Abstract
Background:
The prevalence of coronary artery disease (CAD) in valvular patients is similar to that of the general population, with the usual association with traditional risk factors. Nevertheless, the search for obstructive CAD is more aggressive in the preoperative period of patients with valvular heart disease, resulting in the indication of invasive coronary angiography (ICA) to almost all adult patients, because it is believed that coronary artery bypass surgery should be associated with valve replacement.
Objectives:
To evaluate the prevalence of obstructive CAD and factors associated with it in adult candidates for primary heart valve surgery between 2001 and 2014 at the National Institute of Cardiology (INC) and, thus, derive and validate a predictive obstructive CAD score.
Methods:
Cross-sectional study evaluating 2898 patients with indication for heart surgery of any etiology. Of those, 712 patients, who had valvular heart disease and underwent ICA in the 12 months prior to surgery, were included. The P value < 0.05 was adopted as statistical significance.
Results:
The prevalence of obstructive CAD was 20%. A predictive model of obstructive CAD was created from multivariate logistic regression, using the variables age, chest pain, family history of CAD, systemic arterial hypertension, diabetes mellitus, dyslipidemia, smoking, and male gender. The model showed excellent correlation and calibration (R² = 0.98), as well as excellent accuracy (ROC of 0.848; 95%CI: 0.817-0.879) and validation (ROC of 0.877; 95%CI: 0.830 - 0.923) in different valve populations.
Conclusions:
Obstructive CAD can be estimated from clinical data of adult candidates for valve repair surgery, using a simple, accurate and validated score, easy to apply in clinical practice, which may contribute to changes in the preoperative strategy of acquired heart valve surgery in patients with a lower probability of obstructive disease.
Keywords:
Coronary Artery Disease; Heart Valve Disease; Coronary Angiography; Computed Tomography Angiography
