Photon-Counting Detector CT Iodine Maps Versus SPECT/CT: Advancing Lung Perfusion Imaging in Chronic Thromboembolic Pulmonary Hypertension.

Abstract

Objectives: The aim of this study was to compare the accuracy of photon-counting detector computed tomography (PCD-CT) iodine maps of the lung parenchyma with perfusion scintigraphy for detection and extent estimation of pulmonary perfusion defects.

Materials and methods: This institutional review board-approved retrospective study included 26 subjects (11 male, aged 57.2 ± 15.8 years; 15 female, aged 55.2 ± 15.7 years) who underwent clinically indicated PCD-CT and perfusion SPECT/CT to assess for chronic thromboembolic pulmonary hypertension (CTEPH). Two blinded radiologists used CT iodine maps and corresponding sharp-kernel CT reconstructions in lung window to evaluate presence and extent of lobar perfusion defects and detect patients with CTEPH (or CTEPH overlap with other causes of PH). Accordingly, 2 blinded nuclear medicine physicians/radiologists evaluated perfusion SPECT/CT scans. The clinical diagnosis was reviewed in an interdisciplinary clinical setting. Quantitative analyses were calculated for both modalities. Perfusion defect estimation was compared with right heart catheter measurements.

Results: Of the 26 subjects included, 10 were diagnosed with CTEPH or CTEPH overlap, 12 were diagnosed with PH associated with other pathologic mechanisms, 3 had no PH, and 1 had previous acute pulmonary embolism, which resolved. Radiation dose was greatly reduced for PCD-CT compared with SPECT/CT (1.19 [±0.33] mSv; 6.34 [±1.68] mSv, respectively, P < 0.001). Both PCD-CT readers (R1, R2) showed a trend toward higher accuracy, sensitivity, and specificity for CTEPH diagnosis compared with the scintigraphy consensus (SC) (accuracy: R1 0.85, R2 0.88, SC 0.73; sensitivity: R1 0.90, R2 0.90, SC 0.80; specificity: R1 0.81, R2 0.88, SC 0.69), although there was no significant difference observed (P > 0.688). There was good to excellent agreement between both PCD-CT readers for perfusion defect estimation. Moderate intermodality agreement was observed for CTEPH diagnosis certainty and perfusion defect estimation. The quantitative evaluation showed strong to excellent correlation between PCD-CT and SPECT/CT relative perfusion. There was a significant moderate correlation between PCD-CT perfusion defect estimations and mean pulmonary artery pressure (R1: r = 0.49, P = 0.020; R2: r = 0.49, P = 0.021), pulmonary vascular resistance (R1: r = 0.60, P = 0.003; R2: r = 0.52, P = 0.013), and cardiac index (R1: r = -0.45, P = 0.042).

Conclusions: PCD-CT iodine maps allow for accurate CTEPH detection and are comparable to perfusion SPECT/CT with good quantitative correlation, but only moderate qualitative agreement, at greatly reduced radiation dose. Furthermore, visual PCD-CT perfusion defect extent was associated with prognostic right heart catheter measurements.