Format

Send to

Choose Destination
Can J Anaesth. 2000 Jan;47(1):10-9.

ACC/AHA guidelines as predictors of postoperative cardiac outcomes.

Author information

1
Department of Anaesthesia, Queens University, Toronto, Ontario, Canada. mja@post.queensu.ca

Abstract

PURPOSE:

Recently, the American College of Cardiology - American Heart Association (ACC-AHA) published guidelines and an associated algorithm for preoperative cardiovascular evaluation of patients undergoing non-cardiac surgery. Our purpose was to (i) test guideline's ability to predict adverse cardiac events within seven days after surgery, (ii) determine whether medical clinical predictors or surgical risks was a better predictor of cardiac events.

METHODS:

Retrospective review of 119 cardiology and anesthesia consultations over 15 mo, ending March 31, 1998. Patients were classified into their respective medical clinical predictor and surgical risk groups, as outlined in ACC-AHA guidelines. Associations between the medical predictor and surgical risk scores and adverse cardiac outcomes were quantified via multiple logistic regression analysis. Two outcomes were employed. Outcome I, included: myocardial infarction/ischemia; angina; congestive heart failure, arrhythmia or death. Outcome 2 expanded the definition to include "cancellation of surgery due to cardiac risk" as a negative cardiac outcome.

RESULTS:

Diabetes, Canadian Cardiovascular Class (CCS) III or IV angina, and MI within six months before surgery were strongly associated with the two cardiac outcomes. For outcome 1 and 2, medical predictors and surgical risks, considered simultaneously, performed with a sensitivity of 93% and specificity of 46-51%. When considered separately, major clinical medical predictors had a sensitivity of 87-89%, while surgical risks showed a specificity of 89% in predicting the two outcomes.

CONCLUSION:

Medical predictors in ACC-AHA classification scheme were highly sensitive whereas surgical risks were more specific in predicting adverse post-operative cardiac events. Prospective study is needed to confirm these observations.

PMID:
10626712
DOI:
10.1007/BF03020725
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Springer
Loading ...
Support Center