Abstract

Background & Aims: No therapy has been shown to reliably prevent the evolution of postoperative recurrence of Crohn’s disease. The aim of the current trial was to compare 6-mercaptopurine (6-MP) and mesalamine with placebo for the prevention of clinical, endoscopic, and radiographic recurrence of Crohn’s disease after resection and ileocolic anastomosis. Methods: Five centers randomized 131 patients to receive 6-MP (50 mg), mesalamine (3 g), or placebo daily in a double-blind, double-dummy trial. Patients had clinical assessments at 7 weeks and then every 3 months; colonoscopy at 6, 12, and 24 months; and small bowel series at 12 and 24 months. End points were clinical, endoscopic, and radiographic recurrence rates at 24 months. Results: Clinical recurrence rates (intent to treat) by life-table analysis at 24 months were 50% (95% confidence interval [CI], 34%–68%), 58% (95% CI, 41%–75%), and 77% (95% CI, 61%–91%) in patients receiving 6-MP, mesalamine, and placebo, respectively. Endoscopic recurrence rates were 43% (95% CI, 28%–63%), 63% (95% CI, 47%–79%), and 64% (95% CI, 46%–81%), and radiographic recurrence rates were 33% (95% CI, 19%–54%), 46% (95% CI, 29%–66%), and 49% (95% CI, 30%–72%), respectively. 6-MP was more effective than placebo (P 50% of patients with Crohn’s disease undergo surgical resection to treat complications such as stricture, fistula, abscess, or medically intractable disease.1Becker J.M. Surgical therapy for ulcerative colitis and Crohn’s disease.Gastroenterol Clin North Am. 1999; 28 (viii—ix): 371-390Abstract Full Text Full Text PDF PubMed Scopus (118) Google Scholar After ileocolic surgical resection for Crohn’s disease, the disease typically recurs at the site of resection2Greenstein A.J. Sachar D.B. Pasternack B.S. Janowitz H.D. Reoperation and recurrence in Crohn’s colitis and ileocolitis crude and cumulative rates.N Engl J Med. 1975; 293: 685-690Crossref PubMed Scopus (237) Google Scholar, 3Rutgeerts P. Geboes K. Vantrappen G. Kerremans R. Coenegrachts J.L. Coremans G. Natural history of recurrent Crohn’s disease at the ileocolonic anastomosis after curative surgery.Gut. 1984; 25: 665-672Crossref PubMed Scopus (665) Google Scholar and often extends proximally to the same extent4D’Haens G.R. Gasparaitis A.E. Hanauer S.B. Duration of recurrent ileitis after ileocolonic resection correlates with presurgical extent of Crohn’s disease.Gut. 1995; 36: 715-717Crossref PubMed Scopus (65) Google Scholar and pattern.5Greenstein A.J. Lachman P. Sachar D.B. Springhorn J. Heimann T. Janowitz H.D. Aufses Jr, A.H. Perforating and non-perforating indications for repeated operations in Crohn’s disease evidence for two clinical forms.Gut. 1988; 29: 588-592Crossref PubMed Scopus (368) Google Scholar Early evidence of recurrence can be demonstrated endoscopically in the neoterminal ileum in more than 73%–93% of patients at 1 year,3Rutgeerts P. Geboes K. Vantrappen G. Kerremans R. Coenegrachts J.L. Coremans G. Natural history of recurrent Crohn’s disease at the ileocolonic anastomosis after curative surgery.Gut. 1984; 25: 665-672Crossref PubMed Scopus (665) Google Scholar, 6Olaison G. Smedh K. Sjodahl R. Natural course of Crohn’s disease after ileocolic resection endoscopically visualised ileal ulcers preceding symptoms.Gut. 1992; 33: 331-335Crossref PubMed Scopus (433) Google Scholar and the severity of the mucosal lesions can predict the recurrence of clinical symptoms.7Rutgeerts P. Geboes K. Vantrappen G. Beyls J. Kerremans R. Hiele M. Predictability of the postoperative course of Crohn’s disease.Gastroenterology. 1990; 99: 956-963Abstract PubMed Google Scholar To date, no therapies have been proven to prevent the development of postoperative recurrence.8Leiper K. London I. Rhodes J.M. Adjuvant post-operative therapy.Baillieres Clin Gastroenterol. 1998; 12: 179-199Abstract Full Text PDF PubMed Scopus (7) Google Scholar, 9D’Haens G. Rutgeerts P. Postoperative recurrence of Crohn’s disease pathophysiology and prevention.Inflamm Bowel Dis. 1999; 5: 295-303Crossref PubMed Scopus (31) Google Scholar Treatment with corticosteroids,10Steinhart A.H. Ewe K. Griffiths A.M. Modigliani R. Thomsen O.O. Corticosteroids for maintaining remission of Crohn’s disease.Cochrane Database Syst Rev. 2000; 2 (CD000301)PubMed Google Scholar including budesonide,11Greenberg G.R. Feagan B.G. Martin F. Sutherland L.R. Thomson A.B. Williams C.N. Nilsson L.G. Persson T. Canadian Inflammatory Bowel Disease Study GroupOral budesonide as maintenance treatment for Crohn’s disease a placebo-controlled, dose-ranging study.Gastroenterology. 