Candidemia is a significant cause of mortality among hospitalized patients, both worldwide and in Brazil. Prompt and appropriate treatment are essential to mitigate mortality, and clinical practice guidelines aim to optimize patient care based on the best scientific evidence. This study aims to examine the management of candidemia, assessing adherence to the guidelines of the Brazilian Society of Infectious Diseases in a single center located at São Paulo, Brazil. All adult patients hospitalized from 2016 to 2018 who presented one positive blood culture for Candida spp. were included. Electronic medical records were retrospectively reviewed to collect information relevant to the treatment for candidemia, in order to assess the adherence to the Brazilian guideline for the management of candidemia in relation to nine defined outcomes, and we correlated those findings with 30-day mortality by using uni- and multivariate analyses. A total of 115 patients were included; 68 patients (59.1%) were male, with a mean age of 55 years. C. albicans, C. tropicalis and C. glabrata were the most prevalent species. In total, 80 patients (69.5%) received antifungal treatment. The adherence to Brazilian guideline recommendations was determined as described in the following: initial treatment with echinocandin in 48 (60%); step-down to fluconazole in 21 (26.2%); collection of first control blood culture in 43 (58.9%); collection of second control blood culture, if the first one had been positive, in 14 (73.6%); treatment for 14 days after the first negative blood culture in 53 (65.4%); central venous catheter (CVC) removal in 66 (82.5%); CVC removal if the first control blood culture had been positive in 17 (89.4%); performance of a transthoracic echocardiogram in 51 (63.7%) and performance of a fundoscopy in 59 (73.7%). Univariate analysis showed that CVC removal and initial echinocandin therapy were more prevalent in the surviving group, but with no statistically significant difference. On the other hand, step-down to fluconazole demonstrated higher survival rate in the multivariate analysis OR 0.15 (95% CI 0.03–0.8); p = 0.02. The analysis of these nine recommendations demonstrates that it is necessary to improve adherence to specific recommendations and also disseminate strategies of the initial use of echinocandin as the drug of choice and addressing length of treatment and follow-up and complementary exams. Our study provides reassurance that the step-down to fluconazole is safe and may be recommended, if the preexisting conditions are present.