Patients and study design
This is a retrospective study of 284 consecutive patients referred for rest CMR at our institution (CHWAPI Tournai, Belgium) between August 2017 and November 2021. Patients with congenital heart disease (n=2 pulmonary venous returns and one atrial septal defect), known intracardiac shunts (known to affect methods based on the indicator dilution principles [12]), and unsuitable contrast bolus (n=2) tracking in LV or RV were excluded. As per their dominant pathology documented in clinical records or CMR reports, patients were assigned to one of the following groups:
- “Normal” (no abnormalities were found)
- “Myocarditis” (features suggestive of acute myocarditis are present such as inflammatory hyperemia and edema, necrosis/scar, contractile dysfunction and accompanying pericardial effusion)
- “Infarct” (there are components of myocardial damage such as edema, intramyocardial hemorrhage, microvascular obstruction, and fibrosis)
- “Hypertrophy” (hypertrophic cardiomyopathy is mentioned in the conclusion of the examination)
- “Valvular” (the exam is done to assess significant valvular heart disease)
- “Dilated Cardiomyopathy” (pathological dilatation of the ventricles in the absence of ischemic and structural heart disease)
- “Other” (group in which patients do not have any pathology from another group but whose CMR cannot be considered normal).
The study protocol was approved by the institutional ethics committee and patients gave their informed consent.
CMR
The patients underwent a standard myocardial MR imaging protocol on a 1.5 Tesla MRI system (Siemens Avanto-Fit, Erlangen, Germany) including ECG-gated, breath-hold, short-axis (SAX) and long-axis (LAX) cine single shot fast precession sequence (TR = 34 ms / TE = 1.1 ms / flip angle = 73°) for assessment of myocardial function, and a pulse-gated, free breathing, dynamic first-pass perfusion scan of 80 or 100 phases using a saturation recovery turbo-FLASH pulse sequence (TR = 172 ms / TE = 1.1 ms / flip angle = 5°) with at least 3 short-axes images every heartbeat. Motion correction was applied inline on the dynamic first-pass perfusion images. All patients were injected with a 1 ml/10 kg of body weight bolus dose of gadoterate meglumine (Dotarem® [Guerbet, Villepinte, France] for imaging performed until the end of 2019 and Clariscan® [GE Healthcare, Chicago, United States] thereafter) at a speed of 4 ml/s, followed by a saline flush of 20 ml, using an automatic injector (Ulrich Medical, Tennessee, United States).
Myocardial volumes and pulmonary transit time
MR images were analyzed with Syngovia software (Siemens, Erlangen, Germany). Left ventricular ejection fraction (LVEF), left ventricular end diastolic (EDV), end systolic (ESV) and stroke (SV) volumes were calculated from cine images. Left ventricle (LV) and right ventricle (RV) first-pass signal through time was also analyzed and recorded using Syngovia software. A circular ROI was centered in the cavity of RV and LV on the most basal slice.
The enhancement time curves for each cavity signals were subsequently fitted to a time-sliding LogNormal function using an in-house python script based on the lmfit package software (Levenberg-Marquardt method). PTT was calculated as LV peak enhancement time of LV – RV peak enhancement timeof RV, a. This methodology was adapted from the article by Mischi et al. [13]. nPTT was reported calculated according to thein number of cardiac cycles (= PTT [in sec] x 60 / heart rate (HR)HR [Heart rate [in beats per minute]]). Cardiac output (CO) is the product of stroke volume (SV) and heart rate (HR). Pulmonary blood volume is the product of pulmonary transit time (PTT) and cardiac output (CO). Myocardial volumes and pulmonary blood volumes are indexed to body surface area (BSA).
Statistical analysis
All data were analyzed using Jeffreys’s Amazing Statistics Program (JASP) software. All tests were 2 sided, and a P value<0.05 was considered statistically significant. Continuous variables were summarized as mean ± SD. Categorical variables were reported as counts and percentages. The categorical bipolar sex variable was tested using a student’s T test unless the normality assumption failed. In this case, a Mann-Whitney test is reported. Categorical pathology data were analyzed with analysis of variance. As the data failed parametric assumption, a Kruskal-Wallis H test was used. Post-hoc analysis with a Dunn test is provided. The correlation between variables was assessed using Pearson or Spearman coefficients.