1996; 110: 45-51Abstract Full Text PDF PubMed Scopus (284) Google Scholar, 12Hellers G. Cortot A. Jewell D. Leijonmarck C.E. Lofberg R. Malchow H. Nilsson L.G. Pallone F. Pena S. Persson T. Prantera C. Rutgeerts P. The IOIBD Budesonide Study GroupOral budesonide for prevention of postsurgical recurrence in Crohn’s disease.Gastroenterology. 1999; 116: 294-300Abstract Full Text Full Text PDF PubMed Scopus (241) Google Scholar has not been successful in preventing relapse of Crohn’s disease. Postoperative maintenance studies have shown both positive13McLeod R.S. Wolff B.G. Steinhart A.H. Carryer P.W. O’Rourke K. Andrews D.F. Blair J.E. Cangemi J.R. Cohen Z. Cullen J.B. et al.Prophylactic mesalamine treatment decreases postoperative recurrence of Crohn’s disease.Gastroenterology. 1995; 109: 404-413Abstract Full Text PDF PubMed Scopus (248) Google Scholar and negative14Lochs H. Mayer M. Fleig W.E. Mortensen P.B. Bauer P. Genser D. Petritsch W. Raithel M. Hoffmann R. Gross V. Plauth M. Staun M. Nesje L.B. Prophylaxis of postoperative relapse in Crohn’s disease with mesalamine European Cooperative Crohn’s Disease Study VI.Gastroenterology. 2000; 118: 264-273Abstract Full Text Full Text PDF PubMed Scopus (292) Google Scholar evidence for mesalamine in large, placebo-controlled trials. Both controlled trials and uncontrolled data have supported the potential of the immunomodulators azathioprine or 6-mercaptopurine (6-MP) for the maintenance of medical remissions of Crohn’s disease15Pearson D.C. May G.R. Fick G. Sutherland L.R. Azathioprine for maintaining remission of Crohn’s disease.Cochrane Database Syst Rev. 2000; 2 (CD000067)Google Scholar or healing of recurrent postoperative lesions.16D’Haens G. Geboes K. Ponette E. Penninckx F. Rutgeerts P. Healing of severe recurrent ileitis with azathioprine therapy in patients with Crohn’s disease.Gastroenterology. 1997; 112: 1475-1481Abstract Full Text PDF PubMed Scopus (239) Google Scholar The purpose of this clinical trial was to evaluate the relative benefits of mesalamine and 6-MP compared with placebo for the delay or prevention of postoperative recurrence of Crohn’s disease. The study was designed as a 5-center, double-blind, placebo-controlled trial comparing fixed dosages of mesalamine or 6-MP with placebo over a 2-year postoperative period and was conducted from 1992 to 1996. Patients undergoing a first or subsequent ileocolic resection with a primary anastomosis for disease confined to the ileum and adjacent colon were eligible for enrollment. Patients were excluded if there was evidence of gross Crohn’s disease at the operative margins or in proximal or distal segments of intestine (excluding perianal disease) at the time of surgery or at pathologic examination. Patients with minimal evidence of Crohn’s disease at other sites (aphthoid erosions or microscopic inflammatory changes) were not excluded. Patients were randomized by a central computer by permuted blocks of 6 (unknown to investigators) per center to receive mesalamine (Pentasa; Marion Merrill Dow, Kansas City, MO) 3 g daily, 6-MP (Purinethol; Burroughs Wellcome, Research Triangle Park, NC) 50 mg daily, or placebo. Medications were prepared and dispensed by an assigned pharmacist at each site’s investigational pharmacy who was not directly involved in the care of the patients. Therapy was initiated before postoperative hospital discharge (“baseline visit”), and drugs were administered as mesalamine (250 mg; 4 capsules 3 times daily), 6-MP (50 mg once daily), or identical matching placebo (supplied by Marion Merrill Dow or Burroughs Wellcome). After obtaining written informed consent, all patients received 13 tablets daily containing active drug and/or placebo. Presurgical therapy, including aminosalicylates, antibiotics, or immunomodulators, was discontinued before surgical resection and was not allowed during the postoperative trial. Preoperative treatment with corticosteroids was completely tapered by 3 months after hospital discharge at a rate determined by the treating physician. No concurrent treatment for Crohn’s disease, aside from topical therapy for perianal disease, was allowed during the duration of the trial. Continuous use of nonsteroidal anti-inflammatory drugs was not allowed during the study. An evaluating (treating) physician followed up each patient and was blinded as to the study drug and laboratory results. A