Oesophageal Doppler and calibrated pulse contour analysis are not interchangeable within a goal-directed haemodynamic algorithm in major gynaecological surgery

• Published in: British journal of anaesthesia : BJA Vol. 113; no. 5; pp. 822 - 831
• Main Authors: Feldheiser, A., Hunsicker, O., Krebbel, H., Weimann, K., Kaufner, L., Wernecke, K.-D., Spies, C.
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.11.2014
• Subjects: Algorithms; Calibration; Echocardiography, Transesophageal - methods; Female; goal-directed therapy; Gynecologic Surgical Procedures - methods; Gynecologic Surgical Procedures - mortality; haemodynamic monitoring; Hemodynamics - physiology; Humans; Middle Aged; Monitoring, Intraoperative - instrumentation; Monitoring, Intraoperative - methods; oesophageal Doppler; Ovarian Neoplasms - surgery; Prospective Studies; Pulse - statistics & numerical data; pulse contour analysis; Stroke Volume - physiology; trending; trending; oesophageal Doppler; goal-directed therapy; haemodynamic monitoring; pulse contour analysis
• ISSN: 0007-0912; 1471-6771
• DOI: 10.1093/bja/aeu241

Evidence for the benefit of an intraoperative use of a goal-directed haemodynamic management has grown. We compared the oesophageal Doppler monitor (ODM, CardioQ-ODM™) with a calibrated pulse contour analysis (PCA, PiCCO2™) with regard to assessment of stroke volume (SV) changes after volume administration within a goal-directed haemodynamic algorithm during non-cardiac surgery. The data were obtained prospectively in patients with metastatic ovarian carcinoma undergoing cytoreductive surgery. During surgery, fluid challenges were performed as indicated by the goal-directed haemodynamic algorithm guided by the ODM. Monitors were compared regarding precision and trending. Clinical characteristics associated with trending were studied by extended regression analysis. A total of 762 fluid challenges were performed in 41 patients resulting in 1524 paired measurements. The precision of ODM and PCA was 5.7% and 6.0% (P=0.80), respectively. Polar plot analysis revealed a poor trending between ODM and PCA with an angular bias of −7.1°, radial limits of agreement of −58.1° to 43.8°, and an angular concordance rate of 67.8%. Dose of norepinephrine (NE) (scaled 0.1 µg kg−1 min−1) [adjusted odds ratio (OR) 0.606 (95% confidence interval, CI: 0.404–0.910); P=0.016] and changes in mean arterial pressure (MAP) to a fluid challenge (scaled 10%) [adjusted OR 0.733 (95% CI: 0.635–0.845); P<0.001] were associated with trending between ODM and PCA, whereas there was no relation to type of i.v. solution. Despite a similar precision, ODM and PCA were not interchangeable with regard to measuring SV changes within a goal-directed haemodynamic algorithm. A decrease in interchangeability coincided with increasing NE levels and greater changes of MAP to a fluid challenge.