An emerging body of evidence relates lupus anticoagulant to COVID-19. Both of these topics are enormously confusing on their own, so let’s try to ease into this…
background on lupus anticoagulant
what is lupus anticoagulant?
- Lupus anticoagulant isn’t any single thing, but instead it refers to a variety of different antibodies which may behave differently in different patients. The defining feature of these antibodies is that they bind to phospholipids, causing interference in in vitro clotting tests that require phospholipid (most notoriously the aPTT).
- However, in vivo, these antibodies are often associated with hypercoagulability (via mechanisms which are poorly understood).
- So yeah – lupus anticoagulant makes the coagulation labs look like the patient is bleedy, but in reality the patient may be clotty.
testing for lupus anticoagulant
- This isn’t something which we will generally be doing for our patients. However, understanding this process is critical to interpreting emerging studies.
- Step #1: Lupus anticoagulant prolongs in vitro clotting reactions that involve phospholipid. However, no single test is sensitive for all of the different types of lupus anticoagulants. Therefore, testing for lupus anticoagulant typically begins with two different tests. Commonly used tests include the dilute Russell viper venom assay and a modified aPTT assay (designed to be more sensitive for lupus anticoagulant).
- Step #2: If prolongation in one or both of these assays is found, we need to differentiate between a clotting factor deficiency versus the presence of a clotting inhibitor. If the problem is a clotting factor deficiency, this will be resolved by adding normal serum to create a 50-50 mix (clotting factors are enzymes, so 50% of normal levels is enough to normalize the in vitro test). Alternatively, if there is an inhibitor present (such as lupus anticoagulant), then the addition of normal serum will not cause normalization.
- Step #3: Lupus anticoagulant binds to phospholipid. If we add large mounts of exogenous phospholipid, this will compete for lupus anticoagulant and effectively eliminate it. Thus, the ability of exogenous phospholipid to normalize clotting time confirms the presence of a lupus anticoagulant.
- Elevated C-reactive protein (CRP) may prolong aPTT reactions, mimicking lupus anticoagulant.1 However, this effect doesn’t seem to occur with the dilute Russell viper venom assay.2
different types of lupus anticoagulant
- Perhaps the most notable type of lupus anticoagulant is one which binds to beta-2-glycoprotein. This can be detected in ELISA assays which detect antibodies that bind to beta-2 glycoprotein and/or cardiolipin.
- Patients who are “double positive” for lupus anticoagulant plus antibodies against beta-2 glycoprotein may be particularly likely to form clots (moreso than patients with isolated lupus anticoagulant alone).
- Other lupus anticoagulants exist which bind to other targets (e.g., phosphatidylserine).
what is the clinical significance of lupus anticoagulant?
- This depends on the clinical context. For example, lupus anticoagulant can rarely be found in asymptomatic people.
- Viral infections can often cause a transient, infrequent production of lupus anticoagulant. This is of unclear significance, but in some situations these have been correlated with increased risk of thrombosis.3
- Lupus anticoagulant can occur transiently in the context of patients with acute thrombosis. It’s not entirely clear what role exactly the lupus anticoagulant might be playing here.
- Persistent presence of lupus anticoagulant for >12 weeks combined with evidence of thrombosis may satisfy diagnostic criteria for antiphospholipid antibody syndrome. This is a pro-thrombotic condition which requires therapeutic anticoagulation.
evidence on lupus anticoagulant in SARS & COVID-19
Chow EK et al. Severe Acute Respiratory Syndrome and Lupus Anticoagulants in Children, 2003
This is a case series of 21 pediatric patients with SARS.4 These children weren’t very sick (there were no deaths, thrombotic events, or bleeding events). Labs incidentally revealed an elevated aPTT in 8/21 patients (38%). Further investigation revealed the presence of a lupus anticoagulant in these children.
The authors concluded that prolonged aPTT shouldn’t be interpreted to reveal the presence of disseminated intravascular coagulation, but instead that aPTT elevations may result from lupus anticoagulant (a conclusion which appears to be valid with COVID-19 as well).
This study is provocative, because it demonstrates that lupus anticoagulant antibodies can occur in relatively mild illness due to coronavirus. This opens up the possibility that lupus anticoagulant antibody formation is a relatively earlier player in disease pathogenesis.
Harzallah I et al. Lupus anticoagulant is frequent in patients with COVID-19
This study tested 56 patients with COVID-19 for the presence of lupus anticoagulant using a combination of both dilute Russell viper venom test and sensitive aPTT.5 They found that:
- 25/56 (45%) of patients were positive for lupus anticoagulant.
- 5/50 (10%) of patients were positive for anti-cardiolipin antibody or anti-beta-2 glycoprotein antibody using IgM and IgG detection.
- 3/50 (6%) of patients were positive for both lupus anticoagulant plus either anti-cardiolipin antibody or beta-2-glycoprotein antibody.
This study is lacking some granular details (e.g., the titers and exact patterns of anti-cardiolipin and anti-beta-2 glycoprotein antibody).