second, nontreating physician was assigned at each site to monitor for drug toxicity. Monitoring of complete blood cell count and platelet count was performed weekly for the first 3 weeks. If the white blood cell and platelet counts did not fall below 4500/μL and 150,000/μL, respectively, the patient was continued on the same dosage of 6-MP and the interval between complete blood cell counts was extended to every 12 weeks. If the white blood cell count and platelet counts fell below 4500/μL or 150,000/μL, respectively, the dosage of 6-MP was reduced by one half and the complete blood cell count was repeated at weekly intervals until the white blood cell and platelet counts rose above the thresholds. After 2 additional weeks, if the white blood cell or platelet counts remained below 4,500/μL or 150,000/μL, respectively, the patient was terminated from the trial. A safety monitoring board composed of 2 senior, experienced gastroenterologists from nonenrolling institutions assessed the overall safety and patient withdrawals at 6-month intervals. Blood chemistries, including glucose, electrolytes, liver enzymes, and a kidney profile, were obtained at baseline and at 3-month intervals. Serum amylase and lipase levels were analyzed if the patient developed any clinical signs of possible pancreatitis (abdominal pain, nausea, vomiting). Any evidence of elevated liver enzyme levels or an increase in amylase or lipase levels greater than 2 times that of normal were indications for termination from the trial. Patient evaluation consisted of assessments of clinical, endoscopic, and radiographic disease activity at each study site by the blinded physician. Clinical evaluations were assessed at week 7 after initiating treatment with the study drug and then at 3-month intervals for 24 months. Clinical assessments were determined by the blinded physician and rated as detailed in Table 1. Formal Crohn’s Disease Activity Indices were not calculated, because they had not been validated for study of postoperative recurrence. Colonoscopic examinations with endoscopic descriptions and photography of the anastomosis and preanastomotic ileum were performed by the blinded investigators (all gastroenterologists) at months 6, 12, and 24. The endoscopic scoring system of Rutgeerts et al.3Rutgeerts P. Geboes K. Vantrappen G. Kerremans R. Coenegrachts J.L. Coremans G. Natural history of recurrent Crohn’s disease at the ileocolonic anastomosis after curative surgery.Gut. 1984; 25: 665-672Crossref PubMed Scopus (665) Google Scholar was used to score endoscopic recurrence (Table 2). Radiographic examinations with dedicated small bowel barium studies were performed at 12 and 24 months. Radiographic interpretations were performed by the blinded inflammatory bowel disease radiologist at each institution and were graded as detailed in Table 1. Because there were no radiographic scales validated at the time of the study, this grading scale needed to be created and had not undergone validation before its use.Table 1Clinical and Radiographic Recurrence Grading ScalesSeverity grade assignedClinicalRadiographic1RemissionNormal2Mild symptomsMucosal edema/aphthoid ulcers3Moderate symptomsLinear ulcers/cobblestoning4Severe symptomsStrictures/fistulas/inflammatory massNOTE. Assessments were determined by a physician blinded to patient therapy. Open table in a new tab Table 2Endoscopic Recurrence Severity Grading Scale for Lesions at the Ileocolic Anastomosis or the Preanastomotic IleumScoreDescription of lesions0No lesions15 aphthous lesions with normal mucosa between lesions or lesions confined to the ileocolic anastomosis3Diffuse aphthous ileitis with diffusely inflamed mucosa4Diffuse inflammation with larger ulcers, nodules, and/or narrowingAdapted from Rutgeerts et al.3Rutgeerts P. Geboes K. Vantrappen G. Kerremans R. Coenegrachts J.L. Coremans G. Natural history of recurrent Crohn’s disease at the ileocolonic anastomosis after curative surgery.Gut. 1984; 25: 665-672Crossref PubMed Scopus (665) Google Scholar Open table in a new tab NOTE. Assessments were determined by a physician blinded to patient therapy. Adapted from Rutgeerts et al.3Rutgeerts P. Geboes K. Vantrappen G. Kerremans R. Coenegrachts J.L. Coremans G. Natural history of recurrent Crohn’s disease at the ileocolonic anastomosis after curative surgery.Gut. 1984; 25: 665-672Crossref PubMed Scopus (665) Google Scholar The prespecified end points of the study were the percentage of patients with a clinical, radiographic, or