Coagulopathy of major haemorrhage

In all clinical situations which lead to major haemorrhage there are several causes of worsening coagulation, and these include:

• •
  

  Consumption. Coagulation factors and platelets are consumed during the formation of clots. This happens to prevent loss of blood through damaged vessels.

  
  
• •
  

  Dilution. This is a consequence of replacement of the whole blood that is lost with fluids that do not contain plasma, e.g. crystalloid, colloid and red cell transfusion. Coagulation factors are therefore ‘diluted’ and the levels of clotting factors fall rapidly. High-ratio fresh frozen plasma:red blood cells (FFP:RBC) therapy aims to address this effect.

  
  
• •
  

  Endothelial cell activation. In health the endothelium plays an important role in maintaining blood flow. However, in the setting of significant vascular injury and hypotensive shock, and the resultant increases in vasoactive hormones (e.g. epinephrine), inflammatory cytokines and thrombin, lead to marked and often sustained endothelial cell activation. Initially there is a rapid and immediate release of tissue plasminogen activator (t-PA) from endothelial cells causing fibrinolysis and an increased risk of bleeding and then, over time, as endothelial cell activation continues, there is a shift to a more prothrombotic cell phenotype, increasing subsequent risks of small and large vessel thrombosis.

  
  
• •
  

  Hypoxia, acidosis, hypothermia. This triad predisposes to further bleeding. Hypothermia and acidosis impair the functional ability of both the platelets and coagulation proteases. Haemostatic defects are most evident once the pH falls below 7.1 and body temperature falls below 33 °C.

  
  
• •
  

  Hypocalcaemia. Calcium ions play an important role in the initiation and maintenance of the coagulation cascade. Large volume transfusions increase the risks of hypocalcaemia due to the calcium-chelating effects of the citrate used in blood components.

  
  
• •
  

  Ongoing bleeding. Anaemia has a major effect on primary haemostasis. A low haematocrit reduces axial flow (normally the red cells flow in the middle of a vessel and platelets are pushed towards the endothelium). An inverse relationship has been demonstrated between the haematocrit and in vitro bleeding time. (The bleeding time is a test no longer used in haematology, but which evaluated primary haemostasis, i.e. the haemostatic capacity of platelets and von Willebrand factor (VWF)).

Trauma coagulopathy

In addition to the clotting changes that accompany all forms of major bleeding, there are changes that are thought to be specific to certain patient groups. Trauma-induced coagulopathy (TIC) is, at present, the most thoroughly understood. It is more commonly seen in association with major trauma and hypoperfusion from shock, and is a dynamic process characterized by an early state of hypocoagulability caused by i) degradation of the endothelial glycocalyx, ii) excessive fibrinolysis from a combination of endothelial cell activation and resultant t-PA release, iii) platelet dysfunction, and iv) hypofibrinogenaemia ((likely due to a combination of consumption, fibrinogenolysis, fibrinolysis and reduced production). The importance of fibrinolysis (which causes rapid clot breakdown) in TIC has been known for several years and was confirmed by the CRASH-2 study. This randomized controlled trial (RCT), of over 20,000 trauma patients, reported reduced mortality from all causes and from bleeding in the active arm where participants received tranexamic (TXA). Survival was most improved in those who received TXA early, within 3 hours of injury.

Following resolution of haemorrhage, patients with early TIC and who survive, are at risk of transitioning into late TIC (in the first 24 hours after injury) characterized by a prothrombotic state. This is caused by the evolving changes to the endothelium which leads to an increase in PAI-1 (plasminogen activator inhibitor-1 – a potent serine protease inhibitor that directly inhibits t-PA) which suppresses fibrinolysis. This results in both macrothrombotic complications like venous thromboembolism (VTE) and microthrombosis and organ dysfunction. Consideration of early VTE prophylaxis in this group of patients, when haemostasis has been achieved, is vital.

Post-partum coagulopathy

Other patient groups with major haemorrhage have not been as extensively evaluated, although data are emerging. For this reason, it is important to recognize that transfusion strategies such as empiric high-ratio FFP:RBC which have been shown to benefit trauma patients may not be the optimal therapy for other patient groups. More research is required to unpick differences (or indeed similarities) between, and inform transfusion strategies for, different patient groups.

