From Emergency Care in Sarajevo to the ICU at Mayo

Ognjen Gajic, M.D., was born in Sarajevo, Yugoslavia and was just beginning his residency there when the Bosnian War began. He endured two and a half years of living and working in a city under siege before he was able to rejoin his young family. In 1995, the Gajics emigrated to the U.S.A. as war refugees. Looking for opportunities to reenter his profession, Dr. Gajic volunteered as an observer in a Rhode Island hospital until he was accepted into an internal medicine residency program in New York City. Tenacity is also a characteristic of Dr. Gajic's career in research, which has its roots in the lab of Mayo researcher, Rolf Hubmayr, M.D.

"I was immediately struck by Dr Gajic's tenacity, spirit of inquiry, and energy when I met him on rounds," recalls Dr. Hubmayr. "I made a comment about heart-lung interactions and the next thing I knew, he had reviewed the literature on the topic, made appropriate measurements in patients, and presented me with a preliminary study proposal."

The Basic Research Connection

Having discovered his passion for investigation, Dr. Gajic changed his initial career plans during an extended elective in the Hubmayr lab.

"I came here to be a clinician but my mentor turned my whole world around," says Dr. Gajic. "Rolf encouraged me to push the limits of my abilities by learning and applying scientific methodology."

During his elective, Dr. Gajic established a novel, microscopy-based method to evaluate cell injury and repair mechanisms in ventilator-injured lungs. The first of several papers related to this work made the cover of the American Journal of Respiratory and Critical Care Medicine (Am J Respir Crit Care Med. 2003 Apr 15;167(8):1057-63). By injecting a fluorescent marker that could only penetrate through a torn cell membrane, the group was able to count the number of injured cells under confocal microscopy.

"Single stretch experiments showed no signs of injury," says Dr. Gajic. "But within 30 cycles, the whole lung lit up--almost every cell was injured."

The studies built upon a seminal multi-center clinical trial that demonstrated a significant improvement in the mortality rate of people with the most serious form of acute lung injury--acute respiratory distress syndrome (ARDS). The improvement was made by simply decreasing the size of each breath on the ventilator.

The Hubmayr lab's contribution was to show that when previously healthy lungs are repeatedly overstretched by large tidal volumes (the volume of gas in an individual breath), lung cells break, suggesting that mechanical ventilation may cause lung injury in people with formally healthy lungs--a condition now termed ventilator-induced lung injury.

Dr. Gajic conducted follow-up studies headed by Nicholas Vlahakis, M.D., (a post-doc in the Hubmayr lab at the time) which showed that lungs injured by mechanical ventilation are able to repair themselves. The team effort was bolstered by the recruitment of Bekele Affessa, M.D., a critical care specialist who began analyzing large amounts of patient data and pushing for greater efforts to standardize care across all Mayo ICUs.

At the same time, Dr. Hubmayr was also directing the second largest critical care system in the U.S. -- Mayo has 192 intensive care beds in ten ICUs. His position reinforced the enormous potential of bringing this quality improvement to the ICU bedside. Dr. Hubmayr understood that the largest group of patients treated with mechanical ventilation in the nation's ICUs is the post-surgery group -- most of whom have normal lungs, and some of whom die of acute lung injury. The team's research showed these were lives that might be saved by enforcing lower tidal volumes. But that would take a person with special skills.

The Making of a Critical Care Quarterback

In 2002, with the help of a co-mentors Dr. Afessa and Chair of Biomedical Informatics, Christopher Chute, M.D., Ph.D., Dr. Gajic began studying for a Master's Degree in Mayo's Clinical Research Training Program. His thesis stemmed from a laborious retrospective chart review of 400 patients who had no previous lung disease, and were mechanically ventilated in a Mayo ICU in 2001. Investigators from the Thoracic Diseases Research Unit and the Department of Anesthesiology's Division of Intensive Care and Respiratory Care, collaborated with him in the study. His findings: when acute lung injury follows lung surgery, it is related to mechanical injury by the ventilator during anesthesia and surgery.

In the course of combing the data, Dr. Gajic uncovered a second major treatment-induced risk factor for acute lung injury--chemical injury caused by the transfusion of blood products.

Transfusion-related acute lung injury (TRALI)

Twenty years ago, a Mayo Clinic transfusion specialist, S. Breanndan Moore, M.D., was the first to describe transfusion-related acute lung injury (TRALI)--the sudden accumulation of fluid in the lungs after blood transfusion. TRALI is now recognized as the most important, and potentially preventable, severe complication of blood transfusion.

"You can think about a blood transfusion as a cell transplant," explains Dr. Gajic. "It can cause an immune reaction that causes tissue damage. Our studies revealed that even small amounts of transfused blood products can injure the lungs."

The discovery, which supplements a 1999 Canadian study, is considered a sentinel report--one that changed Mayo practice immediately and is changing the standard of transfusing blood products in critically ill patients nationwide.