endoscopic relapse through 24 months. Clinical relapse was defined as a score of 2 or greater on the clinical recurrence grading scale (Table 1). Radiographic relapse was defined as a score of 2 or greater on the radiographic recurrence grading scale, and endoscopic relapse was defined as a score of 2 or greater on the Rutgeerts endoscopic recurrence grading scale. Clinical relapse was an indication for termination from the study; radiographic or endoscopic relapse were not, unless they resulted in a change in inflammatory bowel disease therapy. Patients who developed symptoms suggesting recurrent Crohn’s disease confirmed by clinical examination and endoscopic or radiographic evaluation necessitating a treatment change according to the treating (blinded) physician were withdrawn from the study. Patients were also withdrawn if they developed a serious adverse event potentially related to study medications (such as abdominal pain consistent with pancreatitis, leukopenia, or potential hypersensitivity to study drug). The institutional review boards of each of the 5 treating institutions reviewed and approved the study before its commencement. A single primary end point was not prespecified in this study. Sample size calculations were performed for the endoscopic criteria, using 2-sided α of 0.05 and 80% power, based on a predicted endoscopic recurrence of 75% at 1 year in the placebo group. A sample size of 50 in each group allows sufficient power to detect a 40% reduction in mild Crohn’s disease lesions and a 75% reduction in more severe lesions at 1 year. The primary analysis was based on the intent-to-treat principle. All patients randomized were analyzed according to the original treatment to which they were assigned. The protocol specified that each of the active study agents would be compared with placebo with respect to clinical, endoscopic, and radiographic relapse. Stata statistical software (release 7.0, 2001; Stata Corp., College Station, TX) was used to perform efficacy analyses. To properly account for patient withdrawal, life tables were used to obtain actuarial estimates and confidence intervals (CIs) for each end point (clinical, endoscopic, and radiographic recurrence at 24 months). Comparisons between therapies used the Cox proportional hazards model.17Matthews D. Farewell V. Using and understanding medical statistics. 3rd ed. Karger AG, Basel, Switzerland1996Google Scholar Adherence (compliance) was assessed by pill counts at each study visit; patients taking 2 years with 2 years with >1 attack/year); 5, long-standing (>2 years continuously active).3.1 ± 1.13.0 ± 1.03.0 ± 1.0NOTE. Values shown are mean ± SD unless otherwise indicated.a Percentage perforating vs. nonperforating. 121 evaluable patients.b Disease activity before surgery: 1, mild disease (short or long-standing); 2, active (short or long-standing); 3, long-standing (>2 years with 2 years with >1 attack/year); 5, long-standing (>2 years continuously active). Open table in a new tab NOTE. Values shown are mean ± SD unless otherwise indicated. Intent-to-treat analysis showed that 24-month clinical recurrence rates (with 95% CIs) were as follows: 50% of patients receiving 6-MP (95% CI, 34%–68%), 58% of patients receiving mesalamine (95% CI, 41%–75%), and 77% of patients receiving placebo (95% CI, 61%–91%). Figure 1 shows the Kaplan-Meier estimates for clinical recurrence, with a benefit for 6-MP versus placebo (hazard ratio [HR], 0.52; P = 0.045) and a trend for mesalamine versus placebo (HR, 0.62; P = 0.123). Perforating-type disease was a predictor of clinical recurrence, independent of treatment modality (HR, 2.0; P = 0.04). The actuarial estimate of the proportion of patients with endoscopic evidence of relapse at 24 months as defined as a Rutgeerts’ endoscopic score >1 was 43% (95% CI, 28%–63%) with 6-MP, 63% (95% CI, 47%–79%) with mesalamine, and 64% (95% CI, 46%–81%) with placebo. Figure 2 shows the Kaplan-Meier estimates for endoscopic recurrence (Rutgeerts’ endoscopic score >1), with only a benefit for 6-MP versus placebo (6-MP vs. placebo: HR, 0.48; P = 0.030; mesalamine vs. placebo: HR, 0.80; P = 0.458). If the criteria for relapse are increased to a Rutgeerts’ endoscopic score >2, the 24-month relapse rates are 16% (95% CI, 7%–35%) for 6-MP, 48% (95% CI, 30%–70%) for mesalamine, and 42% (95% CI, 21%–70%) for placebo, with a trend toward benefit with 6-MP (HR, 0.48; P = 0.13) but not mesalamine (HR = 1.10; P = 0.82) versus placebo (Figure 3). Subgroup analysis showed a benefit of 6-MP versus placebo (P = 0.03) in the avoidance of endoscopic recurrence (Rutgeerts’ endoscopic score >1) in the perforating-type disease group.Figure 3Endoscopic relapse-free survival. Likelihood of avoiding moderately severe endoscopic relapse (Rutgeerts’ endoscopic score >2) following ileocolic resection with primary anastomosis (“baseline”) in patients with Crohn’s disease randomized to 6-MP, mesalamine, or placebo. 