Stroke volume determination using transcardiopulmonary thermodilution and arterial pulse contour analysis in severe aortic valve disease

Abstract

Transcardiopulmonary thermodilution (TPTD, SVTD) as well as calibrated (SVPC CAL) and uncalibrated (SVPC UNCAL) arterial pulse contour analysis (PC) are increasingly promoted as less-invasive technologies to measure stroke volume (SV) but their reliability in aortic valve disease was unknown. The objective of this prospective study was to investigate the validity of three less-invasive techniques to assess SV in conditions involving aortic stenosis (AS) and valvuloplasty-induced aortic insufficiency (AI) compared with transesophageal echocardiography. In 18 patients undergoing transcatheter aortic valve implantation, SVTD and SVPC CAL were determined using a central pressure signal via the brachial artery and SVPC UNCAL using a peripheral radial signal. In aortic valve dysfunction TPTD achieved adequate reproducibility (concordance correlation coefficient (CCC): AS 0.84; AI 0.82) and agreement (percentage error (PE): AS 26.3 %; AI 26.2 %) with the reference technique. In severe AS, SVPC CAL (PE 25.7 %; CCC 0.85) but not SVPC UNCAL (PE 50.4 %; CCC 0.38) was reliable. Neither calibrated nor uncalibrated PC (SVPC CAL: PE 51.5 %; CCC 0.49; SVPC UNCAL: PE 61.9 %; CCC 0.22) was valid in AI. Trending ability to hemodynamic changes, quantified by the ΔSV vector and the angle θ, was acceptable for each measurement modality. Transcardiopulmonary thermodilution is valid in aortic valve dysfunction. Calibration of PC substantially improves reliability in aortic valve disease. Calibrated PC is valid in severe AS. Valvuloplasty-induced AI seriously confounds PC measurements. In uncalibrated PC approaches, the relative SV trend is superior to single absolute values.