Abstract
BACKGROUND: Febrile neutropenia (FN) is a serious chemotherapy-related complication in patients with solid tumors. Identifying simple and accessible biomarkers that can predict prolonged hospitalization may support early risk stratification and clinical decision-making. METHODS: This retrospective study included 169 adults hospitalized with FN between January 2023 and January 2025. Immunonutritional indices, including the Hemoglobin-Albumin-Lymphocyte-Platelet (HALP) score, the Prognostic Nutritional Index (PNI), and the C-reactive protein/albumin ratio (CAR), as well as the Clinical Index of Stable Febrile Neutropenia (CISNE) score were calculated. HALP and PNI were categorized using ROC-derived cut-offs based on the Youden Index. Prolonged hospital stay was defined as a binary variable based on the cohort median (>9 days). Spearman correlation, univariate and multivariate logistic regression were performed to identify predictors of prolonged hospitalization. RESULTS: HALP showed a significant negative correlation with hospitalization duration (r = -0.469; p < 0.001), as did serum albumin (r = -0.184; p = 0.017) and PNI (r = -0.273; p < 0.001). CAR (p = 0.617) and neutrophil count (p = 0.955) demonstrated no correlation. In univariate logistic regression, low HALP (p < 0.001), low PNI (p = 0.001), intermediate CISNE (p = 0.002), high CISNE (p < 0.001), microbiological culture positivity (p < 0.001), and sex (p = 0.015) were significantly associated with prolonged hospitalization. Age, comorbidity status, metastatic stage, and CAR were not significant. In the multivariate model, low HALP (p < 0.001), intermediate CISNE (p = 0.007), high CISNE (p < 0.001), and culture positivity (p < 0.001) remained independent predictors. PNI (p = 0.400) and sex (p = 0.176) did not retain significance. CONCLUSIONS: A Low HALP score, higher CISNE risk categories, and microbiological culture positivity independently predicted prolonged hospitalization in FN. HALP, as a simple and inexpensive immunonutritional marker, may enhance early FN risk assessment when used alongside validated clinical tools such as CISNE or MASCC.