| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL ARTICLE |
Department of Cardiac Surgery, Valladolid University Hospital, Valladolid, Spain
Javier Gualis, MD, Tel: +34 987234900 Ext 2382, Fax: +34 987243924, Email: javgua{at}hotmail.com, Department of Cardiac Surgery, Valladolid University Hospital, Valladolid, Spain.
ABSTRACT
A prospective open-cohort study was performed in 838 adults undergoing coronary revascularization or valve surgery to define the risk factors for development of surgical site infections. Patients diagnosed with mediastinitis or endocarditis during follow-up were compared with patients with no such infection. After 1 year of follow-up, 22 (2.6%) patients had developed mediastinitis or endocarditis. No preoperative or intraoperative variables were identified as risk factors. By multivariate analysis of postoperative variables, respiratory insufficiency, microorganisms in blood cultures, and intensive care unit stay were independent risk factors for the development of these complications. The type of antibiotic prophylaxis had no influence on the incidence of organ or space infections after cardiac surgery.
Key Words: Anti-Bacterial Agents • Cefazolin • Endocarditis • Mediastinitis • Surgical Wound Infection
INTRODUCTION
After open heart procedures, the incidence of surgical site infections (SSI) ranges from 0.4% to 8%, with an overall mean of 1.5%.1 Endocarditis and mediastinitis after cardiac surgery are serious complications that increase short- and long-term morbidity, mortality, and costs.2,3 They also have a negative impact on postoperative quality of life.4 Many risk factors (preoperative, intraoperative, and postoperative) have been associated with these complications in previous reports.5 However, a few prospective randomized studies with adequate sample sizes have focused on perioperative antibiotic prophylaxis regimens, especially with intravenous cefazolin, the most common agent used in cardiac surgery.5 The aim of this study was to analyze the impact of risk factors classically associated with an increased incidence of these postoperative complications in the setting of a uniform and randomly assigned antibiotic prophylactic strategy.
PATIENTS AND METHODS
The study was conducted in the Valladolid University Hospital Cardiac Surgery Department, a tertiary center performing approximately 500 open heart procedures per year. Inclusion criteria were patients over 18 years of age scheduled for elective valvular surgery (repair or replacement) and/or coronary artery bypass grafting through a median sternotomy. Exclusion criteria were preoperative active infection, antibiotic administration 48 h prior to surgery, emergency surgery, beta-lactamic antibiotic allergy, and heart transplantation. The study was approved by the institutional ethical committee, and every patient understood and signed the informed consent document before participating in the study. Patients selected for inclusion were prospectively randomized to 2 groups with different antibiotic pro-phylaxis regimens. In group A, a single intravenous dose of cephazolin 2 g was administered 30 min before skin incision, and an additional intravenous dose of 1 g was given when the blood loss was considered greater than 1 L or when cardiopulmonary bypass time exceed 3 h. In group B, the same preoperative dose of cefazolin 2 g was administered 30 min before skin incision followed by 1 g of cefazolin every 8 h during the first 24 h postoperative hours. During hospitalization, every patient was examined daily by an experienced cardiac surgeon. In cases of suspected or established SSI, an infectious diseases specialist supervised the treatment of the patient. Blood, sputum, urine, and wound drainage samples were cultured in patients with suspected infections. These samples were analyzed in the microbiology laboratory, using standard techniques. All patients who received prosthetic heart valves were followed up for 1 year after the operation, for the detection of early prosthetic endocarditis.
The definition of SSI was based on the Centers for Disease Control and Prevention guidelines.6 The diagnosis of endocarditis was determined according to the Duke criteria.7 To identify risk factors for the development of SSI, 2 cohorts of patients were defined. The first cohort (study group) comprised patients with a confirmed postoperative diagnosis of SSI. The second cohort (control group) included patients with no infective complications. Preoperative clinical characteristics and previously accepted preoperative, intraoperative, and postoperative variables associated with increased incidence of SSI were prospectively recorded for each patient.8
To assess risk factors for SSI, we used one-way analysis of variance for univariate continuous variables and the chi-squared test for categorical variables. In addition, we conducted Fisher’s exact test when the chi-squared expected value of at least one cell was less than 5.9 Due to the large number of potential risk factors studied, multicollinearity among the explanatory variables was avoided by performing collinearity diagnostic analyses. Stepwise selection of variables from the models was undertaken with the following criteria: tolerance >0.4 or variance inflation <2.5, condition number <10, and a variance of 2 or more variables no greater than 0.5.10–12 The dependent variable was a nominal response within the 2 groups: patients without SSI and those with SSI. To evaluate risk factors for SSI and adjust for potential confounders, we carried out multiple logistic regression analyses with a generalized logit link function and stepwise selection. Factors with values of p <0.25 were initially tested in the models, and factors with a value of p <0.05 were considered further. For all models, the adjusted odds ratio (OR) indicated the association of interest. A p value <0.05 was considered significant. Data were stored and analyzed using SPSS software version 14.0 (SPSS Inc., Chicago IL, USA).
