Cardiac calculator predicts risk for perioperative MI and cardiac arrest
MedWire News: Researchers have created a cardiac risk calculator that has higher predictive power for perioperative myocardial infarction (MI) and cardiac arrest (CA) than the Revised Cardiac Risk Index (RCRI).
Although the RCRI is the most commonly used cardiac risk stratification tool in surgical patients, it lacks predictive ability and fails to account for the surgical advances that have been made since its development, explain Prateek Gupta (Creighton University, Omaha, Nebraska, USA) and colleagues in the journal Circulation.
They therefore decided to develop and validate a new cardiac risk calculator.
The study included 211,410 patients from the American College of Surgeons' 2007 National Surgical Quality Improvement Program (NCQIP) database.
In all, 1371 (0.65%) patients developed perioperative MI or CA. Multivariate analysis revealed that surgery type, dependent functional status, abnormal creatinine level (more than 1.5 mg/dl), American Society of Anesthesiologists' class, and increasing age were predictors of perioperative MI or CA.
These five variables were used to create a new risk model, which the researchers then validated on the 2008 NCQIP dataset (n=257,385). They found that the model performed similarly well on the 2007 and 2008 datasets, with areas under the receiver-operating characteristic curve (AUC) of 0.884 and 0.874, respectively.
However, when they applied the RCRI to the 2008 NCQIP database, the resulting AUC was lower than that of the new risk model, at 0.747.
"[This] implies a superior predictive power," they explain.
"Furthermore, instead of classifying patients as high/intermediate/low risk for perioperative cardiac complications, we provided an exact model-based estimate of MI [and] CA probability for a patient."
They conclude: "We anticipate the risk calculator will aid in surgical decision making and the informed consent process."
MedWire (www.medwire-news.md) is an independent clinical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2011
By Piriya Mahendra