The coagulation profiles of pregnant women at term differ from those of other adults – with a shift towards a prothrombotic state. Thrombin generation rises markedly throughout gestation, secondary to the increase in nearly all the coagulation proteases (except factors V, XI and XIII) and the concomitant reduction of anticoagulants such as protein S. Fibrinogen, VWF and factor VIII levels rise the most, with the latter contributing to the acquired activated protein C resistance and the shortened aPTT seen by term. The average fibrinogen level by the end of pregnancy is 4–6 g/litre (normal range for adults: 2–4 g/litre) and for this reason viscoelastic tests (such as thromboelastography (TEG) and rotational thromboelastometry (ROTEM)) show a global increase in clot strength parameters. It is important to recognize this difference, since a normal viscoelastic trace at term will be markedly different from a normal trace for a non-pregnant patient.

These increases in pro-coagulant factors offer patients a large physiological buffer in the event of post-partum haemorrhage (PPH) which means that factor levels, and thrombin generation, are maintained in the majority of PPH cases. This is different to TIC and as such early FFP replacement is not commonly warranted in most PPH cases. One of the most important parameters to recognize during PPH is the presence (or absence) of a low fibrinogen. Post-partum patients with bleeding appear to divide themselves into two groups, either by maintaining a relatively normal fibrinogen or a small proportion who develop a rapid and markedly low fibrinogen. ^, Recently, detailed coagulation analysis of patient with PPH has shown that around 1 in every 1000 patients develop severe hypofibrinogenaemia, as part of an ‘Acute Obstetric Coagulopathy’ (AOC) syndrome. AOC is characterized by low fibrinogen, dysfibrinogenaemia (impaired fibrinogen function), and marked fibrinolysis, and is associated with specific obstetric conditions e.g. placental abruption, and amniotic fluid embolism. Early identification of low fibrinogen (<2 g/litre) and therapeutic replacement is key.

Regardless of the underlying cause of PPH, the use of tranexamic acid has been shown to improve outcomes. The WOMAN trial reported on just over 18,000 women randomized to receive TXA or placebo during PPH. Like the CRASH-2 trial, this study reported a reduction in mortality from bleeding in the TXA arm, with most benefit seen early (within 3 hours of bleed onset). The dose of tranexamic acid used in the WOMAN trial is 1 g by intravenous bolus as soon as possible; a second dose is given if bleeding persists after 30 minutes or recurs within the first 24 hours.

Predicting major haemorrhage

Diagnosis of major bleeding is difficult and is often made using clinical measures (e.g. shock index incorporating heart rate and blood pressure) but these measures can be insensitive, particularly in younger patients in whom blood loss can be masked and haemodynamic stability preserved. Detection and correction of coagulopathy therefore is an important aspect of management of severe haemorrhage. British Society for Haematology (BSH) guidelines recommend the use of serial standard laboratory tests (SLTs) taken every 30–60 minutes to monitor major haemorrhage. European trauma guidelines recommend viscoelastic haemostatic assay (VHA) testing, i.e. TEG and ROTEM.

PT and aPTT: although prolonged PT and aPTT results are clearly associated with poorer clinical outcomes in major bleeding these tests are poor predictors of transfusion requirement across all specialty groups. For example, in a cohort of 300 trauma patients 60% of those requiring massive transfusion had a normal PTr (prothrombin time ratio). Therefore, the ability of SLTs to guide pre-emptive treatment is limited. However, the value of these tests is still evident when used as a repeated measurement to show serial trends in coagulopathy.

Fibrinogen is a better predictor of bleeding in PPH than PT/aPTT in the early stages of bleed onset. A large observational study showed that a fibrinogen level of 2 g/litre gave a 100% positive predictive value for severe (vs non-severe) PPH, and for each 1 g/litre decrease in fibrinogen the risk for severe PPH increased by 2.63-fold. In trauma, patients with severe bleeding do have lower fibrinogen levels than their non-bleeding counterparts, however, there is not such a clear predictive effect for bleeding.

