Preemptive therapy in nonneutropenic patients with Candida infection using the Japanese guidelines.

BACKGROUND

The Japanese Guidelines for the Diagnosis and Treatment of Deep-Seated Mycosis were established in 2003. Proven Candida infection (CI) is defined as at least one positive blood culture yielding a Candida species. Clinically documented CI requires documentation of more than 2 sites of colonization and a positive plasma beta-D-glucan test. Possible CI is diagnosed by one of the above criteria in febrile, nonneutropenic critically ill patients.

OBJECTIVE

To assess the use of definitions of clinically documented and possible CI for guiding preemptive antifungal therapy in critically ill patients.

METHODS

The patients treated in our intensive care unit (ICU) for at least 48 hours between 2000 and 2004 were investigated. The administration of antifungal agents and ICU mortality were compared among proven, clinically documented, and possible CI groups for age, sex, APACHE II score, diagnosis, length of ICU stay, treatment, number of colonization sites, and plasma beta-D-glucan level.

RESULTS

Six patients were diagnosed with proven CI, 25 were diagnosed with clinically documented CI, and 104 with possible CI. The patients with clinically documented CI were compared with those with possible CI, and statistically significant differences were found in the following variables: APACHE II score (p = 0.018), length of ICU stay (p < 0.01), use of ventilator (p = 0.027), tracheotomy (p = 0.027), number of colonization sites (p < 0.001), plasma beta-D-glucan level (p < 0.001), and administration of antifungal agents (p < 0.001); incidence of mortality was not statistically significant (p = 0.33). The shorter length of ICU stay, use of ventilator, and continuous hemodiafiltration were risk factors for death after adjusting for APACHE II score, admission before/after 2003, antifungal therapy, and other factors. Although the frequency of the administration of preemptive antifungal therapy was higher after 2003 than before, the mortality rate did not differ significantly.

CONCLUSIONS

The use of the definitions of clinically documented and possible CI may be beneficial for determining when it is appropriate to initiate preemptive antifungal therapy. However, use of the guidelines did not lead to prevention of possible CI proceeding to clinically documented CI or to improved mortality.