PURPOSE: To use magnetic resonance (MR) imaging to document the appearance of perianal infection in patients with a hematologic malignancy (HM) compared with that in immunocompetent control patients. MATERIALS AND METHODS: After an ethical waiver was obtained, 38 patients with an HM were matched by age and sex to 38 control patients with no history of immunocompromise or Crohn disease. Both groups had undergone MR imaging for perianal symptoms and/or systemic sepsis. Two radiologists who were blinded to the diagnosis independently reviewed the MR images and recorded the size and distribution of abscesses and/or fistula tracts, the extent of perianal edema, and the likely diagnosis. Groups were compared by using the Mann-Whitney-Wilcoxon, χ(2), or Fisher exact test. Receiver operating characteristic (ROC) curves were constructed to estimate the ability of MR imaging to help distinguish patients with an HM from control patients. RESULTS: Patients with an HM had significantly greater perianal edema than did control patients (mean arc angle of anal canal involved, 220° vs 60°; P < .001). However, they had significantly lower rates of fistula (15 [39.5%] vs 35 [92.1%] of 38; P < .001). Abscesses were similar in frequency (10 [26.3%] vs 17 [44.7%] of 38; P = .15) and were unrelated to the degree of neutropenia (P = .71) or the use of chemotherapy (P = .10). Surgical treatment was rarely required in patients with an HM, either during the acute illness (four [10.5%] of 38) or thereafter (three [7.9%] of 38). MR imaging had an excellent ability to help discriminate patients with HM from immunocompetent patients (areas under the ROC curve, 0.91 and 0.97). CONCLUSION: Perianal infection in patients with an HM is more likely to cause diffuse perianal edema and is less likely to cause fistulas than in immunocompetent patients. MR imaging can help distinguish patients with an HM from those without immunocompromise.
PURPOSE: To use magnetic resonance (MR) imaging to document the appearance of perianal infection in patients with a hematologic malignancy (HM) compared with that in immunocompetent control patients. MATERIALS AND METHODS: After an ethical waiver was obtained, 38 patients with an HM were matched by age and sex to 38 control patients with no history of immunocompromise or Crohn disease. Both groups had undergone MR imaging for perianal symptoms and/or systemic sepsis. Two radiologists who were blinded to the diagnosis independently reviewed the MR images and recorded the size and distribution of abscesses and/or fistula tracts, the extent of perianal edema, and the likely diagnosis. Groups were compared by using the Mann-Whitney-Wilcoxon, χ(2), or Fisher exact test. Receiver operating characteristic (ROC) curves were constructed to estimate the ability of MR imaging to help distinguish patients with an HM from control patients. RESULTS:Patients with an HM had significantly greater perianal edema than did control patients (mean arc angle of anal canal involved, 220° vs 60°; P < .001). However, they had significantly lower rates of fistula (15 [39.5%] vs 35 [92.1%] of 38; P < .001). Abscesses were similar in frequency (10 [26.3%] vs 17 [44.7%] of 38; P = .15) and were unrelated to the degree of neutropenia (P = .71) or the use of chemotherapy (P = .10). Surgical treatment was rarely required in patients with an HM, either during the acute illness (four [10.5%] of 38) or thereafter (three [7.9%] of 38). MR imaging had an excellent ability to help discriminate patients with HM from immunocompetent patients (areas under the ROC curve, 0.91 and 0.97). CONCLUSION: Perianal infection in patients with an HM is more likely to cause diffuse perianal edema and is less likely to cause fistulas than in immunocompetent patients. MR imaging can help distinguish patients with an HM from those without immunocompromise.