To evaluate the reliability of the vital signs to evaluate circulatory stability as reflected by cardiac index. Descriptive analysis based on data gathered prospectively, using a predetermined protocol. University-run county hospital, with a large trauma service. Sixty-one high-risk trauma patients with accidental injury who were studied immediately after admission to the Emergency Department, and subsequently, 163 critically ill postoperative ICU patients. Standard fluid therapy, usually crystalloids, but occasionally packed red cell transfusions and colloids, as indicated by clinical criteria. Arterial BP was measured by pressure transducer and arterial catheter; heart rate (HR) was measured by electrocardiograph signal, and cardiac output was measured by thermodilution. In sudden severe hypovolemic hypotension, the mean arterial pressure (MAP) nadir (lowest) roughly correlated (r2 = .25) with flow, but there was poor correlation (r2 = .0001) when all pressure and flow values were evaluated. The pressure and flow values were obtained throughout the course of the hypotensive episodes during the initial resuscitation in ICU patients and during terminal illnesses. Observations at the time of acute severe hypotensive crises that show rough correlation of MAP and cardiac index should not be extrapolated throughout the entire hypotensive period or to other less extreme clinical situations. The stress response to hypovolemia, with endogenous catecholamines and neural mechanisms, tends to maintain arterial pressure in the face of decreasing flow for a variable period of time. However, when these mechanisms are overwhelmed by prolonged hypovolemia, the pressure decreases precipitously, but not synchronously, with flow. We conclude that blood flow cannot reliably be inferred from arterial pressure and heart rate measurements until extreme hypotension occurs.