RESULTS
Between September 2003 and June 2007, 838 consecutive patients were studied, and 419 were randomized to each group of antibiotic prophylaxis. The distribution of preoperative, intraoperative and postoperative variables is shown in Table 1
. At the end of follow-up (100% complete), 22 organ or space infections were diagnosed. The mean time from operation to diagnosis was 9.9 ± 4.8 days. The distribution of SSI between groups is shown in Table 2. No relationship between the antibiotic prophylaxis regimen and the incidence of SSI was found (group A, 3.3%; group B, 1.9%; p >0.05). On univariate analysis, 11 variables were significantly associated with an increased incidence of SSI (Table 3). However, multivariate analysis demonstrated that only 3 postoperative variables could be considered independent risk factors for the development of SSI: respiratory insufficiency, positive blood cultures, and intensive care unit (ICU) stay (Table 3).
|
|
|
Mediastinitis and early prosthetic endocarditis are devastating complications after heart surgery, with a combined incidence of 0.25%–25%, and predicted mortality rate of approximately 30% after diagnosis.13 The reported incidence of postoperative mediastinitis varies from 0.4% to 2%, and it is usually diagnosed between the 4th and 30th postoperative day. Previously established risk factors for this disease are use of bilateral internal mammary arteries (especially in obese diabetic women) and chronic obstructive pulmonary disease.14 Besides a high mortality rate (7%–20%), this complication requires frequent multiple diagnostic tests, repeated surgical procedures, and prolonged and costly hospital stays. Furthermore, postoperative mediastinitis has a significant negative impact on long-term survival.1,15
Our study found that patients treated with an ultra-short-acting antibiotic prophylaxis regimen show a trend towards an increased rate of postoperative mediastinitis. The low incidence of this complication in both groups (1.9% in single-dose patients vs. 1.2% in multidose patients) may explain the lack of statistical significance.
Early prosthetic endocarditis is acquired during the perioperative period. As a nosocomial disease, resistant microorganisms predominate: mainly coagulase-negative staphylococci; less frequently Staphylococcus aureus; and more rarely staphylococci of the viridian group, gram-negative bacilli, Corynebacterium, and fungal endocarditis (Candida albicans and Aspergillus) in 0.8%–3.9% of cases. Usually, early prosthetic endocarditis becomes clinically relevant 2 to 6 weeks after surgery, but follow-up must be extended up to one year. Antibiotic prophylaxis has effectively decreased the incidence of this feared complication from 10% to 2%.13 In our study population, the rate of early endocarditis was low (1.2% in patients treated with a single prophylactic dose, 0.2% in those receiving pro-phylaxis for 24 h). No significant differences could be demonstrated.
The results of our analysis of other risk factors for SSI reveal that a positive postoperative blood culture is an independent risk factor for the development of organ or space infections. Similarly, San Juan and colleagues16 found that early isolation of microorganisms in blood cultures from postoperative febrile cardiac surgical patients was a potent and early predictor of postoperative infective complications. The coexistence of infections in other systems during the postoperative course (lung, urinary tract, etc.) may increase the risk of SSI. Aggressive therapeutic strategies in these situations, specially in patients with positive blood cultures, could probably decrease the incidence of these complications. Our results also suggest that prolonged ICU stay is an independent risk factor for the development of SSI. This finding has been reported previously.5 Obviously, patients with infective complications require longer stays, and it is difficult to discern whether infection is the cause or a consequence of this. However, previous evidence of long ICU stays being associated with a late increased rate of infection suggests that fast-track strategies could help to avoid this complication.5,17 In addition, our data show that respiratory insufficiency during the postoperative course is also an independent risk factor for SSI. Respiratory insufficiency prolongs mechanical ventilation time and ICU stay, and increases the incidence of respiratory tract infection. Respiratory failure occurs more frequently in patients with history of chronic obstructive pulmonary disease, which is a potent predictor of surgical infections including mediastinitis.18 However, we did not find a significant relationship between chronic obstructive pulmonary disease and postoperative SSI in this study. Preventive strategies, such as systematic perioperative intensive respiratory physiotherapy and bronchodilator and mucolytic therapy optimization, could help to reduce respiratory complications and subsequent SSI.
It was concluded that the duration of antibiotic prophylaxis regimen with the dose of cephazolin tested has no significant impact on the incidence of postoperative organ or space infections after open heart surgical procedures. Postoperative respiratory failure, positive blood cultures, and less probably, prolonged ICU stay are independent risk factors for such complications. Aggressive therapy for concomitant infections, optimization of perioperative respiratory function, and ICU fast-track policies could be useful measures to decreasing the rate of these serious complications.
REFERENCES
Asian Cardiovasc Thorac Ann 2009;
17:612-616
© 2009 by SAGE Publications
DOI: 10.1177/0218492309349071
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
