Clinical presentation and diagnosis of intestinal adenocarcinoma in Crohn's disease: analysis of clinical predictors and of the life-time risk.

BACKGROUND

Late diagnosis of cancer in CD often occurs, and the prognosis is poor. The primary aim of this study was to assess the relationship between clinical presentation and diagnosis of intestinal adenocarcinoma in CD; the secondary aim was to evaluate the timing of cancer occurrence in CD patients.

PATIENTS AND METHODS

Medical records of 12 consecutive patients with intestinal adenocarcinoma in CD and of 79 consecutive CD patients undergoing bowel surgery were reviewed. Presentation symptoms were analyzed as possible predictors. Timing of intestinal adenocarcinoma occurrence in patients with CD was analyzed including all the 347 consecutive patients that had undergone surgery for CD in our institute from January 1984 to June 2008. Life table analysis and uni/multivariate analyses were performed.

RESULTS

Ten men and two women underwent surgery for intestinal cancer in CD with a median age of 50 years (31-68). Carcinomas were localized in the terminal ileum in four cases, right colon in three, transverse colon in one, sigmoid colon in one, rectum in two, and an anorectal fistula in one. Only three patients were pre-operatively diagnosed with cancer. At multivariate analysis only age (OR 1.057 (95% CI 0.999-1.107), p = 0.05) and obstruction (OR 6.530 (95% CI 1.533-27.806), p = 0.01) significantly predicted cancer diagnosis. The risk rate (RR) for cancer occurrence started to rise at the end of the third decade of life (RR = 0.005). The analysis of risk rate for cancer occurrence during overt CD showed that it is initially high at onset (RR = 0.001) and after two other peaks at 150 months from onset, it began to rise again. The presence of Crohn's colitis was associated to a significant risk of cancer (HR = 4.790, p = 0.009) while the use of 5-ASA resulted to be a protective factor against cancer occurrence (HR = 0.122, p = 0.013).

DISCUSSION

In CD, rectal bleeding, the most common alarm symptom for intestinal cancer, is not useful for an early diagnosis. CD patients presenting with an older age and obstruction should be thoroughly investigated to rule out neoplastic lesions. There is probably no safe interval of CD where surveillance for intestinal cancer can be omitted. In the meantime, even in absence of active disease, all CD patients should undergo therapy with 5-ASA.