SAN DIEGO – Wait at least 2 days before replacing central venous catheters (CVC) in patients with catheter-associated candidemia, according to the results of a single-center retrospective cohort study of 228 patients.
Waiting less than 2 days to replace a CVC increased the odds of 30-day mortality nearly sixfold among patients with catheter-related bloodstream infections due to candidemia, even after controlling for potential confounders, Takahiro Matsuo, MD, said at an annual scientific meeting on infectious diseases. No other factor significantly predicted mortality in univariate or multivariate analyses, he said. “This is the first study to demonstrate the optimal timing of central venous catheter replacement in catheter-related [bloodstream infection] due to Candida.”
Invasive candidiasis is associated with mortality rates of up to 50%, noted Dr. Matsuo, who is a fellow in infectious diseases at St. Luke’s International Hospital, Tokyo. Antifungal therapy improves outcomes, and most physicians agree that removing a CVC does, too. To better pinpoint optimal timing of catheter replacement, Dr. Matsuo and his associates examined risk factors for 30-day mortality among patients with candidemia who were treated at St. Luke’s between 2004 and 2015.
Amy Karon/Frontline Medical News
Dr. Takahiro Matsuo
Among 228 patients with candidemia, 166 had CVCs, and 144 had their CVC removed. Among 71 patients who needed their CVC replaced, 15 died within 30 days. Central venous catheters were replaced less than 2 days after removal in 87% of patients who died and in 54% of survivors (P = .04). The association remained statistically significant after the researchers accounted for potential confounders (adjusted odds ratio, 5.9; 95% confidence interval, 1.2-29.7; P = .03).
Patients who died within 30 days of CVC replacement also were more likely to have hematologic malignancies (20% versus 4%), diabetes (13% vs. 11%), to be on hemodialysis (27% vs. 16%), and to have a history of recent corticosteroid exposure (20% versus 11%) compared with survivors, but none of these associations reached statistical significance. Furthermore, 30-day mortality was not associated with gender, age, Candida species, endophthalmitis, or type of antifungal therapy, said Dr. Matsuo, who spoke at the combined annual meetings of the Infectious Diseases Society of America, the Society for Healthcare Epidemiology of America, the HIV Medicine Association, and the Pediatric Infectious Diseases Society.
An infectious disease consultation was associated with about a 70% reduction in the odds of mortality in the multivariate analysis, but the 95% confidence interval crossed 1.0, rendering the link statistically insignificant.
Given the small sample size and single-center design of this study, its findings ideally should be confirmed in a larger randomized controlled trial, Dr. Matsuo said. The investigators also did not track whether patients were fungemic at the time of CVC replacement, he noted.
The researchers reported having no conflicts of interest.