Aim: Data on recurrent fever in children is limited. The aim of this study is to evaluate the children with this common symptom. Material and Method: We enrolled 138 patients with frequent fever. Twelve febrile episodes/year was determined as the cut-off for "recurrent fever". Children with <= 12 and >12 febrile episodes/year were included in Group I and Group II, respectively. Their demographic data, symptoms, and clinical and laboratory findings were compared. All children were followed for two years for definitive diagnosis. Results: Fifty-four (39.1%) children experienced recurrent fever according to our criteria. Group I children experienced more frequent sneezing, whereas Group II children experienced more frequent vomiting compared to the other group (p=0.05 and p=0.02, respectively). Febrile episodes were seen more frequently in the winter season in Group I compared to Group II (p=0.03). Age, sex, other clinical symptoms, physical examination findings, and attendance at day care center/school in two groups were not different. Group II children had higher C-reactive protein levels (p=0.001). There was no difference in other laboratory parameters between the two groups. After two years, two patients in Group II were diagnosed with Familial Mediterranean Fever syndrome and immunoglobulin-G2 subgroup deficiency, while one patient in Group I was diagnosed with periodic fever, aphtous stomatitis, pharyngitis, and adenitis syndrome. Discussion: To investigate only those children presenting with fever episodes of at least once a month may not be discriminative. To construct an algorithm for diagnosis and treatment, it would be better to follow recurrent fever patients for a longer duration rather than initially categorizing them according to the number of episodes.