Factors Associated With Complications in Patients With Hematological Malignancies and Febrile Neutropenia: A Cohort Study

Introduction Febrile neutropenia (FN) in patients with hematological malignancy (HM) is associated with multiple hospital complications including mortality. Although different strategies for early detection and prompt treatment have been established, it is a heterogeneous population with risk factors that are difficult to detect. The data available on the prediction of such complications is limited and there lies the importance of characterizing this type of patients in our environment and evaluating the factors related to the adverse outcomes. Methods The study is a retrospective cohort study conducted at San Vicente Foundation University Hospital (HUSVF) and Alma Mater Hospital of Antioquia (HAMA) in Medellín, Colombia, between January 2018 and December 2020, including patients diagnosed with FN who presented FN at the time of diagnosis or up to 30 days after receiving chemotherapy. The main objective was to determine the factors related to mortality and severe complications (ICU admission, need for vasopressors, or need for mechanical ventilation), while the secondary objective was the microbiological characterization of this population. Results Of the 190 FN episodes, 134 (70.5%) had a clinical focus of infection. A causal agent was identified in 125 episodes (65.8%), with the majority being bacteria in 112 cases (92.6%) of the isolates. The most frequently identified bacteria were <i>Klebsiella pneumoniae</i>, <i>Escherichia coli</i>, <i>Pseudomonas aeruginosa</i>, and <i>Staphylococcus aureus</i>. Gram-negative bacilli were isolated in 85 (86%) cases, and resistance was present in 38 cases (44.7%), with both extended-spectrum beta-lactamase (ESBL) and <i>Klebsiella pneumoniae</i> Carbapenemase (KPC) detected in nine (10.5%). In 53 (34.4%) episodes, some complications occurred during FN. The 30-day all-cause mortality was 53 (34.4%), with 27 (50.9%) of these cases associated with complications. Of the 45 (86.5%) patients who died from any cause, all did so during their first episode of FN. In the bivariate analysis, the following factors were associated with higher mortality: hypertension (OR 2.58, 95% CI 1.19-5.58; p=0.014), chronic obstructive pulmonary disease (COPD) (OR 10.2, 95% CI 1.11-93.8; p=0.013), chronic kidney disease (OR 4.27, 95% CI 0.975-18.7; p=0.038), prolonged neutropenia (OR 2.34, 95% CI 1.1-4.95; p=0.024), and lactate dehydrogenase (LDH) levels greater than two times the upper normal limit (UNL) (OR 3.24, 95% CI 1.35-7.75; p=0.007). In contrast, normal albumin levels before chemotherapy were associated with lower mortality (OR 0.381, 95% CI 0.15-0.95; p=0.036). In the multivariate analysis, none of the identified factors were statistically significant in predicting complications or mortality. Conclusion No factors related to complications or mortality were found in the multivariate analysis. However, the heterogeneity of the population suggests that these outcomes are not determined by a single factor, and a study with a larger sample may be needed to confirm them.