Investigating an incidental finding of thrombocytopeniaBMJ 2013; 346 doi: https://doi.org/10.1136/bmj.f11 (Published 11 January 2013) Cite this as: BMJ 2013;346:f11
- Correspondence to: J Murray
When thrombocytopenia is found, repeat the blood count and request a blood film. This will confirm whether thrombocytopenia is genuine and will help direct subsequent investigations
Take a careful drug history, ask about risk factors for HIV and hepatitis C, and assess for features of liver disease
If the patient is well, has no abnormal clinical findings, and has isolated thrombocytopenia with no other abnormalities on blood count or film, immune thrombocytopenia is the most likely cause
The risk of bleeding is not based on the platelet count alone; also consider age, comorbidity, mandated anticoagulation, risk of trauma, and any need for surgery
An asymptomatic 64 year old woman presented to her general practitioner with fatigue and weight gain. Full blood count indices were haemoglobin 125 g/L (reference range 115-165), mean cell volume 90 fL (80-99; 1fL=1 µm3), platelets 54×109/L (150-400), white cell count 6.3×109/L (4-11), and neutrophils 4.8×109/L (2-7.5). A blood film confirmed the low platelet count but was otherwise normal.
What is the next investigation?
Thrombocytopenia may result from impaired production of platelets (for example, as a result of marrow dysfunction), increased destruction (immune or non-immune), abnormal distribution, or a combination thereof. Table 1⇓ lists common causes and examples.
If isolated thrombocytopenia is picked up incidentally in an apparently asymptomatic patient with no relevant drug treatment and a normal blood film, the diagnosis is usually immune thrombocytopenia (ITP).1 2 No accurate data on the relative frequencies of different causes of thrombocytopenia are available in the literature, although there is now an international paediatric and adult registry.3 ITP is an acquired disorder characterised by an isolated thrombocytopenia of less than 100×109/L, and new evidence shows that this is the result of both increased platelet destruction and impaired production.4 Adult chronic ITP has an incidence of about 6 per 100 000 per year in the developed world, with an equal sex ratio, although it is more common in women in the age range 30-60 years. In children the incidence has been estimated at 2-6 per 100 000 per year. It is a diagnosis of exclusion, with no specific tests that can positively diagnose it. ITP may be primary or secondary to other conditions (such as systemic lupus erythematosus, lymphoproliferative disorders, HIV, and hepatitis C). Although the evidence base to guide investigation is limited, international consensus reports on the investigation and management of primary ITP have recently been published.1 2 5
History and clinical examination can point to the underlying cause.
Myelodysplasia is more common in older patients and an important differential diagnosis in younger patients is congenital thrombocytopenia (for example, May-Hegglin anomaly and Bernard-Soulier syndrome). Incidental thrombocytopenia is also more likely to be immune than inherited in children, and ITP is usually short lived after a viral infection, with most children recovering spontaneously. By contrast, in adults it tends to develop insidiously and follow a more chronic course.
Bruising or bleeding
If present, these symptoms suggest that the thrombocytopenia is genuine and should prompt screening for a more generalised coagulopathy. Specifically ask about epistaxis, haematuria, menorrhagia, and excessive bleeding with previous haemostatic challenges (such as surgery, dental extraction, and childbirth). A recent history (especially if previously normal counts are documented) suggests an acquired cause, whereas a lifelong history (with or without a family history) suggests a mild inherited cause.
Symptoms such as fevers, night sweats, and weight loss should prompt investigation for lymphoma, infection, or cancer. Patients with ITP are usually well and have no specific symptoms.
Infection and immune history
Occasionally, thrombocytopenia is the first manifestation of HIV or hepatitis C infection in an otherwise well patient, and it is important to assess risk factors and not to overlook screening.6 Helicobacter pylori infection has also been associated with ITP and should be considered in patients with dyspepsia. The response to eradication treatment is variable.7 Other recent acute or chronic infections (viral, bacterial, or protozoan), some vaccinations (with live attenuated viruses), or a history of autoimmunity (for example, rashes or arthralgia) may also be relevant.
Drug and alcohol history
Consider all recently started drugs (prescription or non-prescription) as potential causes. Some (such as cytotoxics) result in a predictable thrombocytopenia, whereas others result in idiosyncratic thrombocytopenia. Quinine consumption (tonic water) can result in profound thrombocytopenia through an immune mediated mechanism.8 An increasing number of patients are being discharged into the community with thromboprophylaxis. Consider heparin induced thrombocytopenia in patients recently started (<14 days) on heparin; these patients may or may not have thrombotic manifestations. Low molecular weight heparin is about 10 times less likely than unfractionated heparin to have this effect.9 Excessive alcohol intake is a common cause of modest thrombocytopenia (often with macrocytosis), with the platelet count usually at 75-100×109/L.
A history of cancer may cause thrombocytopenia by many mechanisms, including bone marrow infiltration and chronic microangiopathic haemolysis.
Pregnancy broadens the differential diagnosis and should be considered in women of child bearing age. Gestational thrombocytopenia is mild (>80×109/L), common, occurs in the third trimester, and has no adverse clinical consequences. However, the hypertensive disorders (pre-eclampsia and HELLP (haemolysis, elevated liver enzymes, and low platelets) syndrome) may also occur in the third trimester and need urgent obstetric review. Disseminated intravascular coagulation may be precipitated by obstetric emergencies such as placental abruption, postpartum haemorrhage, amniotic fluid embolism, or sepsis. ITP may first present incidentally in pregnancy because women of this age group are commonly affected. The serious and rare associated condition, thrombotic thrombocytopenic purpura, can occur during any trimester.
