Coagulation answers

Question 1.

Consult the BCSH guidelines on peri-operative bridging (October 2016) which offer suggestions on management rather than strict prescriptions.

In bullet point fashion (it may depend on local practice), however discontinuing rivaroxaban the day before should be entirely acceptable. We assume that there is no prior history of HIT in this scenario.

• The bridging plan will depend on
• The risk of ceasing anticoagulation
• The risk of bleeding
• The exact renal function

She is moderate to high risk for ceasing anticoagulation for risks of stroke, and the risk bleeding is moderate (especially since the procedure is being done laparoscopically).

• Day -5 onwards to Day -2 – recommend patient start moving time of taking rivaroxaban to morning dose if not already doing so, so that the patient is taking it at 8am in the morning.
• Day -1 – no rivaroxaban (so last dose will have been two days prior to surgery).
• Day 0 – day of surgery, no rivaroxaban. Give prophylactic dose of LMWH (eg 5000units dalteparin) 6 hours after surgery after haemostasis has been achieved
• Day 1 – prophylactic dose of LMWH in the evening
• Day 2 – reintroduce rivaroxaban at usual dose assuming renal function is still unchanged and discontinue LMWH, or give the option to the surgical team of giving an additional day of just prophylactic LMWH if the patient is high risk for bleeding.

Reiterate to the surgical team that they should provide their contact details for the patient in case there is an issue with DVT symptoms – the patient should attend and if a DVT is suspected the patient should change to LMWH and discontinue rivaroxaban. The patient should continue on the same dose of anticoagulation as previously and monitor for symptoms of fatigue and bleeding from wound sites.

Question 2.

The key to answering the question is whether you know about the pharmacokinetics and half-life of DOACs, and their relationship to the abnormalities present on the clotting screen, and demonstrate an appreciation that different thromboplastins will offer different PT results. DOACs in general because of their anti-Xa activity will prolong the times of PT typically, and APTT to a lesser extent, but the effect is non-linear and highly variable, and is also very much dependent on the reagents used in the lab.

Apixaban maximal concentration is achieved after 1 hour post ingestion and the half-life is approximately 12 hours (rivaroxaban being 5-13 hours). Total elimination of the drug from the body is about 4-5 half-lives so we would expect complete resolution within 48 hours to be certain, although since we do not know the renal function of this patient it is uncertain what the half-life will be in this case. At normal renal function in healthy subjects, approximately 50% of the dose remains at 12hours and <25% at 24 hours (Frost et al., Brit J Clin Pharmacol 2013; 75:476-87).

The likelihood of being able to obtain an apixaban level by means of an anti-Xa chromogenic assay (properly calibrated for interpretation of apixaban) or another assay such as an immunological assay, under the conditions of the vignette are slim. Even if the results of the clotting screen were normal one would still be suspicious of continuing presence of apixaban (see J Douxfils et al. Thrombosis and Haemostasis 2013; 110:283-94).

Because a) the coagulation screen is abnormal and b) we are unaware of the elimination of apixaban in this patient, it would be appropriate to offer reversal of apixaban in this case using PCC. BCSH guidelines only suggest that PCC might be effective – a more recent publication in Blood (Majeed et al., Blood 2017; 130:1706-1712) suggests that reversal with PCC is effective in 69% of cases, with a median dose of 25U/kg used, and 2/84 patients suffering an ischaemic stroke post PCC. Striking a balance of the dose may be important here as there is a small chance of thrombosis in the region of about 2% when PCC is used to reverse warfarin (Dentali et al., Blood Coag Fibrin Cell Haemost 2011; 106:429-438), although most units use a 50U/kg as standard in an intracranial haemorrhage setting.

Dialysis is not an option here as apixaban is largely protein bound, and andexanet is not at present available for routine use.

