Risk factors of deep sternal wound infections in coronary artery bypass graft surgery

Citation
Fd. Wang et Ch. Chang, Risk factors of deep sternal wound infections in coronary artery bypass graft surgery, J CARD SURG, 41(5), 2000, pp. 709-713
Citations number
22
Categorie Soggetti
Cardiovascular & Respiratory Systems
Journal title
JOURNAL OF CARDIOVASCULAR SURGERY
ISSN journal
00219509 → ACNP
Volume
41
Issue
5
Year of publication
2000
Pages
709 - 713
Database
ISI
SICI code
0021-9509(200010)41:5<709:RFODSW>2.0.ZU;2-J
Abstract
Background. Major infections of sternal wounds after coronary artery bypass graft surgery (CABG) occur infrequently, but when they do, they contribute substantial morbidity and mortality, We identified significant risk factor s of deep sternal wound infection (DSWI) following CABG and hoped to plan p rophylactic measures for high risk patients in order to reduce the incidenc e of infection. Methods. From 1996 to 1997, a total of 620 patients received CABG at a medi cal center in Taiwan. The surgical wound was examined every day, Wound infe ctions were defined and classified according to Centers for Diseases Contro l (CDC) definitions, DSWI, were defined as those involving the mediastinum, bone or cartilage, and infections beneath the subcutaneous tissue. Several risk factors were analyzed. Results, We identified 17 (2.7%) DSWI). Univariate analysis indicated that ASA scores, surgical risk index, surgeon, postoperative low cardiac output, re-operation for bleeding, re-wiring of sternum, length of postoperative s tay in the intensive care unit, postoperative duration of mechanical ventil ation, operation time, duration of cardiopulmonary bypass. independent pred ictors by multivariate logistic regression analysis were re-operation for b leeding and operation time, A total of 16 organisms were isolated among 14 patients. Staphylococcus accounted for most isolates (93.8%) and the most o f them were methicillin-resistant (80%). Conclusions. Re-operation for bleeding and operation time were the independ ent predictors.