Semi-automatic quantification of mosaic perfusion of lung parenchyma and its correlation with haemodynamic parameters in patients with chronic thromboembolic pulmonary hypertension
Language English Country England, Great Britain Media print-electronic
Document type Journal Article
PubMed
37661531
DOI
10.1016/j.crad.2023.08.008
PII: S0009-9260(23)00365-3
Knihovny.cz E-resources
- MeSH
- Chronic Disease MeSH
- Hemodynamics MeSH
- Humans MeSH
- Vascular Diseases * complications MeSH
- Perfusion MeSH
- Lung diagnostic imaging MeSH
- Pulmonary Embolism * complications diagnostic imaging MeSH
- Hypertension, Pulmonary * diagnostic imaging MeSH
- Retrospective Studies MeSH
- Aged MeSH
- Check Tag
- Humans MeSH
- Aged MeSH
- Female MeSH
- Publication type
- Journal Article MeSH
AIM: To investigate the feasibility of semiautomatic quantification of mosaic perfusion and the associations between mosaic perfusion on computed tomography (CT; the ratio of hypoperfused parenchyma to the whole lung volume) and haemodynamic parameters through linear regression analysis. MATERIALS AND METHODS: Fifty-eight consecutive patients (mean age 66 years, 28 females) diagnosed with chronic thromboembolic pulmonary hypertension (CTEPH) in General University Hospital, Prague, in 2021 were evaluated retrospectively and underwent both right heart catheterisation and CT pulmonary angiography. The parameters derived from the CT examinations were correlated with the recorded haemodynamic parameters. RESULTS: A method was developed for semiautomatic detection of hypoperfused tissue from CT using widely available software and a statistically significant correlation was found between the proportion of hypoperfused parenchyma and the mean pulmonary artery pressure (mPAP; R2 0.22; p<0.01) and pulmonary vascular resistance (PVR; R2 0.09; p<0.05). CONCLUSIONS: The developed method facilitates the quantification of mosaic perfusion, which is associated with important haemodynamic parameters (mPAP and PVR) in patients with CTEPH.
References provided by Crossref.org