COVID will eventually lead to many improvements in our approach to critical illness. I’ve been hoping that one of these would be an improved understanding of venous thromboembolism (VTE) prophylaxis, but perhaps that was overly optimistic.
We provide VTE prophylaxis to nearly every patient in the ICU. Consequently, even tiny changes in our VTE prophylaxis regimen could have major consequences (when leveraged across billions of patient-days of exposure). Unfortunately, our VTE prophylaxis game remains amateurish. For example:
- The optimal VTE regimen in morbid obesity remains unclear. Different hospitals vary widely in their protocols. Many of the available algorithms don’t make much sense (e.g., patients are roughly categorized into two or three groups based on body mass index and treated with fixed-dose regimens – rather than receiving weight-based therapy).
- Renal function is poorly accounted for (generally ignoring that there’s a difference between a GFR of 31 ml/min versus 91 ml/min).
- The optimal pharmacokinetic targets for VTE prophylaxis remain unclear.
This last point is especially frustrating. Without a sophisticated pharmacokinetic understanding of VTE prophylaxis, we are doomed to forever be empirically comparing different regimens in large RCTs, in a nightmarish EBM purgatory. This is like shopping for pants without knowing your size. Every time you go to the store, you’ll waste tons of time trying on pants which aren’t close to fitting. Eventually, you’ll give up and conclude that your original pants were actually fine. Totally fine.
On the flip side, if we had a better concept of the optimal pharmacokinetic targets for VTE prophylaxis, then we could confidently tailor dosing for unusual patient populations based on measuring anti-Xa levels (e.g., patients with unusual weight, heparin resistance, or abnormal renal function). Pharmacokinetic studies could be performed on 50-100 patients, allowing for accurate dosing in subpopulations which cannot be studied using RCTs (an RCT on VTE prophylaxis requires hundreds of patients in order to have sufficient power to detect rare thrombotic or bleeding events, so RCTs cannot be done on small patient subpopulations).
My guess regarding the optimal pharmacokinetic target for VTE prophylaxis is as follows:
- The risk of bleeding may relate to the peak heparin concentration (or, practically speaking, the 4-hour anti-Xa level after a dose of low molecular-weight heparin).
- The efficacy for preventing VTE may relate to the trough heparin concentration.
- The best way to optimize efficacy while minimizing risk may be to use a strategy employing smaller doses given more frequently (or, ideally, a continuous infusion). For example, a regimen of 0.5 mg/kg q12hr or 0.3 mg/kg q8hr would allow for a gentler curve which simultaneously maximized safety (avoiding high peaks) and efficacy (avoiding deep valleys).
Thus, a reasonable intermediate-dose strategy for VTE prophylaxis might be ~0.5 mg/kg enoxaparin q12hr. This dose is already supported by a moderate amount of evidence in critically ill patients prior to COVID, as previously explored here. The Anticoagulation Forum made similar recommendations for critically ill patients with COVID:1
Currently, a study was published in JAMA evaluating the use of 1 mg/kg enoxaparin daily as a form of “intermediate” intensity VTE prophylaxis.2 Giving a big dose of enoxaparin once daily makes little pharmacokinetic sense, as this will produce relatively high peak heparin levels (thereby increasing the risk of bleeding), yet still generate low trough levels prior to the next dose (thereby reducing the efficacy).
The JAMA study is substantially flawed. Rather than comparing two enoxaparin regimens, they actually compared eight different regimens involving two different medications (table below). Without a sophisticated understanding of heparin pharmacokinetics or monitoring of anti-Xa levels, it’s impossible to be certain whether any of these eight regimens are equivalent to one another. Yet again, we find ourselves at the pants store, desperately trying on different pants without knowing our size. Ultimately, the study attempts to empirically compare eight different regimens employing two different drugs (enoxaparin and unfractionated heparin), using only two different patient groups. That’s not the way statistics works.
The results of the trial were neutral, with some concerning trends (table below). This weakly suggests that 1 mg/kg enoxaparin daily is a suboptimal regimen – a conclusion which could have been reached based on pharmacokinetic principles, without performing the study at all.
Publication of this trial in JAMA will endow it with an unmerited degree of influence. This will lead many to interpret this study to mean that intermediate-dose prophylaxis should be discarded entirely. This is a convenient conclusion, because there’s always comfort in doing things the way we have done them before.
However, I don’t believe that scuttling back to our conventional (and highly flawed) practices of VTE prophylaxis is the answer. If anything, the fact that we can’t even agree on what constitutes intermediate-dose prophylaxis should demonstrate how dim our understanding of VTE prophylaxis is. This should be an illustration of how much work needs to be done to understand the optimal pharmacokinetics and pharmacodynamics of VTE prophylaxis. Although embracing the status quo is comfortable, our patients might benefit more if we could admit our ignorance on the topic and move forwards from there. Because, let’s be honest – we really do need better pants.
- 1.Barnes G, Burnett A, Allen A, et al. Thromboembolism and anticoagulant therapy during the COVID-19 pandemic: interim clinical guidance from the anticoagulation forum. J Thromb Thrombolysis. 2020;50(1):72-81. doi:10.1007/s11239-020-02138-z
- 2.Sadeghipour P, Talasaz AH, et al. Effect of Intermediate-Dose vs Standard-Dose Prophylactic Anticoagulation on Thrombotic Events, Extracorporeal Membrane Oxygenation Treatment, or Mortality Among Patients With COVID-19 Admitted to the Intensive Care Unit. JAMA. Published online March 18, 2021. doi:10.1001/jama.2021.4152