This site is intended for health professionals only

At the heart of general practice since 1960

Read the latest issue online

GPs buried under trusts' workload dump

Practices must define their own thrombocytosis threshold or risk missing cancer

Dr Helen Egglestone 

Thrombocytosis, or raised platelet count, has been recognised as a significant positive predictor for malignancies, especially the LEGO cancers (lung, endometrium, gastric and oesophageal) in the NICE guidance for suspected cancer.

The NICE guidance does not suggest a definite cut-off for the platelet count, though emerging evidence suggests that even a relatively minimal rise or even high normal counts can be suggestive of cancer. A recent study by the British Journal of General Practice showed a nearly 12% positive predictive value for men, and 6.5% for women - much higher than many other markers of suspected cancer.

However, this significance is often likely to be overlooked in routine general practice, because this is a relatively recent soft marker for cancer, particularly on routine screening blood tests, which is when appropriately investigating an elevated platelet count could allow early cancer diagnosis and, most importantly, early treatment.

An initial search of our practice population of about 10,000 patients identified that only 29% of ‘at risk’ patients had a clear-documented plan for their raised platelet count in their medical records, and clinicians needed more clarity and empowerment as to how these fit in with referral criteria and the processes locally.

To address this problem, we decided to create a practice-based guideline for reviews of platelet counts over 450, as this is the current local laboratory cut-off for thrombocytosis.

We decided to create a practice-based guideline for reviews of platelet counts over 450, as this is the current local laboratory cut-off for thrombocytosis

Patients were then reviewed for red-flag symptoms and referred for further investigation as appropriate. Any patient without red-flag symptoms had a repeat count at an interval of six to eight weeks. If thrombocytosis persisted, further imaging was suggested, including a chest X-ray followed by consideration of CT scans.

Six months after the implementation of the guidelines, 75% of patients with a high count had a clear plan - up from the previous 29%. These plans resulted in either repeat testing, imaging, direct referral to a specialist or no action, depending on the circumstances.

All members of the team, including nurse practitioners, doctors and administrative staff, played a role in reviewing results, requesting that patients attend for review, and thorough screening for red-flag symptoms.

Though this is a small cohort and not statistically significant, it demonstrates that implementing a simple practice-based guideline can significantly improve patient outcomes with minimal increase in practice workload.

We also found that approximately 44% of patients with a high platelet count were found to have cancer, once those unsuitable or with a known haematological disorder and chronically raised platelet count were excluded.

This shocking figure clearly hammers home the point that the significance of high platelet count cannot be overemphasised.

Dr Helen Egglestone, a GP trainee in County Durham, conducted the audit, with input from Dr Kamal Sidhu, a GP trainer, and the team at Blackhall and Peterlee Practice

Rate this article  (4.67 average user rating)

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Readers' comments (6)

  • Cobblers

    If I recall one of the criteria for referral along the 2ww cancer pathway was a 'risk' of 3%.

    Your small study suggests that those with a persistently raised platelet count should be referred as 'Cancer ?site (LEGO)'

    Indeed you could spend quite a lot of time working these up before referral but is there that much spare time in your surgery?

    Unsuitable or offensive? Report this comment

  • Excellent article- perhaps the most interesting primary care clinical piece that I have read this year.

    I suspect that there is far more mileage in evidence based data mining of our existing records than in trying to replace GPs with an AI driven interface.

    Unsuitable or offensive? Report this comment

  • I agree, a very interesting read and will definitely influence my practice. Out of the 10,000 patients there is no mention of how many had high platelet counts and what was the cut off? The sample size could be quite small? A platelet count over 450 is very alarming?

    Unsuitable or offensive? Report this comment

  • Absolutely excellent piece of work!

    Unsuitable or offensive? Report this comment

  • Well done Helen and Kamal

    I did a retrospective audit of lung cancers in our practice (following a diagnosis in a young patient who had a normal CXR 8 months earlier)

    High platelet count was one of the risk factors

    I wonder if ethnicity would impact on these parameters though?

    Unsuitable or offensive? Report this comment

  • A very very sensible piece if work.

    Much more useful than incentives help smokers quit and better GP funding improves the patient experience

    Unsuitable or offensive? Report this comment

Have your say