Comparison of cardiac preoperative risk prediction indices

Comparison of the revised cardiac risk index and the NSQIP tool. Watch my YouTube video describing these slides: http://youtu.be/vzWXQI_DFr0

1. Comparison of RCRI (Lee Criteria) and NSQIP (Gupta Criteria) Indices for Predicting Cardiac Risk of Major Noncardiac Surgery Terry Shaneyfelt, MD, MPH Associate Professor, UAB Department of Medicine

2. Predicting Perioperative Cardiovascular Risk Cardiovascular stress testing has limited predictive value 2 most commonly used prediction tools Revised Cardiac Risk Index (RCRI) AKA Lee Criteria National Surgery Quality Improvement Program Risk Calculator (NSQIP) AKA Gupta Criteria

3. What factors are predictive of risk in these indices? Circulation 1999;100:1043 Circulation 2011;124:381

4. RCRI NSQIP Methodology Prospective cohort Historical cohort using NSQIP database Patient Population 4315, ≥ 50 yrs old , 1 hosp. 468,795, > 16 yrs old, 200 hospitals Date of development 1989-1994 2007-08 Outcomes In-hospital MI (CK based), pulmonary edema, V fib, cardiac arrest, complete heart blk MI (trop based) and cardiac arrest thru 30d postop Surgery specific No Yes C-statistic 0.806 (valid cohort) 0.750 (Ford Ann Intern Med 2010) 0.747 (Gupta valid all surg ) 0.591 (Gupta valid vasc surg ) 0.874 (valid cohort all surg ) 0.746 (vascular surgery) Comparison of RCPR and NSQIP Indices

5. NSQIP index has better predictive performance, is surgery specific, and more contemporary RCRI has been used for over a decade Summary Become familiar with one model and use it regularly