Predictive value of carotid ultrasound in combination with passive leg raising on fluid responsiveness in critically ill patients

Abstract

To investigate the value of monitor carotid velocity time integral (VTI) and corrected flow time (FTc) by bedside ultrasound before and after passive leg raising (PLR) in predicting fluid responsiveness in critically ill patients. A prospective observational study was conducted. Fifty patients with critical illness admitted to the First People's Hospital of Fuyang Hangzhou from January 2020 to March 2021 were enrolled. The clinical data including the gender, age, body mass index (BMI), acute physiology and chronic health evaluation II (APACHE II) score, and the duration of mechanical ventilation were recorded. The changes of carotid VTI and FTc were measured by bedside ultrasound, and the values of heart rate, mean arterial pressure (MAP), central venous pressure (CVP), stroke volume index (SVI), and intrathoracic blood volume index (ITBVI) were measured by pulse indicated continuous cardiac output (PiCCO) monitor before and after PLR in all patients. According to the changes of SVI before and after PLR, the patients were divided into fluid responsiveness positive group with the change rate of SVI ≥ 15% and fluid responsiveness negative group with the change rate of SVI < 15%. The differences in the values of VTI, FTc, CVP, and ITBVI obtained before and after PLR (ΔVTI, ΔFTc, ΔCVP and ΔITBVI) were calculated and then compared between the two groups. The predictive values of these indicators on fluid responsiveness in critically ill patients were analyzed by receiver operator characteristic curve (ROC curve), and their relationship with the difference in SVI (ΔSVI) obtained before and after PLR was evaluated by Pearson correlation analysis. Fifty patients were all enrolled in this study, in which 27 patients were fluid response and 23 patients were fluid nonresponse. Basic clinical data were not different between the two groups. The values of ΔVTI, ΔFTc, ΔCVP, and ΔITBVI in fluid response were all significantly higher than those in fluid nonresponse [ΔVTI (cm): 2.07±1.16 vs. 0.67±0.86, ΔFTc (ms): 4.00±6.10 vs. 0.01±2.26, ΔCVP (cmH2O, 1 cmH2O = 0.098 kPa): 1.67±1.14 vs. 1.00±1.17, ΔITBVI (mL/m2): 98±69 vs. 48±70, all P < 0.05]. ROC curve analysis showed that ΔVTI, ΔFTc, ΔCVP and ΔITBVI were all positive for predicting fluid responsiveness, their area under ROC curve (AUC) and 95% confidence interval (95%CI) were 0.870 (0.769-0.972), 0.694 (0.547-0.841), 0.684 (0.535-0.832) and 0.709 (0.564-0.855), respectively. When using ΔVTI 0.92 cm, ΔFTc 1.45 ms, ΔCVP 1.50 cmH2O and ΔITBVI 44.50 mL/m2 as the threshold values, the sensitivities were 96.3%, 63.0%, 44.4% and 81.5%, and the specificities were 65.2%, 78.3%, 82.6% and 56.5%, respectively, in which the predictive value of ΔVTI was the largest. Pearson correlation analysis indicated that ΔVTI, ΔFTc, ΔCVP, and ΔITBVI were positively associated with ΔSVI (r values were 0.971, 0.334, 0.440, 0.650, P values were 0.000, 0.018, 0.001, 0.000, respectively). Carotid ΔVTI and ΔFTc monitored by bedside ultrasound before and after PLR could be as effective as conventional indicators in predicting fluid responsiveness in critically ill patients, and the predictive value of ΔVTI was better than others.