Critically ill patients are admitted to hospitals (ICUs) across the country every day, yet physicians typically have little idea about the clinical trajectory of a given sick patient. Some quickly improve, some quickly deteriorate and die, and others linger in apparent “suspended states” for weeks to months. We believe it is now possible to get some preliminary data as these patients come through the door, and with a few repeated measures, answer the questions "what is going on?" and "how are they going to do?" This will be important to families because we will identify the "hope" from "little to no hope" patients. These data also will guide physicians to personalize therapy, optimizing outcomes.
We obviously are not the first to address these questions. There are a host of probabilistic tools available based upon clinical examination and physiological parameters (for example, APACHE III). These tools are good at defining outcomes of groups of patients, but as reported in numerous studies, they are poor at defining or classifying individuals. What the current tools lack is an ability to measure or gauge at a molecular level (1) heritable (genetic) predisposition to a given clinical trajectory, and (2) the patient's response to current therapy (in other words, is what we are doing with the patient working?).
Investigators from the Schools of Medicine, Engineering, Arts and Science, and Business at Washington University in St. Louis and the School of Nursing at the University of Missouri-St. Louis are collaborating to explore the application of physiological genomics to define the host response to critical illness and injury. Our hypothesis is that modeling the critically ill state as a comprehensive set of mathematical models (“the virtual human”) will lead to improved diagnostics, prognostics and therapeutic targets. This research is supported by awards from the Barnes-Jewish Hospital Foundation, Washington University and the National Institutes of Health.
What does our group provide that is new? Today it is possible to both gauge heredity and measure the response to therapy at a genome-wide level. Thus, the biological assumptions underlying our efforts are that (1) predisposition to recovery is determined to a greater or lesser degree by our genetics and (2) response to therapy can be measured/predicted/individualized at the level of RNA and protein. For example, our data indicate that RNA and protein abundance in circulating blood from animals or patients can be used to classify and model the response to infection and systemic inflammation.
Finally, modeling of the inflammatory response is important to scientists who study a wide variety of illnesses that alter or involve immunological responses, including cancer, Alzheimer's disease, atherosclerosis, autoimmune illnesses, sepsis, AIDS, injury from trauma or weapons of mass destruction (WMD) and many others. The new threat of critical illness on a city-wide or national scale (whether due to influenza, anthrax, WMD or other threats to public health) has created a federal imperative to build infrastructure and gain expertise, as recently reported in the lay press.*
*McNeil, Jr., D.G., Hospitals Short on Ventilators if Bird Flu Hits, The New York Times, Mar 12, 2006. http://www.nytimes.com/2006/03/12/national/12vent.html?emc=eta1