Noninvasive continuous detection of arterial hypotension during induction of anaesthesia using a photoplethysmographic signal: proof of concept

• Published in: British journal of anaesthesia : BJA Vol. 122; no. 5; pp. 605 - 612
• Main Authors: Coutrot, Maxime, Joachim, Jona, Dépret, François, Millasseau, Sandrine, Nougué, Hélène, Matéo, Joaquim, Mebazaa, Alexandre, Gayat, Etienne, Vallée, Fabrice
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.05.2019; Oxford University Press (OUP)
• Subjects: arterial hypotension; general anaesthesia; haemodynamics; intraoperative monitoring; Life Sciences; perfusion index; photoplethysmography; pulse oximetry; intraoperative monitoring; pulse oximetry; photoplethysmography; general anaesthesia; perfusion index; arterial hypotension; haemodynamics
• ISSN: 0007-0912; 1471-6771
• DOI: 10.1016/j.bja.2019.01.037

During general anaesthesia, intraoperative hypotension (IOH), defined as a mean arterial pressure (MAP) reduction of > 20%, is frequent and may lead to complications. Pulse oximetry is mandatory in the operating room, making the photoplethysmographic signal and parameters, such as relative dicrotic notch height (Dicpleth) or perfusion index (PI), readily available. The purpose of this study was to investigate whether relative variations of Dicpleth and PI could detect IOH during anaesthesia induction, and to follow their variations during vasopressor boluses. MAP, Dicpleth, and PI were monitored at 1-min intervals during target control induction of anaesthesia with propofol and remifentanil in 61 subjects. Vasopressor infusion (norepinephrine or phenylephrine) was performed when hypotension occurred according to the decision of the physician. The delta in Dicpleth and PI accurately detected IOH, with areas under the receiver operating characteristic curves (AUC) of 0.86 and 0.83, respectively. The optimal thresholds were –19% (sensitivity 79%; specificity 84%) and 51% (sensitivity 82%; specificity 74%) for ΔDicpleth and ΔPI, respectively. There was no difference between the ROC of ΔDicpleth and ΔPI (P=0.22). Combining both ΔDicpleth and ΔPI further improved the hypotension detection power (AUC=0.91) with a sensitivity and specificity of 84%. MAP variations were correlated with ΔDicpleth and ΔPI during vasopressor infusion (r=0.73 and –0.62, respectively; P<0.001). The relative variation in Dicpleth and PI derived from the photoplethysmographic signal can be used as a non invasive, continuous, and simple tool to detect intraoperative hypotension, and to track the vascular response to vasoconstrictor drugs during induction of general anaesthesia. NCT03756935.