Temporal concordance between pulse contour analysis, bioreactance and carotid doppler during rapid preload changes

by Jon-Émile S. Kenny, Igor Barjaktarevic, Andrew M. Eibl, Matthew Parrotta, Bradley F. Long, Mai Elfarnawany, Joseph K. Eibl

We describe the temporal concordance of 3 hemodynamic monitors.

Healthy volunteers performed preload changes while simultaneously wearing a non-invasive, pulse-contour stroke volume (SV) monitor, a bioreactance SV monitor and a wireless, wearable Doppler ultrasound patch over the common carotid artery. The sensitivity and specificity for detecting preload change over 3 temporal windows (early, middle and late) was assessed.

40 preload changes were recorded in total (20 increase, 20 decrease). Immediately, the wearable Doppler had high sensitivity (100%) and specificity (100%) for detecting preload change with an area under the receiver operator curve (AUROC) of 0.98 for both velocity time integral (VTI, 10.5% threshold) and corrected flow time (FTc, 2.5% threshold). The sensitivity, specificity and AUROC for non-invasive pulse contour were equally good (9% SV threshold). For bioreactance, a 13% SV threshold immediately detected preload change with a sensitivity, specificity and AUROC of 60%, 95% and 0.75, respectively. After two SV outputs following preload change, the sensitivity, specificity and AUROC of bioreactance improved to 70%, 90% and 0.85, respectively.

Carotid Doppler ultrasound and non-invasive pulse contour detected rapid hemodynamic change with equal accuracy; bioreactance improved over time. Algorithm-lag should be considered when interpreting clinical studies.