Article Text

Download PDFPDF

Lung cancer awareness, early diagnosis and staging
P64 Review of CXRs Requested in the Community: Can We Improve Cancer Detection?
  1. V Wells,
  2. A Stanton,
  3. P Osborne,
  4. W Hicks,
  5. A Beale,
  6. N Ridley
  1. Great Western Hospital, Swindon, UK


Introduction In our trust all patients with a CXR suggestive of cancer have their CXR reviewed in the lung MDT. Following a serious incident investigation in our trust the question was raised as to whether this practise should be extended to all patients with “red flag symptoms” but not necessarily an abnormal CXR, in an attempt to improve early detection of lung cancer.


  1. To determine frequency of red flag symptoms (i.e. hoarseness/haemoptysis) in patients undergoing community chest radiography

  2. To determine frequency of subsequent lung cancer diagnosis in such patients

  3. To determine if routine MDT review of such patients would improve time to lung cancer diagnosis.

Method Review of community CXR request forms and images over a 1 month period (January 2010) was performed. Any CXR requested for a red flag symptom and reported as “normal”, was re-reviewed by a radiologist. Review of medical records over the succeeding 27 months to determine if there was any subsequent detection of lung cancer.

Results 549 community CXR’s were requested in the review period. A random sample of 229 requests were reviewed. 8 requests included haemoptysis in clinical details and none included hoarseness. 5/8 CXR’s were reported as normal. 3/8 were reported as abnormal, with one suggestive of cancer (subsequently diagnosed) and discussed at MDT anyway. The remaining 2 patients with abnormal CXR’s had subsequent unremarkable imaging. None of the remaining 5 patients with haemoptysis and a normal CXR had any subsequent presentation with lung cancer.

Conclusion The prevalence of red flag symptoms in patients being referred for community CXR’s is low. In those with red flag symptoms and a CXR that does not suggest cancer, subsequent presentation with cancer is at worst highly unlikely. The practise of routinely reviewing images of such patients in a lung cancer MDT is unlikely to improve detection rate or time to diagnosis of lung cancer.

Statistics from

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.