VHA tests have also not been shown to be able to predict future bleeding if a test is taken before bleeding starts in an at-risk patient. However, in both PPH and trauma, low clot strength measures (and in trauma an increased TEG lysis of 3%) indicate that a patient is at higher risk of bleeding. An EXTEM measure at 5 minutes (EXTEM CA5) of 35–40 mm has been shown to be more sensitive than a PTr of 1.2 for predicting the need for massive transfusion in trauma. The advantage of many of the VHA measures over SLTs is speed of results and that they can be undertaken near to the patient – the devices can provide clinically useful clot strength measures at 5 minutes and 10 minutes from test start. Similarly to SLTs, serial measurements hold more clinical value than standalone tests.

Treating coagulopathy during major haemorrhage

The aim of treatment for major haemorrhage is to stop the bleeding. Although coagulopathy confers a negative impact on outcomes during bleeding, the inverse (i.e. full reversal of coagulopathy by normalizing PT and aPTT) has not been definitively shown to improve outcomes.

SLT thresholds for transfusion treatment are generally based on historical, not evidence-based, strategies – i.e. ‘give FFP when the PTr is 1.5× the upper limit of normal (ULN)’ or ‘give cryoprecipitate when the fibrinogen is 1.0 g/litre’. These thresholds have been challenged recently and certainly for trauma haemorrhage empiric transfusion therapy, which does not rely on SLTs, has become standard of care.

However, the era for empiric transfusion therapy being the optimal method for treating major trauma haemorrhage has come to an end. Several trauma RCTs, e.g. PROPPR, COMBAT, and CRYOSTAT-2, have failed to show improved mortality when using an empiric approach. These studies used clinical gestalt for study entry, without an additional coagulation measure to either confirm coagulopathy or guide transfusion therapy. The negative results have led researchers towards investigating a more individualized approach – using SLT or VHA measures. For example, the FEISTY2 study is comparing cryoprecipitate versus fibrinogen concentrate in trauma patients with low fibrinogen (defined as: ROTEM FIBTEM A5 ≤ 10 mm or TEG FF A10 ≤ 15 mm or FibC ≤ 2 g/litre).

VHA-guided transfusion therapy for major bleeding is often used but, outside cardiac and liver transplant surgery settings, there is currently insufficient evidence for the UK National Institute for Health and Care Excellence (NICE) to support their use. However, RCT data are emerging which provide support for the use of VHA-guided transfusion thresholds. In an RCT of women with moderate to severe PPH, a FIBTEM A5 >12 mm indicated a fibrinogen level that was adequate for haemostasis and no supplementation was needed. In trauma, a recent large multi-centre RCT (iTACTIC) comparing clinical outcomes in patients treated using VHA- and SLT-guided transfusion algorithms, concluded that VHA testing is able to identify coagulation issues rapidly and direct therapy sooner than SLT alone but this does not translate to a reduction in mortality or reduced requirement for blood component in this study. Although the primary endpoint was negative for the whole cohort, there was a difference in mortality in brain-injured patients – one of several pre-defined subgroups, and this will need further evaluation. These data show a lack of superiority of VHA over SLT but do demonstrate in a large clinical trial that VHA can be used to guide transfusion therapy in this critically ill patient population.

In summary – most clotting tests, whether they are SLT or VHA based, are poor predictors of future bleeding in at-risk patients. Both tests are most useful when taken multiple times and compared serially. If VHA tests are used in your hospital, they should be used within a well-tested and validated trans-fusion algorithm. Transfusion therapy should continue to be given according to algorithms until bleeding has stopped. Once haemostasis has been achieved, transfusion therapy should not be used simply to correct deranged clotting test results.

Vitamin K antagonists (VKA)

Warfarin is the most used VKA in the UK. It acts by inhibiting the vitamin K dependent production of several clotting factors (namely factors II, VII, IX, X as well as proteins C, S and Z). Warfarin is measured using the PT and the result is standardized by the international normalized ratio (INR). There is a strong correlation between INR and bleeding (or thrombotic) risk. The therapeutic window for warfarin is narrow and regular monitoring is necessary. The half-life of warfarin is approximately 36 hours, which means that the effects of warfarin will be seen generally for about 5 days after the last dose is taken.

