Metabolic pathways associated with right ventricular adaptation to pulmonary hypertension: 3D analysis of cardiac magnetic resonance imaging

Mark Attard, Timothy J. W. Dawes, Antonio de Marvao, Carlo Biffi, Wenzhe Shi, John Wharton, Christopher J. Rhodes, Pavandeep Ghataorhe, J. Simon R. Gibbs, Luke Howard, Daniel Rueckert, Martin R. Wilkins, Declan P. O’ReganEuropean 2018 · DOI

TL;DR

This study uses advanced 3D cardiac MRI analysis to link metabolic profiles with the right ventricle's adaptation to pulmonary hypertension (PH). They found that elevated wall stress in the right ventricle is associated with specific metabolic changes related to energy metabolism and cellular stress, offering new insights into PH progression and potential therapeutic targets.

What Problem Does This Paper Address?

Pulmonary hypertension (PH) is a serious condition where the blood pressure in the lungs is abnormally high. This puts a significant strain on the right ventricle (RV), the heart chamber responsible for pumping blood to the lungs. While the RV initially adapts to this increased workload, it can eventually fail, leading to poor outcomes. Understanding how the RV adapts and what factors contribute to its failure is crucial for developing better treatments. Previous research has hinted that metabolic changes might play a role, and that circulating metabolites can predict survival in PH patients. However, it was unclear how these metabolic profiles specifically relate to the physical changes (remodeling) and mechanical stress experienced by the RV under PH. This paper aims to bridge that gap by quantitatively linking RV biomechanics and metabolic signatures in a large PH patient cohort.

Methods

The researchers employed a sophisticated, automated 3D analysis pipeline for cardiac magnetic resonance imaging (CMR). This involved:

  • Automated Image Segmentation: Using a multi-atlas approach and machine learning, they created detailed 3D models of the RV from CMR scans, accounting for shape and motion. This is an advancement over manual segmentation, allowing for more consistent and scalable analysis.
  • 3D Biomechanical Modeling: They developed a method to map regional wall stress across the RV. This goes beyond global measures by identifying areas of high mechanical load.
  • Statistical Parametric Mapping: This technique was used to map the association between circulating metabolite levels and regional RV wall stress, as well as other RV phenotypic characteristics (like wall thickness, excursion, and dilatation). This allows for a spatially resolved understanding of these relationships.
  • Metabolomics: They analyzed blood samples from patients to quantify levels of 10 specific metabolites previously linked to PH outcomes.

Datasets

  • PH Patient Cohort: 312 patients with PH were included. A subset of 182 patients also had metabolomics data available, with CMR and metabolomics collected within a median of 11 months.
  • Healthy Control Data: CMR data from 1985 healthy volunteers was used as a normal reference for ventricular geometry.

Key Insights and Ideas

  • Wall Stress as a Driver of Metabolic Change: The study strongly suggests that increased mechanical stress on the RV wall is not just a consequence of PH but actively influences metabolic pathways. This highlights the RV's active response to its environment.
  • Specific Metabolic Signatures of RV Stress: Certain metabolites were consistently linked to elevated wall stress. Notably, decreased levels of the steroid dehydroepiandrosterone sulfate (DHEA-S) and increased levels of modified nucleosides (like N2,N2-dimethylguanosine) were associated with higher wall stress.
  • Metabolites Reflecting RV Remodeling: The identified metabolic profiles were associated with patterns of RV remodeling, particularly eccentric remodeling (where the ventricle dilates and thins). This suggests that metabolic pathways are involved in how the RV physically changes in response to PH.
  • Energy Metabolism and Stress Response Pathways: The identified metabolites are involved in crucial cellular processes like energy metabolism (e.g., fatty acid metabolism) and cellular stress responses. This provides a mechanistic link between the physical strain on the RV and its cellular machinery.
  • DHEA-S as a Potential Protective Factor: The finding that lower DHEA-S levels are associated with higher wall stress and poorer RV adaptation is particularly interesting, given DHEA's known anti-inflammatory and potential cardioprotective effects.

Key Results

  • RV Wall Stress Predicts Survival: RV wall stress was found to be an independent predictor of all-cause mortality in PH patients (hazard ratio = 1.27).
  • Metabolite Associations with Wall Stress: Six metabolites were significantly associated with elevated wall stress, including increased levels of modified nucleosides and fatty acid acylcarnitines, and decreased levels of sulfated androgen.
  • Metabolite Associations with RV Phenotype: Metabolites were associated with various RV phenotypic measures, such as RV mass, end-diastolic volume, ejection fraction, and ventricular mass index, indicating their role in RV adaptation and remodeling. For instance, DHEA-S showed a negative association with RV ejection fraction and wall stress.
  • Regional Wall Stress Variations: The study demonstrated significant regional variations in RV wall stress, with the free wall in the inlet region experiencing the greatest stress.

Limitations and Future Work

  • Causality: The study identifies associations, but cannot establish direct causal relationships between metabolites and RV dysfunction.
  • Tissue vs. Blood Metabolites: The study used blood metabolites, and direct tissue or venous sampling would provide a more precise understanding of local metabolic activity.
  • Therapeutic Diversity: The patient cohort received a range of standard therapies, which could influence metabolic profiles.
  • Future Directions: The authors suggest that these findings identify tractable pathways for therapeutic intervention, focusing on reducing wall stress, improving oxygen delivery, restoring mitochondrial function, and modifying neurohormonal modulation. Further research could refine the computational models with fiber orientations and explore the therapeutic potential of targeting these identified metabolic pathways.

Summary updated

Abstract

Using computational image phenotyping, we identify metabolic profiles, reporting on energy metabolism and cellular stress-response, which are associated with adaptive RV mechanisms to PH.
BibTeX
@inproceedings{attard2018metabolic,
  author = {Mark Attard and Timothy J. W. Dawes and Antonio de Marvao and Carlo Biffi and Wenzhe Shi and John Wharton and Christopher J. Rhodes and Pavandeep Ghataorhe and J. Simon R. Gibbs and Luke Howard and Daniel Rueckert and Martin R. Wilkins and Declan P. O’Regan},
  title = {{Metabolic pathways associated with right ventricular adaptation to pulmonary hypertension: 3D analysis of cardiac magnetic resonance imaging}},
  booktitle = {European},
  year = {2018},
  doi = {10.1093/ehjci/jey175},
}
Figure