Comparison of cardiac index measurements in intensive care patients using continuous wave vs. pulsed wave echo-Doppler compared to pulse contour cardiac output

Cardiac index (CI) assessments are commonly used in critical care to define shock aetiology and guide resuscitation. Echocardiographic assessment is non-invasive and has high levels of agreement with thermodilution assessment of CI. CI assessment is derived from the velocity time integral (VTI) assessed using pulsed wave (PW) doppler at the level of the left ventricular outflow tract divided by body mass index. Continuous wave (CW) doppler through the aortic valve offers an alternative means to assess VTI and may offer better assessment at high velocities.We performed a single centre, prospective, observational study in a 15-bed intensive care unit in a busy district general hospital. Patients had simultaneous measurements of cardiac index by Pulse Contour Cardiac Output (PiCCO) (thermodilution), transthoracic echocardiographic PW-VTI and CW-VTI. Mean differences were measured with Bland-Altman limits of agreement and percentage error (PE) calculations.Data were collected on 52 patients. 71% were supported with noradrenaline with or without additional inotropic or vasopressor agents. Mean CIs were: CW-VTI 2.7 L/min/m2 (range 0.78-5.11, SD 0.92). PW-VTI 2.33 L/min/m2 (range 0.77-5.40, SD 0.90) and PiCCO 2.86 L/min/m2 (range 1.50-5.56, SD 0.93). CW-VTI and PiCCO mean difference was - 0.16 L/min/m2 PE 43.5%. PW-VTI and PiCCO had a mean difference of - 0.54 L/min/m2 PE 38.6%. CW-VTI and PW-VTI had a mean difference of 0.38 L/min/m2 PE 46.0%.CI derived from both CW-VTI and PW-VTI methods underestimate CI compared to PiCCO, with the CW-VTI method having closer values overall to PiCCO. CW-VTI may offer a more accurate assessment of CI. If using Critchley's PE cutoff of 30%, none of the doppler methods may accurately reflect the actual cardiac index.