Dual-energy CT to estimate clinical severity of chronic thromboembolic pulmonary hypertension: Comparison with invasive right heart catheterization

Purpose To evaluate whether the extent of perfusion defects assessed by examining lung perfused blood volume (PBV) images is a stronger estimator of the clinical severity of chronic thromboembolic pulmonary hypertension (CTEPH) compared with other computed tomography (CT) findings and noninvasive parameters. Materials and methods We analyzed 46 consecutive patients (10 men, 36 women) with CTEPH who underwent both dual-energy CT and right-heart catheter (RHC) examinations. Lung PBV images were acquired using a second-generation dual-source CT scanner. Two radiologists independently scored the extent of perfusion defects in each lung segment employing the following criteria: 0, no defect, 1, defect in <50% of a segment, 2, defect in ≥50% of a segment. Each lung PBV score was defined as the sum of the scores of 18 segments. In addition, all of the following were recorded: 6-min walk distance (6MWD), brain natriuretic peptide (BNP) level, and RHC hemodynamic parameters including pulmonary artery pressure (PAP), right ventricular pressure (RVP), cardiac output (CO), the cardiac index (CI), and pulmonary vascular resistance (PVR). Bootstrapped weighted kappa values with 95% confidence intervals (CIs) were calculated to evaluate the level of interobserver agreement. Correlations between lung PBV scores and other parameters were evaluated by calculating Spearman's rho correlation coefficients. Multivariable linear regression analyses (using a stepwise method) were employed to identify useful estimators of mean PAP and PVR among CT, BNP, and 6MWD parameters. A p value < 0.05 was considered to reflect statistical significance. Results Interobserver agreement in terms of the scoring of perfusion defects was excellent (κ = 0.88, 95% CIs: 0.85, 0.91). The lung PBV score was significantly correlated with the PAP (mean, rho = 0.48; systolic, rho = 0.47; diastolic, rho = 0.39), PVR (rho = 0.47), and RVP (rho = 0.48) (all p values < 0.01). Multivariable linear regression analyses showed that only the lung PBV score was significantly associated with both the mean PAP (coefficient, 0.84, p < 0.01) and the PVR (coefficient, 28.83, p < 0.01). Conclusion The lung PBV score is a useful and noninvasive estimator of clinical CTEPH severity, especially in comparison with the mean PAP and PVR, which currently serve as the gold standards for the management of CTEPH.