults (.9 vs 5.1 months, P 5 .002), and often within 4 weeks of surgery (53% vs 22%, P 5 .002). Forty-seven percent of the total cohort underwent a postoperative endoscopy, and of those, 72% had endoscopic recurrence at the neoterminal ileum, while 32% Table 1
593 had clinical recurrence (Table 2). Adjusted for length of follow-up, the overall recurrence rate (including endoscopic, clinical, or surgical) was 37% at 1 year, 65% at 3 years, and 78% at 5 years. Time to any ileocolic recurrence did not differ significantly between the 2 groups. The censored endoscopic recurrence rate was 45% at 1 year, 73% at 3 years, and 88% at 5 years. Clinical recurrence rates were 28% at 1 year, 52% at 3 years, and 65% at 5 years. The log-rank test comparing time with endoscopic recurrence based on age did not find a statistically significant difference between the younger and older adult groups (P 5 .918). Similarly, there was no difference in time to clinical recurrence (P 5 .914) or any recurrence (P 5 .972) between the 2 age groups (Fig. 1). On univariate and multivariate analysis of time to endoscopic or clinical recurrence (Tables 3 and 4), there were no significant predictors of decreased time to endoscopic recurrence or clinical recurrence in the pediatric onset group. In the adult onset group, positive margins on pathology was predictive of a shorter time to endoscopic recurrence (HR 3.3, P 5 .036), while initiation of postoperative prophylaxis within 4 weeks of surgery was correlated with longer time to clinical recurrence in multivariate analysis (HR .3, P 5 .043). Preoperative disease duration, presence of multifocal disease, or multiple areas of resection were not associated with disease recurrence in either group.
Baseline demographic, clinical, and perioperative characteristics
Male Mean age at diagnosis (range) Mean age at surgery (range) Preoperative disease duration (years) Prior resection Preoperative medications Steroids Aminosalicylates Immunomodulators Biologics Antibiotics Urgent surgery Emergent surgery Laparoscopic procedure Diverting stoma Multifocal disease Positive margins (n 5 84) Length of TI resected (cm) Length of right colon resected (cm) Total bowel resected (cm) Length of stay (days) Readmission within 30 days Postoperative complications
All patients (n 5 142)
Age at diagnosis % 16 years (n 5 34)
Age at diagnosis . 16 years (n 5 108)
P value
68 (48) 28 (5–73) 39 (13–86) 12 6 10 54 (38) 119 (84) 85 (60) 43 (30) 44 (31) 28 (20) 37 (26) 35 (25) 11 (8) 81 (57) 24 (17) 46 (32) 29 (35) 23 6 17 10 6 7 36 6 20 5.9 6 3.3 15 (11) 51 (36)
19 (56) 13 (5–17) 23 (13–60) 9.7 6 8.6 11 (32) 31 (91) 23 (68) 14 (41) 22 (65) 8 (24) 8 (24) 5 (15) 4 (12) 21 (62) 5 (15) 12 (35) 7 (24) 15 6 8 12 6 9 28 6 12 6.2 6 3.5 2 (6) 11 (32)
49 (45) 32 (17–73) 44 (17–86) 12.2 6 10.5 43 (40) 88 (81) 62 (58) 29 (27) 22 (21) 20 (19) 29 (27) 30 (28) 7 (7) 60 (56) 19 (18) 34 (32) 22 (40) 26 619 967 38 6 22 5.8 6 3.3 13 (12) 40 (37)
.328 ,.001 ,.001 .193 .544 .285 .322 .135 ,.001 .621 .824 .171 .461 .557 .797 .680 .571 ,.001 .093 ,.001 .609 .522 .685
Values are given as mean 6 standard deviation or number (percentage) unless otherwise specified. TI 5 terminal ileum.
