14 Oct 2010
- Study first in the world to investigate use of real-time heart rate variability for cardiac arrest prediction
- Research to be presented at the inaugural SingHealth Duke-NUS Scientific Congress
Singapore, 13 October 2010 – A new tool using heart rate variability to assess patients’ risk of cardiac arrest is 50% more accurate than relying on traditional vital signs alone, researchers at Singapore General Hospital (SGH) and Nanyang Technological University (NTU) found. These new findings will be presented at the inaugural SingHealth Duke-NUS Scientific Congress, held on 15 and 16 October 2010 at Suntec International Convention & Exhibition Centre.
By combining heart rate variability with patients’ age and vital signs, the researchers were able to predict cardiac arrest within 72 hours with a sensitivity of 66.7% and a specificity of 98.1%; compared to only 41.7% sensitivity and 99% specificity when using traditional vital signs alone.
“While this new triage tool cannot replace a clinician’s judgment, it does give us an objective tool to assess patients’ risk, which will greatly help in early identification of high risk patients,” said Associate Professor Marcus Ong, Consultant, Director of Research and Senior Medical Scientist with the Department of Emergency Medicine at SGH, who led the study.
This is the first time in the world that heart rate variability is being evaluated for real-time use in cardiac arrest prediction at the emergency department. “Using our expertise in advanced signal processing and computational intelligent techniques, we worked with the clinical team at SGH to develop a new software. This advanced software is able to analyse heart rate variability in real-time to effectively support clinical decisions in the time-sensitive environment of an emergency department,” said Associate Professor Lin Zhi Ping from the School of Electrical and Electronic Engineering at NTU, and co-lead investigator.
In the study, the new tool was tested using the records of patients who presented with chest pain, shortness of breath, heart failure or suspected heart attacks. The study used data from 425 cardiovascular patients who were monitored with defibrillator monitors at SGH Emergency Department from November 2006 and December 2007.
The team is now working with a commercial partner to develop the prototype software further and initiate a clinical trial to validate this research. Eventually, the team hopes that this software will be included as a standard component in monitoring equipment used in emergency departments. In addition, intensive care units, hospital wards and emergency relief missions also stand to benefit from this new tool.
This study was supported by a grant from SingHealth Foundation and NTU.
Click for pdf format
« Back to previous
page