Chest radiograph

From Ganfyd

Jump to: navigation, search
See Category: Chest radiographs for image gallery of chest radiographs.
Normal CXR

A radiographic investigation which produces a plain view of the entire thoracic cavity.

Ideally chest radiographs should be taken in the erect postero-anterior (PA) projection as this causes the least magnification of the heart. In a very unwell individual this may not be possible and a supine antero-posterior (AP) film may have to be accepted.

Films are routinely taken in inspiration, with the exception of a case where a subtle pneumothorax needs to be excluded and occasionally at the reporting Radiologist's request to see if a lesion moves in different phases of respiration, helping to site it.

A lateral film may be taken, but is not routine. The more interesting lung should be closest to the film in order to appear sharpest in the image.


It is very rare that a decision to xray a chest which has not been recently xrayed causes criticism or upset, and if it takes a noticeable time to decide whether to request one, the answer is almost certainly to do it. In areas or times when tuberculosis is more common, this applies doubly so. Nevertheless, we must aim to minimise the radiation dose to the population as a whole.


The best way of learning how to spot abnormalities on a chest xray is to look at a lot of them, normal and abnormal. If there is an obvious tumour or pneumonia examine the rest of the film more thoroughly. There may well be another abnormality and it can be embarrassing to have this pointed out by someone else. (Does the “tumour” have a fluid level?)

Going over the image in a systematic way reduces the chances of missing difficult to see features. Here is one such.

  • general review – is the film well penetrated and symmetrical. Are breast shadows (if present) equal? Are nipple shadows (if present) equal? Are there artefacts such as wire sutures after cardiac surgery?
  • is the image centred? – inner clavicles should be the same distance from the midline
  • is the trachea central?
  • follow the mediastinal outline downwards. Are their any bumps that shouldn’t be there? What might they be?
  • does the lowest part of the heart shadow meet the diaphragm at a sharply defined angle? If not why not?
  • does the dome of the diaphragm have a normal sweep? Why not?
  • is the heart width less than twice the chest width (cardiothoracic ratio over 50% is abnormal e.g. in heart failure)
  • does the outer edge of the diaphragm meet the pleura at a sharp acute angle? If not why not?
  • is the sweep from the cardiophrenic angle to the lung apex symmetrical and equal? If not, why not?
  • do the lung markings go from the mediastinum to all areas of the lung normally?
  • check the skeleton. Fractured ribs are difficult to see but there may be signs of old fractures. They may be at different stages or places suggesting previous trauma. Bony metastases or osteoporotic signs may be seen.
LogoKeyPointsBox.pngA Glasgow radiologist caused great excitement some years ago by diagnosing rickets in the wrist (inadvertently shown) of a child referred for a chest xray which was otherwise normal

With practice and pattern recognition this will eventually take only a few seconds. Try these [1] [2] [3]