Epidemiology of candidaemia and antifungal susceptibility patterns in an Italian medical-surgical intensive care unit between July 2003 and June 2008

Abstract number: P1439

Azzini A.M., Boccafoglio F., Lo Cascio G., Concia E.

Objective: To evaluate epidemiological trends of candidaemia (CA) and to identify risk factors for it in critically ill patients; to define antimycotic susceptibility patterns inside our intensive care unit (ICU) in order to optimise both chemotherapy and prevention of drug resistance of this infection.

Methods: We retrospectively identified all cases of CA and collected their demographic/clinical/laboratory data, in particular the presence of medical devices and antibiotics/antimycotics administration within the 3 weeks before a CA episode. We also considered if CA was CVC-related. Antifungal susceptibility patterns were collected and empirical treatment's adequacy was evaluated as well.

Results: 58 CA were identified but only 51 were evaluable (all nosocomial, 80% ICU-acquired) The overall incidence of CA was 29.5/10000 patient-days, with a peak during 2004 (37.9/10000 patient-days) and 2005 (32.8/10000 patient-days). C. parapsilosis was isolated in 47% of CA, followed by C. albicans (29%), C. glabrata (12%), C. tropicalis (7%), C. guilliermondii (3%) and C. sake (2%). Non-albicans species remained the more frequent pathogens during 2004–2005–2006, only in 2007 we observed a reversal of trend. 100% of C. albicans and C. tropicalis were fluconazole-sensible (S), on the contrary C. parapsilosis resulted fluconazole-S in only 33% of cases, in 60% SDD and in 7% resistant (R). Most of R or SDD C. parapsilosis were isolated during 2005–2006. C. glabrata was fluconazole-S in 29% of cases and SDD in 71%. Most of patients had a medical device (i.e. 98% CVC), 96% of them was on antibiotic therapy within the 3 weeks before. 65% underwent surgery, particularly abdominal (79%). CVC-related CA were 59%, mainly due to C. parapsilosis. 76% of CA were treated. Fluconazole was employed in 54%, any amphotericin B formulation in 41% and caspofungin in 5%. Treatment was adequate in 74% of CA (inadequacy due to a delay in starting treatment or to posology). 30-days crude mortality was 62.5%, and 61.5% in patients adequately treated (p = 0.27). We didn't find risk factors statistically significant for mortality (table1).

Conclusions: We described an epidemic outbreak by C. parapsilosis in 2004–05, as demonstrated by DNA-fingerprinting. This epidemic was faced by the implementation of more strict infection control measures (hand-hygiene and CVC-management guide-lines) Because of the frequence of fluconazole-R or SDD C. parapsilosis we recommended to start CA treatment with an amphotericin B formulation.

Risk factorsOR95% CIP
Age (years)  0.7518
  geqslant R: gt-or-equal, slanted601.3000.256–6.610 
Gender  0.5995
Underlying disease  0.7435
  Solid neoplasia3.0000.642–14.023 
  Acute pancreatitis3.0000.423–21.297 
  Haematological neoplasia1.5000.106–21.312 
  Cardiopathy + Rheumatic disease + Polytrauma1.2000.164–8.799 
Ward of origin  0.3723
  Haematology + Emergency + Other hospital1.0670.183–6.212 
Diagnosis  0.7155
  Respiratory failure1.1430.250–5.224 
  Septic shock2.3570.485–11.452 
  Cardiogenic shock + Major trauma + Cardiac arrest1.7140.228–12.890 
CVC removed  0.0516
Concomitant bacteraemia  0.1005
Corticosteroids  0.5947
Candida albicans  0.3003

Session Details

Date: 16/05/2009
Time: 00:00-00:00
Session name: 19th European Congress of Clinical Microbiology and Infectious Diseases
Location: Helsinki, Finland, 16 - 19 May 2009
Presentation type:
Back to top