When faced with a patient with critical bleeding who is known to be taking warfarin, it is important to reverse the anticoagulant effect rapidly. Outcomes for trauma patients are known to be worse in patients taking warfarin. Reversal of warfarin should be conducted using combined therapy, namely vitamin K and pro-thrombin complex concentrate (PCC). PCC immediately replaces the missing/low clotting factors; however, the clotting factors in PCC have a shorter half-life than the effects of the warfarin and IV vitamin K must also be given. Intravenous vitamin K has its peak effect approximately 6 hours after administration, just at the point when PCC is beginning to lose its effect.

How to reverse warfarin: patients with major bleeding on warfarin need reversal of the anticoagulant effect, regardless of INR. Vitamin K can be given at a dose of 5–10 mg IV and the PCC dose should be rounded to the nearest complete vial. For ease, at our hospital we offer weight band dosing: patients weighing <60 kg get 1500 units PCC; 60–75 kg require 2000 units; 76–90 kg require 2500 units and those >90 kg require 3000 units. The dose of PCC is capped at 3000 units. The INR should be rechecked immediately after administration of PCC and again at 6 hours.

In patients with non-major bleeding events, reversal of warfarin often requires vitamin K only and dose depends on the patient’s INR. For an INR of 5, warfarin should be omitted and 1–3 mg IV vitamin K given. For an INR <5, consider IV vitamin K 1–3 mg (depending on clinical situation) and modify the warfarin dose. INR needs to be rechecked in 6 hours after administration of IV vitamin K.

Perioperative management of warfarin: perioperative anti-coagulation management depends on the balance between bleeding and thrombosis and largely depends on the bleeding risk of the procedure and the thrombotic risk of the patient.

For elective surgery, warfarin should be held for 5 days if reversal of anticoagulation is needed. INR should be repeated the day before surgery and vitamin K given if the INR is 1.5. A further INR needs to be repeated on the morning of the surgery in these instances. Warfarin can be restarted on the day or one day post operation with the patient’s normal maintenance dose if haemostasis is secured.

Bridging with full treatment dose heparin is only required in patients with high thrombotic risk (consider the following patient groups for this: VTE within previous 3 months or a complex VTE patient with a high target INR; recent stroke/transient ischaemic attack (TIA) within previous 3 months or multiple risks and atrial fibrillation; or mechanical heart valve). Patients who require bridging with treatment dose heparin usually start low-molecular-weight heparin (LMWH) on the morning of day 3 before surgery and LMWH is continued until 24 hours before surgery. Full-dose anticoagulation should be withheld for the first 24 hours in low bleeding risk surgery and 48 hours in a high bleeding risk surgery. Do use prophylactic doses of LMWH both for prevention of hospital-acquired thrombosis and reducing the patient’s overall thrombotic risk, according to your local hospital policy.

DOACs

DOACs that are currently used in clinical practice include anti-Xa drugs such as: apixaban, rivaroxaban and edoxaban; and the direct anti-thrombin agent dabigatran. In patients experiencing major bleeding or life-threatening bleeding while taking a DOAC the following management is advised. Stop the DOAC. Manage the bleeding according to your local major haemorrhage policy, including the use of tranexamic acid. Take a full coagulation screen including PT, aPTT, fibrinogen and thrombin time. A drug-specific test can also be taken (i.e. a ‘rivaroxaban level’), but not all hospitals are able to offer this service. Do remember that a patient can have significant anticoagulant on board and have normal clotting screens, especially apixaban. Standard ROTEM tests (EXTEM/INTEM CT) can detect DOACs (dabigatran, edoxaban, rivaroxaban) at therapeutic levels, but appear insensitive to apixaban. More recently, DOASENSE, which is a urine DOAC dipstick test for detecting direct oral anticoagulants, has been approved by NICE. It can detect both factor Xa inhibitors (apixaban, edoxaban and rivaroxaban) and thrombin inhibitors (dabigatran) in urine.

