The results of each method were then compared against precise measurements at autopsy. The size of an m.i. was measured as a percentage of left ventricular volume. All measurements using MRI were taken within 24 hours after heart attack to simulate "real life" conditions.
Results showed that the full-width-at-half-maximum method was superior to all other methods, at 94 percent accuracy, when the least strict computer modeling formula was used, set at only one standard deviation between density of damaged and undamaged tissue. Other, stricter formulas in current use were accurate to 85 percent or less, if at all. When visual cues, the most commonly used method, were used, infarct size was measured with 69 percent accuracy.
"Our hope is that these results will help establish clear and effective guidelines for measuring the size of an m.i., and this will improve a physician's ability to make an accurate prognosis," said lead study author and Hopkins cardiology research fellow Luciano Amado, M.D. "Cardiologists also need to accurately gauge the amount of damaged muscle as part of our preparations for future treatments that could possibly repair it, most notably stem cell therapies."
This six-month study was conducted between December 2001 and June 2002, with funding provided by the National Institutes of Health and Datascope Corp. Other Hopkins investigators involved in this research were Bernhard Gerber, M.D.; Gilberto Szarf, M.D.; Khurram Nasir, M.D., Ph.D.; and Dara Kraitchman, V.M.D., Ph.D., who helped design the study. Additional assistance was provided by Dan Rettm
Contact: David March
Johns Hopkins Medical Institutions