Predictors of postoperative ventricular dysrhythmias: a multivariate study
Ferraris, V.A.; Ferraris, S.P.; Gilliam, H.S.; Berry, W.R.
Journal of Cardiovascular Surgery 32(1): 12-20
1991
ISSN/ISBN: 0021-9509 PMID: 2010441 Document Number: 368183
Postoperative ventricular dysrhythmias were studied to document their incidence after coronary bypass grafting and to identify risk factors for their development with the hope of finding a subgroup of patients who might benefit from postoperative, prophylactic drug therapy. One-hundred-nine patients who were undergoing urgent or elective coronary bypass grafting were studied, prospectively. Twenty-five of 109 patients (23%) developed significant postoperative ventricular dysrhythmias that required counter-shock or drug intervention. Seven of eight instances of sustained ventricular tachycardia, the most serious dysrhythmia, occurred within 36 hours of operation. There was no postoperative mortality related to these dysrhythmias. Serious postoperative complications, such as stroke, hemorrhage, or myocardial infarction, were decreased in patients with ventricular dysrhythmias versus those without (8% versus 16%, p = 0.053 for the Fisher's exact test statistic). Univariate statistical analysis was performed using 15 patient variables and revealed that advanced age (p = 0.008 for the unpaired t test), failure to use an internal mammary artery conduit (p = 0.03 for the two-tailed Fisher's exact test), and development of postoperative atrial dysrhythmias (p = 0.02 for the two-tailed Fisher's exact test) were significantly more common in patients with postoperative ventricular dysrhythmias. Variables such as previous myocardial infarction, ejection fraction less than 50%, prolonged operative time, perioperative myocardial infarction, or fewer number of vessels bypassed were not significantly increased in patients with dysrhythmias (the statistical power for these "negative" results was greater than 0.8). Logistic regression analysis revealed that the only independent predictors of development of ventricular dysrhythmias were advanced patient age (improvement in log-likelihood = 10.03, p = 0.002) and preoperative use of calcium-channel blocking drugs (improvement in log-likelihood = 3.52, p = 0.06). Study conclusions are as follows: (1) postoperative ventricular dysrhythmias in this patient population have minimal effects upon, and may actually be associated with decreased, hospital morbidity or mortality, (2) these dysrhythmias occur more commonly in older patients, (3) factors that predispose patients to postoperative ventricular dysrhythmias may also predispose them to atrial dysrhythmias in the early postoperative period, and (4) multivariate analysis indicates that advanced patient age and preoperative use of calcium-channel blocking drugs are predictors of development of these dysrhythmias. These results suggest that postoperative drug prophylaxis against ventricular dysrhythmias is not indicated.