Abstract
Background
Cancer is one of the leading causes of death in children in Mexico. Infections are the main cause of morbidity and mortality in these patients. Febrile neutropenia (FN) constitutes an infectious emergency and early aggressive antibiotic treatment is the standard of care. Recent guidelines suggest discontinuing empirical antibiotics in patients who have negative blood cultures at 48 hours, who have been afebrile for at least 24 hours, and who have evidence of marrow recovery. Nevertheless, recommendations about discontinuing antibiotics and discharging patients while they are still neutropenic are less clear. We aimed to evaluate the safety of early hospital discharge of FN patients who are still neutropenic.
Methods
Observational, case–control study nested in a prospective cohort of pediatric oncology patients with FN at Hospital Infantil de México Federico Gómez (HIMFG) in Mexico City from May 2015 to September 2017. We defined early discharge as when a patient is discharged while neutropenic (ANC <500 cell/mm3) and has completed at least 7 days of antibiotics. Patients with FN who were discharged with neutropenia were defined as cases and patients with FN who were discharged after recovering from neutropenia were controls. To assess the safety of hospital early discharge, the following outcomes were analyzed until 7 days after discharge: new onset of fever, hospital readmission, need to restart antibiotic treatment, septic shock, and death. Descriptive statistics were performed with measures of central tendency. Variables of interest were compared with Pearson’s χ 2 or Student’s test.
Results
In total, 929 febrile neutropenia episodes were analyzed. The mean age was 7.5 years, 55.3% were female. Hematologic malignancies were the most frequent type of malignances in 50.8%. Acute lymphoblastic leukemia (ALL) was the underlying disease in 41%. Of the 929 FN episodes, 180 (19.3%) were discharged with neutropenia. Patients with ALL were the most frequent in 49.4%, followed by acute myeloid leukemia 18.8% and rhabdomyosarcoma 6.6%. Thirty-five percent were in maintenance therapy, 22% in remission induction therapy, and 9% in consolidation. 19.4% of discharged patients received granulocyte-colony stimulating factor. Ten patients (5.5%) were re-admitted during the 7 days following discharge. Six patients returned for chemotherapy administration and one was scheduled for liver biopsy. Three patients were re-admitted due to infectious complications (1.6%), none of them were under oral antibiotic treatment; two patients due to FN without microbiological isolation and one patient with septic shock due to multi-drug-resistant Pseudomonas aeruginosa. Older patients had a higher risk of readmission, with a mean age of 14.6 years (SD 4.6 years, 95% CI 7.7–21.6) (P = 0.01), compared with the mean of 7.7 years (SD 2.7 years, (95% CI 7.0, – 8.4) of patients who were not re-admitted.
Conclusions
In our population of pediatric patients with FN who were discharged before neutrophil recovery, readmission due to infectious complications was low (1.6%). Discharging patients with persistent neutropenia who are afebrile and had completed a course of antibiotics seems an acceptable practice with a low risk of readmission.
Oral antibiotics with early hospital discharge compared with in-patient intravenous antibiotics for low-risk febrile neutropenia in patients with cancer: a prospective randomised controlled single centre study