Ciljevi: Identificirati čimbenike rizika za ponovljenu epizodu neutropenijske vrućice (NV) u djece s malignim bolestima te izraditi prediktivni model za ponovljenu epizodu NV s adekvatnom specifičnošću i senzitivnošću u djece s malignim bolestima. Ispitanici i metode: U retrospektivnu studiju uključene su 182 epizode NV u 45 djece s malignim bolestima koja su liječena na Odjelu za hematologiju i onkologiju Klinike za pedijatriju Kliničkog bolničkog centra Rijeka u razdoblju od 1.1.2008. do 31.12.2016. Kriterij uključenja u studiju bila je barem jedna epizoda NV. Povezanost ponovljene epizode NV i njene težine s relevantnim kliničkim i laboratorijskim parametrima analizirana je putem logističke regresije. Rezultati: Trideset osam (84%) pacijenata je imalo ponovljenu epizodu NV, od toga dvadeset pet (56%) tešku epizodu NV. Značajni čimbenici predikcije za ponovljenu epizodu NV su trajanje prve epizode NV 9 ili više dana te broj eritrocita ≤3.0×1012/L. Prediktivni model deriviran iz ova dva čimbenika ima točnost od 87% (95% interval pouzdanosti [CI] 73%-94%), senzitivnost od 82% (95% CI 53%-97%) i specifičnost od 88% (95% CI 79%-93%). Zaključak: Trajanje prve epizode NV i anemija su značajno povezani s rizikom za ponavljanje epizode NV. Ovi prediktori su korisni u identifikaciji pacijenata visokog rizika za neželjene događaje koji imaju korist od brze intervencije Objectives: To identify risk factors of a repeated episode of severe febrile neutropenia (FN) in children with malignant diseases and to create an accurate and easy-to-use predictive model. Materials and methods: This retrospective cohort study conducted at the Division of Hematology and Oncology, Department of Pediatrics, Clinical Hospital Center Rijeka, from January 1, 2008 to December 31, 2016. included 182 episodes of FN in 45 children with malignant diseases who experienced at least one FN episode. The association of the second severe FN episode appearance with relevant clinical and laboratory data was analyzed by logistic regression. Results Thirty eight patients (84%) had repeated FN, of which twenty five (56%) had severe FN. Significant predictors of a repeated FN episode were duration of the first FN episode of 9 or more days and red blood cells ≤3.0×1012/L. The predictive model constructed by crossing these two indicators had the accuracy of 87% (95% confidence interval [CI] 73%-94%), sensitivity of 82% (95% CI 53%-97%), and specificity of 88% (95% CI 79%-93%). Conclusion The duration of the first FN episode and anemia are significantly associated with the risk for FN re-occurrence. These factors may be useful in the identification of children with cancer who are at high risk for adverse outcome and may benefit from rapid intervention.