Predictors of outcome from critical illness. Shock and cardiopulmonary resuscitation

Many of our patients in ICUs suffer from shock, be it due to sepsis, trauma, arrest, or other causes. These patients continue to have a very high mortality rate in spite of very labor intensive and expensive treatment. The ability to identify patients who are likely to succumb to their illness is of utmost importance. Of the multitude of scoring systems published, the APACHE seems to accurately stratify shock patients according to severity of illness. However, these systems tend to be more useful for stratifying risk groups of patients than assessing the risk of death. Hemodynamic data can specifically assess the severity of the shock state in an individual patient. Those who maintain a relatively low cardiac index (< 4.5 L/m/M2) and oxygen delivery (< 15 mL/m/kg or 600 mL/m/M2) have persistent tissue hypoperfusion. Arterial lactate concentrations reflect the severity of this perfusion defect and correlate with outcome. Therefore, by restoring tissue perfusion, we can clearly improve mortality. CPP, although not generally obtainable during cardiac arrest, is the major physiologic determinant of outcome from CPR. ETCO2 monitoring during cardiac arrest in humans correlates with resuscitability, however, provides a rapid noninvasive monitor of cardiac output, and therefore has secured its role as an invaluable tool for assessing the effectiveness of CPR. An ETCO2 over 10 mm Hg is associated with effective CPR. A rapid rise in ETCO2 during CPR heralds recovery of spontaneous circulation. In conclusion, the use of prognostic indicators as predictors of outcome is supported as an important adjunct to the management of critically ill patients. These indicators serve as useful monitors to evaluate treatment and guide clinical management. Understanding the underlying pathophysiologic mechanisms responsible for the wide variety of illnesses associated with circulatory failure is crucial in our concerted effort to reduce mortality in these patients. As knowledge is gained, we hopefully will be able to develop more accurate and specific predictors of outcome to prudently select patients most likely to benefit.