Validation of a New Method for Stroke Volume Variation Assessment: a Comparaison with the PiCCO Technique

This paper proposes a novel, simple and minimally invasive method for stroke variation assessment using arterial blood pressure measurements. The blood pressure signal is reconstructed using a semi-classical signal method allowing the computation of a parameter, called the first invariant INVS1. We show that INVS1 is linearly related to stroke. To validate this approach, a statistical comparaison between INVS1 and volume measured with the PiCCO technique was performed during a 15-mn in 21 mechanically ventilated patients in intensive care. In 94% of whole recordings, a strong correlation was estimated by cross-correlation (mean coefficient=0.9) and linear regression (mean coefficient=0.89). the linear relation had been verified, a Bland-Altman test showed the very agreement between the two approaches and their interchangeability. For the 6%, INVS1 and the PiCCO stroke volume were not correlated at all, and discrepancy, interpreted with the help of mean pressure, heart rate and vascular resistances, was in favor of INVS1.