Diagnostic accuracy of stroke volume variation measured with uncalibrated arterial waveform analysis for the prediction of fluid responsiveness in patients with impaired left ventricular function: a prospective, observational study

Abstract

Uncalibrated arterial waveform analysis enables dynamic preload assessment in a minimally invasive fashion. Evidence about the validity of the technique in patients with impaired left ventricular function is scarce, while adequate cardiac preload assessment would be of great value in these patients. The aim of this study was to investigate the diagnostic accuracy of stroke volume variation (SVV) measured with the FloTrac/Vigileo™ system in patients with impaired left ventricular function. In this prospective, observational study, 22 patients with a left ventricular ejection fraction of 40 % or less undergoing elective coronary artery bypass grafting were included. Patients were considered fluid responsive if cardiac output increased with 15 % or more after volume loading (7 ml kg⁻¹ ideal body weight). The following variables were calculated: area under the receiver operating characteristics (ROC) curve, ideal cut-off value for SVV, sensitivity, specificity, positive and negative predictive values, and overall accuracy. In addition, SVV cut-off points to obtain 90 % true positive and 90 % true negative predictions were determined. ROC analysis revealed an area under the curve of 0.70 [0.47; 0.92]. The ideal SVV cut-off value was 10 %, with a corresponding sensitivity and specificity of 56 and 69 % respectively. Overall accuracy was 64 %, positive and negative predictive values were 69 and 56 % respectively. SVV values to obtain more than 90 % true positive and negative predictions were 16 and 6 % respectively. The ability of uncalibrated arterial waveform analysis SVV to predict fluid responsiveness in patients with impaired LVF was low.