Impact of the Implementation of a Fever and Neutropenia Management Algorithm in the Detection of Microbiologically Confirmed Infections in Mexican Children

Abstract

Abstract Background Fever and neutropenia is a medical emergency in children with cancer. Microbiological identification of the febrile focus is estimated to occur in 20-30% of cases. Algorithms that implement evaluation and treatment strategies help institutions to standardize the appropriate clinical and microbiological diagnosis and management of these episodes, thus improving the rate of confirmed infections. In this study we aimed to report the impact in the identification rates of microbiologically confirmed infections before and after implementing a management algorithm for children with fever and neutropenia which included obtaining blood and urinary cultures, antibiotic administration in the first hour upon arrival, and the evaluation of patients by a pediatric infectious diseases team. Methods We performed an ambispective cohort study including children aged 1 to 16 years, with ≥1 episodes of fever and neutropenia admitted to Hospital Universitario “Dr. José Eleuterio González” to assess changes in microbiological identification before and after the implementation of a management algorithm. Demographic and clinical data of episodes from February 1st, 2017 to December 31st, 2021 were gathered information from clinical records. Qualitative variables were expressed in frequency and percentages, quantitative variables in mean and standard deviation or median and interquartile range. Results A total of 343 episodes were recorded, 122 (35.6%) pre-algorithm, and 221(64.4%) post-algorithm. Microbiologically confirmed infection occurred in 18 (14.7%) episodes pre-algorithm vs 79 (35.7%) post-algorithm. The most frequently microbiologically confirmed infection pre-algorithm was bacteremia (n=10, 8.2%), and viral respiratory tract infections (n=36, 16.3%) in the post-algorithm group. Of those children with bacteremia in the pre-algorithm group, 50% (n=13) were Gram-positive bacteria, and Coagulase-negative Staphylococci were the most prevalent (n=4, 40%). Bacteremia in the post-algorithm group was reported in 23 (10.4%) episodes, of which 56.3% (n=13) were Gram-negative bacteria, and Escherichia coli was the most prevalent (n=5, 21.7%). Respiratory viral infections were reported in 1 (0.8%) in the pre-algorithm vs 36 (16.3%) in post-algorithm group. Influenza viruses accounted for most respiratory infections in the post-algorithm group 33.3% (n=12), and co-infection with 2 or more viruses was reported in 8 (22.2%) isolations. Conclusions The implementation of a fever and neutropenia management algorithm increased the rate of microbiologically confirmed infections in children, also increasing the rate of detection of clinically confirmed infections other than bacteremia, such as urinary, respiratory, and invasive fungal infections. Viral respiratory infections accounted for most of post-algorithm episodes, and its detection was greatly improved by the introduction of the management algorithm. TABLE 1. RATES OF MICROBIOLOGICALLY CONFIRMED INFECTIONS PRE AND POST-ALGORITHM