We evaluated the feasibility of perfusion SPECT/CT for providing quantitative data for estimation of perfusion defect extent in chronic thromboembolic pulmonary hypertension (CTEPH).Methods
Thirty patients with CTEPH underwent 99mTc–human serum albumin lung perfusion SPECT/CT. Perfusion defects were quantified using 3 different methods: (1) visual, semiquantitative scoring of perfusion defect extent in each lung segment, (2) threshold-based segmentation of perfused lung volumes, and (3) threshold-based segmentation of perfused lung volumes divided by segmented lung volumes at CT (perfusion index). Imaging findings were correlated with right-sided heart catheterization results and N-terminal pro–B-type natriuretic peptide. Receiver operating characteristic analysis was performed to identify SPECT thresholds for mean pulmonary arterial pressure (PAPm) greater than 50 mm Hg.Results
Assessment of lung perfusion provided similar results using all 3 methods. The perfusion defect score correlated with PAPm (rs = 0.60, P = 0.0005) and was associated with serum levels of N-terminal pro–B-type natriuretic peptide (rs = 0.37, P = 0.04). Perfused lung volume (40% threshold, rs = −0.48, P = 0.007) and perfusion index (40% threshold, rs = −0.50, P = 0.005) decreased as PAPm increased. Receiver operating characteristic analysis showed that perfusion defect score (sensitivity, 88%; specificity, 77%; area under the curve [AUC] = 0.89, P = 0.001), perfused lung volume (sensitivity, 88%; specificity, 64%; AUC = 0.80, P = 0.01), and perfusion index (sensitivity, 88%; specificity, 64%; AUC = 0.82, P = 0.009) could identify patients with PAPm of greater than 50 mm Hg.Conclusions
Quantitative analysis of perfusion defects at SPECT is feasible, provides a measure of disease severity, and correlates with established clinical parameters. Quantitation of perfusion SPECT may refine the diagnostic approach in CTEPH providing a quantitative imaging biomarker, for example, for therapy monitoring.