Intended for healthcare professionals

Practice Uncertainties

Should we screen extensively for cancer after unprovoked venous thrombosis?

BMJ 2017; 356 doi: (Published 22 March 2017) Cite this as: BMJ 2017;356:j1081
  1. Faizan Khan, MSc epidemiology student1 2,
  2. Christian Vaillancourt, associate professor of emergency medicine, emergency physician1 2 3,
  3. Marc Carrier, associate professor of medicine, thrombosis specialist1 2 4
  1. 1Faculty of Medicine, University of Ottawa, Ottawa, Ontario, Canada
  2. 2Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa
  3. 3Department of Emergency Medicine, Ottawa Hospital, Ottawa
  4. 4Department of Medicine, Ottawa Hospital, Ottawa
  1. Correspondence to: M Carrier mcarrier{at}

What you need to know

  • The prevalence of occult cancer in patients with a first unprovoked venous thromboembolism seems to be lower (~4%) than previously reported (10%)

  • There is limited evidence to recommend extensive cancer screening with computed tomography in such patients

  • Consider history and physical examination, basic laboratory tests, and results from routine age-specific cancer screening to guide further testing as an alternative to extensive screening

How far to go in screening patients with an unprovoked venous thromboembolism (VTE) for an occult cancer is a clinical dilemma. Unprovoked VTE, either deep vein thrombosis or pulmonary embolism, can be the first manifestation of an undiagnosed cancer. Until recently, the literature suggested that up to 10% of such patients would be diagnosed with a cancer in the year after their diagnosis of VTE.1 However, the incidence of occult cancer in patients studied in two recent, high quality, randomised controlled trials was only about 4%.23 This drop in the proportion of people with occult cancer may require an adjustment in the clinical approach.

Fig 1 outlines a conservative approach and a more detailed approach to investigating such patients. Extensive screening has become the standard of care, though it is based on limited data.

Fig 1 Occult cancer screening strategies in unprovoked venous thromboembolism (from NICE guidelines 20124)

However, high quality data from recently completed trials discussed below suggest that extensive screening strategies may not provide additional value over routine cancer screening in the frequency of cancer detection in these patients.

What is the evidence of uncertainty?

Search strategy and study selection

We searched PubMed (from inception to 31 December 2016) for randomised controlled trials and systematic reviews using the search terms “cancer screening,” “venous thromboembolism,” “unprovoked,” “meta-analysis,” and “randomized controlled trial.” We reviewed articles published in English between 2012 (publication of NICE guidelines) and 2016. We also searched the Cochrane Library and …

View Full Text

Log in

Log in through your institution


* For online subscription