Invasive Mould Infections (IMI) In Acute Leukemia Patients on Chemotherapy - Can We Win the Battle?

Abstract 4871

Aim: To determine risk factors for, and clinical features of, invasive mould infections in patients with acute leukemia on chemotherapy.

Methods: A matched case control study where acute leukemia patients on chemotherapy were identified and clinical data pertaining to their disease, clinical course, IMI characteristics and treatment outcomes were evaluated. This was conducted in a tertiary hospital from 1 January 2004 to 31^st March 2007.

Results: All acute leukemia patients on chemotherapy receive itraconazole prophylaxis. Nineteen cases of invasive mould infections (IMI) were found in the period, giving an incidence of 11.2%, using only patients undergoing curative chemotherapy as denominator. The incidence in patients receiving the first cycle of chemotherapy was 8.9%; in patients receiving subsequent cycles of chemotherapy was 4%. Thirteen cases were proven by EORTC/MSG criteria, 8 of them by histology without positive cultures. There were 3 proven cases of fusariosis, and 1 each of aspergillosis and nodulisporiosis. Cases were more likely than controls to have had an absolute neutropenia for at least 14 days in the chemotherapy episode in which the IMI was diagnosed (p=0.045), to have been bacteremic prior to the diagnosis of IMI (p=0.045), and to have had fever not responding to a carbapenem (p=0.016). They also had a longer length of stay (51.8 vs 27.2 days, p=0.01). The use of itraconazole as anti-fungal prophylaxis was not less common in cases than controls. Common clinical features noted in cases included fever not responding to a carbapenem (84.7%), cough (42.1%), a rise in alkaline phosphatase (36.9%), breathlessness (26.3%), and hemoptysis (21.1%). A pleural effusion was noted on CXR or CT chest in 52.8% of cases, and a characteristic CT thorax in 47.4% of cases. Sixty-eight percent of cases received amphotericin (conventional or lipid preparation) as part of a febrile neutropenia protocol, though only 15.8% received this agent when an IMI was considered. Clinical improvement was attributed to an increase in ANC in 26.3% of cases, and to the introduction of voriconazole in 10.5%. The next cycle of chemotherapy was delayed for 2 – 4 weeks in 15.8% of cases, and for >4 weeks in 31/6% of cases. Three patients (15.8%) died within three months of IMI diagnosis, 47.4% survived >1 year.

Conclusions: Anti-fungal prophylaxis with itraconazole did not reduce the likelihood of IMI and this should be changed. Absolute neutropenia is a risk factor for IMI in patients with acute leukemia. Fever not responding to carbapenems and a rise in ALP may be considered as red flags for clinicians to consider the possibility of an IMI and institute prompt treatment.