Clinical inspection for petechiae, bruising, mucosal bleeding, and fundal bleeding may help define the severity of thrombocytopenia and risk of bleeding.
Lymphadenopathy or hepatosplenomegaly should prompt concern about lymphoproliferative disease, autoimmune disease, cancer, or infection.
Features of chronic liver disease are relevant. Even in the absence of hypersplenism, liver disease can cause thrombocytopenia by many mechanisms, including impaired production of thrombopoietin.10
Investigations in primary care
When a low platelet count is picked up incidentally, the full blood count must be repeated and a blood smear performed. Comparison with previous results may show changes over time. Timing, degree of thrombocytopenia, clinical context, and the presence or absence of bleeding will help direct investigations and the need for referral.11
In all patients, a blood film is important to confirm whether thrombocytopenia is genuine or factitious (platelet clumping owing to EDTA antibodies or giant platelets failing to be counted by automated analysers; fig 1⇓). In the presence of platelet clumping, the full blood count will be more accurate if performed on a citrated sample. In ITP, red and white cells have normal morphology and platelets may be large but otherwise normal. Other abnormalities in the film may offer clues to the underlying cause of thrombocytopenia. Features of other conditions include dysplastic changes (in myelodysplasia), abnormal circulating cells (such as leukaemic blasts), and red cell fragments (suggesting microangiopathic haemolysis in cancer, disseminated intravascular coagulopathy, or thrombotic thrombocytopenic purpura). Red cell fragments and a raised lactate dehydrogenase should prompt urgent referral to haematology to exclude the potentially fatal diagnosis of thrombotic thrombocytopenic purpura, which requires prompt plasmapheresis. The other classic features of thrombotic thrombocytopenic purpura (neurological symptoms, fever, and renal dysfunction) are not always present.12 Table 2⇓ lists other suggested investigations.
When to refer?
Patients with modest isolated thrombocytopenia (platelet count 100-150×109/L) without atypical features (such as lymphadenopathy or fever) do not require referral to hospital, particularly if the counts are stable. It is prudent to occasionally recheck the full blood count in primary care to ensure that counts do not deteriorate or another condition become evident. For example, rarely, isolated thrombocytopenia with no circulating blasts in the peripheral blood is the initial presentation of acute leukaemia. If the results of the full blood count are unchanged when repeated six weeks later, it is usually safe to extend the follow-up interval to several months. Patients must be aware that they should re-present if new symptoms such as bruising or bleeding occur.
Indications for urgent referral include severe thrombocytopenia (<20×109/L), severe bleeding, and red cell fragments or blasts on the blood film. Referral is also warranted if the patient has constitutional symptoms, bruising, minor bleeding, or abnormalities on examination (such as lymph nodes or splenomegaly) or the blood film (such as dysplastic changes). Referral to or discussion with a haematologist is reasonable if the platelet count is less than 100×109/L or the patient also has anaemia, neutropenia, or other changes in the blood count, such as macrocytosis. Many such situations can be discussed on an advisory basis without formal review in a hospital clinic.
Consensus guidelines recommend bone marrow examination (after referral) in patients over 60 years of age (mainly to exclude dysplasia) and in those with systemic symptoms or signs suggestive of haematological cancer.1 2
After referral to haematology and a period of careful observation, the patient’s platelet count dropped further (<30×109/L). Subsequent investigations included a bone marrow examination, which confirmed the presence of megakaryocytes with no abnormal features, therefore supporting the diagnosis of ITP.
The decision to start treatment should balance the individual patient’s bleeding risk against the side effects of treatment. Bleeding risk is influenced not only by platelet count, but also by age and comorbidity, such as liver or renal dysfunction. Platelet counts greater than 30×10⁹/L rarely require treatment in the absence of bleeding, trauma, surgery, or mandated anticoagulation.1 2 If the platelet count is less than 50×10⁹/L, avoid antiplatelet agents such as aspirin and non-steroidal anti-inflammatory drugs if possible.
If the decision is to treat (as in this case), the mainstay of treatment for ITP is immunosuppressive therapy, usually starting with corticosteroids.1 2 13 Tranexamic acid can be useful for the management of mucosal bleeding but must be avoided if haematuria is present because of the risk of clot retention.
Our patient initially responded well to steroid treatment but repeatedly relapsed on steroid withdrawal. Second line treatment options include rituximab, which has good evidence of efficacy although it is not licensed for this condition,14 or laparoscopic splenectomy for long term remission. Other agents with good quality data on effectiveness from randomised controlled trials include the thrombopoietin receptor agonists, romiplostim or eltrombopag, although cost is a problem.15 16 Romiplostim is now approved by NICE for patients with ITP that is severely symptomatic and refractory to standard treatment.
Cite this as: BMJ 2013;346:f11
This series of occasional articles provides an update on the best use of key diagnostic tests in the initial investigation of common or important clinical presentations. The series advisers are Steve Atkin, professor, head of department of academic endocrinology, diabetes, and metabolism, Hull York Medical School; and Eric Kilpatrick, honorary professor, department of clinical biochemistry, Hull Royal Infirmary, Hull York Medical School. To suggest a topic for this series, please email us at.
Both authors helped plan and draft the article, and both approved the final revision. JM is guarantor.
Competing interests: All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.
Provenance and peer review: Commissioned; externally peer reviewed.
Patient consent not required (patient anonymised, dead, or hypothetical).