Skip to Content

Michael Pinsky Awarded an R01 to Develop a Model for Detecting and Predicting Shock in ICUs

Sat, 10/22/2016

Septic shock is the cause of the majority of ICU admissions in the U.S., and carries a mortality rate of more than one in five patients. Because the signs of shock, including cardiorespiratory instability (CRI), occur long after the initial injury began and are difficult to detect, a better diagnostic or even predictive tool could be of great use.

This is the goal of an R01 recently funded by the National Institute of General Medical Sciences (NIGMS), part of the National Institutes of Health (NIH), for which Michael Pinsky, MD, CM, Dr hc, is the principal investigator. The grant (#R01GM117622), titled, “Machine learning of physiological variables to predict, diagnose, and treat cardiorespiratory instability,” will seek to develop a model that can be used in ICUs to quickly identify when shock is occurring in a patient, and to predict in which patients it may soon occur.

Previous research by Pinsky, Professor of Critical Care Medicine, Bioengineering, Anesthesiology, Cardiovascular Diseases, and Clinical & Translational Sciences, and his collaborators has yielded a monitoring system that can quickly identify patients at risk for CRI in step-down units (SDUs). Pinsky has also applied machine learning (ML) in the lab setting to predict when and in which subjects CRI would occur. This new grant will enable the investigators to apply these findings to patients in intensive care units. The investigators will first gather data from ICUs at the University of Pittsburgh, the University of California, Irvine, and the University of California, San Diego, after which they will develop a user interface that can be employed at the bedside to identify those patients most at risk for developing CRI.

Pinsky currently has one other R01 on which he is a principal investigator, “Predicting Patient Instability Noninvasively for Nursing Care-Two,” or PPINNC-2, which is exploring similar topics. While PPINNC-2 is held in the School of Nursing and focuses on improving nurses’ ability to detect when CRI is occurring or may soon occur, this particular R01, on the other hand, is concerned with patients in the ICU, a much more vulnerable population.

Co-investigators on the grant include Gilles Clermont, MD, MSc, Professor of Critical Care Medicine, Industrial Engineering, Mathematics, Chemical Engineering, and Clinical and Translational Science; Artur Dubrawski, PhD, MSc, Director of the Auton Lab at Carnegie Mellon University; and Marilyn Hravnak, PhD, RN, Professor in the Acute and Tertiary Care Department of the University of Pittsburgh School of Nursing.