594
The American Journal of Surgery, Vol 208, No 4, October 2014
Table 2
Postoperative outcomes and recurrence characteristics
Length of follow-up (years) Postoperative time to medication prophylaxis (months) Medication restarted within 4 weeks of surgery Patients with postoperative endoscopy Time to postoperative endoscopy (months) Postoperative time to any recurrence (years) Any recurrent TI disease Endoscopic recurrence (n 5 67) Clinical recurrence Surgical recurrence
All patients (n 5 142)
Age at diagnosis % 16 years (n 5 34)
Age at diagnosis . 16 years (n 5 108)
P value
2.4 6 2.5 3.9 6 8.6
3.1 6 2.7 .9 6 1.5
2.2 6 2.4 5.1 6 9.8
.063 .002
42 (30) 67 (47) 11.5 6 7.8 1.3 6 1.6 61 (43) 48 (72) 45 (32) 7 (5)
18 (53) 15 (50) 10.7 6 6.3 1.6 62 17 (50) 13 (87) 13 (38) 2 (6)
24 (22) 50 (45) 11.2 6 7.2 1.2 6 1.5 44 (41) 35 (70) 32 (30) 5 (5)
.002 .574 .793 .212 .427 .760 .400 .673
Values are given as mean 6 standard deviation or number (percentage). TI 5 terminal ileum.
Comments Predicting postoperative recurrence of CD after ileocecal resection remains a major challenge. Because of the variable course of CD, it has proven difficult to establish strict postoperative management guidelines to prevent recurrence. In fact, there are 3 different algorithms described by De Cruz et al1 as ‘‘valid approaches’’ in the management of postoperative CD: (1) postoperative drug treatment stratified according to the risk of recurrence; (2) aggressive approach of placing all patients on biologics postoperatively; and (3) a tailored prophylactic approach based on findings at
Figure 1 Time to recurrence based on age at diagnosis %16 or .16 years.
postoperative endoscopy.1 Prospective studies to validate which approach should be used are lacking. The association between early postoperative immunoprophylaxis and a decreased association with recurrence is an important finding of our study in the era of unclear postoperative management algorithms to prevent recurrence. In the only other study to evaluate the timing of prophylaxis in preventing recurrence after surgery for ileocecal CD, Bordeianou et al5 reported no difference between immediate (within 1 month of surgery) and tailored prophylaxis (after first endoscopy). Our results may differ because our cohort included patients who had previous resections for CD, while the study by Bordeianou et al only included patients undergoing their first operation for CD. Further studies to investigate this important clinical question should be carried out, preferably as a large prospective randomized trial. Studies in adults have established multiple predictors of disease recurrence after surgical resection, which include smoking, history of prior resection, genetic markers, perforating disease type, ileocolonic location, and short preoperative disease duration.1,6,7 Younger age at disease onset has also been associated with higher risk of recurrence, but results have been less conclusive because of the potential confounding effect of longer follow-up among younger patients.1,8–10 A fundamental question of inflammatory bowel disease remains unanswered: whether pediatric onset CD has a different pathogenesis leading to a different phenotype or whether it is the same disease process but merely occurring at an earlier time point.11,16 Some experts believe that genetic contribution has a greater role than environmental factors in pediatric and adolescent CD.16 Several consistent differences between pediatric onset CD and its adult counterpart, including disease type, disease location, disease behavior, sex preponderance, and genetically attributable risks, support the theory that pediatric onset CD represents a distinct disease phenotype.11,16,20 Thus, one would expect
I.D. Bobanga et al.
Crohn’s recurrence after ileocecectomy
595
Table 3 Cox proportional hazards regression of possible predictors of endoscopic recurrence in two age groups with univariate and multivariate analysis Age at diagnosis % 16 years (n 5 15)
Age at diagnosis . 16 years (n 5 50)
Univariate analysis Predictor
Crude HR
95% CI
P value
Crude HR
95% CI
P value
Age at diagnosis Age at surgery Preoperative disease duration Emergent procedure Prophylaxis within 4 weeks Positive margins
.9 1.1 1.1 1.4 .6 .6 Multivariate analysis
.73–1.21 .96–1.17 .97–1.25 .42–4.77 .01–.59 .10–3.39
.626 .277 .133 .573 .016 .558
1.0 1.0 1.0 2.2 .4 3.9
.99–1.03 .99–1.03 .99–1.06 .83–5.67 .16–1.13 1.46–10.37
.992 .496 .183 .116 .086 .007
Adjusted HR
95% CI
P value
1.1 .5 3.3
.27–4.20 .13–1.82 1.08–10.14
.937 .288 .036
95% CI
P value
.4
.74–1.14
.439
.9
.01–5.303
.917
Adjusted HR Preoperative disease duration Emergent procedure Prophylaxis within 4 weeks Positive margins
CI 5 confidence interval; HR 5 hazards ratio.
