National and Kapodistrian University of Athens
Λοιμώξεις από ενδοφλέβιους καθετήρες. Αξιολόγηση διαγνωστικών μεθόδων. Διερεύνηση συνθηκών μόλυνσης
Abstract
dc:descriptionIntroduction: The aim of the present study was to assess the rate of colonization and infection related with CVC (Central Venous Catheter) in a tertiary Greek hospital in order to identify the causative microbial agents and sources of CVC colonization, to evaluate the diagnostic methods and to determine the risk factors associated with bacteremia related to CVC usage. Methods: Three- hundred-and-forty- CVCs from three-hundred inpatients of a tertiary care hospital of Athens were evaluated during the period 12/2001-6/2005. All specimens were processed using: the “roll plate” semiquantative method (Maki technique), quantitative method (Cleri technique), broth culture, skin cultures from insertion site of each catheter and cultures from Hub. Blood cultures were obtained from patients in suspicion for catheter-related bloodstream infection. Time of catheter removal, as well as patients’ possible risk factors were recorded. The isolates were identified by PASCO (Difco, USA) and API (bioMerieux, France) system. Their antibiotic susceptibility patterns were studied by using the Kirby-Bauer and E-test method, if it was necessary. The isolates were also genotypically analyzed by Pulsed Field Gel Electrophoresis (PFGE). Results: The study enrolled 340 central venous catheters. In 53 (15.6%) cases Catheter Related Bloodstream Infection (CRBI) was found while bacteraemia not related to catheter was found in 6 (1.8%) cases. Catheter colonization was found in 90 (26.5%) cases. The sensitivity and specificity of Maki method was equal to 94.3% and 99.0%respectively. With reduction of cut-off value of Maki method to ≥8 the sensitivity increased to 100%. In 19 of the 53 cases CRBI, the Hub culture had a positive outcome. In 29 of the 53 cases CRBI, the skin culture had a positive outcome and the most common organism was CoNS. The mean duration of catheter in individuals with CRBI and positive Hub culture was 21.2 days (SD=8.0 days) and was greater than the mean duration of catheter equal to 15.3 days (SD=2.0 days) in individuals with CRBI and positive skin culture (p<0.001). The most commonly isolated organisms in blood were gram-positive cocci; the same bacteria were isolated from catheter tip and patients skin. Most of the organisms isolated from Hub cultures were gram-negative bacteria. S. aureus was the most commonly isolated organism in patients with CRBI and renal insufficiency. As defined from the results of multiple logistic regression analysis the odds of CRBI was significant greater in patients who received corticoids (OR=7.56, 95% CI: 2.78-20.53), in patients with diabetes (OR=3.24, 95% CI: 1.27-8.27) and in those with solid organ cancer (OR=6.19, 95% CI: 1.24-30.98). Patients in which the insertion site dressing were being changed every 48 hours or later, had greater risk for CRBI compared to those which the insertion site dressing were being changed every 24 hours. Also, it was found that the risk of CRBI increases as the duration of catheter increases (OR=1.16, 95% CI: 1.09-1.23). The incidence rate of CRBI per 1000 catheter days is equal to 11.47 [95% CI: 8.77 – 15.02]. The incidence rate is significantly greater in patients who received corticoids (IRR=1.92, CI: 1.08-3.42), when used transparent dressing (IRR=3.49, CI: 1.86-6.53) and in cases that IV delivery system was replaced every 48 hours or later (IRR= 1.92, CI: 1.08 3.42). Lower incidence rate per 1000 catheter days was found in individuals who had subclavian access (IRR=0.51, CI: 0.27-0.98) or Hickman (IRR=0.14, CI: 0.02-1.00). PFGE analysis in several cases of CRBI revealed that microorganisms isolated from different patients had the same pulsotype. Conclusions: We suggest the reduction of cut-off value of Maki method to ≥8 colonies in order to increase the sensitivity. Determination of risk factors associated with CVC infection will contribute to the development of preventive strategies and impact the morbidity and mortality of patients with CVC. We found a high incidence of CRBI as well as a high proportion of CRBI caused by skin flora during catheter insertion. A great number of catheters were colonized by Hub. Microorganisms isolated in several cases had the same pulsotype among patients. This fact demonstrates the nosocomial spread of bacterial strains. Removal of the catheter proved to be futile in many cases and showed the need for the establishment of high efficiency diagnostic methods in the modern clinical microbiology laboratories.
Degree
thesis:*- Grantor dc:publisher
- National and Kapodistrian University of Athens
- Year dc:date
- 2007
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Ταρπατζή, Αικατερίνη
- Tarpatzi, Aikaterini
Subjects
dc:subject × 14Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/18353