Bleeding secondary to dabigatran can be reversed with idarucizumab (a direct anti-dabigatran monoclonal antibody that binds and clears dabigatran). Idarucizumab is given as two 2.5 g IV slow bolus injections. Andexanet alfa is a recombinant form of human factor Xa protein which binds specifically to anti-Xa DOACs such as apixaban or rivaroxaban, reversing their anticoagulation effects. The ANNEXA-4 study showed good clinical outcome in patients with acute major bleeding. Andexanet alfa is administered as an intravenous bolus at a target rate of 30 mg/min over 15 minutes (low dose) or 30 minutes (high dose), followed by administration of a continuous infusion of 4 mg/min (low dose) or 8 mg/minute (high dose) for 120 minutes. The decision to use low or high dose andexanet depends on the patient’s dose of DOAC and the timing of the last dose. Thrombotic complications were observed in 10% of patients in the trial, and so re-anticoagulation should be considered after andexanet treatment, once bleeding is secured.

Currently, andexanet is only licensed in England for life-threatening or uncontrolled GI bleeding. This is because there was no robust evidence to suggest that the use of andexanaet reduces long-term disability after intracerebral haemorrage, so more data are required. PCC (50 units/kg) is also commonly used in the UK to treat significant bleeding in the presence of an anti-Xa medication and is given principally to improve thrombin-generating capacity. There is no role for IV vitamin K in patients with bleeding secondary to DOACs. For non-major bleeding, delay the next dose of a DOAC or discontinue treatment according to the clinical situation. Intravenous or oral tranexamic acid 1 g can be considered.

Perioperative management of DOACs: due to their short half-lives, bridging with heparin is not used. Discontinuation of DOACs for elective procedures largely depends on a patient’s renal function. Dabigatran, apixaban, rivaroxaban and edoxaban should be held 24 hours before low bleeding risk procedures and 48 hours before for high-risk procedures in patients with normal renal function. In patients with poor renal function, the duration of discontinuation of a DOAC will need to be increased. DOACs can be restarted 24 hours after low bleeding risk procedures, if haemostasis is secured. For high-risk bleeding procedures, DOACs should not be restarted until at least 48 hours after the procedure/surgery. Remember, restarting a DOAC means that a patient will be fully anticoagulated within a few hours of administration of the drug. Their use should not be compared to warfarin, where even if this is started on the day of surgery, the INR is not therapeutic until 2–5 days later.

For emergency surgery, whenever possible, aim to delay surgery until plasma levels of DOACs fall. Remember the risk of bleeding perioperatively is dependent on the time since the last DOAC dose was taken, the type of surgery and the patient’s renal function. If an anticoagulant effect cannot be ruled out, neuraxial anaesthesia should be avoided. The ability to make predictions regarding haemostasis at surgery in patients taking DOACs is limited by the uncertainty in the concentration of each drug that is associated with haemostatic safety.

Practically, tranexamic acid can be used to reduce bleeding in most surgical cases. In patients taking dabigatran, the thrombin time (TT) is acutely sensitive to presence of drug and if it is normal there is likely to be only very low levels of drug on board. If the TT is prolonged, indicating effects of anticoagulant still present, idarucizumab can be used as a reversal agent where bleeding risk is significant. In patients taking anti-Xa DOACs who require emergency surgery, PCC may be used, but it must be remembered that this is not a direct antidote and may lead to an increased thrombotic risk. In some low-risk surgical procedures, it may be prudent to only use a dose of PCC if the patient shows signs of bleeding perioperatively, rather than using it prophylactically. This decision is usually best made with a haematologist and the operating surgeon.

Anti-platelets

Bleeding on antiplatelet therapy: bleeding in patients during treatment with aspirin, P2Y ₁₂ antagonists or GPIIa/IIIb inhibitors should be managed in the first instance with general haemostatic measures. Platelet transfusion (two to three adult doses) can be as given for critical bleeding or prevention of bleeding before emergency high-risk surgery. Although desmopressin can shorten bleeding time in patients exposed with anti-platelets such as aspirin and clopidogrel, safety concerns in patients with cardiovascular disease often prevents use. Use of platelet transfusion in patients with spontaneous cerebral haemorrhage on antiplatelet therapy is not routinely recommended. The PATCH study which is a multicentre randomized trial concluded that platelet transfusion is inferior to standard care for people taking antiplatelet therapy before intracerebral haemorrhage. Death and dependence at 3 months were higher in the platelet transfusion group.

