Citations 0
Invited Critique
July 2002

Perianal Crohn Disease—Invited Critique

Author Affiliations

Copyright 2002 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2002

Arch Surg. 2002;137(7):778. doi:10.1001/archsurg.137.7.778

Anorectal problems afflict one fifth of all patients with Crohn disease. Some series report 80% to 90% surgical success rates in selected patients,14 but postoperative complications substantially impair some patients.57 Thus, patient selection is critical in deciding whether to be aggressive or to perform conservative procedures such as simple drainage or seton placement. Pikarksy and coauthors propose that a simple score might predict surgical outcome. Prospectively collected data and an a priori scoring system yielded a dichotomous distribution that correlated with postoperative outcome when the results were analyzed retrospectively. The study group was small and heterogeneous regarding surgical indications and procedures. Furthermore, half of the patients underwent "conservative procedures" and therefore do not help test the hypothesis that these scores predict whether an aggressive procedure is likely to be successful. (Although potentially corrective, fibrin glue instillation does not carry the same level of postoperative risk as the more aggressive procedures described.) Concomitant medical therapy for Crohn disease is not discussed but is of particular interest because anti–tumor necrosis factor α antibody facilitates closure of some fistulas. Outcomes were categorized subjectively by surgeons, not by patients or by objective criteria. Nevertheless, half of the patients selected clinically for more aggressive procedures had poor outcomes that might have been predicted by their scores. Conversely, 6 of 16 patients selected for conservative procedures had low scores. We do not know whether these patients would have profited from more aggressive procedures or whether the score simply supplements clinical intuition. Because the proposed score thresholds were derived retrospectively, they should be validated prospectively in a larger series, and their superior predictive value to clinical intuition and other scoring systems should be tested. If this testing succeeds, this score may prove useful in reporting and comparing published clinical trials or in cueing surgeons that certain patients might or might not profit from aggressive therapy despite contrary clinical intuition.

First Page Preview View Large
First page PDF preview
First page PDF preview