6-MP vs. placebo: HR, 0.48; P = 0.13. Mesalamine vs. placebo: HR, 1.10; P = 0.82.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Radiographic recurrence at 24 months was estimated to be present in 33% (95% CI, 19%–54%) of patients receiving 6-MP, 46% (95% CI, 29%–66%) receiving mesalamine, and 49% (95% CI, 30%–72%) receiving placebo. Kaplan-Meier estimates (Figure 4) failed to show a benefit of either 6-MP (HR, 0.57; P = 0.15) or mesalamine (HR, 0.61; P = 0.19) versus placebo. Seventy-four (56%) patients withdrew from the trial before the 24-month end point, although 33 (45%) of these patients withdrew only after clinical recurrence, thus providing full information regarding the clinical end point for 69% of the total patient population. The reasons for withdrawal are detailed in Table 4. Nineteen patients were withdrawn due to adverse events (6-MP, 9; mesalamine, 6; placebo, 4); the adverse event rates were similar across treatment groups (P = 0.498). The adverse events experienced in the patients treated with 6-MP were leukopenia (2 patients), alopecia (2 patients), diarrhea (2 patients), and flatus, gastrointestinal bleeding, and phlebitis (one patient each). Adverse events noted in patients taking mesalamine were diarrhea (2 patients) and allergic reaction, bowel obstruction, elevated liver function tests, and arthralgias (1 patient each). Two patients receiving placebo experienced abdominal pain, 1 developed a headache, and 1 had diarrhea. No confirmed pancreatitis was observed, and the 2 patients withdrawn because of abdominal pain (suspected but not confirmed pancreatitis) were subsequently found to have been receiving the placebo. The only serious adverse event requiring hospitalization was observed in 1 patient randomized to mesalamine who developed a postoperative bowel obstruction that was unlikely related to study medication and responded to conservative therapy.Table 4Patient AccountabilityPlacebo6-MPMesalamineTotalPatients randomized404744131Lost before clinical recurrence12151441Withdrew consent122Surgical complication201Adverse experience374Noncompliance243Lost to follow-up424Lost after clinical recurrence1581033Adverse experience122Noncompliance111Pregnancy002Lost to follow-up1355Completed trial13242057In clinical remission5151131 Open table in a new tab The purpose of the current study was to assess the potential of 6-MP and mesalamine in delaying or preventing postoperative recurrence of Crohn’s disease at an ileocolic anastomosis. At the time of trial initiation, prior studies had suggested potential benefit from mesalamine for prevention of relapse in quiescent Crohn’s disease but primarily for the prevention of postoperative recurrence.18Camma C. Giunta M. Rosselli M. Cottone M. Mesalamine in the maintenance treatment of Crohn’s disease a meta-analysis adjusted for confounding variables.Gastroenterology. 1997; 113: 1465-1473Abstract Full Text Full Text PDF PubMed Scopus (426) Google Scholar Dosages used in these maintenance trials ranged between 1.5 and 3 g daily. However, a recently reported European study of mesalamine for prevention of postoperative recurrence of Crohn’s disease using a 4-g dose failed to show an overall benefit compared with placebo,14Lochs H. Mayer M. Fleig W.E. Mortensen P.B. Bauer P. Genser D. Petritsch W. Raithel M. Hoffmann R. Gross V. Plauth M. Staun M. Nesje L.B. Prophylaxis of postoperative relapse in Crohn’s disease with mesalamine European Cooperative Crohn’s Disease Study VI.Gastroenterology. 2000; 118: 264-273Abstract Full Text Full Text PDF PubMed Scopus (292) Google Scholar although some benefit was observed for patients with disease confined to the terminal ileum. The choice of a 3-g dose of mesalamine was based on the prior maintenance trials that had preceded the trial by Lochs et al.14Lochs H. Mayer M. Fleig W.E. Mortensen P.B. Bauer P. Genser D. Petritsch W. Raithel M. Hoffmann R. Gross V. Plauth M. Staun M. Nesje L.B. Prophylaxis of postoperative relapse in Crohn’s disease with mesalamine European Cooperative Crohn’s Disease Study VI.Gastroenterology. 