background: what is heparin resistance?
Heparin works by binding to antithrombin III and thereby activating antithrombin III, an endogenous anticoagulant which inhibits clotting factors (especially Xa).
Heparin + Anti-thrombin III → [Heparin-Antithrombin-III complex] → Inhibition of Xa activity
Heparin resistance refers to situations where unusually large doses of heparin are required to achieve anticoagulation. This is generally defined as requiring >35,000 units per day of heparin to achieve anticoagulation (e.g., an infusion of more than ~25 units/kg/hour).
There are three causes of heparin resistance:
- Pseudo heparin resistance: This occurs only when heparin is being titrated using PTT. High levels of factor VIII and/or fibrinogen tend to drag down the PTT value. Artificially low PTT values make it seem like heparin isn’t working – when it actually is.
- Antithrombin III deficiency can be caused by many things (e.g., liver disease, acute thrombosis, disseminated intravascular coagulation). Since heparin works via binding to antithrombin III, this deficiency leads to true physiological heparin resistance.
- Low heparin concentration due to binding by acute-phase proteins. Systemic inflammation increases the production of proteins that bind heparin (e.g., platelet factor 4).
Heparin resistance is common in the ICU, especially among sicker patients with more profound systemic inflammation. So it should come as little surprise that this would be a problem with COVID-19 patients. That being said, let’s take a look at the evidence.
Interpretation of these studies frequently refers to anti-Xa levels (a functional assay of the efficacy of heparin to block coagulation). This is a bit tricky, as there is some disagreement in precise anti-Xa targets. It's also confusing because target levels vary depending on drug and indication. The table below should provide a reasonable idea of anti-Xa targets:
Dutt et al. – 40 mg enoxaparin daily fails badly for prophylaxis
A dose of 40 mg enoxaparin daily is widely believed to be adequate for DVT prophylaxis among ICU patients. However, review of pharmacokinetic data suggests that this dose is usually inadequate (more on this here). As ICU patients become sicker, they develop more heparin resistance and the 40-mg enoxaparin dose just doesn’t cut it.
So, it shouldn’t come as any surprise that this finding is borne out in COVID-19 patients as well. These authors examined anti-Xa levels among ICU and ward patients treated with 40 mg enoxaparin daily.1 In the ICU, this dose produced inadequate anti-Xa levels 95% of the time! And these authors were shooting for a lowish target level of 0.2-0.5 IU/ml. Many other studies target 0.3-0.5 IU/ml or 0.3-0.7 IU/ml when using daily enoxaparin for DVT prophylaxis – using these targets, the 40-mg enoxaparin regimen failed 100% of the time.
Consistent with other data, patients who are less ill exhibit less heparin resistance. So, ward patients with COVID-19 were less likely to be underdosed when recieving 40 mg enoxaparin (although underdosing remained a problem there as well).
Vlot et al. – Four-fold standard dosing of nadroparin works reasonably well
These authors described DVT prophylaxis for sixteen COVID patients in the ICU using a regimen of nadroparin which is roughly four times the standard dosage (the usual dose is 2850 IU daily; these authors used 5700 IU twice daily, or 7600 IU twice daily among patients >120 kg).2
Peak anti-Xa levels are shown above. These were reasonably on target. The optimal anti-Xa level remains unclear, but for twice-daily dosing an anti-Xa level >0.2 is probably fine. So most of the patients were receiving reasonably therapeutic and safe doses of anticoagulation. Nonetheless, there is a three-fold variation in the anti-Xa levels – indicating significant interindividual variation.
Generalizing from nadroparin to enoxaparin is difficult, because nadroparin has a shorter half-life than enoxaparin (~3 hr vs. ~4.5 hours, respectively). However, this study may indirectly suggest that an augmented dose of enoxaparin BID could work reasonably well.
Stattin et al. – Worrisome variation between LMWH dose vs. anti-Xa levels
This study involved 31 ICU patients with COVID-19.3 Most patients were treated with dalteparin at ~75-100 IU/kg, somewhat higher than the usual prophylactic dose. (For comparison: a full therapeutic dose of dalteparin is 150-200 IU/kg, whereas a typical prophylactic dose is fixed at 5,000 IU). A few patients developed thrombosis and were subsequently switched to therapeutic dose dalteparin.
Above is a graph that relates the dalteparin dose with the anti-Xa levels. A few points are notable:
- Even when prophylactic-dose dalteparin was dosed moderately higher than usual (~75-100 IU/kg), many patients achieved subtherapeutic levels (<0.2 IU/ml).
- There was a surprising amount of variation between patients regarding responsiveness to dalteparin.
A major weakness of this study is that it doesn’t consider renal function. Borderline renal function may be one factor leading to excessive anti-Xa levels.
Trunfio et al: Titration of prophylactic enoxaparin dose against anti-Xa levels
This study describes 56 COVID patients in an ICU treated with enoxaparin for DVT prophylaxis.4 Most patients initially received 40 mg enoxaparin, but a variety of agents and doses were also used. The initial anti-Xa level was out of target range (defined as 0.3-0.7 IU/ml) about 40% of the time, more often subtherapeutic than supratherapeutic.
