healthmedicinet

The study protocol was approved by the hospital’s ethics commitee (Comissão de ética
CentroHospitalar de Gaia/espinho EPE) and complies with the declaration of Helsinki.
Written informed consent was obtained from all participants.

Thirty-six patients with cirrhosis followed in a hepatology outpatient clinic, able
to comply with the instructions during the exam, were recruited and referred to CMR.
The diagnosis of cirrhosis was based on clinical, laboratory and ultrasonographic
criteria and was also confirmed by liver biopsy in 22 % of the cases. Patients with
a known history of hypertension, diabetes, cardiac disease or relevant ECG abnormalities
were excluded. Patients with large volume ascites and/or unable to tolerate breath-holding,
renal insufficiency (creatinine clearance???60 ml/min/1.73 m
²
) or standard contraindications to CMR or gadolinium were also excluded.

A group of eight subjects, with similar age and sex distribution as the patient group,
without known cardiovascular risk factors, referred to CMR for a different indication
(mostly atypical chest pain evaluation) and with a completely normal scan, was used
as control.

CMR acquisition

Patient preparation

Patients were instructed to refrain from smoking, coffee, tea, aminophylline, for
24 h before the scan. Beta-blockers were suspended 48 h before the study.

CMR protocol

Images were acquired using a 1.5 T scanner (Siemens Symphony TIM; Siemens, Erlangen,
Germany) with a 6-channel anterior chest coil and spinal coils within the gantry table.

Cine imaging

After scout images, cine images using a retrospective ECG-gated balanced steady state
free-precession sequence (TR 3.0 ms, TE 1.3 ms, flip angle??90°) were acquired during
brief periods of end-expiratory breath-hold. Two-, four and three-chamber orientations,
as well as multiple equidistant short-axis planes (slice thickness 8 mm; gap 2 mm)
allowing coverage of the entire cardiac volume were performed. Thirty phases were
obtained per cardiac cycle.

For dobutamine stress imaging, three long-axis and three short-axis slices (basal,
mid-ventricular and apical) were acquired, in order to cover 16 myocardial segments
16]. Dobutamine was infused intravenously at 3-min stages at doses of 10 and 20 ?g/kg/min.
Repeat short-axis images as well as long-axis images were acquired at the end of each
stage. During dobutamine infusion, patient symptoms, heart rate, blood pressure, and
electrocardiogram were monitored.

Aortic flow imaging

Aortic flow was measured using phase contrast gradient echo pulse sequence with one-direction
“through-plane” motion-encoding (slice thickness 5 mm; FOV 320?×?320 mm
²
, in-plane resolution ?1 mm, TR/TE = 5.9/3.0 ms, flip angle 22°, bandwidth ~350 Hz/pixel),
centered in ascending aorta and aligned orthogonally to the expected main blood flow
direction in two spatial directions, at the level of the pulmonary bifurcation. Velocity
encoding sensitivity (Venc) was adapted to the expected velocities (typically 150
for the rest images and 300 during the dobutamine-stress acquisitions). Thirty frames
were acquired per cardiac cycle using a free-breathing technique with three excitations
per k-space line.

Perfusion imaging

Our protocol for stress perfusion imaging has been previously described 17]. Maximal hyperemia was achieved with intravenous adenosine (140 ?g.kg???1.min???1)
infusion for 5 min. Within the last 2 min of infusion, an intravenous bolus of 0.07 mmol/Kg
of gadobutrol (Gadovist, Bayer HealthCare Pharmaceuticals, Berlin, Germany), was injected.
Three short-axis slices (basal, mid-ventricular and apical) were imaged during the
first pass of the bolus of gadolinium using a gradient echo pulse sequence with a
single saturation pre-pulse per R–R interval shared over the three slices. Typical
sequence parameters were: echo time, 1.18 ms; repetition time, 192 ms; inversion time,
110 ms; flip angle, 12°; slice thickness, 10 mm; field of view, 290–460 mm; matrix,
192?×?128 mm; in-plane spatial resolution, 1.5–2.4 mm 2]; bandwidth, 789 Hz per pixel. Patients were asked to hold their breath on full expiration
for the duration of the first pass of the gadolinium bolus.

Late gadolinium enhancement

Late gadolinium-enhancement (LGE) was assessed using a gradient-recalled phase-sensitive
inversion-recovery (PSIR) sequence (TR 46 ms, TE 3.4 msec, flip angle 15°, IR time
280–360 msec) ?10 min after the administration of 0.2 mmol/kg of gadobutrol.

CMR analysis

Images were anonymized and analysis was performed by operators blinded to clinical
data.

A commercially available software (CMR42; Circle Cardiovascular Imaging Inc., Calgary,
Canada) was used to assess left and right ventricular volumes and function, from the
short-axis cine images stack. Left ventricular ejection fraction (EF) during stress
was derived from two long axis and one short axis; for comparison the same method
was also used to calculate resting EF. Phase-contrast pulse sequences at rest and
peak dobutamine dose were analyzed with the same software, to determine cardiac output.

Feature tracking (FT), a technique analogous to echocardiographic speckle tracking,
which allows tracking of tissue voxel motion of CMR cine images 18]–21] was used to assess left ventricular strain. Four-, two- and three-chamber views were
used to calculate longitudinal strain. Radial strain and circumferential strain were
derived from the three short-axis planes. For each parameter three repeated measurements
were performed and subsequently averaged. Global longitudinal strain (GLS), global
radial strain (GRS) and global circumferential strain (GCS) were defined as the mean
strain of the three individual planes. Measurements were performed at rest and at
each stage of dobutamine infusion.

For the stress perfusion analysis, perfusion defects were defined as subendocardial
or transmural visually dark myocardial areas when compared with remote healthy myocardium,
persisting for at least 10 frames.

Reproducibility

Reproducibility of FT derived strain was assessed in 10 randomly selected subjects.
For intraobserver variability, the same operator repeated the measurements, more than
4 weeks after the initial analysis. For interobserver variability, a second operator
re-analysed the images.

Statistical analysis

Data were stored and analyzed using IBM SPSS Statistics, Version 20.0 (IBM Corp.,
Armonk, NY, USA). Results are presented as median (25th–75th percentile) for quantitative
variables and as n (%) for categorical variables. A significance level of 5 % was
used.

The Mann–Whitney test was used to evaluate differences in continuous variables between
groups. The Chi-squared test was used to compare proportions. Spearman’s coefficient
was used to test correlations. Bland–Altman analysis was performed for reproducibility
testing.