Cardiopulmonary resuscitation: a retrospective review.

The outcome of patients admitted to intensive care after a cardiac arrest was determined by reviewing intensive care unit records at four hospitals for 1993 and 1994. Of the 112 patients identified, 49 survived intensive care of whom 28 were discharged from hospital. In January 1996, 26 of the 28 patients could be traced; 22 of these were still alive. Seven factors were significantly different between survivors and nonsurvivors. At the cardiac arrest these were the number of direct current shocks (p < 0.05) and adrenaline doses (p < 0.01) given. In intensive care the factors were the presence of reactive pupils (p < 0.01), Glasgow Coma Score (p < 0.001), APACHE II score (p < 0.05), arterial standard bicarbonate (p < 0.05) and the use of inotropes (p < 0.05). It was not possible to use individual variables to predict outcome at the time of intensive care unit admission. The results suggest that neurological function is an important determinant of outcome and more sensitive neurophysiological testing might be a useful prognostic tool.