This study is lacking in some granular details, but it does overall support the feasibility of titrating enoxaparin dose against anti-Xa levels. Provocatively, the authors found a correlation between prompt adjustment of enoxaparin dose and superior clinical outcomes.
White et al and Huisman et al: Heparin infusion reveals heparin resistance in 8/10 patients
In everyday medical practice, we don’t routinely check anti-Xa levels for patients on low molecular-weight heparin. Consequently, heparin resistance may fly under the radar among these patients. Alternatively, heparin resistance rapidly becomes extremely obvious among patients with heparin infusions – because we are struggling to give the patient enough heparin to achieve therapeutic anticoagulation.
White et al. describes ten patients with COVID in the ICU who required heparin infusion.5 8/10 patients had heparin resistance (requiring of >35,000 units/day heparin to achieve therapeutic anticoagulation). Three patients required >50,000 units/day. This confirms that heparin resistance is common among COVID patients in the ICU.
Beun et al. describes a series of four patients displaying the same phenomenon.6 All patients in the series required very high doses of heparin to achieve therapeutic anticoagulation:
Mechanism(s) of heparin resistance
As discussed above, there are three causes of heparin resistance:
- Pseudo heparin resistance: High levels of factor VIII and/or fibrinogen artificially lower the PTT level.
- Antithrombin III deficiency.
- Low heparin concentration due to binding to acute-phase proteins.
What is the mechanism of heparin resistance seen in COVID patients?
1) pseudo heparin resistance
Studies consistently show that COVID patients have high levels of factor VIII and fibrinogen, so some amount of pseudo resistance might be expected. Only one study compared PTT values to anti-Xa levels among COVID patients (figure below).5 A few patients did seem to exhibit a bit of pseudo resistance (subtherapeutic APTT ratio despite a therapeutic anti-Xa level). It’s unclear how often this would be clinically relevant in a larger patient population.
Increasingly, hospitals are switching from using PTT to using anti-Xa levels to titrate heparin infusions. Use of anti-Xa levels avoids the issue of pseudo resistance entirely.
2) antithrombin III deficiency
Numerous studies have reported on antithrombin III levels. These levels were often slightly reduced, but not usually reduced enough to cause heparin resistance (e.g., not <50% activity). It’s likely that antithrombin deficiency could occur in scattered patients, but overall this doesn’t seem to be a major driver of heparin resistance.
3) low heparin concentration due to acute phase proteins
Based on the exclusion of #1-2, this is probably the primary cause of heparin resistance, when using anti-Xa assays for clinical management. Of course, many patients could have multifactorial heparin resistance (due to a combination of mild antithrombin III deficiency and low heparin concentration).
In vitro studies using blood from COVID patients have shown that adding heparin produces lower anti-Xa activity than expected.5 This again confirms the presence of heparin resistance. It also implies that heparin may be bound by proteins circulating in the blood (rather than, for example, being adsorbed onto the endothelial glycocalyx).
management of heparin resistance
Given that antithrombin III levels are usually preserved, administration of exogenous antithrombin III probably won’t help much. That’s fine, because antithrombin III is an expensive and potentially high-risk medication.
The easiest approach is generally just to use a higher heparin dose. Heparin resistance can usually be overpowered by higher doses of heparin. There is no specific “maximal” dose of heparin which may be used. For example, Beun et al. reported using 64,500 units/day heparin in one patient with COVID-19.6 Heparin doses may need to be aggressively escalated to achieve a therapeutic effect.
Another strategy for management of heparin resistance is to transition to an argatroban infusion.7 This is a viable strategy, but it has a few drawbacks. First, argatroban infusions must be titrated against a PTT level – which may create issues with pseudo resistance (similar to heparin pseudo resistance)! Second, argatroban may lack some of the anti-inflammatory, anti-complement, and endothelial-protective effects of heparin.8
limitations to our knowledge
After months of speculation, it is nice to see some data. However, important caveats need to be borne in mind.
anti-Xa level is a surrogate endpoint
Anti-Xa levels do seem to correlate reasonably well with the effect of heparin across various situations. But there is no denying that this is solely one measurement of coagulation, among a myriad of other parameters.
To make matters a bit fuzzier, there is some disagreement about precisely what anti-Xa levels we should be targeting. For example, for once-daily prophylactic low molecular weight heparin some sources recommend targeting 0.2-0.6 IU/ml, whereas others recommend targeting 0.3-0.7 IU/ml. The precise target will shift numbers slightly, but it doesn’t change the ultimate interpretation of the above studies.
It would obviously be better to perform a prospective RCT comparing different enoxaparin doses using a clinically relevant endpoint. Such studies are underway. Hopefully they will be completed and come to publication rapidly.
Is COVID truly different from other critical infections?
It’s unclear how unique the coagulopathy of COVID actually is. It ultimately may end up that COVID isn’t all that special, but rather that severe coagulopathy is commonly seen in patients with profound infection. As such, it’s possible that the huge influx of COVID patients has forced us to to reexamine a problem which has been underneath our noses all along.
