Perianal infections in patients with leukemia: importance of the course of neutrophil count.

PURPOSE

This study was performed to evaluate relations among neutrophil count (including its course), type of lesion, treatment, and prognosis in patients with leukemia and perianal infection.

METHODS

Medical records of patients with acute and chronic leukemia who were followed during the last five years were reviewed retrospectively.

RESULTS

The incidence of perianal infections was found to be 7.3 percent in 259 patients with acute leukemia. Only 1 of 108 patients with chronic leukemia suffered from this problem. Twenty percent of all patients with this complication died as a result of sepsis. Perianal abscess was the sole and obligatory indication for surgical treatment in our patients. There were ten patients in each treatment group. The operative group had better results (9 cures, 1 complication vs. 3 cures, 7 complications). However, median neutrophil count at diagnosis was notably higher in the operative group 1,280/mm3 vs. 96/mm3; P = 0.075). Also, significantly more frequent abscess formations and, consequently, operative treatments were performed in patients with a period of normal neutrophil counts during the infection compared with continuously neutropenic patients (9 operative, 4 nonoperative vs. 1 operative, 6 nonoperative; P = 0.057). Ten cures, three complications vs. two cures, five complications (3 mortalities) were present in patients with and without normal neutrophil counts, respectively (P = 0.062). When only severely neutropenic patients were considered, four patients in the surgery group had normal neutrophil counts before or shortly after surgery. However, only two of eight patients with perianal cellulitis had normal counts during full-course infection (P = 0.06).

CONCLUSIONS

The course of the neutrophil count during infection was an important factor affecting the perianal lesion, and indirectly, choice of treatment and prognosis. A period of normal counts during infection usually led to well bordered and fluctuant lesions, and the prognosis was acceptable with operative treatment. However, continuously neutropenic patients developed nonfluctuating indurations. We found disappointing results with nonoperative treatment of such patients. In all studies, regarding treatment of perianal infections in neutropenic patients, the course of the neutrophil count and indications for surgery should be clarified to get reliable results.