SESSION TITLE: Imaging
SESSION TYPE: Original Investigation Slide
PRESENTED ON: Sunday, October 27, 2013 at 10:45 AM - 11:45 AM
PURPOSE: Mosaic pattern of lung attenuation on chest CT may be due to various etiologies such as small airway, and vascular lung diseases. We studied the frequency of mosaic pattern between pulmonary hypertension (PH) subgroups as well as the correlation between CT findings mosaic pattern, main pulmonary artery size, and pulmonary artery/aorta (PA/Ao) ratio with hemodynamic data, pulmonary function tests (PFT), and lung perfusion scan.
METHODS: We retrospectively reviewed the medical records of patients referred to the PH Clinic, Mayo Clinic Florida from January 1992 to December 2006. The clinical data, CT mosaic pattern, hemodynamic data, PFT’s, and lung perfusion scans were collected. The main pulmonary artery size and pulmonary/aortic ratio (Pa/Ao) were measured.
RESULTS: 306 patients with confirmed PH were included. 65% were women. The mean age was 63 years (range17-88, SD 14).159 (52%) patients were classified as Group 1, 43 (14%) patients in Group 2, 87(28%) patients in Group 3, and 17 (6%) in Group 4. Overall, 67 patients (22%) had CT mosaic pattern: 24/159 (15%) in Group 1, 17/43(40%) in Group 2, 18/87 (21%) in Group 3, 8/17 (47%) in Group 4. Mosaic pattern was significantly higher in patients with Group 4 PH (8/17, 47%; p=0.01). From 267 patients with available imaging to review, the mean size of the main pulmonary artery was 3.40 cm.(SD 0.65), mean aorta size was 3.15 cm. (SD 0.41) and mean PA/Ao ratio was 1.09 (SD 0.25). There was a significant increased main PA size, PA/Ao ratio in patients with mosaic pattern (p=0.012, and 0.009). Mosaic pattern was not significantly correlated to hemodynamic data included right atrial pressure, mean pulmonary artery pressure, pulmonary vascular resistance, and cardiac output (p= 0.56, 0.70, 0.48, and 0.24 respectively) nor segmental defects on lung perfusion scan (p=0.10). 244 patients had complete PFT, 89/244. We found no correlation between obstructive lung disease (FEV1/FVC < 70%; p=0.68), restrictive lung disease (TLC < 80%;p=0.38) and mosaic pattern.
CONCLUSIONS: In this single-center retrospective review, a mosaic pattern was observed on chest CT in 1/5 of patients evaluated for PH and was more prevalent in patients with Group 4 CTEPH. Both the main pulmonary artery size and PA/Ao ratio were positively associated with mosaicism. Conversely, there was no correlation between the mosaic pattern and hemodynamic data, PFT, and lung perfusion scan.
CLINICAL IMPLICATIONS: A mosaic pattern on chest CT in patients with PH should raise concern for the Group 4 CTEPH diagnosis.
DISCLOSURE: The following authors have nothing to disclose: Kamonpun Ussavarungsi, Augustine Lee, Charles Burger
No Product/Research Disclosure Information