Background/Objective: Continuous blood pressure (BP) monitoring is essential in clinical settings, where rapid hemodynamic changes influence patient management. While intra-arterial measurement remains the reference standard, non-invasive volume-clamp systems offer a potential alternative. We assessed the accuracy of finger-cuff-based continuous BP monitoring compared to invasive measurement in patients with pulmonary hypertension (PH). Methods: This post hoc analysis from a crossover RCT included PH patients who underwent repetitive hemodynamic assessments at rest and during exercise. The participants had simultaneous invasive BP monitoring via the radial artery and a non-invasive finger-cuff device (Finapres® NOVA Basic). The mean blood pressure (mBP) was compared at rest, 50% of the maximal workload, and at the end of exercise using Bland–Altman and Taffé analysis. Results: In the study, 24 patients (seven female; 59 ± 14 years) contributed 385 paired mBP measurements. The invasive and non-invasive methods showed similar values at rest (96.1 ± 16.7 vs. 96.4 ± 17.2 mmHg) and during maximal exercise (106.8 ± 18.6 vs. 111.8 ± 21.6 mmHg). The overall Bland–Altman bias was 2.8 mmHg with wide limits of agreement (−39.6 to 45.3 mmHg), which remained broad across all exercise intensities. The Taffé analysis revealed a non-uniform, directionally dependent bias: the non-invasive system overestimated the mBP at low pressures and underestimated it at higher pressures. The measurement variability was substantially greater for the non-invasive method than for the invasive reference. Conclusions: In PH patients, finger-cuff-based continuous BP monitoring demonstrated acceptable group-level agreement but insufficient individual-level accuracy for clinical decision-making.
Titz et al. (Thu,) studied this question.