Article Text

PDF

Commentary
  1. David M Hansell
  1. Royal Brompton Hospital, London SW3 6NP, UK
  1. Dr D M Hansell.

Statistics from Altmetric.com

The common feature of the reports by Franco et al 1 and Oliver et al 2 is the use of spiral (or volumetric) computed tomography to demonstrate features which would not be readily identifiable on conventional computed tomographic (CT) scanning. The advantages of spiral CT over conventional CT scanning are twofold: increased speed of data acquisition and volumetric (rather than slice by slice) data acquisition. The attribute of speed means that most thoracic examinations can be performed within a single breath hold and the timing of intravenous contrast administration can be precisely tailored, thus allowing reproducible enhancement of any desired part of the vasculature—for example, the pulmonary arteries in cases of suspected pulmonary embolism. Because an entire volume of data is acquired (with almost equal spatial resolution in the three axes) it is possible to reconstruct images in any plane, including three-dimensional (3-D) reconstructions.3 Most examinations acquired with spiral CT scanning are presented as a series of transaxial slices, reflecting the traditional presentation of conventional CT images.

In the report by Franco et al 1 the clarity with which the anomalous arteries feeding the sequestrated lung are shown on the 3-D reconstructions is striking. In the past a separate preoperative examination (either aortography or possibly magnetic resonance angiography) to identify the vascular supply would have been regarded as mandatory. Other imaging tests such as radionuclide scintigraphy or ultrasound may answer specific questions in cases of pulmonary sequestration, but the wealth of information now available from a single spiral CT examination is remarkable. Quite apart from their aesthetic appeal, the main benefit of these readily produced 3-D reconstructions is an easy appreciation of what can be complex anatomy. Nevertheless, claims for the increased diagnostic gain from these 3-D reconstructions should not be too extravagant: the anomalous vessels would be identifiable on images presented in the standard transaxial format, although without such immediacy. Furthermore, demonstration of the venous drainage into the pulmonary circulation (for the classic intralobar sequestrations) may not be so readily obtained with a single spiral CT examination. However, the ability to extract so much information from a spiral CT examination represents a substantial advance on conventional CT scanning.

Spiral CT pulmonary angiography is an effective way of demonstrating pulmonary embolism in segmental and larger arteries.4 The basic sign of a filling defect within a well opacified pulmonary artery is straightforward enough. The case report by Oliver et al 2 highlights the fact that there may be ancillary signs of pulmonary embolism on spiral CT scanning—in this case shift of the interventricular septum—which corroborates the diagnosis and, more controversially, provides prognostic information. Shift of the interventricular septum and other signs of right ventricular dysfunction are readily demonstrated on echocardiography, but in cases of suspected pulmonary embolism echocardiography does not provide the breadth of information of a spiral CT examination. For example, additional signs of pulmonary embolism, including a mosaic perfusion pattern of the lung parenchyma and radiographically cryptic pleural effusions or small pulmonary infarcts, can be readily picked up on spiral CT scanning. Conversely, because spiral CT scanning provides the “big picture”, an alternative diagnosis may be shown by spiral CT scanning in up to one third of patients investigated for suspected pulmonary embolism.5

The application of image processing to volumetric spiral CT data can be broadly divided into graphic 3-D realisations—for example, virtual reality bronchoscopy6—and the rendering of data so that it is suitable for quantitative analysis. However, progress towards routine volumetric (3-D) depictions of spiral CT data is likely to be slow.7 Even at this early stage of development it is possible to extract very precise volumetric measures of abnormal lung; the most obvious application is in the quantification of low attenuation lung (corresponding to emphysema) on inspiratory and expiratory spiral CT scans. Early results have shown remarkably good correlation between the extent of low attenuation lung derived from 3-D reconstructions of the lungs with functional indices of air flow obstruction and air trapping.8 With this new technique the entire lungs are evaluated, unlike the conventional “density mask” approach which can be applied only to individual CT sections, (which introduces problems with sampling). With the powerful combination of volumetric data from spiral CT scanning and advanced image processing, the excitement has only just begun.

References

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.