Candida species have only rarely been isolated from the cerebrospinal fluid. Some of these isolates are true pathogens with a high morbidity and mortality, while others are only colonizers or contaminants. Spontaneous recovery from Candida meningitis has also been reported. The purpose of this study was to identify factors indicative of patients positive for Candida from cerebrospinal fluid who should receive antifungal therapy. A total of 36 patients with a positive cerebrospinal fluid culture for Candida were included in this retrospective analysis. Seventeen of these patients had received antifungal therapy under the impression of Candida meningitis. The differences in the case fatality rate and the duration of hospitalization between the antifungal treatment and untreated groups were statistically significant (p<0.04 and p<0.005, respectively). Statistical analysis showed a significant difference in the percentage of patients with previous ventricular shunt (14/17 vs 8/19, p<0.05), central venous line in situ (9/17 vs 1/19, p<0.005), multiple positive cerebrospinal fluid culture (11/17 vs 1/19, p<0.0005), isolation of Candida at least 20 days after hospitalization in adults (6/7 vs 1/13, p<0.005), and cerebrospinal fluid white blood cell count (median 77 vs 2 /mm3, p<0.005). These results suggest that antifungal agent should be initiated promptly in patients whose cerebrospinal fluid is positive for Candida and who have one or more of the identified risk factors.