Beyrouti R et al. Characteristics of ischaemic stroke associated with COVID-19
This is a case series of six patients with large-vessel occlusive stroke due to COVID-19 in London.6 These patients appeared fairly hyper-coagulable (with markedly elevated D-dimer and ferritin values). 5/6 of these patients had a positive lupus anticoagulant, without anti-cardiolipin or anti-beta-2 glycoprotein antibodies. The remaining patient was double-positive with both lupus anticoagulant and anti-cardiolipin/anti-beta-2-glycoprotein antibodies; interestingly this double-positive patient had the highest D-dimer value (>80,000 ug/L).
It’s interesting that only one of these patients had an elevated aPTT level at 41 seconds, despite five patients' having lupus anticoagulant on more detailed testing. This illustrates that a standard aPTT level is an insensitive test for lupus anticoagulant. One explanation for this is that COVID-19 patients tend to have elevated factor VIII levels, which will tend to pull the aPTT level down, potentially hiding the presence of lupus anticoagulant.
Zhang et al. Coagulopathy and antiphospholipid antibodies in patients with Covid-19
These authors present three patients with multiple cerebral infarcts and antiphospholipid antibodies:7
The authors report that lupus anticoagulant wasn’t detected, but they don't describe how they evaluated for lupus anticoagulant (e.g., whether they used two different assays including dilute Russell viper venom tests). Interestingly, the aPTT in all of these patients was prolonged. I wonder if these patients may have actually had lupus anticoagulant that wasn’t detected by their assays.
Bowles et al. Lupus anticoagulant and abnormal coagulation tests in patients with COVID-19
These authors in the Royal London Hospital investigated the etiology of prolonged PTT in patients with COVID-19.8 They started with a cohort of 216 patients, among whom PTT was elevated in 44 patients (20%).
35 of these patients with prolonged PTT were evaluated further. Lupus anticoagulant assays were performed in 34 patients, and were positive in 31/34 of these patients (91%)(table below). The presence of lupus anticoagulant was confirmed with 50:50 mixing studies and the addition of phospholipid:
The clotting factors showed an interesting pattern, with elevated levels of factor VIII and reduced levels of factor XII:
In practice, this pattern of clotting factors will cause hypercoagulation.
- Factor XII levels are low. This is probably irrelevant, though. Factor XII does very little in vivo. For example, deficiency in factor XII has few clinical consequences.
- Factor VIII levels are elevated. This is a more important coagulation factor, so elevated levels may tend to lead to clinical hypercoagulability. Furthermore, elevated levels of Factor VIII may tend to decrease the aPTT and thereby obscure the presence of a lupus anticoagulant.
Unfortunately, these authors didn’t test for specific antibody types (e.g. anti-cardiolipin or anti-beta-2-glycoprotein antibodies).
Hematology.org (Drs. Wool et al.)
This is a website by several hematologists hosted by the American Society of Hematology. These authors describe a series of 27 COVID-19 patients from two institutions evaluated for lupus anticoagulants:
- 20/27 (74%) of patients had positive dilute Russell viper venom assays, often significantly so.
- 4/27 (15%) had a positive Staclot Lupus Anticoagulant (aPTT-based system).
- 0/27 had IgM or IgG antibodies against cardiolipin or beta-2 glycoprotein.
Some salient details are missing, and the series is underpowered (at n=27).
Helms J et al. High risk of thrombosis in patients with severe SARS-CoV-2 infection: a multicenter prospective cohort study
This is a prospective study describing 150 COVID-19 patients within four intensive care units.9 These patients were compared to a similar number of critically ill ARDS patients without COVID, shedding some light on whether COVID causes unique coagulation abnormalities (or whether these are merely abnormalities which are widely seen in critically ill patients). Patients with COVID-19 were found to develop more thrombotic complications than non-COVID patients (e.g. with a PE rate of 12% vs. 2%).
Lupus anticoagulation was evaluated in 57 patients based upon clinical suspicion (e.g., due to elevated PTT or clinical thrombosis). Lupus anticoagulant was found in 50/57 of these patients. Evaluation included a full workup as described above, including two screening tests (a modified aPTT assay and a dilute Russell viper venom assay) as well as mixing studies and confirmation with the addition of phospholipid.
This study raises the possibility that lupus anticoagulant may be present in a large fraction of critically ill COVID patients. However, since only 57 of 150 patients were evaluated for lupus anticoagulant, this likely introduced selection bias.
Synthesis of this data
Combining data from these seven sources might start to give us a rough concept of what’s going on here.
- Lupus anticoagulant is present in lots of COVID-19 patients if it is carefully sought after using dedicated tests, such as the Russell viper venom assay and modified aPTT (perhaps roughly 50-75%, depending on disease severity).
- Fewer COVID-19 patients have a prolonged aPTT level using standard assays (perhaps ~20%). Among these patients, the great majority (~90%) might have lupus anticoagulant.
- Most patients with lupus anticoagulant do not have anti-cardiolipin or anti-beta-2 glycoprotein antibodies. However, a few patients do have a combination of lupus anticoagulant plus anti-cardiolipin and/or anti-beta-2 glycoprotein antibodies. This small subset of “double positive” patients might display a propensity towards thrombosis (e.g., ischemic stroke).
clinical implications of lupus anticoagulant in COVID-19?