Suggested topics for revision:

• DOAC dosing and administration, half-life and reversal
• Routine bridging plans and renal impairment/age criteria for administration
• Basic understanding of effects on coagulation screens and measurement

Question 3.

a) There are several things to comment on – a prolonged APTT, slightly elevated vWF/FVIII levels (which might be because of the malignancy), and more subtly the FXI level which is slightly low – note the range for abnormal FXI is <0.7IU/mL, and not 0.5IU/mL as with the other factors. Note the ranges for this were deliberately not included in this question.

However, the level of the FXI is not sufficient to explain that degree of prolongation of the APTT (and in fact may well have been normal). The probability is that you have a lupus anticoagulant for this patient – note that a partial correction is possible with a 50:50 mix. You could suggest two options here in terms of tests – if available, to retest the APTT using a lupus-insensitive reagent or request a dRVVT to confirm the presence of a lupus anticoagulant.

Also don’t forget to repeat the FXI level – to confirm that it is in fact a borderline result.

b) The key here is to recognise that the level of the FXI is inconsequential here especially in the context of the clinical history which gives you a previous history of dental extractions which have been unremarkable. In the context of a near normal FXI level and absence of bleeding history, it would be safe to offer standard thromboprophylaxis. The antiphospholipid antibody positivity is again of no consequence here in the absence of any VTE history and therefore standard thromboprophylaxis as per standard urological guidelines would be appropriate.

c) The reason is that the patient is now on LMWH which will prolong the results of the tests. Possibly being in a renal unit they are using tinzaparin which has relatively more anti-IIa activity compared to the other LMWHs (such as dalteparin and enoxaparin, which has the highest specific anti-Xa activity). Correspondingly, it is the anti-IIa effect which prolongs the APTT more proportionately (see O Thomas et al., PLoS One 2015; 10:e0116835, figures 1 and 2).

d) In a simple case of a provoked DVT (which this clearly is), for an above knee DVT the treatment time would be three months and discontinuing. However, there is the complication of the malignancy and we are not told whether there is metastatic spread, and the additional issue of the positive lupus anticoagulant comes into play. If there is a good justification the duration and rationale should be acceptable.

On balance, if the tumour is completely resected and since we have only one positive lupus anticoagulant, it would be reasonable to call this provoked and anticoagulate for three months with warfarin (or a DOAC), then bring her back to clinic 6 weeks after discontinuation for a D-dimer and repeat lupus anticoagulant screen to decide on whether to further anticoagulate her.

If still suggestion of active malignancy, currently DOACs have not yet been licensed (as of 2017) for cancer VTE, although there has been a trial comparison between edoxaban and LMWH which demonstrates non-inferiority of a DOAC. Currently LMWH remains the licensed treatment of choice in cancer associated VTE.

Suggested topics for revision:

• Reasons for particular choices of LMWHs between the three main options (dalteparin, tinzaparin, enoxaparin) – consider obesity, renal function, ease of dosing
• Understand the appreciation between lupus sensitive and lupus insensitive reagents
• Briefly run through normal coagulation ranges for factor assays and the features of FXI deficiency.

Question 4.

a) 'Mild haemophilia' could mean a number of things, and since he has just arrived with his first presentation in hospital in the UK it is unlikely that we will have any results to hand. The probability is that he has Haemophilia A (but don't forget Haemophilia B which is also another cause, as is FXI deficiency which used to be known as Haemophilia C). The presumption is that most centres do 1-stage FVIII assays and may incorrectly report mild or moderate haemophilia A result as higher than expected, which certain missense mutations may do (in severe haemophilia there is normally little discrepancy expected between 1-stage and chromogenic assays).

Since the APTT is a one-stage assay, again the normal APTT is expected to reflect that of the 1-stage FVIII assay.

If there were five tests then potentially you could include FXIII but there is likely to be a history of this in the vignette, because if patients had FXIII deficiency they would have either had some complications around the peridelivery period or that they may have prophylactic FXIII replacement.

