Interpretation of HRCT (High Resolution Computed Tomography) is based on recognition and description of pathological patterns. Numerous flowcharts are presented in the available litera- ture on HRCT, often one for each pattern. Consequently, HRCT can be perceived as somewhat inaccessible. In this book we have organized the flowcharts for the various patterns into one easy-to-use algorithm, and hopefully have created a simple tool to aid diagnosis.
The first part of the book describes the various patterns that can be found in HRCT. The second section comprises an overview of differential diagnosis, including sample scans and descriptions of typical findings. At the end of the book there is a list of more extensive literature on HRCT that we strongly recommend.
Technical aspects of HRCT
In the early days of HRCT the examination involved taking sequential thin sections (1-1.5 mm) with intervals of 10-20 mm. Using existing volumetric scan techniques both thin and thicker (5 mm) sections can be reconstructed from the same scan. In HRCT a high spatial frequency or “sharp” reconstruction algorithm is used.
What an HRCT protocol should include is open to debate. In our experience an HRCT examination should include both thick and thin transverse sections and coronal and sagittal reformats. Diagnosis of air trapping requires 4-5 transverse sections in expiration. An inspi- ratory phase (full inspiration) series with the patient in the prone position can help rule out ”normal” dependent density.
Preferably, the scan should be performed without contrast medium. Note that high-con- trast content in the lung parenchyma, e.g. in CT-angio for pulmonary embolism, can result in increased patchy parenchymal attenuation, which can be falsely interpreted as ground-glass opacity.
Lobule (secondary lobule)
Structures within the secondary lobule are the key to HRCT interpretation. A secondary lob- ule is the smallest functional lung unit identifiable on CT (1-2.5 cm). Each secondary lobule contains up to a dozen acini, each of which contains groups of alveoli (primary lobules). In a healthy secondary lobule an artery can be identified centrally and veins peripherally. Next to the artery is a bronchiole, the walls of which are normally too thin to visualize. There are also lymph vessels – some peripheral and some radiating out from the center – which can only be visualized on CT when they are thickened (pathological). There is also a connective tissue stroma (both in the lobular periphery and intralobular parenchyma), which cannot usually be visualized radiographically.
Since the term primary lobule is no longer in common use; from this point on the terms lobule and secondary lobule should be considered synonymous.
The HRCT algorithm
The algorithm in this book is intended to assist in finding probable diagnoses based on recogni- tion of patterns. At the highest levels (1 and 2) of the algorithm, the patterns are grouped under respectively high or low attenuation as follows:
High attenuation: nodules, ground-glass, reticular pattern, consolidation
Low attenuation: cystic pattern, mosaic perfusion
At levels three and four of the algorithm some of the patterns are divided into sub-groups. Finally, at the lowest level of the algorithm a number of differential diagnoses are listed for each pattern/sub-pattern. Note that some of diagnosis may also be secondary manifestations of rheumatoid systemic diseases or part of a drug reaction. Typical findings for these and other differential diagnosis are described and illustrated in the second part of the book (chapter 7-20).
Useful diagnostic and procedural tips
Obtaining a detailed patient history including laboratory and respiratory physiology tests is mandatory. Are there signs of infection or heart failure? Is there indication of rheumatoid sys- temic disease? Is the patient on any medication? Have the patient been exposed to dusts, birds etc? Smoking history?
Many of the diseases with pulmonary engagement are long-standing and dynamic. Always check for previous radiological examinations for comparison.
When approaching a HRCT examination do not start by suggesting a diagnosis based on a general impression. Instead identify the involved pattern or patterns and use the algorithm to find some potential diagnosis.