Wideband myocardial T2 mapping in patients with implantable cardiac devices at 1.5 T

Presenting Author(s)

• 

  Pauline Gut, PhD

  PhD
  University Hospital (CHUV) and University of Lausanne (UNIL)
  Lausanne, Vaud, Switzerland

Primary Author(s)

• 

  Pauline Gut, PhD

  PhD
  University Hospital (CHUV) and University of Lausanne (UNIL)
  Lausanne, Vaud, Switzerland

Co-Author(s)

• HC

  Hubert Cochet, MD, PhD

  Professor
  Bordeaux University - INSERM U1045
  Bordeaux, Aquitaine, France
• 

  Thomas Küstner, PhD

  Professor
  University Hospital of Tübingen
• PA

  Panagiotis Antiochos, MD

  Cardiologist
  Cardiac MR Center of the University Hospital Lausanne, CHUV, Switzerland
  Lausanne, Vaud, Switzerland
• AM

  Ambra Masi, MD

  Cardiologist
  University Hospital (CHUV) and University of Lausanne (UNIL), Switzerland
• 

  Juerg Schwitter, MD

  Full Professor
  University Hospital and University of Lausanne
  Lausanne, Vaud, Switzerland
• FS

  Frederic Sacher, MD, PhD

  Professor
  Department of Cardiovascular Imaging, Hôpital Cardiologique du Haut-Lévêque, CHU de Bordeaux
  Pessac, Aquitaine, France
• PJ

  Pierre Jaïs, MD, PhD

  PROF/PhD
  Hôpital Cardiologique du Haut-Lévêque, CHU de Bordeaux
  Bordeaux, Aquitaine, France
• MS

  Matthias Stuber, PhD

  Professor/Director
  CIBM/CHUV/UNIL
  Lausanne, Switzerland
• 

  Aurelien Bustin, FSCMR

  Research Associate
  Department of Cardiovascular Imaging, Hôpital Cardiologique du Haut-Lévêque, CHU de Bordeaux, Avenue de Magellan, Pessac, France; IHU LIRYC, Electrophysiology and Heart Modeling Institute, Université de Bordeaux – INSERM U1045, Avenue du Haut Lévêque, Pessac, France; Department of Diagnostic and Interventional Radiology, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland
  Bordeaux, Aquitaine, France

Background: Myocardial T2 mapping enables non-invasive detection of inflammation and oedema¹, but in patients with implantable cardioverter devices (ICDs), off-resonance effects cause severe image artefacts and inaccurate T2 values. This work aimed to develop a wideband² gradient echo (GRE) T2 mapping sequence and combine it with an advanced patch-based denoising approach (HD-PROST)⁴. This novel technique was tested in healthy volunteers and patients and was compared against the clinical reference balanced steady-state free-precession (bSSFP) sequence³.

Methods: Eight healthy volunteers were scanned (MAGNETOM, Avanto) without and with an ICD externally taped on the left clavicle 10 cm away from the heart. Twenty patients were included: thirteen without ICD with known or suspected oedema (MAGNETOM, Sola), and seven with cardiac devices (5 ICDs, 2 pacemakers) (MAGNETOM, Aera). Three short-axis slices were acquired during end-expiration in mid-diastole at 1.5T.

Three T2-prepared modules (0,27,55ms)5 (bandwidth=1.6 kHz for conventional, 5.0 kHz for wideband²) were implemented in a GRE T2 mapping (Fig1). Sequence parameters: resolution=1.4mmx1.4mm, slice thickness=8mm, FOV=360mmx287mm, FA=15°, GRAPPA x2, TE/TR=1.17/2.91ms, readout bandwidth=1221Hz/pixel, 7/8 phase partial Fourier, strong asymmetric echo.

Myocardial T2 values were measured in AHA-segments in volunteers and in oedema/remote regions in patients, and were compared using repeated-measures ANOVA. Artefact severity in device-patients was rated with a 4-point Likert scale (1=non-diagnostic, 4=minimal artefact).

Results: In healthy volunteers, mean T2 values without ICD were similar between conventional (43.5±2.2ms) and wideband GRE (42.9±2.0ms, P=0.600) (Fig2). With ICD, conventional GRE underestimated T2 (36.4±5.9ms) compared to wideband values (43.9±2.1ms, P< 0.001) which were similar to those obtained without ICDs (P=0.560).

In patients without devices (Fig3A), oedema T2 values were higher than remote values in all sequences (P< 0.001). HD-PROST denoising did not alter mean T2 values (P >0.05), but reduced T2 standard deviations (P=0.006), thereby decreasing noise and improving image visualization. Remote and oedema T2 values were comparable between conventional (remote: 35.2±3.4ms, oedema: 48.2±1.9ms) and wideband GRE (remote: 34.5±2.2ms, oedema: 49.5±1.3ms) (P >0.05). Compared with bSSFP (remote: 44.1±1.6ms, oedema: 60.5±3.8ms), GRE yielded lower T2 values (P< 0.001) but preserved relative oedema elevation.

In patients with devices (Fig3B), bSSFP maps were more artefacted (score=2.6±1.8) than conventional GRE (3.0±1.9). Wideband GRE significantly reduced ICD-artefacts (4.3±1.0)

Conclusion: Denoised wideband GRE T2 mapping mitigated ICD-artefacts, enabled accurate myocardial T2 quantification and reliable oedema detection, offering a promising diagnostic tool for myocardial quantification in ICD patients.

Fig1: Sequence diagram of the investigated wideband myocardial GRE T2 mapping.
Fig2: Conventional and wideband T2 maps obtained in a healthy subject (A) with and without an ICD, and mean T2 values in all healty volunteers (B) with global mean T2 values statistical analyses.
Fig3: (A) Columns 1-3: Example of presence and absence of oedema in reference bSSFP and wideband GRE in patients without implantable devices. (B) T2 maps with pacemakers (PMs), TV-ICDs (transvenous ICDs) and S-ICD (subcutaneous ICD). ARVC, arrhythmogenic right ventricular cardiomyopathy; DCM, dilated cardiomyopathy; MINOCA, myocardial infarction with non-obstructive coronary arteries; n.a., not available.