We also need to exclude vWD as a possibility so the four tests we would request are:

Chromogenic FVIII (2-stage assays are no longer performed), FIX, FXI (if only three tests were allowed I would remove this one) and vWF activity (either latex or RCo would be acceptable).

b) The results show that the FVIII level in a chromogenic assay confirms mild haemophilia A - and with other factors being normal (including von Willebrand factor), a FVIII concentrate is appropriate in this instance (recommend Advate or Refacto AF as the current first-line, third generation concentrates).

The dose would be to raise the level to 1IU/mL so it would be:

1IU/ml = 100IU/dL

0.2IU/ml = 20IU/dL

In IU/dL: (target level - actual level) x patient's weight (kg) / 2

Thus if the patient is 70kg, then (100-20) x 70kg /2) = (80 x 70)/2 = 2800units (or 40units/kg)

(Recall that 1 unit FVIII raises the endogenous FVIII level by 0.02IU/mL (or 2IU/dL); and that 1 unit FIX raises the endogenous FIX level by 0.01IU/mL (or 1IU/dL).) So if this were FIX concentrate, you would use 5600 units instead.

If the clinical scenario differed slightly (with the clinical history including a sister who is affected, for example) and a relative lack of 1-stage vs chromogenic discrepancy, then the consideration of a type 2N vWD would arise, which is a vWF abnormality of the binding region of vWF with FVIII, which on its own will be relatively unstable. Tests for type 2N vWD require vWF:FVIII binding studies. Haemophilia A/B are X-linked inheritance patterns, while type 2N vWD is autosomal recessive.

Suggested topics for revision:

• Appreciation of 1-stage vs chromogenic assay discrepancies in certain haemophilia patients
• Be aware of the distinctions in diagnosis between Haemophilia A and type 2N vWD

Question 5.

This question is slightly obscure - but it tests whether you have assessed the situation thoughtfully and given appropriate advice in the context of the patient according to the vignette.

a)The differentials with the FBC could be simply reactive (raised WCC/Plt) post operatively, but the clue is that the platelet count was previously raised two years and it was presumably not acted upon. It’s possible he may also be losing blood at this point with a slight anaemia so you need to keep an eye on this and discuss with the surgical team. Simple things to check would be his haematinics, along with a von Willebrand screen (antigen and activity plus multimers), and possibly including a PFA-100 if this was available to screen for platelet function disorders. We don’t have any information about his past history to suggest a bleeding tendency.

b) This is likely to be an acquired von Willebrand syndrome secondary to a myeloproliferative neoplasm (with elevated platelet count).

c) The management plan for this man is tricky. This appears to be a thrombocythaemia which is probably mediated by a myeloproliferative disorder, although we don’t yet have the confirmatory genetic analyses (ie JAK2V617F/exon 12 mutation, MPL or CALR results), but would be suggestive with the platelet count, blood film result and the von Willebrand screen, especially in the context of someone who has previously had an elevated platelet count about two years ago – and can’t be put down solely to an acute phase reaction.

Cytoreductive therapy in the context of recent surgery is contraindicated, but will be useful in the medium term when his surgical wound is healed. The immediate situation would be to offer vWF concentrate calculated to the activity at 30-100IU/kg, and monitoring the vWF levels plus FVIII to ensure that you are not overshooting with the FVIII and subsequently causing thrombosis (depending on the concentrate that you use, which may vary in the FVIII quantity). Tranexamic acid should also be used, and also to discontinue any LMWH prophylaxis, while noting that the risk of thrombosis in this post-op patient is high (post operative and probable MPN, even in the context of an aVWS).

At a later stage hydroxycarbamide or other treatments such as pegylated interferon may be useful, as well as staging investigations such as a bone marrow and abdominal ultrasound as per routine myeloproliferative neoplasm workup. In lymphoproliferative or plasma cell dyscrasias the treatment of the condition should also be taken into account – IVIg is more effective in the situation with IgG paraproteins although caution should be taken in the context of high paraprotein levels and concern about hyperviscosity.

