Harnessing Big Data to Aid ICU Patients

Next-gen monitoring approaches may yield early markers of impending clinical decline

Innovations in Pulmonology & Sleep Medicine - Fall 2018

 Frank Jacono, MD

A hospital Intensive Care Unit’s high-intensity monitoring produces a veritable font of critical data about patients. In fact, sometimes there is so much data that it’s hard to use it in a way that helps clinicians improve care, says Frank Jacono, MD, Division of Pulmonary, Critical Care and Sleep Medicine, UH Cleveland Medical Center; Chief, Pulmonary and Critical Care Medicine Section, Louis Stokes Cleveland VA Medical Center and Associate Professor, Medicine, Case Western Reserve University School of Medicine.

Dr. Jacono and his colleagues are conducting research to see if they can change this – especially for patients at high risk of decompensating.

INNOVATING ADVANCED DIAGNOSTICS

“In the ICU, we monitor patients for all sorts of things, such as cardiac rhythm, breathing patterns and blood pressure,” Dr. Jacono says. “There’s obviously valuable information about the health status of patients contained in all of these vital signs. However, clinicians focus on mean values recorded over minutes to hours. We’ve hypothesized that the natural variability in these physiologic patterns presented over very short time scales can be analyzed to yield subtle, early markers of impending clinical decline.”

Patient decompensation can happen fast, says Dr. Jacono, who also serves as chief of the Pulmonary and Critical Care Medicine Section at the Louis Stokes Cleveland Department of Veterans Affairs Medical Center.

“If we could pick up on critical changes in variability in the acute setting,” he says, “we might be able to alert physicians earlier than currently possible that the patient is about to get worse, so they can resuscitate, diagnose and treat them sooner.”

KEY AREAS OF FOCUS

Two areas of research interest are sepsis and acute respiratory distress syndrome (ARDS).

Sepsis. Dr. Jacono and his team are conducting an observational study of sepsis, a blood stream infection in which the body’s normal immune response becomes overwhelmed and dysregulated.

“The inflammation becomes so overt, it starts feeding back in a negative way and starts injuring organs in the body,” he says of the hallmark transition of sepsis. “We know that in patients with sepsis with shock, every hour of delay in initiating appropriate antibiotics can double the mortality rate.”

In preclinical models, Dr. Jacono and team are developing measures and metrics of variability – specifically heart rate variability, breathing pattern variability and the combination of the two – to see if a change early in the course of sepsis and inflammation might prove helpful.

ARDS. “ARDS is a very severe lung injury,” Dr. Jacono says, “So severe that patients end up on a ventilator having lots of trouble getting oxygen into their blood.” The complex condition is difficult to treat, leading to a mortality rate of 30 to 40 percent.

“In preclinical models of severe lung injury, we’re seeing how changes in breathing patterns, for example, may be markers of patients with mild lung injury who might go on to develop ARDS versus just getting better as part of their natural course of disease,” he says. Breathing pattern variability may be an augmented vital sign that could help clinicians in the critical care unit make therapeutic and diagnostic decisions about their patients.

PARTNERING IN OTHER RESEARCH

Dr. Jacono and team also collaborate in other research projects, such as the recently published, multi-center trial with patients who experience respiratory failure on mechanical ventilation. “We determined that an analytical approach focused on breathing pattern and heart rate variability could improve clinical decision processes about whether patients are ready to come off mechanical ventilation and start breathing on their own,” he says.

In another study, the team partnered with biotech startup Byteflies to trial and test a wearable health application capable of continuously monitoring physiologic signals. Building on related research in critically ill patients, Dr. Jacono envisions that the development of clinical-grade wearable technologies like the Byteflies Sensor Dot will someday “allow physicians to better monitor patients after discharge from the ICU to prevent subsequent decompensations that might require re-admission, improving patient outcomes and saving healthcare dollars.”

UH Cleveland Medical Center Medical Intensive Care Unit is also the top enrolling site for a phase 1/2 clinical trial of stem cell treatment (MultiStem) in patients with severe ARDS. The study is sponsored by Athersys, a Cleveland-based biotech firm.

If you’re part of the UH network of hospitals and your patients ends up in the ICU, they may be eligible to be included in this trial, Dr. Jacono says. “We’re investigating what could be the next new treatment for ARDS.”

“UH is on the cutting edge of developing novel approaches to the early identification of patients who are likely to get sick, and for whom known, effective therapies could be initiated in a timely fashion,” Dr. Jacono says. "The hope is that what we learn from these studies will eventually re-write the standard of care for patients with sepsis, ARDS and other serious illnesses.”

For questions about the latest advancements in the treatment of patients in the ICU, email Dr. Jacono at fjj@case.edu.