Anesth Analg 2001; 92: 596-601

Reviewed by: R. Prasad, MD

Summary:

• HF is the most common cause of death after CABG
• Most of the variation seen in mortality after CABG can be explained by fatal HF
• Therefore, being able to stratify pts by risk of death from HF may be useful in directing aggressive preemptive therapy.
• Several factors (female sex, h/o CABG, EF<40%, urgent or emergent surgery, age >70 or >80, PVD, DM, dialysis-dependent RF, 3-v CAD) with varying weights were incorporated into a clinical risk assessment tool (CRAT - my abbreviation).
• Here's the complete formula (not the simplified CRAT) for MathPad users.

• Is this really important? That is, will it really make any difference in care? Hard to say, since I don't do CABG's ...
• Even their "risk assessment tool" is a bit cumbersome. It's straight forward, of course, but unless you calculate it many times you would need to have a summary card (they have such a "pocket-sized card" for their 8 hospitals). Of course, Palm (MathPad) saves the day!

• Northern New England Cardiovascular Disease Study Group: 8 medical centers in the the north east US, 7/1/87 - 4/1/91.
• HF identified as mode of death if management of hypotension or low CI was the "seminal event" leading to death.
• Looked at preop patient characteristics associated with death from HF in univariate and multivariate logistic regression models.

• 8641 patients, 387 (4.48%) deaths, 249 (64.8%) of these due to HF.
• CRAT had good discrimination ability and confidence intervals similar to the logistic regression model.

• Limitations include:
• This is a regional study.
• Even in a large database, few with the highest risk scores. Their recommendation: consider anyone with score >10 as having 15% risk
• The CRAT is less precise than using the logistic regression equation (which is used in the MathPad formula, above).
• They use it to help direct TEE, invasive monitoring, IABP more towards high-risk patients. Also, using it in research studies.