Question 3.1

A 45-year-old patient was admitted to the ICU with refractory hypoxic respiratory failure requiring veno-venous extra corporeal membrane oxygenation (VV-ECMO).

The day 8 coagulation profile is below:


Patient Value

Adult Normal Range

Prothrombin time

16 sec


INR (International Normalized Ratio)



APTT (activated partial thromboplastin time)

76 sec*


Platelet count

69 x 109/L*



2.0 g/L



18 mg/L*

< 0.5

3.1.1    List four potential causes of the thrombocytopenia. (2 marks)

3.1.2    List four investigations that would help differentiate between causes.    (2 marks) 

[Click here to toggle visibility of the answers]

College Answer

Syllabus topic/section:

2.1.21 Applied Pharmacology in Intensive Care.

To identify, assess and manage common coagulation abnormalities of critically ill patients.

Candidates should be commended for the depth of knowledge displayed here. Causes of thrombocytopenia were correctly noted and prioritised, Investigations including BMAT, SRA, PF4 ELISA test and were correctly identified.
Coagulation tests, Heparin resistance and the corresponding treatment strategies were explained well.


Let us deconstruct this answer from basic principles.

1) the PT is normal, so the liver is probably doing the appropriate synthetic liver thing.

2) the APTT is abnormal, but VV ECMO is conventionally anticoaculated with heparin, which means this finding on its own is not remarkable. But the finding of a normal PT would at suggest that the APTT is in fact due to the heparin, and not due to some liver dysfunction.

3) The fibrinogen is reasonably normal, which suggests the platelets are probably not being consumed by some kind of coagulative intravascular disseminated process (let's call it DIC).

4) the D dimer is raised, which may or may not have meaning in the VV ECMO patient. 

So: of the many diverse causes of thrombocytopenia, this one would ideally have to be one which is not DIC, is compatible with a normal liver, and which is in a patient probably on heparin. 

These constraints cause the range of possibiities to contract somewhat. For maximum points, the candidates were expected to produce their list "correctly noted and prioritised",  which could look like this:

  • Consumption: most likely, because:
    • the ECMO circuit has tendency to grind platelets, and this is especially so in VV-ECMO, where the flow rate often has to be quite high to contribute enough oxygenated blood (Jirintano et al, 2020)
    • the ECMO circuit is anticoagulated, the heparin can give rise to HITS, and the timeframe (eight days) is correct.
  • Autoimmune destruction unrelated to heparin (i.e. ITP or TTP) needs to also be ruled out, but there is no specific evidence to support this in the stem.
  • Decreased production: less likely, because where in the stem is there any evidence of bone marrow suppression, but this is something that can happen during the sort of critical illness that lands you on VV-ECMO
  • Dilution: possible, as the bleeding risks in this population are high, but the stem does not mention anything about any recent massive transfusions, and moreover there is no specific investigation to 
  • Sequestration: even less likely, as there should be no reason for the liver or spleen to be doing this.

Four investigations, therefore, would have to include:

  • ELISA for anti-PF4 , to test for HITS
  • Platelet aggregation tests (serotonin release assay)
  • Blood film (looking for other features of intravascular haemolysis that would suggest that the circuit is eating the blood components)
  • The college examiners seemed to regard a bone marrow aspirate and trephine (BMAT) as an appropriate investigation here, which suggests  that anything equally implausible (eg. an ADAMTS-13 level, or antiphospholipid antibodies) would have also been accepted.


Jiritano, Federica, et al. "Platelets and extra-corporeal membrane oxygenation in adult patients: a systematic review and meta-analysis." Intensive care medicine 46 (2020): 1154-1169.

Warkentin, Theodore E., et al. "The platelet serotonin‐release assay.American journal of hematology 90.6 (2015): 564-572.