Functional genomics of critical illness and injury

Despite the increasing wealth of available patient information, diagnosis in the intensive care unit (ICU) remains paradoxically and frustratingly difficult. This is caused, in part, by the complicated nature of critical illness coupled with the frequent inability of ICU patients to communicate. All critical care physicians can attest to the frustration of suspecting that there is a pathologic process lurking somewhere within, leaving only the subtlest clues. For example, consider the critically ill patient with a mild leukocytosis and a new low-grade fever. Is there an infection? Is it sepsis? If so, where is the source? The computed tomographic scan of the abdomen or chest is frequently nonspecific and has not made obsolete the diagnostic utility of careful, serial physical examinations. Cultures of sputum, blood, urine, or cerebrospinal fluid are neither fast nor sensitive. The diagnostic armamentarium abounds with other laboratory tests, such as serum electrolytes, liver function tests, lactic acid levels, and coagulation profiles, that by themselves serve only as markers of organ dysfunction. Should broad-spectrum antibiotics be started empirically? As a result of their indiscriminate use, nosocomial infections with multidrugresistant organisms continue to rise with alarming frequency. In the back of the clinician’s mind, other haunting questions are whispered: Is this the start of multiple organ dysfunction syndrome (MODS)? If so, will it lead to death? How can these issues be resolved? Recent advances in biotechnology and the sequencing of the human genome offer unprecedented opportunities to increase our understanding of critical illness and injury. More importantly, the exciting new field of functional genomics will provide the scientific basis for dramatic improvements in the accuracy of diagnostic and prognostic tests and for identifying new therapeutic targets (1). In the examples cited above, for instance, whole-genome changes in circulating leukocyte transcription and translation can be examined to determine what these cells “know” that the patients, numerous available blood tests, and sophisticated radiologic studies cannot tell. This review summarizes the new investigative paradigms, technology, and analysis that make broad-scale genomics possible, along with some recent data from studies of critical illness and injury in patients and clinically relevant animal models (2– 6).