Siri Kurland: Invasive Candida infections: Treatment of the critically ill and patients with infective endocarditis

Abstract

Invasive Candida infections have a major impact on morbidity and mortality in critically ill patients. Prompt initiation of effective antifungal therapy and source control are cornerstones in management. The echinocandin caspofungin is first-line treatment in invasive Candida infections and dosage is based on the Child-Pugh liver scoring system, developed for patients with chronic liver disease. Pathophysiological changes in the critically ill, e.g. hypoalbuminemia, may affect antifungal pharmacokinetics (PK) and drug exposure, calling into question if dosage regimens are adequate for this group. Candida infective endocarditis (CIE) is a rare but serious form of deep-seated Candida infection, burdened with high mortality rates. The standard of care has been antifungal therapy combined with cardiac valve surgery. 

The overall aim of this thesis was to study aspects of invasive Candida infections in relation to caspofungin pharmacokinetics and pharmacodynamics (PK/PD) and the management of patients with CIE.

In a prospective study of critically ill patients, we investigated the prevalence of hepatic impairment, the effect of Child-Pugh score on caspofungin PK and whether caspofungin PK/PD targets were achieved. The prevalence of patients with pathological liver function tests was high, but the Child-Pugh score did not significantly impact caspofungin PK. The PK/PD target was reached, but with small margins. No dose reduction is warranted in critically ill patients with hepatic impairment, in the absence of chronic liver disease.

The impact of plasma protein levels on the antifungal activity of caspofungin at clinically relevant concentrations was studied in time-kill experiments. Reduced plasma protein levels in vitro increased caspofungin’s fungicidal effect on Candida glabrata. 

In a retrospective observational study, we compared the effect of echinocandins, mainly caspofungin, on the outcome in patients with CIE with those of other antifungal regimens. We found no significant differences in outcome between regimens. 

Further investigation of the CIE cohort showed that Candida parapsilosis was overrepresented in CIE compared with candidemia, indicating a higher propensity to cause CIE. In the overall CIE cohort, surgery did not improve outcome. Therefore, the recommendation of surgery in all patients with CIE may be questioned. Long-term suppressive antifungal therapy reduced the number of relapses.