Mechanical ventilation with positive end-expiratory pressure in critically ill patients: comparison of CW-Doppler ultrasound cardiac output monitoring (USCOM) and thermodilution (PiCCO).

BACKGROUND Aggressive mechanical ventilation can markedly and unpredictably affect cardiac function. The fall in cardiac output (CO) is due to a reduction in left ventricular stroke volume (SV). The aim of the present pilot study was to assess the effects of different positive end-expiratory pressure (PEEP) levels on circulatory function and to compare them with continuous wave (CW)-Doppler ultrasound cardiac output monitoring (USCOM) and a thermodilution-based haemodynamic monitoring system (PiCCO). METHODS Twenty mechanically ventilated (PEEP < or = 10 mbar) adult patients (female n = 6, male n = 14, mean age 62 years, mean SAPS II-score 48.5), the majority with pneumonia and septic shock) were followed with USCOM and PiCCO at stepwise increased PEEP-levels from 0-10 mbar (1 mbar steps). The changes in CO/SV were recorded. RESULTS With both methods, an increase of PEEP resulted in a decrease of SV and CO. Although the absolute decrease was consistently higher by USCOM, the changes of the parameters were qualitatively comparable. CO fell from 8.83 L/min (+/- 2.39) by 0.4 L/min to 8.49 L/min (+/- 2.48) with PiCCO and from 9.3 L/min (+/- 3.43) by 1.0 L/min to 8.3 L/min (+/- 3.2) with USCOM. The median CO/SV fell by 4.5%/5.2% with PiCCO and 10.8%/9% with USCOM, respectively. Correlation of CO values with the two methods by Bland-Altman yielded comparable results (mean percentage error at PEEP 0 mbar 13%, PEEP 10 mbar 18%). An adequate flow signal with USCOM was achieved in all patients. CONCLUSIONS A significant influence of mechanical ventilation with PEEP on haemodynamic parameters was evident both with USCOM and PiCCO. While thermodilution methods like PiCCO are well established but time-consuming and invasive, CW-Doppler based USCOM constitutes an important tool for easy, rapid and reliable diagnosis and haemodynamic monitoring of critically ill patients.