Suggested topics of revision:

• von Willebrand disease diagnostic criteria and management
• acquired von Willebrand syndrome with diagnosis and management (Tiede A. Diagnosis and treatment of acquired von Willebrand syndrome. Thromb Res 2012; S2-S6.)

Question 6.

This is probably an exceptional case that will not appear in the exam, but will demonstrate the thinking process required and steer your reading and revision for the exam.

a) The FBC is largely normal, however the PT and APTT are prolonged, with a low fibrinogen. All these point towards a hypofibrinogenaemia/dysfibrinogenaemia (which may have a bleeding phenotype or thrombotic phenotype).

- suggested investigations might include fibrinogen antigen, and clot weight by gravimetric assays (but this is time consuming and not necessarily available). Fibrinogen antigen will help determine the subtype of dysfibrinogenaemia. Typically normal antigen levels with reduced activity are those which are associated with higher levels of thrombosis.

b) Until the delivery period, she needs to be on anticoagulation with a LMWH of your choice (twice daily in split dosing). This dose can be based on an early pregnancy weight, and increased as the gestation progresses and pregnancy weight increases. At extremes of weight (ie <50kg or >90kg), antiXa monitoring is recommended, or for those whom bleeding risk is higher, which it may be in this lady. In terms of fibrinogen replacement this will depend somewhat on family history and other bleeding/thrombotic issues: fibrinogen replacement will be indicated more if the patient has recurrent miscarriages and otherwise complicated history. For a case of low fibrinogen and thrombosis, a risk/benefit decision may be to aim for a fibrinogen level of >0.6-1 or 1-1.5g/L throughout the pregnancy depending on clinical history.

Recommended levels at delivery/LSCS: fibrinogen 1g/L - offer 20-30mg/kg or cryoprecipitate 1pool/10kg, and recheck levels after replacement, aiming not to exceed 1.5g/L in view of the thrombotic history.

Timings will need to be carefully considered, especially if the patient is wishing to have a normal vaginal delivery which will make planning of anaesthetic considerations (eg spinal/epidural anaesthesia) difficult, as will the giving of fibrinogen concentrate at the right time. Caesarean section would be the ideal situation for the planning but the patient's wishes should be taken into account for a normal vaginal delivery as far as possible, and induction of labour would definitely be suitable. Early planning in advance of delivery is essential in conjunction with the anaesthetists and obstetricians, especially the consideration of neuraxial anaesthesia in this instance if there is any evidence of a bleeding phenotype in this patient.

At onset of labour, IoL or prior to LSCS: fibrinogen concentrate given to boost fibrinogen to 1g/L, and recheck every 12 hours during labour/delivery until one day post delivery, keeping a tight range

If LSCS: LMWH split dose - last dose 24h prior to LSCS.

A spontaneous vaginal delivery would still be possible but patient should be advised that neuraxial anaesthesia may not be feasible depending on when the last dose of LMWH was administered. The patient should stop LMWH injections at the time of onset of labour, and if found to be false contractions, then she should resume taking these.

If neuraxial anaesthesia is used, then post partum thromboprophylaxis should be given 4-6h after removal of the epidural catheter

Post delivery the anticoagulation should continue for six weeks, or until at least six months after the initial event. Resumption of full dose anticoagulation should be at the discretion of the obstetrician as to when the bleeding risk has settled and it is safe to resume as soon as possible.

Suggested reading:

Khalil A et al. Systemic thromboembolism in pregnancy: venous thromboembolism. In: Cohen H, O'Brien P (Eds). Disorders of thrombosis and haemostats in pregnancy. Springer. pp51-70.

Bournikova L. Fibrinogen replacement therapy for congenital fibrinogen deficiency. JTH 2011; 1687-1704.

Question 7.

a) This is a bit of a gift question so you can write plenty here, but don't get bogged down in detailing the all the minutiae. The examiners appreciate you have eight questions to answer!

I suggest you review the Practical Haemostasis website, specifically the page on HIT assays. However, primarily the way your lab is likely to diagnose this is two methods:

- latex based assay (like the gel cards used in transfusion for forward and reverse grouping.

