Candidemia Treatment is Improved by Infectious Disease Consultation: A Population-Based Cohort Study.

INTRODUCTION

Candidemia is a common bloodstream infection with morbidity and mortality. Both the European Society of Clinical Microbiology and Infectious Diseases and Infectious Diseases Society of America have evidence-based recommendations for its management. While infectious diseases consultation (IDC) reduced mortality, effects on guideline concordance remain unclear.

METHODS

We conducted a multicentre retrospective cohort study (2010-2018) in Canada to characterize the role of IDC in attaining evidence-based recommendations for adults with candidemia. Such recommendations include follow-up blood cultures to document blood clearance, prompt antifungal initiation and selection, source control, treatment duration determination, and ancillary investigations. We measured associations between IDC and adherence to evidence-based recommendations.

RESULTS

Patients with candidemia who received IDC were more likely not to be left without antifungal therapy (26% versus 4.8%). Echocardiograms were more common in noncritically ill patients with IDC (64% versus 44%; P < 0.01) and in all critically ill patients, irrespective of IDC status (75% versus 63%; P = 0.08). In patients with uncomplicated candidemia, treatment duration ≥14 days was similar with or without IDC (78% versus 70%; P = 0.15). Patients with complicated candidemia and IDC were more likely treated for ≥42 days (an appropriate duration; 100% versus 47%; P = 0.06). Fluconazole (transition therapy) was not affected by IDC for critically ill (41% and 29%; P = 0.34) or noncritically ill patients (34% versus 24%; P = 0.51). Mortality rates at 30 and 60 days after candidemia diagnosis were lower in patients receiving IDC (P < 0.05).

CONCLUSIONS

Best practice guidelines were issued for patients with candidemia; adherence varied, but involvement of an infectious disease physician increased implementation and was associated with considerably decreased mortality.