Open AccessCan bioimpedance cardiography assess hemodynamic response to passive leg raising in critically ill patients
Author(s) -
Li Li,
Yuhang Ai,
Li Huang,
Meilin Ai,
Qianyi Peng,
Lina Zhang
Publication year - 2020
Publication title -
medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.59
H-Index - 148
eISSN - 1536-5964
pISSN - 0025-7974
DOI - 10.1097/md.0000000000023764
Subject(s) - medicine , impedance cardiography , cardiology , receiver operating characteristic , predictive value , cardiac output , shock (circulatory) , hemodynamics , area under the curve , critically ill , predictive value of tests , stroke volume , ejection fraction , heart failure
Passive leg raising (PLR) is a convenient and reliable test to predict fluid responsiveness. The ability of thoracic electrical bioimpedance cardiography (TEB) to monitor changes of cardiac output (CO) during PLR is unknown. In the present study, we measured CO in 61 patients with shock or dyspnea by TEB and transthoracic echocardiography (TTE) during PLR procedure. Positive PLR responsiveness was defined as the velocity-time integral (VTI) ≥10% after PLR. TTE measured VTI in the left ventricular output tract. The predictive value of TEB parameters in PLR responders was tested. Furthermore, the agreement of absolute CO values between TEB and TTE measurements was assessed. Among the 61 patients, there were 28 PLR-responders and 33 non-responders. Twenty-seven patients were diagnosed with shock and 34 patients with dyspnea, with 55.6% (15/27) and 54.6% (18/34) non-responders, respectively. A change in TEB measured CO (ΔCO) ≥9.8% predicted PLR responders with 75.0% sensitivity and 78.8% specificity, the area under the receiver operating characteristic curve (AUROC) was 0.79. The Δd 2 Z /d t 2 (a secondary derivative of the impedance wave) showed the best predictive value with AUROC of 0.90, the optimal cut point was −7.1% with 85.7% sensitivity and 87.9% specificity. Bias between TEB and TTE measured CO was 0.12 L/min, and the percentage error was 65.8%. TEB parameters had promising performance in predicting PLR responders, and the Δd 2 Z /d t 2 had the best predictive value. The CO values measured by TEB were not interchangeable with TTE in critically ill settings.