Background: Despite a significant increase in fungal blood stream (BSI) infections in India, there is paucity of data on regional prevalence of major fungal species, and risk factors for this infection. We describe the epidemiology and predictors of fungal BSI in a tertiary care center in Kerala, India with a novel antimicrobial stewardship program in place.
Methods: Data on adult inpatients who had at least one positive fungal culture from blood samples were collected from electronic medical records over a period of 48 months (January 2012- December 2015). Year wise epidemiology and risk factor characterization of fungal BSI were done using ꭓ2 method.
Results: 219 fungal BSI were identified with incidence of 1.08 cases/1000 patients and there was a 15% decrease over the 4 year period. There was a 300% increase in fungal BSI in patients older than 80 years.Candida was the most common cause of fungal BSI (92%), with a 100% increase in incidence of C.glabrata and C.haemolunii, and a 45% decrease in C.parapsilosis seen over the 4 year period. Community-acquired fungal BSI increased by 700% while hospital-associated infections dropped by 29%. 23% decrease in inappropriate antifungal treatment was observed from 2012 to 2015. Isolates reflected a 71% increase in resistance to amphotericin B and a 114% increase in fluconazole resistance. 31% reduction in all-cause mortality was seen in the cohort over the study period. Among the risk factors for fungal BSI, ICU stay, use of urinary catheter, surgery, neutropenia and diabetes decreased while prior antibiotic use and steroid use significantly increased over the years (p<0.05). Predictors of mortality included male gender, prior use of antibiotics, ICU stay, use of ventilator, chemotherapy, chronic liver disease, hypertension, presence of Candida parapsilosis, and inappropriate therapy (p<0.05).
Conclusion: A significant shift in fungal BSI epidemiology was observed in our center with increase in overall antifungal resistance. Antimicrobial stewardship and infection control programs may have contributed to reduced mortality and reduced hospital-associated infections.
F. E T, None
S. Singh, None
K. S. Kaye, None
P. Patel, None