Supported by two Specialized Centers of Clinically Oriented Research (SCCOR) grants from the National Institutes of Health (NIH) and another from the National Blood Foundation, the team is now collaborating with researchers, both within Mayo and from other universities, to discover the mechanisms, treatment and prevention of TRALI.

METRIC

By 2004, Dr. Gajic had acquired a unique set of skills and knowledge in critical care physiology, clinical epidemiology, and informatics. His thesis opened the door to many opportunities for improving critical care. It highlighted the need for better communication between the various Mayo databases, it initiated new research to study interventions that might reduce the risk of developing acute lung injury, and it indicated how informatics could help bring best practices to the ICU bedside--all efforts now undertaken by Mayo Epidemiology and Translational Research in Intensive Care, or METRIC; an appropriate acronym for a lab that measures outcomes.

METRIC is an interdisciplinary clinical research lab launched by Dr. Gajic to improve quality and safety in the ICU. It has since initiated many projects besides those that limit tidal volumes and blood transfusions. Other studies include identifying triggers for sepsis, managing antibiotics, and assessing sedation and glucose control. The team has also developed computerized decision systems to ensure that the new standards of practice are followed in Mayo ICUs. For example, it is no longer possible to order a blood transfusion in writing at Mayo. By using only computerized orders, the system alerts physicians when the patient's hemoglobin is outside an acceptable range. Having physicians justify the need for transfusion has reduced its use by almost one third.

"The computerized system turns a traditional response into a thoughtful review of all of the implications of a particular intervention," says Dr. Gajic. "The importance of linking epidemiological research and systems engineering in order to standardize the delivery of care becomes clear when you consider the complexity of treating critically ill patients."

Intensivists (critical care specialists) deal with acute management of shock, cardiopulmonary resuscitation, cardiac failure, respiratory failure, multisystem organ failure, sepsis and other infections, central nervous system crises, and organ transplantation. In addition to considering how the underlying disease or injury impairs the physiology of all organ systems, they analyze how the physiology is changed by life support interventions, drugs and other treatments, and weigh the patient's genetic predispositions and the family's preferences. Now, factor in the thought processes of hundreds of individual caregivers, and the organization of individual ICUs and you'll begin to understand the need for computer-assisted decision support.

"We collect thousands of pieces of information about a single patient every day," says Dr. Hubmayr. "It's not humanly possible to assimilate all of it and make the best practices decision every time."

One project aims to reduce errors by standardizing patient rounds. Anesthesiologist Daniel Brown, M.D., Ph.D., heads up a team that is developing an automated check list for best ventilator practices that is reviewed, on every patient, every day.

"The system also generates a document of management decisions and a daily to-do list," says Dr. Brown. "And it feeds the data base with information that has been validated by the entire team of physicians, nurses, respiratory therapists, pharmacists and the patient's family--data that is vital for continual improvement."

The combination of multiple projects has rewarded the team with remarkable success. The hospital mortality of mechanically ventilated patients has been reduced by an astounding 33 percent.

Critical Teamwork: Critical Success

A perusal of METRIC faculty hints at the breadth and depth of teamwork involved in the effort.

"METRIC represents an important forum for communication and multidisciplinary collaboration between scientists, clinician scholars, health care administrators and allied health care providers--especially nursing, respiratory therapy and pharmacy--engaged in the delivery of critical care services," says Dr. Hubmayr. "For example, we are indebted to Marcy Harris for her leadership role in nursing research."

Marcelline Harris, RN, Ph.D., is a biomedical informatics specialist for the Divisions of Medical Informatics Research and Nursing Research. Her partnership is critical to METRIC's goals because the data that physicians and nurses need to access, in order to improve practice, is very similar. Funded by three extramural grants, Dr. Harris is leading the formation of a clinical informatics unit focusing on infrastructure such as data marts and terminology systems. Her cross collaborations between nursing, biomedical informatics, and critical care have resulted in more specialty nurses being embedded in Mayo ICUs as quality coaches.

Honing Quality Improvements in Mayo ICUs

METRIC is now building an automated infrastructure to measure management and outcomes in critical care. The upgraded system can instantly retrieve data from 1998 to the present, allowing researchers to assess the huge spectrum of risk factors involved in critical care medicine. It now takes one person a few minutes to complete a 400-patient search compared to the one for Dr. Gajic's thesis that consumed an entire team for a full year.

"Ogie has built the equivalent of a Google browser to uncover constellations of findings that we don't currently recognize as triggers," says Dr. Hubmayr. "We are also using it to help us make better decisions about how we staff our ICUs and how we educate our residents."

Mayo has one of the most sophisticated medical information systems in the nation. By transitioning information technology into the clinical practice, Mayo continuously observes the clinical course and outcomes of critically ill patients. By using basic epidemiologic principles to identify potentially beneficial or harmful practices, and medical informatics and electronic medical records to help implement best practices, Mayo is improving the quality of care in the ICU in a cost-effective manner every day.

Says Dr. Hubmayr, "When it comes to critical care, the delivery of care and the scholarly evaluation of practice cannot be uncoupled."