2000; 118: 264-273Abstract Full Text Full Text PDF PubMed Scopus (292) Google Scholar and was consistent with the dose used in the positive trial reported by McLeod et al.13McLeod R.S. Wolff B.G. Steinhart A.H. Carryer P.W. O’Rourke K. Andrews D.F. Blair J.E. Cangemi J.R. Cohen Z. Cullen J.B. et al.Prophylactic mesalamine treatment decreases postoperative recurrence of Crohn’s disease.Gastroenterology. 1995; 109: 404-413Abstract Full Text PDF PubMed Scopus (248) Google Scholar The rationale for using 6-MP was based on the data showing maintenance benefits from the purine analogues for maintenance therapy of Crohn’s disease15Pearson D.C. May G.R. Fick G. Sutherland L.R. Azathioprine for maintaining remission of Crohn’s disease.Cochrane Database Syst Rev. 2000; 2 (CD000067)Google Scholar and the personal experience of 2 of the investigators (B.I.K.19Korelitz B.I. Adler D.J. Mendelsohn R.A. Sacknoff A.L. Long-term experience with 6-mercaptopurine in the treatment of Crohn’s disease.Am J Gastroenterol. 1993; 88: 1198-1205PubMed Google Scholar and D.H.P.20Present D.H. Korelitz B.I. Wisch N. Glass J.L. Sachar D.B. Pasternack B.S. Treatment of Crohn’s disease with 6-mercaptopurine A long-term, randomized, double-blind study.N Engl J Med. 1980; 302: 981-987Crossref PubMed Scopus (993) Google Scholar). Although there were no dose-ranging data to predict response, the collaborating investigators selected a dosage that was likely to be tolerated without inducing adverse events in a large proportion of patients who were “disease free” undergoing “prophylactic” therapy.21Present D.H. Meltzer S.J. Krumholz M.P. Wolke A. Korelitz B.I. 6-Mercaptopurine in the management of inflammatory bowel disease short- and long-term toxicity.Ann Intern Med. 1989; 111: 641-649Crossref PubMed Scopus (715) Google Scholar There remains no standard of therapy for the treatment of patients with postoperative Crohn’s disease.22Hanauer S.B. Sandborn W. Management of Crohn’s disease in adults.Am J Gastroenterol. 2001; 96: 635-643Crossref PubMed Google Scholar The endoscopic recurrence of Crohn’s disease after ileocolic resection is predictable,3Rutgeerts P. Geboes K. Vantrappen G. Kerremans R. Coenegrachts J.L. Coremans G. Natural history of recurrent Crohn’s disease at the ileocolonic anastomosis after curative surgery.Gut. 1984; 25: 665-672Crossref PubMed Scopus (665) Google Scholar, 7Rutgeerts P. Geboes K. Vantrappen G. Beyls J. Kerremans R. Hiele M. Predictability of the postoperative course of Crohn’s disease.Gastroenterology. 1990; 99: 956-963Abstract PubMed Google Scholar with numerous factors contributing to the likelihood of recurrence. These include age, location and duration of disease, perforating subtype of Crohn’s disease, and smoking history.4D’Haens G.R. Gasparaitis A.E. Hanauer S.B. Duration of recurrent ileitis after ileocolonic resection correlates with presurgical extent of Crohn’s disease.Gut. 1995; 36: 715-717Crossref PubMed Scopus (65) Google Scholar, 8Leiper K. London I. Rhodes J.M. Adjuvant post-operative therapy.Baillieres Clin Gastroenterol. 1998; 12: 179-199Abstract Full Text PDF PubMed Scopus (7) Google Scholar, 23Lautenbach E. Berlin J.A. Lichtenstein G.R. Risk factors for early postoperative recurrence of Crohn’s disease.Gastroenterology. 1998; 115: 259-267Abstract Full Text Full Text PDF PubMed Scopus (167) Google Scholar, 24Moskovitz D. McLeod R.S. Greenberg G.R. Cohen Z. Operative and environmental risk factors for recurrence of Crohn’s disease.Int J Colorectal Dis. 1999; 14: 224-226Crossref PubMed Scopus (53) Google Scholar Perforating-type disease at the time of surgery was shown in the current study to be an independent predictor of subsequent clinical recurrence, and 6-MP was found to be protective against endoscopic recurrence (Rutgeerts’ endoscopic score >1) in patients who had the perforating subtype. Sample-size restrictions limited adequate evaluation of the impact of disease type on other outcomes. Unfortunately, the data collected on smoking in the current trial were insufficient to allow for meaningful statistical analysis. The end points selected for this trial included clinical, endoscopic, and radiographic analysis due to the recognition that mucosal findings typically precede and predict clinical recurrence.7Rutgeerts P. Geboes K. Vantrappen G. Beyls J. Kerremans R. Hiele M. Predictability of the postoperative course of Crohn’s disease.Gastroenterology. 