Objective: To describe outcomes of septic shock and multiple organ failure arising from candidemia.
Design: Secondary cohort analysis of data from the placebo arm of the North American Septic Shock Trial (NORASEPT II), the largest prospective, randomized, double-blind, controlled multiple center study of septic shock conducted to date, with predetermined end point analysis of outcomes.
Setting: Adult intensive care units in 105 hospitals in the United States and Canada.
Subjects: A cohort of ten purely candidemic patients in septic shock were compared with a cohort of 376 purely bacteremic patients in septic shock. Patients were not immunocompromised, because patients on corticosteroids, with neutropenia, or posttransplantation were excluded from enrollment in NORASEPT II.
Measurements and main results: Demographic variables, baseline characteristics, 28-day mortality rates, and multiple organ failure were compared for the two cohorts. Candidemic patients were more likely to have a history of underlying renal failure at baseline and to require dialysis at onset of septic shock. Both causes of septic shock are associated with an extremely high severity of illness (Acute Physiology and Chronic Health Evaluation II: candidemic septic shock, 32 +/- 10; bacteremic septic shock, 30 +/- 8; p =.44). More than 70% of patients with candidemia and septic shock were in multiple organ failure at days 3, 7, and 14; patients with candidemic septic shock sustained persistent multiple organ failure and showed delayed recovery from multiple organ failure compared with patients with bacteremic septic shock. Mortality rate at 28 days was 60% in candidemic septic shock and 46% in bacteremic septic shock (p =.38).
Conclusions: Candidemia with septic shock is infrequent in nonimmunocompromised patients but has a very high mortality rate, a high likelihood of associated multiple organ failure, and possibly a delayed recovery from multiple organ failure. Patients with candidemic septic shock are more likely to have underlying renal failure at baseline.