Volumetric High-resolution Computed Tomography of the Lung—a Comparison to Standard Acquisition in Diffuse Lung Disease

JJ Kasthoori, G Kumar, ML Wastie

Hong Kong J Radiol 2005;8:233-43

Aim: The purpose of this study was to compare volumetric high-resolution computed tomography of the lungs, which can be reconstructed in different planes and subjected to various post-processing techniques, to standard high-resolution computed tomography of the lungs.

Patients and Methods: Sixty consecutive patients referred for high-resolution computed tomography of the lungs were scanned using a 16-slice computed tomography scanner. Images of the entire chest were obtained with a slice thickness of 0.625 mm. A kV of 120 and exposure varying from 100-320 mAs was used. Two sets of images were obtained by reconstructing the raw data using a high spatial frequency algorithm. These comprised 1.25-mm sections at 10 mm intervals (considered the standard data) and contiguous 0.625-mm thick sections (considered the volumetric data). The volumetric images were reviewed with multiplanar reformation on a dedicated workstation. The axial projections were also post-processed as maximum intensity projection and minimum intensity projection. Analysis was done in terms of abnormality detection, characterisation and extent and results for the standard and volumetric imaging were compared.

Results: Ten (7 with poor image quality, 3 with normal findings) out of 60 patients were excluded from the study. Volumetric high-resolution computed tomography was found to provide additional findings in 80% of the study population with documented lung disease in terms of detection (12%), characterisation (22%) and depiction of the extent (46%) of the abnormality. Volumetric high-resolution computed tomography when post-processed to maximum intensity projection/minimum intensity projection images was significantly superior to the standard acquisition in terms of characterising and depicting the extent of the abnormalities (p < 0.05). Multiplanar volume reformats were significantly better than the standard high-resolution computed tomography in terms of determining the true extent of abnormalities (p < 0.05).

Conclusion: Volumetric high-resolution computed tomography with multiplanar reconstruction and post-processing gives considerably more information than standard high-resolution computed tomography.