1990; 99: 956-963Abstract PubMed Google Scholar, 25McLeod R.S. Wolff B.G. Steinhart A.H. Carryer P.W. O’Rourke K. Andrews D.F. Blair J.E. Cangemi J.R. Cohen Z. Cullen J.B. Chaytor R.G. Greenberg G.R. Jaffer N.M. Jeejeebhoy K.N. MacCarty R.L. Ready R.L. Weiland L.H. Risk and significance of endoscopic/radiological evidence of recurrent Crohn’s disease.Gastroenterology. 1997; 113: 1823-1827Abstract Full Text PDF PubMed Scopus (86) Google Scholar The timing of end-point assessment is crucial because rates of recurrence vary between end points. We selected a 24-month trial to assess the endoscopic changes that were known to occur early and also to determine whether there were changes in clinical recurrence that were independent of mucosal lesions. An endoscopic recurrence rate of 64% was seen in the placebo group at 24 months, similar to the 61% actuarial “endoscopic/radiographic” relapse rate in postoperative patients receiving placebo at 24 months reported by McLeod et al.25McLeod R.S. Wolff B.G. Steinhart A.H. Carryer P.W. O’Rourke K. Andrews D.F. Blair J.E. Cangemi J.R. Cohen Z. Cullen J.B. Chaytor R.G. Greenberg G.R. Jaffer N.M. Jeejeebhoy K.N. MacCarty R.L. Ready R.L. Weiland L.H. Risk and significance of endoscopic/radiological evidence of recurrent Crohn’s disease.Gastroenterology. 1997; 113: 1823-1827Abstract Full Text PDF PubMed Scopus (86) Google Scholar Similarly, in the trial by McLeod et al.,13McLeod R.S. Wolff B.G. Steinhart A.H. Carryer P.W. O’Rourke K. Andrews D.F. Blair J.E. Cangemi J.R. Cohen Z. Cullen J.B. et al.Prophylactic mesalamine treatment decreases postoperative recurrence of Crohn’s disease.Gastroenterology. 1995; 109: 404-413Abstract Full Text PDF PubMed Scopus (248) Google Scholar symptomatic relapse (symptoms plus endoscopic or radiologic confirmation of disease) was observed in 31% and 41% of patients randomized to 3 g of mesalamine or placebo, respectively, over a maximum of 72 months. These rates are lower than the 58% and 77% clinical relapse rates we observed at 24 months with a different mesalamine formulation. Clinical relapse in the current study was also more common than rates reported by Lochs et al.,14Lochs H. Mayer M. Fleig W.E. Mortensen P.B. Bauer P. Genser D. Petritsch W. Raithel M. Hoffmann R. Gross V. Plauth M. Staun M. Nesje L.B. Prophylaxis of postoperative relapse in Crohn’s disease with mesalamine European Cooperative Crohn’s Disease Study VI.Gastroenterology. 2000; 118: 264-273Abstract Full Text Full Text PDF PubMed Scopus (292) Google Scholar who diagnosed clinical relapse rates at 18 months in 25% of patients receiving 4 g of mesalamine and 31% of patients randomized to placebo. It is possible that the patients referred to our study centers were more likely to have been tertiary referrals with more aggressive disease. Our use of an intent-to-treat analysis provided a “worst-case” scenario that also may explain differences between the results of this and previously published trials or clinical experience. Our data also showed a higher rate of clinical recurrence at 2 years than seen in trials by Rutgeerts et al. comparing metronidazole (20 mg/kg daily for the first 3 months following surgery) with placebo (26% vs. 43%)26Rutgeerts P. Hiele M. Geboes K. Peeters M. Penninckx F. Aerts R. Kerremans R. Controlled trial of metronidazole treatment for prevention of Crohn’s recurrence after ileal resection.Gastroenterology. 1995; 108: 1617-1621Abstract Full Text PDF PubMed Scopus (721) Google Scholar or with ornidazole (1.0 g daily for the first 12 months postoperatively) versus placebo (37% vs. 45%) (P = 0.036).27Rutgeerts P. Van Assche G. D’Haens G. Baert F. Noman M. Aerden I. Geboes K. D’Hoore A. Penninckx F. Ornidazol for prophylaxis of postoperative Crohn’s disease final results from a double blind placebo controlled trial.Gastroenterology. 