to also find differences in factors predicting disease recurrence in pediatric and adult onset CD. Few studies have evaluated risk factors for postoperative recurrence in younger patients. Baldassano et al13 found high Pediatric Crohn’s Disease Activity Index, colonic or ileocolonic disease, and preoperative use of 6-mercaptopurine to be associated with high rates of postoperative recurrence in children and Griffiths et al21 also showed similar results. We were unable to find any studies in the literature that directly compared factors that decrease the time to recurrence in children compared with adults. The results of this study show that while the time to any recurrence, endoscopic recurrence, or clinical recurrence is not statistically different between the 2 age groups, there are different predictive factors that increase the risk of
recurrence in pediatric and adult onset CD. After subgroup multivariate analysis, we found that positive margins at ileocecectomy were predictive of endoscopic recurrence only in the adult group. Furthermore, restarting or initiating postoperative prophylactic medications within 4 weeks of surgery was predictive of delayed time to clinical recurrence only in the adult onset group. There were several findings that differentiated the patients with pediatric onset disease from the adult onset patients in our study. The younger patients at diagnosis were more frequently treated with immunomodulators preoperatively, and were restarted on early postoperative immunotherapy more often than older adults. Despite this more aggressive medical management, the incidence and time to postoperative recurrence was not only equal, but
Table 4 Cox proportional hazards regression of possible predictors of clinical recurrence in 2 age groups with univariate and multivariate analysis Age at diagnosis % 16 years (n 5 34)
Age at diagnosis . 16 (n 5 108)
Univariate analysis P value
Crude HR
95% CI
P value
1.1 .87–1.31 1.0 .91–1.04 .9 .86–1.02 .5 .06–3.9 .5 .12–1.81 .6 .12–2.60 Multivariate analysis
.537 .430 .224 .505 .267 .455
1.0 1.0 1.0 3.1 .4 3.4
.96–1.01 .97–1.02 .95–1.01 1.05–8.91 .15–.92 1.42–8.09
.266 .215 .521 .041 .032 .006
Adjusted HR
P value
Adjusted HR
95% CI
P value
Predictor
Crude HR
Age at diagnosis Age at surgery Preoperative disease duration Emergent procedure Prophylaxis within 4 weeks Positive margins
Age at surgery Prophylaxis within 4 weeks Positive margins CI 5 confidence interval; HR 5 hazards ratio.
95% CI
95% CI
1.4 .3 2.0
.35–5.50 .08–.96 .75–5.35
.646 .043 .169
596 also trended toward increased recurrence in the younger group (50% vs 41%). These associations between medical treatment and recurrence suggest a more aggressive disease phenotype in pediatric onset CD. Our results have important implications for the clinician who may choose to advise patients with early onset CD of the more aggressive disease phenotype and increased risk of recurrence after surgical intervention. Further studies should focus on whether a more aggressive surgical approach is warranted in these younger patients. Ideally, the timing and type of pre- and postoperative medication prophylaxis to prevent recurrence in pediatric onset CD should be evaluated in a prospective multi-institutional setting. There are several limitations to our study. Retrospective studies are limited in their ability to show associations and not causative factors. Our study size limited our ability to explore subgroups and trends that may provide important differentiations between unique groups with CD. Another limitation in CD studies is the difficulty in accurately discerning the time of clinical recurrence because of the variety of presentations. While endoscopic and surgical recurrence can be standardized based on objective criteria, the definition of clinical recurrence varies with provider assessment of patient symptoms and is not always accompanied by an objective test. Finally, our study is limited to a single tertiary care multi-hospital institution and may be difficult to generalize the findings to other hospital settings. In conclusion, our retrospective analysis comparing risk factors for recurrence between patients with pediatric and adult onset CD who undergo ileocecal resection found that the rate of recurrence is not different between the 2 age groups. Despite more aggressive management with immunologics preoperatively and earlier resumption of postoperative immunoprophylaxis, younger patients are at equal risk of recurrence. Immediate postoperative immunoprophylaxis was associated with decreased time to clinical recurrence only in patients with adult onset CD. Further studies designed to analyze recurrence of CD in younger patients is warranted to help identify best practice for these patients.
