Assessment of Agreement Between Two Measurement Methods of Left Ventricular Outflow Tract (LVOT) Velocity Time Integral (VTI) (AUTO-VTI)

Assessment of Agreement Between Two Measurement Methods of Left Ventricular Outflow Tract (LVOT) Velocity Time Integral (VTI)

Assessment of Agreement Between Two Measurement Methods of Left Ventricular Outflow Tract (LVOT) Velocity Time Integral (VTI) : Automatic Versus Manual, in Critically Ill Patients in Acute Circulatory Failure

Acute circulatory failure (ACF) is a common cause of admission in intensive care unit (ICU). Echocardiography is a widespread tool nowadays for the initial assessment and the hemodynamic monitoring. An interesting data from this exam is the Left ventricular outflow tract (LVOT) velocity time integral (VTI), reflecting stroke volume, and therefore cardiac output. A new tool for automated recording has been developped on the VENUE GE echograph. This study aims at assessing this automated measurement of LVOT VTI compared with the classic manual method.

Acute circulatory failure is a dramatically common issue in critical care, affecting nearly 1/3 of the patients admitted in ICU, and associated with a high morbi-mortality. Echocardiography enables a quick point of care assessment of the hemodynamic status of the patient. One key parameter is stroke volume, which can be estimated by LVOT VTI. In practice, the guidelines recommend measuring LVOT VTI in a 5 chambers cavity window with pulsed wave doppler, trying to get an angle between aortic flow and the doppler signal as cloth as possible to 0°. Than VTI must be calculated from an average of 3 to 5 cycles in sinus rhythm, or 5 to 10 cycles in arrythmia such as atrial fibrillation. The main issue is that this approach is time consuming and tedious, and most of the practitioners choose to measure only the most representative beat (" Best VTI "), likely to be source of a loss of reproducibility. In order to make it easier, some companies have developed softwares to help the intensivist. Thereby, General Electric has equiped the VENUE echograph with a system that helps first selecting the proper place for sampling, and then automatically acquire, trace, and calculate the average LVOT VTI based on a 4 seconds sample. Finally, this tool gives a VTI trending when several measures are repeated along the patient' stay. This program has already been tested on a swine model in hemorrhagic shock with encouraging results (better agreement with cardiac output by thermodilution with the automated method than with the manual one). However, feasibility was only 60%. This human study aims at assessing the interest of automated LVOT VTI measurement compared to the manual standard one when performing a cardiac echography for hemodynamic assessment of patients hospitalised in ICU. Otherwise, a measure of the " Best-VTI " will be done at the same time in accordance with the current practice, in order to assess this approach. Finally, when a fluid resuscitation will be indicated, a second exam will be done, and agreement of VTI variation between automated and manual method studied.

This study aims at assessing the interest of automated LVOT VTI measurement compared to the manual standard one when performing a cardiac echography for hemodynamic assessment of patients hospitalised in ICU. Otherwise, a measure of the " Best-VTI " will be done at the same time in accordance with the current practice, in order to assess this approach. Finally, when a fluid resuscitation will be indicated, a second exam will be done, and agreement of VTI variation between automated and manual method studied.

An echocardiographic exam is realised after inclusion, measuring especially : LVOT VTI Respiratory variation of inferior veina cava diameter For theses two measurements, the manual method is first recorded, and then the automated one is done. For patients that need fluid resuscitation of at least 250mL in less than 10minutes (indication at the discretion of the physician in charge of the patient), a second echocardiography is done according to the same protocol described before.

Agreement between the average LVOT VTI obtained by the manual method, and with the automatic one, estimated through calculation of Lin's coefficient and Bland-Altman plot.

- Strength of the correlation between the automated and manual method, through estimation of correlation coefficient (Pearson or Spearman)

-Number of the event: One or more automated spectral tracing by the machine, judged as inappropriate by the expert

- Agreement between the " Best-VTI " chosen by the expert, the average VTI by manual method, and the automated VTI, assessed by Lin's coefficient and Bland-Altman plot.

- Agreement between the variation of VTI after fluid loading obtained by automated and manual method, assessed by Lin's coefficient and Bland-Altman plot.

- Agreement between the proportion of patients classified as preload-dependant after fluid challenge (LVOT VTI variation ≥ 15%) by automated and manual method, assessed by calculation of Kappa coefficient.

- Agreement between respiratory variation of inferior vena cava measured by automated and manual method, assessed by Lin's coefficient and Bland-Altman plot

- Rate of inappropriate measurement of respiratory variation of inferior vena cava by the software of the automated method.

Inclusion Criteria: Patient admitted in ICU and presenting an acute circulatory failure, defined by persistent hypotension (SAP < 90mmHg or MAP < 65mmHg) despite fluid resuscitation of 30ml/kg, or the need for vasopressor. Echocardiographic hemodynamic assessment for at least one of the following signs : SAP < 90mmHg; Urine output < 0,5mL/kg/h during more than 2 hours; Blood lactate level > 2mmol/L; Increase of Norepinephrine doses needed The patient or his trusted person / legal representative / member of the family gave his free and informed consent, et have signed the consent form, or patient included in an emergency situation. The patient has to benefit from the French national healthcare insurance. Age ≥ 18 years old. Exclusion Criteria: Participation in another study, or exclusion period from another study assessing the same primary endpoint. Patient placed under judicial protection, or guardianship Patient or his trusted person refuses to sign consent form Pregnant, parturient, or breastfeeding woman Patient with poor echogenic window