A performance comparison of the most commonly used minimally invasive monitors of cardiac output.

Abstract

Shock is common in critically ill and injured patients. Survival during shock is highly dependent on rapid restoration of tissue oxygenation with therapeutic goals based on cardiac output (CO) optimization. Despite the clinical availability of numerous minimally invasive monitors of CO, limited supporting performance data are available. Following approval of the University of Saskatchewan Animal Research Ethics Board, we assessed the performance and trending ability of PiCCOplus™, FloTrac™, and CardioQ-ODM™ across a range of CO states in pigs. In addition, we assessed the ability of invasive mean arterial blood pressure (iMAP) to follow changes in CO using a periaortic transit-time flow probe as the reference method. Statistical analysis was performed with function-fail, bias and precision, percent error, and linear regression at all flow, low-flow (> 1 standard deviation [SD] below the mean), and high-flow (> 1 SD above the mean) CO conditions. We made a total of 116,957 paired CO measurements. The non-invasive CO monitors often failed to provide a CO value (CardioQ-ODM: 40.6% failed measurements; 99% confidence interval [CI], 38.5 to 42.6; FloTrac: 9.6% failed measurements; 99% CI, 8.7 to 10.5; PiCCOplus: 4.7% failed measurements; 99% CI, 4.5 to 4.9; all comparisons, P < 0.001). The invasive mean arterial pressure provided zero failures, failingless often than any of the tested CO monitors (all comparisons, P < 0.001). The PiCCOplus was most interchangeable with the flow probe at all flow states: PiCCOplus (20% error; 99% CI, 19 to 22), CardioQ-ODM (25% error; 99% CI, 23 to 27), FloTrac (34% error; 99% CI, 32 to 38) (all comparisons, P < 0.001). At low-flow states, CardioQ-ODM (43% error; 99% CI, 32 to 63) and Flotrac (45% error; 99% CI, 33 to 70) had similar interchangeability (P = 0.07), both superior to PiCCOplus (48% error; 99% CI, 42 to 60) (P < 0.001). Regarding CO trending, the CardioQ-ODM (correlation coefficient, 0.82; 99% CI, 0.81 to 0.83) was statistically superior to other monitors including iMAP, but at low flows iMAP (correlation coefficient, 0.58; 99% CI, 0.58 to 0.60) was superior to all minimally invasive CO monitors (all comparisons P < 0.001). None of the minimally invasive monitors of CO performed well at all tested flows. Invasive mean arterial blood pressure most closely tracked CO change at critical flow states.