• Siyuan Xie, Ding Han, Gang Chen, and Shoudong Pan.
• Department of Anesthesia, Capital Institute of Pediatrics Affiliated Children's Hospital, Beijing, China.
• Medicine (Baltimore). 2022 Dec 9; 101 (49): e31795e31795.

AbstractOptimal perioperative fluid management is essential for reducing complications in children undergoing thoracoscopic surgery. The study aimed to assess the performance of 2 dynamic preload parameters - pulse pressure variation (PPV) and stroke volume variation (SVV)- either used alone or combined into a multivariable regression model for predicting fluid responsiveness in children undergoing video-assisted thoracoscopic surgery with one-lung ventilation. Children aged 1 to 6 years old undergoing video-assisted pulmonary segmentectomy or lobectomy were enrolled. Volume loading with 5 mL/kg of hydroxyethyl starch was administered over 15 minutes after establishment of artificial pneumothorax. PPV, SVV, cardiac index, cardiac cycle efficiency, and the difference between systolic blood pressure and dicrotic pressure were recorded using the pressure recording analytical method before and after volume loading. Patients with an elevation in cardiac index greater than 10% were defined as responders, and the remaining patients were nonresponders. Of 40 children, 36 were included in the final analysis, containing 13 responders and 23 nonresponders. SVV had an accuracy of 74% (95% confidence interval, 55-93%) for predicting fluid responsiveness, and a best cutoff of 22% showed a sensitivity of 62% and a specificity of 96%. PPV was incapable of discriminating responders from nonresponders. The multivariate regression model did not perform better than SVV alone. We found PPV failed to predict fluid responsiveness, while SVV predicted fluid responsiveness reasonably in the present context. There was no enhancement in predictivity accuracy with multivariable regression models. The accuracy of these approaches was limited, and more discriminative methods need to be found.Copyright © 2022 the Author(s). Published by Wolters Kluwer Health, Inc.

23 keywords...

hide…