TY - JOUR AU - Lee, Christine K. AU - Rinehart, Joseph AU - Canales, Cecilia AU - Cannesson, Maxime PY - 2014 DA - 2014/01/10 TI - Comparison of automated vs. manual determination of the respiratory variations in the EKG R wave amplitude for the prediction of fluid responsiveness during surgery JO - Journal of Computational Surgery SP - 5 VL - 1 IS - 1 AB - Electrocardiogram (EKG) monitoring is a common standard of care across all operating rooms and intensive care units. Studies have suggested that respiratory variations in the EKG R wave amplitude (EKGv) can be used as an indicator of fluid responsiveness in mechanically ventilated patients under general anesthesia, but to date all calculations of variation have been done by hand. The aim of this study was to assess if a computer-automated algorithm could compute and monitor EKGv with the same precision as manual measurement. Batches of 30 s each of EKG lead II waveforms were recorded during surgical procedures with mechanical ventilation. R wave amplitude variability was assessed both manually and by automated algorithm. For both calculations, wave height was defined as R wave peak minus preceding Q wave trough, and the minimum and maximum amplitudes determined for each respiratory cycle. EKGv was calculated as 100 × [(RDIImax − RDIImin)/(RDIImax + RDIImin)/2]. Fifty-seven batches of waveforms were calculated. We found that our computer-automated algorithm calculation of EKGv was significantly correlated to manual measurements (r = 0.968, P < 0.001). Bland-Altman analysis also showed a strong agreement between automated and manual EKGv measurements (bias 0.13% ± 3.06%). The observed correlations between the manually and automatically calculated EKGv suggest that our current computer-automated algorithm is a reliable method for calculating EKGv. Validation in prospective volume expansion studies will be needed to assess the true clinical utility of this automated measurement. SN - 2194-3990 UR - https://doi.org/10.1186/2194-3990-1-5 DO - 10.1186/2194-3990-1-5 ID - Lee2014 ER -