Cardiovascular

Geometric morphological parameters from cardiac magnetic resonance imaging in assessment of chronic pulmonary thromboembolism, a prospective pilot study.

TL;DR

CMR-derived apical right ventricular fractal dimension and perimeter-area ratio reflect early, load-dependent ventricular remodeling in chronic pulmonary thromboembolism and demonstrated potential as complementary non-invasive imaging markers for severity assessment and risk stratification, highlighting the apex as the most sensitive site of RV adaptation.

Key Findings

Apical RVFD and RVPAR robustly differentiated CPTE patients from healthy controls with high diagnostic accuracy.

  • AUC ranged from 0.911 to 0.961 for differentiating CPTE from controls using apical RVFD and RVPAR
  • Parameters were quantified on short-axis cine images at end-diastole and end-systole across basal, mid, and apical levels
  • 53 CPTE patients were compared to 53 age- and sex-matched healthy controls
  • ROC analysis was used to assess diagnostic performance

Apical RVFD and RVPAR differentiated CTEPH from CTEPD with moderate-to-good diagnostic accuracy.

  • AUC ranged from 0.819 to 0.895 for differentiating CTEPH from CTEPD
  • The study included 37 CTEPH patients and 16 CTEPD patients
  • Differences were most pronounced at the apical level, both in end-diastole and end-systole (p < 0.05)
  • The apical level was identified as the most sensitive site of RV adaptation

Right ventricular fractal dimension was significantly higher and RVPAR was significantly lower in CTEPH compared to other groups.

  • Differences in RVFD and RVPAR were most pronounced at the apical level
  • Significant differences were observed both at end-diastole and end-systole (p < 0.05)
  • Higher RVFD indicates greater geometric complexity of the right ventricular endocardial border
  • Lower RVPAR in CTEPH reflects shape changes associated with more severe disease

Both RVFD and RVPAR provided significant incremental diagnostic value over right ventricular ejection fraction (RVEF).

  • Incremental diagnostic value was assessed using multivariable logistic regression, net reclassification improvement (NRI), and integrated discrimination improvement (IDI)
  • Both parameters added diagnostic information beyond conventional RVEF
  • Results highlight that geometric-morphological parameters capture aspects of RV remodeling not reflected by functional indices alone

In univariable analysis, RVFD correlated positively with afterload indices and negatively with perfusion/output metrics.

  • Spearman correlations were used to assess relationships between geometric parameters and hemodynamic measures
  • All correlations were statistically significant (p < 0.05)
  • RVPAR showed inverse associations compared to RVFD with the same hemodynamic indices
  • All patients underwent right heart catheterization within 24 hours of CMR to obtain invasive hemodynamic measurements

In multivariable linear regression, RVPAR retained weak independent negative associations with mean pulmonary arterial pressure and pulmonary vascular resistance.

  • RVPAR showed independent negative associations with mPAP and PVR after multivariable adjustment
  • The associations were described as 'weak' independent effects
  • RVFD did not retain independent associations in multivariable modeling in the same manner as RVPAR
  • Right heart catheterization within 24 hours of CMR provided the invasive hemodynamic reference standard

The study prospectively enrolled 53 CPTE patients and 53 age- and sex-matched healthy controls, with all patients undergoing both CMR and right heart catheterization.

  • 53 CPTE patients included 16 with CTEPD and 37 with CTEPH
  • All patients underwent CMR and right heart catheterization within 24 hours
  • Biventricular FD and PAR were quantified on short-axis cine images at basal, mid, and apical levels
  • The study was designed as a prospective pilot study

CMR-derived apical RVFD and RVPAR were interpreted as reflecting early, load-dependent ventricular remodeling in CPTE.

  • The parameters were described as capturing 'early, load-dependent ventricular remodeling'
  • The apex was highlighted as 'the most sensitive site of RV adaptation'
  • The parameters were proposed as 'complementary non-invasive imaging markers for severity assessment and risk stratification'
  • Cine-derived fractal dimension and perimeter-area ratio were noted to 'characterize right ventricular remodeling' and 'correlate with invasive hemodynamics'

What This Means

This research examined whether two geometric measurements derived from cardiac MRI scans — fractal dimension (FD, a measure of the complexity of the heart chamber's inner surface) and perimeter-area ratio (PAR, a measure of chamber shape) — could help doctors distinguish between different forms of chronic blood clot disease in the lungs and assess disease severity. The study followed 53 patients with chronic pulmonary thromboembolism (a condition where blood clots persistently block lung vessels) and 53 healthy comparison subjects. All patients had both cardiac MRI and an invasive procedure called right heart catheterization performed within 24 hours of each other, allowing the researchers to directly compare the MRI-based measurements against the gold standard hemodynamic measurements. The study found that measurements taken from the tip (apex) of the right ventricle were particularly useful. These apical measurements could distinguish patients with chronic clot disease from healthy individuals with very high accuracy (AUC 0.911–0.961) and could also separate patients with more severe disease causing elevated lung blood pressure (CTEPH) from those with milder disease (CTEPD) with good accuracy (AUC 0.819–0.895). Importantly, these geometric measurements added diagnostic information beyond the commonly used measure of right heart pumping function (right ventricular ejection fraction), and they correlated with invasive measurements of lung blood pressure and vascular resistance. This research suggests that geometric shape analysis of the right heart from standard cardiac MRI scans — without requiring additional special sequences or contrast agents — could serve as a non-invasive tool to help characterize the severity of chronic lung clot disease and monitor how the right heart is adapting to increased workload. The findings from this pilot study indicate that the tip of the right ventricle may be especially sensitive to the early changes caused by this disease, and these measurements could complement existing assessments used in clinical practice.

Have a question about this study?

Citation

Du J, Guo Y, Yu H, Liu A, Xi L, Zhang S, et al.. (2026). Geometric morphological parameters from cardiac magnetic resonance imaging in assessment of chronic pulmonary thromboembolism, a prospective pilot study.. European radiology experimental. https://doi.org/10.1186/s41747-026-00811-1