Risk Stratification in Cardiac Surgery

Cardiac Risk for Cardiac Surgery

Clinical Scoring Metrics


While a multitude of risk-stratification scores have been developed, several risk factors have appeared recurrently – advanced age, female gender, elevated BMI, and decreased LV function are poor prognostic variables, as are emergent and redo operations [Kaplan JA, ed. Essentials of Cardiac Anesthesia. Saunders, 2008 p 7]


The Euroscore was developed base on outcomes in 19,030 patients. It comprises 18 variables, weighted from 1-4, and defines low risk as 0-2 points (0.8% mortality), medium risk as 3-5 points (3.0% mortality), and high risk as > 5 points (11% mortality). Overall hospital mortality in the cohort was 4.7% [Roques F et al. Eur J Cardiothorac Surg 15: 816, 1999]

EuroScore Outcomes (based on 19,030 patients [Roques F et al. Eur J Cardiothorac Surg 15: 816, 1999]) Low Risk (0-2 points): 0.8% in-hospital mortality Moderate Risk (3-5 points): 3.0% in-hospital mortality High Risk (6 or more points): 11.2% in-hospital mortality

Cardiac Anesthesia Risk Evaluation (CARE) Score

The CARE score was developed as an attempt to simplify risk stratification (the EuroScore, which contains 18 variables, is somewhat cumbersome). It contains only 6 variables and was validated using data from over 3500 patients. While not specifically compared to the EuroScore, the CARE score showed and equivalent sensitivity/specificity tradeoff (by receiver operator curve analysis) when compared to more complex cardiac anesthesia risk assessment schematics (developed by Parsonnet, Tuman, and Tu, respectively) [Dupuis JY et al. Anesthesiology 94: 194, 2001; FREE Full-text at Anesthesiology]

CARE Score Rating [Dupuis JY et al. Anesthesiology 94: 194, 2001]) 1: Stable cardiac disease with no additional medical problems (mortality 0.5%) 2: Stable cardiac disease with one or more controlled medical problems (mortality 1.1%) 3: Any uncontrolled medical problems OR complex surgery (mortality 2.2%) 3E: Criteria for 3 (see above), emergent surgery (mortality 4.5%) 4: Any uncontrolled medical problems AND complex surgery (mortality 8.8%) 4E: Criteria for 4 (see above), emergent surgery (mortality 16.7%) 5: Chronic or advanced cardiac disease for whom surgery is a last resort (mortality 29.3%) 5E: Criteria for 5 (see above), emergent surgery (mortality 46.2%)

Functional Tests


Unlike the aforementioned clinical scoring metrics, functional tests seek to obtain information that cannot be gleaned from the patient’s history in order to better assess risk and potentially change outcome. It should be noted that while some of the tests mentioned below have demonstrated significant predictive utility, none of them have been definitively shown to change outcomes. That does not mean that they should not be used, only that their utility lies in the risk/benefit assessment regarding the decision to undergo cardiac surgery – if the decision to undergo cardiac surgery has already been made, these tests may not change outcomes in any appreciable way

Electrocardiogram (ECG)

Because patients having cardiac surgery are already known to have cardiac dysfunction, the purpose of an ECG is not to detect dysfunction, but rather to form a basis for comparison (i.e. to aid in the detection of intraoperative events) as well as to define the “ischemic threshold” (when used in the context of a stress test). ECG-based stress testing has lower sensitivity than TTE-based stress testing

Pharmacologic Stress Testing

Adenosine is a myocardial vasodilator. Dipyridamole blocks adenosine reuptake. Thus, dipyridamole maximally vasodilates coronary arteries, preferentially “stealing” blood away from stenotic vessels and simulating the effects of exercise or stress, an effect that can be visualized following injection of isotopes (ex. thallium, technetium scintigraphy). In order to approximate the amount of “susceptible” myocardium, two injections are required – an initial injection of dipyramidole + isotope (underfilled areas are either dead or susceptible), and a second injection of isotope alone (underfilled areas are dead, areas of myocardium which have re-filled as compared to the initial injection are “at risk”)

Dobutamine Stress Echocardiography (DSE) differs from Dipyridamole-Isotope scintigraphy in that dobutamine actually causes the heart to work harder (as opposed to simulating it via coronary vasodilation) and uses TTE, not scintigraphy, to image the heart. DSE is as sensitive as Dipyridamole-Isotope scintigraphy and has the added advantage of assessing valvular competence and left ventricular function. Furthermore, DSE does not require a second set of imaging, and can thus be accomplished more quickly than Dipyridamole-Isotope scintigraphy


Besides its obvious use in DSE (see above), echocardiography is useful in evaluation of left ventricular and valvular function, as well as being the most sensitive detector of ischemia (regional wall motion abnormalities appear within 10-15 seconds of ischemia)


For recently-published Universal Definition of Myocardial Infarction, please see [Thygesen K et al. Circulation 116: 2634, 2007; FREE Full-text at Circulation]

Cause(s) of Myocardial Injury

The major causes of myocardial injury following cardiac surgery are myocardial ischemia and reperfusion injury. As would be expected, cross-clamp time and CPB duration are significantly related to poor outcomes in cardiac surgery. According to Kaplan, the incidence of additional cardiac events within two years is 4% in patients without perioperative MI, and 49% in those who suffer a perioperative MI. [Kaplan JA, ed. Essentials of Cardiac Anesthesia. Saunders, 2008 p 10]

Importantly, if myocardial ischemic time short, the myocardium may be “stunned” but not permanently injured, ultimately able to recover without any long term dysfunction. Oxygen free radicals, calcium overload, and excitation-contraction uncoupling are likely causes of myocardial stunning [Bolli R. Circulation. 82: 723, 1990; FREE Full-text at Circulation]. Of note, the oft-quoted 20 minute time limit after which stunning becomes ischemia is based on experiments in dogs.

Cardiac Risk for Non-Cardiac Surgery

Please see Cardiac Risk (Anesthesia Text) for an extensive discussion of cardiac risk in non-cardiac surgery.