A prior blog explored the issue of heparin resistance and enoxaparin dosing among critically ill patients in general (not COVID patients). The conclusions from that blog are generally consistent with the above data with COVID patients. In short, we’ve probably been under-dosing enoxaparin among critically ill medical patients for a long time. Consistency between COVID and non-COVID data supports that this is a widespread and real phenomenon.
what do the guidelines say?
Numerous eminence-based guidelines exist, which make a variety of conflicting recommendations. Some examples are shown below.9
Several guidelines acknowledge the issue of heparin resistance, recommending monitoring of anti-Xa levels among patients receiving heparin infusions (table below). However, the guidelines don’t go so far as to recommend monitoring of anti-Xa levels among patients receiving prophylactic doses of low molecular weight heparin.
New data from COVID patients, as well as prior data gathered over decades among non-COVID patients, may have the following clinical implications:
- Heparin resistance is common among COVID patients in the ICU. The degree of heparin resistance increases in parallel to the overall illness severity.
- For patients who require therapeutic anticoagulation (e.g., due to DVT or PE), a heparin infusion may be the optimal treatment. An infusion will allow for immediate detection of heparin resistance and subsequent titration as necessary to achieve therapeutic targets. Monitoring of anti-Xa levels is likely superior to monitoring of PTT levels, if available. An alternative might be therapeutic enoxaparin with monitoring of anti-Xa levels. If enoxaparin is used without anti-Xa level monitoring, it’s possible that patients could have occult heparin resistance and ineffective treatment.
- Usual heparin dosing for DVT prophylaxis (typically 40 mg daily enoxaparin) is inadequate among COVID patients in the ICU. A higher dose is required in nearly all patients. The optimal dose remains unknown, but a dose of ~0.5 mg/kg enoxaparin BID might be reasonable for patients with adequate renal function.
- Until more evidence is available, monitoring anti-Xa levels and titrating enoxaparin accordingly might theoretically be the most evidence-based strategy for DVT prophylaxis (using the patient’s own coagulation studies as their personal source of “evidence”). This may seem like a lot of work, but the truth is that we check lots of labs in our patients and generally do nothing about them. So checking one more lab could be worthwhile, if it will directly drive important treatment decisions. Please note, however, that monitoring anti-Xa levels for DVT prophylaxis is not recommended by any current guidelines.
- Heparin resistance (IBCC)
- Mythbusting 40 mg enoxaparin daily for DVT prophylaxis in critical illness (PulmCrit)
- 1.Dutt T, Simcox D, Downey C, et al. Thromboprophylaxis in COVID-19: Anti-FXa-the Missing Factor? Am J Respir Crit Care Med. 2020;202(3):455-457. doi:10.1164/rccm.202005-1654LE
- 2.Vlot E, Van den, Hackeng C, Sohne M, Noordzij P, Van D. Anti Xa activity after high dose LMWH thrombosis prophylaxis in covid 19 patients at the intensive care unit. Thromb Res. 2020;196:1-3. doi:10.1016/j.thromres.2020.07.035
- 3.Stattin K, Lipcsey M, Andersson H, et al. Inadequate prophylactic effect of low-molecular weight heparin in critically ill COVID-19 patients. J Crit Care. 2020;60:249-252. doi:10.1016/j.jcrc.2020.08.026
- 4.Trunfio M, Salvador E, Cabodi D, et al. Anti-Xa monitoring improves low-molecular-weight heparin effectiveness in patients with SARS-CoV-2 infection. Thromb Res. 2020;196:432-434. doi:10.1016/j.thromres.2020.09.039
- 5.White D, MacDonald S, Bull T, et al. Heparin resistance in COVID-19 patients in the intensive care unit. J Thromb Thrombolysis. 2020;50(2):287-291. doi:10.1007/s11239-020-02145-0
- 6.Beun R, Kusadasi N, Sikma M, Westerink J, Huisman A. Thromboembolic events and apparent heparin resistance in patients infected with SARS-CoV-2. Int J Lab Hematol. 2020;42 Suppl 1:19-20. doi:10.1111/ijlh.13230
- 7.McGlynn F, McGrath J, Varghese C, et al. Argatroban for therapeutic anticoagulation for heparin resistance associated with Covid-19 infection. J Thromb Thrombolysis. Published online August 24, 2020. doi:10.1007/s11239-020-02251-z
- 8.Godoy L, Goligher E, Lawler P, Slutsky A, Zarychanski R. Anticipating and managing coagulopathy and thrombotic manifestations of severe COVID-19. CMAJ. 2020;192(40):E1156-E1161. doi:10.1503/cmaj.201240
- 9.Flaczyk A, Rosovsky R, Reed C, Bankhead-Kendall B, Bittner E, Chang M. Comparison of published guidelines for management of coagulopathy and thrombosis in critically ill patients with COVID 19: implications for clinical practice and future investigations. Crit Care. 2020;24(1):559. doi:10.1186/s13054-020-03273-y
- IBCC – Posterior Reversible Encephalopathy Syndrome (PRES) - June 21, 2021
- PulmCrit – A history of hypothermia for cardiac arrest, 2002-2021 (RIP) - June 17, 2021
- IBCC – Reversible Cerebral Vasoconstriction Syndrome (RCVS) - June 14, 2021