2002; 122 (abstr): A-80Google Scholar These trials also differed in their definition of clinical recurrence. Although the Crohn’s Disease Activity Index is typically used to assess clinical activity in trials of active Crohn’s disease and medical maintenance of remission, it has not been validated in postoperative recurrence. Given the variable time line of onset of postoperative clinical symptoms due to active Crohn’s disease, we and the investigators of many of the other studies on postoperative Crohn’s disease chose instead to use simplified disease symptom indices to define this outcome. The finding that the clinical relapse rate exceeded the endoscopic relapse rate raises questions over the use of purely clinical scales in defining disease activity and reflects the myriad of possible etiologies of bowel symptoms in patients with postoperative Crohn’s disease. Although only 57 of 131 patients in our trial completed all 24 months of the study, 33 of the patient withdrawals (45%) occurred only after clinical relapse had already been documented. Thus, 90 of the 131 patients (69%) could fully be evaluated for the end point of postoperative clinical relapse. Noncompliance or patient request accounted for 23% of withdrawals; patients were required to ingest 13 pills daily and had to endure 3 colonoscopies and 2 small bowel studies over the 24-month study period. There were no differences in compliance among the treatment groups. This is the first blinded trial to assess immunomodulatory therapy for the postoperative prevention or delay of recurrence of Crohn’s disease. Despite the consistent overall benefits of 6-MP compared with placebo (and mesalamine), the data are less compelling than may have been anticipated based on the uncontrolled data of Korelitz et al.19Korelitz B.I. Adler D.J. Mendelsohn R.A. Sacknoff A.L. Long-term experience with 6-mercaptopurine in the treatment of Crohn’s disease.Am J Gastroenterol. 1993; 88: 1198-1205PubMed Google Scholar regarding efficacy of 6-MP for prevention of postoperative recurrence of Crohn’s disease and observations that healing of recurrent Crohn’s ileitis may be achieved with azathioprine.16D’Haens G. Geboes K. Ponette E. Penninckx F. Rutgeerts P. Healing of severe recurrent ileitis with azathioprine therapy in patients with Crohn’s disease.Gastroenterology. 1997; 112: 1475-1481Abstract Full Text PDF PubMed Scopus (239) Google Scholar Potential explanations include the singular and possibly low dose of 6-MP (in contrast to dosing recommendations for 1.0–1.5 mg/kg28Sandborn W.J. Azathioprine state of the art in inflammatory bowel disease.Scand J Gastroenterol Suppl. 1998; 225: 92-99Crossref PubMed Scopus (122) Google Scholar). In the majority of patients, the treatment dose was <1 mg/kg. However, the low and uniform dose was selected to avoid problems with myelosuppression and to simplify the protocol design. Furthermore, in most prior studies of the purine analogues for Crohn’s disease, therapy has been performed with the initial combination of corticosteroids and 6-MP or azathioprine.15Pearson D.C. May G.R. Fick G. Sutherland L.R. Azathioprine for maintaining remission of Crohn’s disease.Cochrane Database Syst Rev. 2000; 2 (CD000067)Google Scholar, 29Sandborn W. Sutherland L. Pearson D. May G. Modigliani R. Prantera C. Azathioprine or 6-mercaptopurine for inducing remission of Crohn’s disease.Cochrane Database Syst Rev. 2000; 2 (CD000545)Google Scholar The efficacy of monotherapy with 6-MP has not been evaluated and, in addition, it is generally accepted that these agents exert maximal effects after 3–6 months.28Sandborn W.J. Azathioprine state of the art in inflammatory bowel disease.Scand J Gastroenterol Suppl. 1998; 225: 92-99Crossref PubMed Scopus (122) Google Scholar Several trials have shown that the majority of recurrent Crohn’s lesions are already present within 3 months after ileocolic anastomosis. It is possible, therefore, that 6-MP does not actually prevent recurrence of new lesions but may help to heal developing lesions. In any event, the consistent statistically positive results for the “low dose” of 6-MP compared with placebo and the numerical (but not statistically significant) results compared with mesalamine suggest that additional studies to compare a dose range with 6-MP (either according to weight or therapeutic monitoring for 6-MP metabolites), pretreatment of patients with 6-MP before elective resection, or possibly combination therapy with mesalamine and 6-MP may improve the overall outcome. In conclusion, in this placebo-controlled, double-blind, double-dummy, multicenter trial, 6-MP (but not mesalamine) was more effective than placebo at reducing the rates of clinical and endoscopic recurrence of Crohn’s disease at the surgical anastomosis following ileocolic resection. This benefit was most apparent for patients with perforating-type disease at the time of surgery. 6-MP therapy was safe and can be administered in a single daily dose. Further trials are necessary to optimize strategies for individual patients at risk for postoperative recurrence and to assess the timing, dosing, and benefits of concomitant therapy.