A familiar scenario:
You are on duty and your standby phone goes off just as you have come back from your coffee break. The ambulance cell informs you that they are bringing in an elderly patient who has fallen a flight of stairs in his own home. He is confused with a fluctuating GCS and is bleeding from a scalp wound the paramedic covered up with a bandage.
On arrival, he is intoxicated with alcohol and has a GCS of 12/15, your assessment made difficult by the bottle of Chardonnay he has had with his dinner. His wife who is with him tells you that he is taking warfarin for a previous deep-venous thrombosis and his last INR check (last week) was 3.2.
It is a no brainer and you request a CT brain starting him on a prothrombin complex infusion to reverse the effect of his warfarin medication. You take him down to your CT scan suite as you are worried about an intracranial haemorrhage.
His CT scan is however reported as normal and he is admitted to your observation ward later during your shift. He is discharged home after an eventful period of observation.
You are pleased and congratulate yourself thinking: “job well done!”. But…is it really?
Prothrombin complex concentrates (PCC) are now widely used in EDs to rapidly reverse warfarin-related coagulopathies in the setting of life-, limb-, or eye-threatening injuries. We are often tasked as emergency physicians with making complex and rapid decisions regarding the reversal of coagulopathy (as in the scenario above) and in doing so we are required to weigh the risks and benefits of PCC use with often no specialist input at the time of decision.
We have known for a while that PCCs are not completely harmless as they carry an associated risk of allergic reaction with them but more importantly a poorly studied risk of thromboembolism ranging from 1.4% to 10%. This is certainly not a negligible risk.
So do we always help our patients by aggressively reversing their iatrogenic coagulopathy? Well, this paper attempted to determine this very risk.
What was the aim of the trial?
This paper from the Journal of Emergency Medicine attempted to determine the risk of thromboembolism after reversal of warfarin-associated coagulopathy by PCC in patients with intracranial haemorrhage.
What kind of study was this?
This was a retrospective chart review at an academic hospital over a period of approximately five years.
We know that retrospective studies are inherently limited by availability of information recorded. We also all know that in moments of panic around emergencies, medical documentation is not at its best and all EPs are probably at fault here to a certain degree. I will let you draw your own conclusions around potential limitations…
Who did they include?
All patients with warfarin-associated head injury with intracranial haemorrhage and who received 3F-PCC (three-factor PCC). These patients furthermore received fresh frozen plasma and vitamin K in addition of the reversal agent. This is interesting because in Virchester, we use a five-factor PCC and do not routinely use FFP concomitantly.
This somewhat potentially limits the generabilisabilty of this study.
What outcomes were considered?
The authors looked at a reasonable range of potential outcomes as consequences of the administration of PCC , namely: DVT, PE, limb ischaemia, TIA, CVA, myocardial infarction (STEMI and NSTEMI) and unexplained death. I am unsure if the last one should have been included really unless secondary to a thromboemblism confirmed by a post-mortem examination. The arbitrary cut-off time was 30 days.
What did they find then?
209 patients received PCC for warfarin-associated intracranial haemorrhage between 2008 and 2013. There was no significant difference in baseline characteristics and types of ICH. The death rate was 26.8%.
In 29 subjects who were taking warfarin for a previous PE or DVT indication, there was an increase in recurrent VTE events (36.8% vs. 11.6%, p=0.007). Subjects with a history of DVT or PE had a significantly increased risk of sustaining VTE complication within 30 days of reversal therapy administration (OR 4.455, 95% CI 1.586 – 12.511, p=0.005).
Age was not found to be a risk factor for VTE complications after 3F-PCC administration on a univariate logistical regression analysis (OR 0.983, 95% CI 0.940 – 1.029, p=0.469)
The authors concluded that their data suggest that the clinician should carefully assess the indication for taking warfarin and any past medical history of PE or DVT to weigh the risks and benefits of administering PCC therapy.
We knew that reversing warfarin therapy was not completely without risks. Guidelines often recommend use of PCC to reverse anticoagulation but rarely discuss or quantify the potential risks of VTE secondary to reversal treatment.
As resuscitationists, we often have to make rapid decisions but this needs to be done with a careful analysis of risk versus benefits. In cases of small ICH, stable conscious level or remote timing of symptoms onset, this decision is not an easy one to take due to time pressures.
Despite its limitations, I believe this paper is food for thought for all emergency physicians.