References 1. De Cruz P, Kamm MA, Prideaux L, et al. Postoperative recurrent luminal Crohn’s disease: a systematic review. Inflamm Bowel Dis 2012;18:758–77.
The American Journal of Surgery, Vol 208, No 4, October 2014 2. Sachar DB. The problem of postoperative recurrence of Crohn’s disease. Med Clin North Am 1990;74:183–8. 3. Rutgeerts P, Geboes K, Vantrappen G, et al. Predictability of the postoperative course of Crohn’s disease. Gastroenterology 1990;99: 956–63. 4. Olaison G, Smedh K, Sjo¨dahl R. Natural course of Crohn’s disease after ileocolic resection: endoscopically visualised ileal ulcers preceding symptoms. Gut 1992;33:331–5. 5. Bordeianou L, Stein SL, Ho VP, et al. Immediate versus tailored prophylaxis to prevent symptomatic recurrences after surgery for ileocecal Crohn’s disease? Surgery 2011;149:72–8. 6. Yamamoto T. Factors affecting recurrence after surgery for Crohn’s disease. World J Gastroenterol 2005;11:3971–9. 7. Borley NR, Mortensen NJ, Jewell DP. Preventing postoperative recurrence of Crohn’s disease. Br J Surg 1997;84:1493–502. 8. Scarpa M, Angriman I, Barollo M, et al. Risk factors for recurrence of stenosis in Crohn’s disease. Acta Biomed 2003;74(Suppl 2):80–3. 9. Ryan WR, Allan RN, Yamamoto T, et al. Crohn’s disease patients who quit smoking have a reduced risk of reoperation for recurrence. Am J Surg 2004;187:219–25. 10. Post S, Herfarth C, Bo¨hm E, et al. The impact of disease pattern, surgical management, and individual surgeons on the risk for relaparotomy for recurrent Crohn’s disease. Ann Surg 1996;223:253–60. 11. Sauer CG, Kugathasan S. Pediatric inflammatory bowel disease: highlighting pediatric differences in IBD. Gastroenterol Clin North Am 2009;38:611–28. 12. Beaugerie L, Seksik P, Nion-Larmurier I, et al. Predictors of Crohn’s disease. Gastroenterology 2006;130:650–6. 13. Baldassano RN, Han PD, Jeshion WC, et al. Pediatric Crohn’s disease: risk factors for postoperative recurrence. Am J Gastroenterol 2001;96: 2169–76. 14. Hojsak I, Pavic AM, Misak Z, et al. Risk factors for relapse and surgery rate in children with Crohn’s disease. Eur J Pediatr 2014;173: 617–21. 15. Lacher M, Helmbrecht J, Schroepf S, et al. NOD2 mutations predict the risk for surgery in pediatric-onset Crohn’s disease. J Pediatr Surg 2010;45:1591–7. 16. Ruel J, Ruane D, Mehandru S, et al. IBD across the age spectrum-is it the same disease? Nat Rev Gastroenterol Hepatol 2014;11:88–98. 17. Law ST, Li KK. Age-related differences in the clinical course of Crohns disease in an Asian population: a retrospective cohort review. Indian Pediatr 2013;50:1148–52. 18. Boualit M, Salleron J, Turck D, et al. Long-term outcome after first intestinal resection in pediatric-onset Crohn’s disease: a populationbased study. Inflamm Bowel Dis 2013;19:7–14. 19. Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)da metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009;42:377–81. 20. Van Limbergen J, Russell RK, Drummond HE, et al. Definition of phenotypic characteristics of childhood-onset inflammatory bowel disease. Gastroenterology 2008;135:1114–22. 21. Griffiths AM, Wesson DE, Shandling B, et al. Factors influencing postoperative recurrence of Crohn’s disease in childhood. Gut 1991;32: 491–5.
