This site is intended for health professionals only

At the heart of general practice since 1960

GPs 'should consider cancer' in patients with upper-normal platelet counts

GPs should consider cancer in patients with an upper-normal platelet count, a new study has suggested.

Researchers linked primary care patient records with national cancer data and found that the incidence of cancer rose in line with the increasing platelet count, with a consistently higher prevalence in males.

While it was previously thought that a platelet count above 400 x 109/l was an indicator for cancer, this new research has suggested that GPs should now consider cancer diagnosis in patients with a platelet count above 325 × 109/l for men and 375 × 109/l for women.

However the paper has called for the findings to be replicated in a much larger sample before a ‘blanket recommendation’ is issued.

The cohort study, published in Oxford Family Practice, linked medical records from the Clinical Practice Research Datalink with English National Cancer Registration Service data, from 2000 to 2013.

A sample of 2,700 patients was split into three groups depending on platelet count: 325–349 × 109/l; 350–374 × 109/l; 375–399 × 109/l.

The research team then obtained the incidence of cancer diagnosis in the year following that platelet count, and saw that it rose with the increasing platelet count from 2.6% in the 325–349 × 109/l group to 5.1% in 375–399 × 109/l band.

Colorectal cancer was most commonly diagnosed across all three groups, and incidence was consistently higher in males than in females.

Men in the lowest group - with platelet counts >325 × 109/l - exceeded NICE’s 3% risk threshold for cancer investigation.

Women surpassed the threshold at a platelet count of 375 × 109/l, although the risk for the 350–374 × 109/l group was just under at 2.8%.

The paper stated that: ‘This suggests that clinicians should consider a cancer diagnosis in patients with a platelet count above these values’.

However it added that until the results are replicated on a larger scale, ‘a blanket recommendation for investigation would be premature’.

NICE's suspected cancer guidance, updated last year, highlighted a 'raised platelet count' as a cancer indicator however did not specify what is classified as 'raised'.

Imperial College London professor of oncology Iain McNeish said: ‘There appears to be a linear gradient of increasing cancer risk with increasing platelet count, so it's important to define where the cut-off should be to define ‘high’, and also the age at which you would start investigating patients for cancer’.

Research from last year found that a platelet count above 400 x 109/l, known as thrombocytosis, is a significant marker of cancer risk.

It said that over 11% of men and 6% of women diagnosed with thrombocytosis were then diagnosed with cancer in the following year.

The risk increased further when an additional high platelet count was measured within six months of the first finding.

Readers' comments (9)

  • The title is wrong. It should say - consider cancer in patients with higher platelet counts.

    Unsuitable or offensive? Report this comment

  • If we had adequate access to endoscopy and CT we might be able to to something useful with this knowledge.

    Unsuitable or offensive? Report this comment

  • I am seeing some new partners run themselves in circles and driving both themselves and the patients crazy with some of these approaches. I am yet to see one early diagnosis. And again if the rest of the system is alligned to this it's just a waste of time.

    Unsuitable or offensive? Report this comment

  • I never quite know what to do with these patients. The haematologist is not interested in these referrals.

    Ideally there should be "a suspected cancer" pathway. I understand that such clinics exist in some parts of the country. We then refer anyone where we thibk there may be a malignnancy but the primary is not at all obvious.

    Raised platelets, eosinophils, weight loss etc

    Unsuitable or offensive? Report this comment

  • FAO Merlin

    We have exactly this service, vague symptoms pathway. People were triaged and often would have a CT CAP and not seen. Was a smart service and well run.

    It's being decommissioned of course.

    Unsuitable or offensive? Report this comment

  • 2 week wait to exclude cancer in everyone with NORMAL platelets but at the higher range? They would not fit the criteria imposed on us and the service will not cope. It shows people who research and write the rules are not clinicians to even come up with such a suggestion.

    Unsuitable or offensive? Report this comment

  • Another excuse to sue......

    Unsuitable or offensive? Report this comment

  • Tricky to know how to usefully use this information. Over-diagnosis and over-treatment are the big challenges in modern medicine today.

    Unsuitable or offensive? Report this comment

  • Again not sure this is any of our business.

    This is one for NICE,why have a dog and bark yourself ?They need to look at how sensitive and specific this is, as a predictor,starting with how common raised platelets are,and a quick glance at the long long list of causes.

    As already mentioned haematology,whom one might expect to know this have no interest.

    If a radiologist sees a big lump on a chest x ray he tends to add, "suggest urgent referral to respiratory".

    NICE have to issue guidelines to haematologists, saying they must add,"normal platelets,advise urgent referral to gastroenterology".

    Personally I have never seen this,which suggests our legions of haematologists and gastroenterologists know something the researchers don't

    Unsuitable or offensive? Report this comment

Have your say