ABSTRACT
PURPOSE: Bedside transthoracic echocardiography (TTEcho) is a noninvasive cardiac output (CO) monitoring method that has grown recently. However, there are questions regarding its accuracy compared to invasive methods. We aimed to evaluate the agreement and correlation of TTEcho and pulse index continuous CO (PiCCO) monitor measurements for CO and systolic volume (SV) in critically ill patients. METHODS: This prospective experimental study included consecutive adult patients who required invasive hemodynamic monitoring admitted at an intensive care unit in the Federal District, Brazil, from January/2019 to January/2021. Correlation and agreement between SV and CO measurements by PiCCO and TTEcho were performed using the Spearman correlation and the Bland-Altman analysis. RESULTS: The study enrolled 29 patients, with adequate TTEcho evaluations in all patients. There were very strong correlations between CO-TTEcho and CO-PiCCO (r = 0.845, P < .001) and SV-TTEcho and SV-PiCCO (r = 0.800, P < .001). TTEcho estimations for CO and SV were feasible within the limits of agreement in 96.6% (28/29) compared to PiCCO. The mean difference between CO-PiCCO and CO-TTEcho was 0.250â L/min (limits of agreement: -1.083 to 1.583â L/min, percentage error: 21.0%), and between SV-PiCCO and SV-TTEcho was 2.000â mL (limits of agreement: -16.960 to 20.960, percentage error: 24.3%). The reduced cardiac index (CI) measurements by TTEcho showed an accuracy of 89.7% (95% IC: 72.6%-97.8%) and an F1 score of 92.7% (95% IC: 75.0%-98.0%), considering the CI-PiCCO as the gold standard. CONCLUSION: Echocardiographic measurements of CO and SV are comparable to measurements by PiCCO. These results reinforce echocardiography as a reliable tool to evaluate hemodynamics in critically ill patients.