For urgent surgery that imposes high bleeding risk in patients who have taken antiplatelet agents, preoperative IV tranexamic acid should be given. If, despite tranexamic acid, there is excessive perioperative or postoperative bleeding or if the bleeding risk is perceived to be very high, administer two pools of donor platelets as well. This may improve haemostasis if given at least 2 hours after the last dose of aspirin. For urgent low bleeding risk surgery in patients on antiplatelet therapy, routine platelet transfusion should not be given.

For elective procedures, aspirin generally can be continued for most invasive non-cardiac surgeries including neuraxial anaesthesia. However, if the risk of bleeding is high, aspirin can be withheld from day 3–7. In patients on dual anti-platelet agents, low-risk bleeding procedures may be conducted without interruption of antiplatelet treatment. For high bleeding risk procedures, elective surgery should be deferred. If deferral is not possible, aspirin should be continued and the second anti-platelet, i.e. clopidogrel or ticagrelor should be stopped 5 days prior to surgery (7 days for prasugrel).

Screening for IBD perioperatively

The best way to explore whether a patient might have an IBD preoperatively is to ask them about their personal bleeding his-tory and that of their family. Do take care with younger patients, particularly children and male teenagers/young adults as they are much less likely to have been exposed to prior haemostatic challenges (such as menstruation, surgery or childbirth). If a patient says that they have a bleeding history, then grading the severity is important and a history of: previously needing to seek medical attention for bleeding (i.e. epistaxis that will not stop); iron deficiency; troublesome bleeding at more than one site (i.e. nosebleeds and heavy menstrual bleeding) and needing hospital admission and transfusion are important flags. Haematologists tend to use a scoring system, most commonly the ISTH BAT (International Society for Haemostasis and Thrombosis Bleeding Assessment Tool) and the greater the score, the higher the chance the patient will bleed when challenged. An online version of the score can be accessed here: https://bleedingscore.certe.nl/

Clotting screens

The PT and aPTT are often taken preoperatively. A set of normal clotting screen results should not be taken to mean that the patient does not have a bleeding risk. A good history of personal and/or family history is much more specific, and weight should be placed on the patient’s history over and above a normal clotting screen.

Haemophilia, which is an X-linked disorder, has two forms – haemophilia A and haemophilia B (due to absent or low factor VIII or factor IX respectively). Haemophilia may be severe, moderate, or mild depending on the baseline clotting factor level of a patient: severe, <1%; moderate, 1–5%; and mild, >5%. Normal clotting factor levels are 50–150%. In all but the very mildest forms of haemophilia, the aPTT will be proportionately prolonged according to the factor VIII/IX level. However, when a factor level is greater than 30% the aPTT can be within the normal range and the patient may still have very mild haemophilia. In addition to this, some forms of mild haemophilia have a structural change that is not picked up by all clotting assays, such that an aPTT may be normal even when the functioning factor VIII level is low. This is known as a one-stage/two-stage clotting factor discrepancy. Here again, the history from the patient is important. Do not forget that carriers for haemophilia may also have a low clotting factor level and may require perioperative haemostatic support.

von Willebrand disease can be divided into three broad types, 1, 2 and 3. Type 1 is the most common and generally has the mildest phenotype and is due to quantitative reduction of normal VWF. Type 2 comes in four forms and is due to the production of qualitatively abnormal VWF such that normal amounts of VWF are made but put simply it does not work very well. Type 3 is the most severe form of von Willebrand disease with levels of VWF (and factor VIII) <5%.

In mild VWF and some forms of type 2 von Willebrand disease, the aPTT can be normal. (Remember – VWF levels do not influence the aPTT, it is the factor VIII level that does).

