Abstract
PURPOSE: The standardized uptake value (SUV) can be measured by single-photon emission computed tomography with computed tomography (SPECT/CT) to provide a quantitative assessment; however, it has not been utilized in lung perfusion imaging (LPI). Since Tc-99m MAA is primarily trapped in the lungs, we hypothesized that the SUV normalized to lung volume using CT attenuation correction scans (SUVLV) might offer a more reliable evaluation than the conventional SUV normalized to body weight (SUVBW). METHODS: This single-center retrospective study included 58 patients (39 derivation cohorts: 25 patients with chronic thromboembolic pulmonary hypertension (CTEPH) and 14 controls; 19 validation cohorts). SPECT/CT-LPI and right heart catheterization were performed at baseline, initial follow-up, and after completing BPA in the derivation cohort. The lung volume fractions below optimal SUV thresholds that optimally correlated with mean pulmonary arterial pressure (mPAP) were investigated. The diagnostic performance of SUVLV was evaluated to predict mPAP >20 mm Hg in the validation cohort. RESULTS: Mean SUVLV remained consistent (0.88-0.90, P=0.565) across populations and treatment stages, whereas mean SUVBW varied significantly (17.8-25.8, P<0.001). Lung volume fraction with SUVLV ≤0.7 exhibited a strong correlation with mPAP (r=0.632, P<0.001) and decreased progressively with successful BPA and mPAP improvement. In the validation cohort, a cutoff value of lung volume fraction of 50.1% with SUVLV≤0.7 demonstrated 100% sensitivity and specificity for predicting mPAP >20 mm Hg. CONCLUSIONS: SUVLV is a reliable quantitative measure for SPECT/CT-LPI compared with SUVBW and may enhance intrapatient and interpatient assessments of disease severity and treatment response in CTEPH.