Cardiopulmonary assessment: is improvement needed?

Clinical parameters alone have repeatedly been proven unreliable in assessing cardiopulmonary status, especially in hemodynamically unstable patients. To learn if we had a diagnostic problem in our hospital, we compared physician assessment of cardiac index (CI) and thoracic fluid content (TFC) to values obtained using impedance cardiography (ICG). We selected the newest available ICG monitor, the BioZ, which employs this noninvasive technology. For CI measurements we have shown it to be equivalent to thermodilution and to be more reproducible (variability: 6.3% vs. 24.7%). Physician assessment of CI and TFC (high, normal, or low) was compared to the BioZ monitor's results in 186 patients, considered to be hemodynamically unstable, from the emergency room, the intensive care units, and the floors. Normal values were defined for CI (2.5-4.2 L/min m(2)) and for TFC (males: 30-50 kohm(-1) and females: 21-37 kohm(-1)). The concordance between physician assessment and the BioZ was 51% for CI with Kappa of 0.14 and 58% for TFC with Kappa of 0.19. Attendings did slightly better than the surgical residents with CI (52% vs. 48%) but slightly worse with TFC (57% vs. 61%). The potentially serious conditions of low CI and high TFC were misdiagnosed 42% and 46% of the time, respectively, by all physicians. Analysis of the data revealed that physician use of clinically available objective hemodynamic data, such as heart rate, blood pressure, and pulse pressure index, would not have been helpful. Furthermore, assistance from the pulmonary artery catheter (PAC) is often not available in our hospital, which has experienced a 90% decrease in its utilization over the past six years. Considering the increasing acuity of our aging patient population, accurate assessment of cardiopulmonary status is needed. The use of ICG could be a valuable addition to the physician's armamentarium.