The effect of lung injury and excessive lung fluid, on impedance cardiac output measurements, in the critically ill.

Abstract

To investigate the relationship between the attenuation of impedance cardiac output (IC(co)) measurements and lung fluid content in critically ill patients. Observational study. Intensive Care Unit of a major teaching hospital in Hong Kong. Twenty-four critically ill patients who required a pulmonary artery catheter. Triplicate thermodilution cardiac output (TD(co)) and BoMed NCCOM3 (IC(co)) measurements were made simultaneously on a single occasion in each patient. Lung fluid accumulation was assessed by: (a) thoracic impedance (Zo), (b) radiological assessment of chest X-rays using an alveolar consolidation score (0-4) and (c) scoring the degree of hypoxia and use of positive end-expiratory pressure (PEEP). Offsets (TD(co)-IC(co))/TD(co), expressed as percentage, were compared with these indices of excess lung fluid. Patients were divided into those with sepsis (n = 13), fluid balance problems (n = 5) and cardiothoracic problems (n = 6). Mean cardiac output values were: 6.7 l/min TD(co) (range 3.6-12.9) and 5.2 l/min IC(co) (range 2.7-9.0). Overall the TD(co) and IC(co) values showed great variance, with a bias and limits of agreement of 1.49 +/- 4.16 l/min, or +/- 69%. In septic patients, increasing offset was correlated with decreases in Zo (r = 0.73, P = 0.005) and increases in alveolar consolidation score (r = 0.72, P = 0.005). The BoMed under-estimates cardiac output in critically ill patients. In septic patients the degree of attenuation of IC(co) can be related to the extent of lung injury and fluid accumulation within the thorax.