Characterization and clinical significance of right ventricular mechanics in pulmonary hypertension evaluated with cardiovascular magnetic resonance feature tracking. Issue 1 (December 2016)

• Record Type: Journal Article
• Authors: de Siqueira, Maria; Pozo, Eduardo; Fernandes, Veronica; Sengupta, Partho; Modesto, Karen; Gupta, Sushilkumar Satish; Barbeito-Caamaño, Cayetana; Narula, Jagat; Fuster, Valentin; Caixeta, Adriano; Sanz, Javier
• Is Part Of: Journal of cardiovascular magnetic resonance. Volume 18:Issue 1(2016)
• Page Start: 1
• Page End: 12
• Publication Date: 2016-12
• Subjects: Pulmonary hypertension -- Strain -- Right ventricle; Cardiovascular system -- Magnetic resonance imaging -- Periodicals; 616.1207548
• DOI: 10.1186/s12968-016-0258-x
• Languages: English
• ISSNs: 1097-6647

Abstract

Abstract Background Prognosis in pulmonary hypertension (PH) is related to right ventricular (RV) function. Quantification of RV mechanics may offer additive value. The objective of our study is to determine the feasibility and clinical and prognostic value of RV strain analysis by cardiovascular magnetic resonance (CMR) based feature tracking (FT) in PH. Methods We retrospectively enrolled 116 patients (age 52.2 ± 12 years, 73.6 % women) referred to CMR for PH evaluation who underwent right heart catheterization within 1 month. Using dedicated FT software, peak global longitudinal and circumferential RV strain and strain rates (GLS, GCS, GLSR, and GCSR, respectively) were quantified from standard cine images. Using multivariate regression analysis, we evaluated the associations of strain with a composite endpoint of death, lung transplantation, or functional class deterioration. Results RV strain analysis was feasible in 110 (95 %) patients. Patients were classified into: Group A (no PH, normal right ventricular ejection fraction [RVEF];n = 17), Group B (PH, normal RVEF;n = 26), or Group C (PH, abnormal RVEF;n = 67). All strain and strain rate values were reduced in Group C. Furthermore, GCSR was significantly reduced in Group B (-0.92 [-1.0/-0.7];p < 0.001) compared to Group A (-1.12 [-1.3/-0.9];p < 0.001). After adjustment for six clinically meaningful covariates, GLS (hazard ratio 1.06;p = 0.026), GLSR (hazard ratio 2.52;p = 0.04), and GCSR (hazard ratio 4.5;p = 0.01) were independently associated with the composite endpoint. GCSR successfully discriminated patients with and without events (p = 0.01). Conclusions Quantification of RV strain with CMR-FT is feasible in the majority of patients, correlates with disease severity, and is independently associated with poor outcomes in PH.