5 - Non-invasive pulmonary vascular resistance estimation by CMR – validation using right heart catheterization

Thursday, February 13, 2020

3:25 PM – 3:35 PM

Location: Salon J1

Background: Pulmonary hypertension (PH) is diagnosed by right heart catheterization through invasive measurement of mean pulmonary arterial pressure (mPAP), and further classified using mean pulmonary capillary wedge pressure (PCWP) and pulmonary vascular resistance (PVR). Cardiovascular magnetic resonance (CMR) 4D flow analysis has been used to estimate mPAP from pulmonary artery vortex duration with excellent accuracy. Also, it has been shown using echocardiography that peak longitudinal left atrial (LA) strain be used to estimate PCWP with good accuracy. We aimed to estimate PVR from CMR compared to invasive measurements from right heart catheterization.

Methods: Patients referred to right heart catheterization (n=20, age 67±16 years, 60% female) underwent 1.5T CMR (MAGNETOM 1.5T Aera, Siemens Healthcare, Erlangen, Germany) 10±10 days after catheterization. Right heart catheterization was performed via the jugular vein with a Swan-Ganz catheter. All CMR analysis was performed blinded to results of catheterization. mPAP was estimated, as previously validated, from pulmonary artery vortex duration using a commercially available time-resolved 2D phase contrast sequence with three directional velocity encoding (20 frames/cardiac cycle, acquisition duration 6-12 minutes). Vortex duration was analyzed using prototype software (4D Flow, Siemens Healthcare, Erlangen, Germany). LA strain was measured by feature tracking in a 2-chamber long axis cine image (Segment R6405, Medviso, Lund, Sweden). CMR PCWP was estimated from the linear regression equation between LA strain and invasively measured PCWP. The transpulmonary pressure gradient (TPG) was defined as mPAP minus PCWP. Cardiac output (CO) by CMR was measured using 2D phase contrast of the pulmonary artery with through-plane velocity encoding. PVR was defined as TPG/CO.

Results: Compared to catheterization, CMR estimated mPAP from pulmonary artery vortex duration (R2=0.79, p< 0.001, bias±SD 0.4±4.9 mmHg) PCWP from left atrial strain (R2=0.62, p< 0.001, bias±SD -0.6±4.8 mmHg), CO from 2D flow imaging (R2=0.61, p< 0.001, bias±SD -0.2±0.8 L/min) and PVR from the combined CMR measures (R2=0.61, p< 0.001, bias±SD 0.3±1.8 WU).

Conclusion: Non-invasive estimation of PVR by CMR is feasible and shows good agreement with PVR derived from invasive catheterization. These results provide a proof of concept of a CMR-only method to estimate PVR, which merits further validation in independent cohorts.

Presenting Author(s)

Joao G. Ramos, MD

PhD Student
Karolinska Institutet and Karolinska University Hospital

Disclosure: Disclosure information not submitted.

Primary Author(s)

Joao G. Ramos, MD

PhD Student
Karolinska Institutet and Karolinska University Hospital

Disclosure: Disclosure information not submitted.

Co-Author(s)

BW

Bjorn Wieslander, MD, PhD

Karolinska Institutet and Karolinska University Hospital

Disclosure: Disclosure information not submitted.
AF

Alexander Fyrdahl, MSc

Medical physicist
Karolinska Institutet and Karolinska University Hospital

Disclosure: Disclosure information not submitted.
GR

Gert Reiter

MR Physicist
Siemens Healthineers

Disclosure: Disclosure information not submitted.
UR

Ursula Reiter

MR Physicist
Medical University of Graz

Disclosure: Disclosure information not submitted.
NJ

Ning Jin, PhD

Staff Scientist
Siemens Medical Solutions

Disclosure: Disclosure information not submitted.
VS

Viktoria Skott, PhD

Researcher
Karolinska Institutet and Karolinska University Hospital

Disclosure: Disclosure information not submitted.
MM

Michael Melin, MD

MD
Karolinska Institutet and Karolinska University Hospital

Disclosure: Disclosure information not submitted.
PS

Patrik Sundblad, MD, PhD

Karolinska Institutet and Karolinska University Hospital

Disclosure: Disclosure information not submitted.
AS

Andreas Sigfridsson, PhD

MR Scientist
Karolinska Institutet and Karolinska University Hospital

Disclosure: Disclosure information not submitted.
Martin Ugander, MD, PhD

Professor
Karolinska Institutet and Karolinska University Hospital

Disclosure: Siemens Medical: Grant/Research Support