- PF4 ELISA assay - which is more sensitive with a low specificity. This is also very time consuming as ELISAs often take at least two hours, if not an afternoon or morning.

- platelet aggregation assay (but not especially sensitive)

Rarely used now but is the gold standard: Serotonin release assay.

b) The key here is you can use whichever agent that you find most comfortable with and you should get full marks. Be careful with the use of fondaparinux in this question because of the renal function here, its half life (about 17-21h) - and that it is NOT licensed for use in HIT. DOACs in theory are also possible but again these are not licensed for use in HIT.

Options are:

- thrombin inhibitors (eg Argatroban, lepirudin, bivalirudin)

- antiXa inhibitors (danaparoid)

If the patient was already on warfarin then it would be discontinued in the same manner, preferably with no bridging.

Admit the patient - commence the infusion of choice and monitor APTR/Xa to desired levels

Ensure allergies are documented to heparin on the drug chart and his wristband

Discontinue the infusion according to the half life and discuss with the vascular team about what levels they would be happy to

Recommence the infusion when the surgical team are happy that there will be no further bleeding complications.

While this is not a prescriptive bridging plan, a reasonable choice would be argatroban in this situation as the half life in patients with normal renal function is about 50-60mins and generally provides a stable APTR when started. Danaparoid has a long half life of 25h so this might be more difficult to use in the perioperative setting.

Question 8.

The question might be pointing towards mucocutaneous bleeding but in someone so young phenotypically it is difficult to tell. A useful guide is that with an APTT that is normal the patient is unlikely to have a significant bleeding tendency but in this age group all possible diagnoses should be considered.

a) Platelet aggregation studies have already been selected below so you are asked about ADDITIONAL tests to request (so don't write platelet ags/nucleotides otherwise you will not get a mark!)

if these are negative then the tests which give you normal clotting and platelet count is normal: FXIII levels, alpha2 antiplasmin deficiency or PAI-1 deficiency. These are exceptionally rare but would be useful exam answers in this setting.

Alternatively if all these are negative then you would test all the extrinsic and intrinsic factors (taking into account the age of the patient, ethnicity and any history of consanguinity).

- also consider non-accidental injury as a possible cause in consultation with the paediatricians.

b) You need to learn pattern recognition with the platelet aggregation studies, as you would with HPLC plots for haemoglobinopathy screening and immunophenotyping for malignant disorders. If you find this difficult, consider that you need to learn a lot more immunophenotyping compared with 5-6 patterns in platelet aggregation!

In general:

Bernard-Soulier: relatively normal traces apart from lack of response to high dose ristocetin (will also come with macrothrombocytopenia)

Glanzmann's: Absence of response to all plt aggregation apart from to high dose ristocetin

Storage pool/release defect disorder: lack of secondary aggregation (primary aggregation is possible but without the release/presence of platelet granules they cannot complete proper aggregation). Thus need to look at platelet nucleotide ATP:ADP ratios.

VWD type 2B/platelet type VWD: increased response to ristocetin (gain of function mutation)

Aspirin defect: absent response to arachidonic acid, with lack of response to collagen

P2Y12 inhibitor defect: lack of response to ADP.

The diagnosis here is a storage pool disorder.

c) Recommendations for when the child requires dental extractions:

- patient should have his procedure done not at the dentist but at a dental hospital with appropriate transfusion cover

- HLA typing should be done in advance as a basis of record

- vaccinate against hepatitis B

- tranexamic acid

- local surgical measures such as topical TXA and splints

- platelets on standby from single donor (often not required as routine for storage pool disorder patients unless they are severe phenotypically). HLA typing is more crucial in patients with BSS (who have absent or defective GPI-IX-V) or GT (where they have defective or absent GPIIb-IIIa on their platelets).

For further reading:

Seligsohn U. Treatment of inherited platelet disorders. Haemophilia 2012; 18 (S4):161-5.