One of the challenges with COVID-19 is distilling emerging and theoretical information into data which is actually relevant for clinical management of patients. For example, testing all COVID-19 patients for lupus anticoagulant is unlikely to be beneficial (because we won’t know the clinical implication of this finding).
For actual management of COVID-19 patients, the following points are relevant:
- If elevated aPTT is found, this will very often reflect lupus anticoagulant. Therefore, an elevated aPTT should not generally be interpreted to be a contraindication to anticoagulation. If time permits, mixing studies could be performed to confirm the presence of a clotting inhibitor (rather than a clotting factor deficiency, as might be seen in disseminated intravascular coagulation).
- For patients being anticoagulated with a heparin infusion, it is probably preferable to follow an anti-Xa activity level (a.k.a. “heparin level”) rather than aPTT. This is especially true if the baseline aPTT level is abnormal.
A major controversy in management of COVID-19 is what dose of anticoagulant to use for different patients. It is interesting to speculate whether patients with prolonged aPTT could represent patients at increased tendency to clot due to lupus anticoagulant. However, this requires further investigation – specifically to determine whether aPTT is an independent predictor of thrombosis above and beyond the D-dimer level.
- “Lupus anticoagulant” refers to antibodies which prolong coagulation tests in vitro (especially aPTT), yet which may be associated with hypercoagulability en vivo. Lupus anticoagulant can be stimulated by various viral infections (e.g., hepatitis C), wherein it sometimes correlates with an increased risk of thrombosis.
- Dedicated testing suggests that lupus anticoagulant may be present in roughly half of patients with COVID-19. However, only ~20% of patients with COVID-19 seem to display elevated aPTT levels on conventional testing assays.
- If elevated aPTT levels are encountered in a patient with COVID-19, they are likely to represent lupus anticoagulant activity (rather than a true hypocoagulable state). As such, elevated aPTT values shouldn’t necessarily be interpreted as a contraindication to anticoagulation.
- When possible, heparin infusions may be best titrated against anti-Xa levels rather than aPTT values (as anti-Xa levels may be more reliable and less affected by lupus anticoagulant).
- It is unclear what role (if any) lupus anticoagulant may have in the pathogenesis of COVID-19. It is conceivable that these antibodies could promote a hypercoagulable state, but it’s also possible that they are merely an associative epiphenomenon of the infection.
- IBCC section on thrombosis in COVID-19
- COVID-19 thrombosis update: anticoagulation data from Mt. Sinai (PulmCrit)
- COVID-19: Thrombosis and hemoglobin (RebelEM)
- D-dimer cutoffs to predict thrombosis in COVID-19 (PulmCrit)
- COVID-19: Dimers, clots, and DIC (ERCast with Rob Orman, MD and Thomas Deloughery, MD)
- 1.Connell NT, Battinelli EM, Connors JM. Coagulopathy of COVID‐19 and antiphospholipid antibodies. J Thromb Haemost. Published online May 7, 2020. doi:10.1111/jth.14893
- 2.Schouwers S, Delanghe J, Devreese K. Lupus Anticoagulant (LAC) testing in patients with inflammatory status: does C-reactive protein interfere with LAC test results? Thromb Res. 2010;125(1):102-104. doi:10.1016/j.thromres.2009.09.001
- 3.Abdel-Wahab N, Talathi S, Lopez-Olivo M, Suarez-Almazor M. Risk of developing antiphospholipid antibodies following viral infection: a systematic review and meta-analysis. Lupus. 2018;27(4):572-583. doi:10.1177/0961203317731532
- 4.Chow E, Chiu W. Severe acute respiratory syndrome and lupus anticoagulants in children. Br J Haematol. 2003;123(2):367-368. doi:10.1046/j.1365-2141.2003.04608.x
- 5.Harzallah I, Debliquis A, Drénou B. Lupus anticoagulant is frequent in patients with Covid‐19. J Thromb Haemost. Published online April 23, 2020. doi:10.1111/jth.14867
- 6.Beyrouti R, Adams ME, Benjamin L, et al. Characteristics of ischaemic stroke associated with COVID-19. J Neurol Neurosurg Psychiatry. Published online April 30, 2020:jnnp-2020-323586. doi:10.1136/jnnp-2020-323586
- 7.Zhang Y, Xiao M, Zhang S, et al. Coagulopathy and Antiphospholipid Antibodies in Patients with Covid-19. N Engl J Med. Published online April 23, 2020:e38. doi:10.1056/nejmc2007575
- 8.Bowles L, Platton S, Yartey N, et al. Lupus Anticoagulant and Abnormal Coagulation Tests in Patients with Covid-19. N Engl J Med. Published online May 5, 2020. doi:10.1056/nejmc2013656
- 9.Helms J, Tacquard C, et al. High risk of thrombosis in patients with severe SARS-CoV-2 infection: a multicenter prospective cohort study. Intensive Care Med. Published online May 4, 2020. doi:10.1007/s00134-020-06062-x