Management

Once the diagnosis of an IBD is recognized, close liaison with the local haemophilia centre is vital. Patients who have a diagnosis of IBD will have a ‘haemostasis/bleeding card’ and if you ask to see it, it will contain a brief description of their diagnosis, their regular treatment and how to contact their haemophilia centre. Sometimes, patients expect that non-haemophilia doctors should know of their bleeding condition and do not mention it, particularly if being treated in the same hospital. It is always worth asking the question!

In general, most patients with a diagnosis of an IBD will require some form of haemostatic support perioperatively. This varies according to the severity of their condition and the type of surgery required. Haemostatic therapy can broadly be thought of in three steps, as follows.

Tranexamic acid inhibits clot breakdown. It is a synthetic analogue of the amino acid lysine and reversibly binds to the lysine receptor sites on plasminogen. This prevents plasminogen from being converted to plasmin, a potent fibrinolytic enzyme. In general, tranexamic acid is a useful medicine for all patients with IBD and can be used alone in minor procedures or in combination with desmopressin (DDAVP) or factor concentrate for more significant surgery.

DDAVP: desmopressin is a synthetic form of vasopressin and works by stimulating the release of VWF from the Weibel-Palade bodies of endothelial cells. VWF levels (and factor VIII levels), in those patients who respond to DDAVP, increase by threefold to fivefold with the peak level expected 90 minutes after a subcutaneous injection. DDAVP is useful in some forms of von Willebrand disease and for mild haemophilia. The dose is much higher than the dose given for renal patients – i.e. 0.3 micrograms/kg and a specially formulated DDAVP is used for subcutaneous injection, called Octim.

Clotting factor concentrate: factor VIII or IX concentrate is given to haemophilia patients perioperatively, when required. In the UK, these factor concentrates are now recombinant. However, haemophilia patients were treated in the past with plasma products and some patients will have been exposed to variant Creutzfeldt–Jakob disease (vCJD). The patient’s haemophilia centre will hold information about the patient’s vCJD status.

von Willebrand disease is treated, when necessary, with VWF concentrate that contains both VWF and factor VIII which is a plasma derived product. Vonicog alfa or Veyvondi is a recombinant VWF is now available in the UK to be used in patients with von Willebrand disease via named patient access programme.

Non-factor therapies: there have been many recent advancements in haemophilia treatment. The most used of these new treatments is emizicumab, which is a bispecific monoclonal antibody with specificity for factors IXa and X. This drug thereby mimicks the normal action of factor VIII in vivo . The overall effect of this monoclonal antibody is to provide the equivalent functional factor VIII level of between 10% and 15%, thereby converting a patient with severe or moderate haemophilia A to a milder phenotype. It is important to recognize that as their functional factor VIII level remains low at around 10–15%, they are still prone to bleeding, and additional factor replacement is needed for overt bleeding episodes. Emicizumab may also affect standard coagulation assays (the aPTT will be shortened) therefore clinical history and information on diagnosis and treatment history from the bleeding card is especially important.

Neuraxial anaesthesia and IBD

In many cases, when a patient is treated perioperatively for their IBD, nerve blocks, epidural or spinal anaesthesia can be used. Care should be paid to the timing of the removal of an epidural – and again close liaison with the haemophilia centre is advised – as factor levels should be greater than 50% for removal. However, for von Willebrand disease, even when treated with VWF concentrate, the UK Haemophilia Centre Doctors’ Organisation (UKHCDO) advises against neuraxial anaesthesia, particularly for type 2 and 3 von Willebrand disease patients.

Duration of treatment

Duration of therapy will mostly be dictated by the type of surgery the patients has undergone. Treatment may therefore be a one-off (i.e. dental surgery) or may be required for 2 weeks (i.e. neurosurgery). IBD patients, particularly haemophilia patients, are at increased risk of late bleeding postoperatively. The haemophilia centre will provide a surgical plan and regular input into the patient’s care postoperatively.

The management of coagulation abnormalities in the context of bleeding can be challenging. Outside haemophilia and warfarin therapy, where the aPTT and the INR are sensitive diagnostic and therapeutic measures, SLT and VHA tests are generally not good predictive measures for risk of bleeding and should not be relied on. SLT and VHA measures are better when used serially to monitor a bleeding patient and future studies may provide more information about thresholds for transfusion and VHA use.