Lung Perfusion with Dual-Energy Multi-detector Row CT: Can It Help Recognize Ground Glass Opacities of Vascular Origin?

Pontana, François; Remy-Jardin, Martine; Duhamel, Alain; Faivre, Jean-Baptiste; Wallaert, Benoit; Remy, Jacques

doi:10.1016/j.acra.2009.12.013

« Previous Next »Academic Radiology
Volume 17, Issue 5 , Pages 587-594, May 2010

Lung Perfusion with Dual-Energy Multi-detector Row CT:

Can It Help Recognize Ground Glass Opacities of Vascular Origin?

François Pontana, MD
- Department of Thoracic Imaging, Hospital Calmette, University Center of Lille (EA 2694), Boulevard Jules Leclercq, 59037 Lille cedex, France
Martine Remy-Jardin, MD, PhD
- Department of Thoracic Imaging, Hospital Calmette, University Center of Lille (EA 2694), Boulevard Jules Leclercq, 59037 Lille cedex, France
- Address correspondence to: M.R.-J.
Alain Duhamel, PhD
- Department of Biostatistics, University Center of Lille, Lille, France
Jean-Baptiste Faivre, MD
- Department of Thoracic Imaging, Hospital Calmette, University Center of Lille (EA 2694), Boulevard Jules Leclercq, 59037 Lille cedex, France
Benoit Wallaert, MD
- Department of Pulmonology, Hospital Calmette, University Center of Lille (EA 2694), Boulevard Jules Leclercq, 59037 Lille cedex, France
Jacques Remy, MD
- Department of Thoracic Imaging, Hospital Calmette, University Center of Lille (EA 2694), Boulevard Jules Leclercq, 59037 Lille cedex, France

Received 6 October 2009; received in revised form 26 November 2009; accepted 8 December 2009. published online 04 March 2010.

Rationale and Objectives

The aim of this study was to evaluate whether the spectral characterization of the iodine content of lung microcirculation could help identify ground-glass opacity (GGO) of vascular origin.

Materials and Methods

Thirty-five consecutive patients with GGO of bronchioloalveolar (group 1; n = 24) and vascular (group 2; n = 11) origin underwent dual-energy multi-detector computed tomographic angiography of the chest using a standard injection protocol. For each patient, two radiologists evaluated by consensus the presence, location, and extent of GGO on diagnostic computed tomographic scans (ie, contiguous 1-mm-thick averaged images from both tubes) and characteristics of the corresponding areas on perfusion scans.

Results

A total of 443 segments with GGO were depicted on the diagnostic scans (group 1, n = 231; group 2, n = 212), always intermingled with areas of normal lung attenuation, with a mean of 12.7 segments with GGO per patient. Areas of GGO were located at the level of the upper lobes (n = 128), middle lobe and/or lingula (n = 81), and lower (n = 234) lobes, involving <25% (n = 165), 25% to 50% (n = 103), 50% to 75% (n = 155), and >75% (n = 20) of the segmental surface. The overall quality of perfusion scans was rated as interpretable in all patients. Perfusion scans depicted areas of hyperattenuation within segments of GGO with a significantly higher frequency in group 2 (211 of 212 [99.5%]) than in group 1 (27 of 231 [12%]) (P < .0001). Hyperattenuated areas of vascular origin were observed to match the areas of GGO in surface (203 of 211 [96%]) and contours (208 of 211 [98.6%]).