Abstract
Background
Invasive mold infections (IMIs) remain a significant cause of morbidity and mortality in patients with acute leukemia (AL) and those undergoing hematopoietic stem cell transplantation (HSCT). We describe the epidemiology of IMIs, the incidence of IMI in patients with acute myelogenous Leukemia (AML) post HSCT, and risk factors for mortality.
Methods
Patients were identified using ICD9 and ICD10 codes using a University of Kansas internal database from 2009-2019, microbiology records, and an AML HSCT database and were followed through May 1st, 2020. Patients’ electronic medical records were reviewed for inclusion. IMI was defined as proven or probable using the 2009 National Institute of Allergy and Infectious Diseases Mycoses Study Group (MSG) guidelines. Incidence was calculated as IMI cases/100-person-years. Risk factors for overall mortality were evaluated using a Cox regression model.
Results
We included 138 patients: 79 developed IMI after HSCT (8 autologous, 71 allogeneic) and 59 developed IMI after AL diagnosis. Seventeen of the AL patients underwent HSCT after IMI diagnosis (12 within 100 days of IMI). Proven IMI occurred in 45 (32.6%) and probable IMI occurred in 93 (67.4%) patients. The most common prophylactic agent prior to IMI diagnosis was fluconazole (31.2%), with 21.0% receiving none. Aspergillus was the most commonly identified mold with 91 (65.9%) cases. The average treatment duration was 101 (range 0 - 799) days. The incidence of IMI in patients with AML who underwent HSCT was 2.35 cases/100 person-years. All-cause mortality among patients with AL or HSCT who developed IMI was 23.1% at 6 weeks, 34.1% at 12 weeks, and 61.2% at 1 year. On univariate Cox model, Karnofsky performance status > 70 was associated with lower mortality (hazard ratio (HR) 0.317, 95% confidence interval (CI) [0.110, 0.914]) among HSCT recipients. ICU admission within 7 days prior to IMI diagnosis (HR 6.469, 95% CI [1.779, 23.530]) and each one point increase in BMI (HR 1.051, CI [1.001, 1.103]) were associated with increased mortality in the AL group.
Figure 1 - Invasive mold infections by pathogen in HSCT-recipients and acute leukemia patients from 2009-2019.
Figure 2 - Antifungal prophylactic agents prescribed for at least one week at time of IMI diagnosis
Table 1 - Univariate survival analysis calculated using a Cox proportional-hazards regression model among patients who developed IMI after HSCT and patients who developed IMI after acute leukemia diagnosis
Conclusion
IMIs are associated with significant mortality in HSCT recipients and AL patients; patients at higher risk for mortality include those with lower baseline Karnofsky scores, recent ICU admissions, and higher BMI at time of IMI diagnosis.
Disclosures
Wissam El Atrouni, MD, ViiV (Advisor or Review Panel member)
Clinical Problem-Solving: Fever and Rapidly Progressive Weakness in an Immunocompromised Patient
