Editorials

Preventing Relapse After Surgery for Crohn's Disease: Where Do We Go From Here?

While surgical bowel excision often produces dramatic relief of symptoms in children with Crohn's disease, the operative success is always dampened by the possibility of disease relapse. Various studies suggest adults and children with Crohn's have a 50% risk of clinical recurrence as early as 3 years after their initial operation ^(1-4). Factors proposed to increase risk of disease recurrence following resection include younger age ^(1,3), female sex ⁽⁵⁾, perforating Crohn's disease ⁽⁶⁾, longer duration of disease ⁽³⁾, extent of ileal involvement ^(5,7), end-to-end anastomosis ⁽⁸⁾, and diffuse colonic involvement ^(2,3). The large number of studies have not consistently identified the same risk factors, and so it is impossible for the clinician to predict whether an individual patient is at high risk for early recurrence.

The study of pharmacologic prophylaxis of postoperative disease recurrence remains in its infancy. Randomized trials in adults suggest that when either mesalamine or metronidazole are begun postoperatively there is a lower rate of clinical relapse than if no therapy is given. Unfortunately, while the results of these studies reach statistical significance, the improvement is modest. Metronidazole- or mesalamine-treated patients exhibit a 25-30% clinical recurrence rate within 2 years, compared to approximately 40% of placebo-treated patients ^(9,10). There is clearly a need for more effective agents to prevent postoperative recurrence. However, since these patients are often asymptomatic and feeling well postoperatively, and since a significant percentage of patients will not experience a relapse in the immediate future, the risk of relapse must be balanced against the potential toxicities of the proposed therapy.

In this issue of the Journal of Pediatric Gastroenterology and Nutrition, Kader et al. ⁽¹¹⁾ suggest that 6-mercaptopurine may decrease the likelihood of postoperative recurrence. The authors retrospectively identified 10 patients who had undergone their initial surgery for a variety of complications of Crohn's disease, including medically unresponsive mucosal disease, fistulae, bowel necrosis, and appendiceal abscess. Five patients were begun on 6-mercaptopurine perioperatively in addition to medical therapy with 5-aminosalicylates and/or prednisone, while the other five control patients received salicylates and/or prednisone, but no mercaptopurine. Three of five patients not treated with 6-mercaptopurine developed clinical relapse within a mean of 4 months, while all five of the 6-mercaptopurine-treated group remain relapse-free for a mean duration of 33 months post-operatively. Although the number of patients in each group is small, survival analysis suggests a significantly improved outcome for the 6-mercaptopurine-treated group.

In a nonrandomized study such as this, it is critical to ascertain whether the untreated control and mercaptopurine groups are sufficiently similar to establish that the 6-mercaptopurine is the likely cause of the improved outcome. Both groups of children described are of comparable ages, and they had received similar medical therapy and had similar disease activity at time of operation. However, four patients in the mercaptopurine group had disease limited to the ileum and ascending colon, whereas three patients in the untreated group had more extensive colonic disease. The authors utilize Fisher's exact test to try to demonstrate no significant differences in disease involvement between groups; however, the small sample size limits the power of this analysis, and the low p value (0.067) certainly raises the question of whether there may be significant differences. Since diffuse ileocolonic disease increases relapse risk, and since two of the five control patients who relapsed had extensive colonic disease, it is possible that the apparently improved outcome of the 6-mercaptopurine-treated group may in part have been related to a more favorable pattern of disease distribution.

In addition, the surgeries performed in the two groups are different. Although all five mercaptopurine-treated patients underwent ileal or ileocecal resections, two of the five control patients underwent segmental colectomies; both these patients relapsed. Although the authors correctly state that there is no evidence in the medical literature supporting a worse outcome for patients receiving segmental colectomies, the number of segmental colectomies in the published studies is so small that direct comparisons of outcome with other surgeries are rarely made. Lastly, the three patients who relapsed in the untreated group had a rapid time to relapse. In the study of Griffiths et al., only 14 of 80 children relapsed within 1 year after surgery, again raising the question of whether the control group in the Kader study is somewhat atypical ⁽³⁾.

In spite of these issues, the observation the authors have reported in this study is potentially very important. The dramatic difference in the outcome of the two patient groups reported by Kader et al. suggests low dose (1mg/kg) 6-mercaptopurine may indeed have utility in postoperative recurrence prophylaxis, and at a minimum is well tolerated. This specific issue has not been directly addressed in the larger long-term follow-up studies of 6-mercaptopurine in adults with Crohn's ⁽¹²⁾. However, one study in adults did note that only one of eight patients treated with azathioprine after bowel resection relapsed ⁽¹³⁾. If 6-mercaptopurine were to indeed prove to be superior to salicylates or metronidazole for prevention of recurrence, the efficacy of the drug may outweigh the well-known risks, which occur in less than 10% of patients. Thus, the work of Kader et al., while preliminary, is provocative enough to warrant a randomized trial comparing 6-mercaptopurine to aminosalicylates in prevention of postoperative recurrence. Given the potential for adverse effects with mercaptopurine use, however, this preliminary study should not serve as a green light for the indiscriminate use of 6-mercaptopurine as postoperative recurrence prophylaxis.

Acknowledgment: Supported in part by a General Clinical Research Centers Clinical Associate Physician Award from the National Institute of Health, MOI RR02172.