PURPOSE: Mediastinitis is an uncommon but serious complication following cardiac surgery, as it leads to increased patient morbidity and mortality, as well as hospital costs. The objective of this retrospective study was to determine pre-operative predictors of post-operative mediastinitis.
METHODS: We examined 6722 consecutive patients who underwent cardiac surgery between September 2000 and September 2009. Descriptive data are expressed as means and standard deviations for continuous variables, median and inter-quartile ranges for non-normally distributed variables, and percentage frequency for categorical variables. Statistical differences between cases and controls were determined using an unpaired t-test, Kruskall Wallis for non-parametric values, or the chi-square test for categorical values. Unadjusted and adjusted logistic regression analysis was used to examine factors (use of IMA) associated with mediastinitis. All statistical analyses were performed using SAS software, version 9.2. All results were considered statistically significant if the two-sided p value was < 0.05. The controls were well matched to cases with respect to age, gender, year of surgery, and disease burden.
RESULTS: The incidence of mediastinitis in our cohort was 1.54% (N=104). Patient age (OR 1.26 per 10 years, 95% CI 1.03-1.55), COPD (OR 2.74, 95% CI 1.73-4.32), diabetes (OR 2.44, 95% CI 1.61-3.70), morbid obesity (OR 2.31, 95% CI 1.36-3.94), as well as hospital stay prior to cardiac surgery (OR 1.18, 95% CI 1.04-1.34) were all independent predictors for the development of mediastinitis. These predictors provided good discrimination, with a model C-statistic of 0.72. Parsonnet score, as a measure of overall disease burden, did not contribute significantly to mediastinitis risk assessment.
CONCLUSION: We confirm that comorbid conditions such as diabetes, COPD, and morbid obesity are important pre-operative predictors of mediastinitis. We also identified prolonged pre-operative hospital stay as an independent predictor of mediastinitis risk.
CLINICAL IMPLICATIONS: IMPLICATION: Length of pre-operative hospital stay, which may impact nutritional or immune status, could represent a modifiable risk factor for mediastinitis. Results may help physicians for preoperative decision making.
DISCLOSURE: Rakesh Chaturvedi, No Financial Disclosure Information; No Product/Research Disclosure Information