We see elderly patients on aspirin come to the ED after a fall nearly every day. The ubiquitous use of antiplatelet agents and the ever-increasing age of our population makes this scenario all too common. The data on antiplatelet agents for head trauma, however, are conflicting, and we lack consensus on how to manage these patients. (Surgery 2011;150:861; World Neurosurg 2010;74[2-3]:279; J Emerg Med 2014;46:410; Acad Emerg Med 2017;24:1258; Am J Surg 2006;192:743.)
Warfarin is typically considered notorious for causing life-threatening bleeding in the elderly. Mounting evidence shows that antiplatelet agents may pose even more of a risk and are associated with increased mortality. (J Trauma 2005;58:518.) Antiplatelet agents also lack a clinically effective reversal strategy once the bleeding occurs.
A recent study suggested that patients taking antiplatelet drugs are at an even higher risk for intracranial hemorrhage (ICH) than those taking warfarin after a ground-level fall. (Acad Emerg Med 2017;24:1258.) The rate of ICH in elderly patients on antiplatelet medications was 4.3 percent compared with 1.7 percent for those on warfarin. These results are consistent with those in a 2012 study that found the incidence of ICH to be 12 percent in patients taking clopidogrel compared with five percent in those on warfarin. (Ann Emerg Med 2012;59:460; http://bit.ly/2OEWMjG.) Another study found that 33.4 percent of elderly patients with a GCS of 15 who suffered a ground-level fall had positive CT scan findings. (Am J Surg 2017;214:105.) This study is controversial and lacks rigorous methodology, but it demonstrates that elderly patients are at increased risk for intracranial pathology following seemingly innocuous mechanisms of trauma. The majority of these patients receive CT scans as part of their ED workup.
The question to ask before thinking about treatment is whether having medication-induced platelet dysfunction adversely affects those with ICH. Intuitively we think that lack of adequately functioning platelets must be detrimental in bleeding, but the evidence suggests otherwise. One study showed no statistically significant difference in hematoma expansion in those taking antiplatelets compared with those who were not, and the same was true for in-hospital mortality. (Neurocrit Care 2010;12:24.)
Another retrospective study found that patients taking antiplatelet agents are more likely to present with a severe bleed and have higher mortality rates, but found on serial CT scans that those with lower-grade injuries did not appear to have any significant progression of bleeding despite being on antiplatelet medications. (J Trauma 2008;65:785.) Those on warfarin, however, can exhibit progression of hematoma if not reversed.
No clear evidence suggests that being medicated with an antiplatelet drug puts a patient with traumatic ICH at a significantly greater risk for mortality. Being on an antiplatelet agent, however, does appear to put one at an elevated risk for a bleed immediately following head trauma. An elderly person on clopidogrel has a greater chance of having a bleed compared with one who does not take antiplatelet medication, but if both develop a bleed, the one taking clopidogrel will not necessarily have a worse outcome.
Therapy is controversial once a bleed has been detected. Common treatment algorithms for ICH in patients taking antiplatelet agents include the use of desmopressin (DDVAP) and platelet transfusion. Platelet transfusion in spontaneous ICH has recently been called into question following the publication of the PATCH trial, which did not include traumatic ICH. (Lancet 2016;387:2605.) A systematic review did not find any randomized controlled trials that examined platelet transfusion for traumatic ICH, but found four cohort studies that demonstrated an odds ratio of 1.77 for in-hospital mortality in those receiving platelet transfusion. (J Emerg Med 2015;49:561.) TBI patients older than 50 on antiplatelet agents who received platelet transfusion in a cohort study had a mortality rate of 17.5 percent, while the group that received no transfusion had a mortality rate of 16.7 percent. (Am Surg 2009;75:1100.) No statistically significant difference was seen between the two groups. More recently, a study found that platelet transfusion seems to reverse the antiplatelet effects of these medications based on lab assay, but did not find evidence of clinical benefit when examining a patient-oriented outcome. (Brain Inj 2018;32:325.) No decidedly positive study has demonstrated any mortality or clinical outcome benefit in administering platelets for those with traumatic ICH on antiplatelet agents.
DDAVP, the other therapeutic option for this group, increases the release of von Willebrand factor and improves the function of platelets, decreasing hematoma expansion. (Neurol Res Int 2014;2014:298767; http://bit.ly/2OMPjik.) Numerous studies have demonstrated the efficacy of DDAVP in in vitro lab measures of platelet activity. DDAVP is commonly used to reduce platelet dysfunction associated with uremia and other conditions such as ITP. (Crit Care Med 2016;44:2251; Neurol Res Int 2014;2014:298767; http://bit.ly/2OMPjik; J Neurotrauma 2015;32:1815; Stroke 2014;45:2451.)
The results are not convincing, however, when it comes to DDAVP for reversal of antiplatelet medication-induced platelet dysfunction. A small study of 14 patients demonstrated a statistically significant improvement in platelet function after DDAVP and a change in hematoma size averaging -0.5 cc with a range of -1.5 cc to 8.4 cc and two patients experiencing hematoma growth. (Stroke 2014;45:2451.) The study, however, lacked a control group and meaningful patient-oriented outcomes. The authors attempted to measure modified Rankin Scale scores on their small sample size but failed. Another study of 10 patients demonstrated similar laboratory markers of improved platelet function, with no clinical outcomes measured. (Neurol Res Int 2014;2014:298767; http://bit.ly/2OMPjik.)
Another study looked at patients with traumatic ICH receiving transfusion of platelets in addition to DDAVP and explored clinical outcomes such as mortality. (J Neurotrauma 2015;32:1815.) Patients were eligible to be treated with DDAVP and platelets if they had platelet counts below 100,000, were taking an antiplatelet medication, had an impending neurosurgery, or were otherwise enrolled at the treating physician's discretion. Neither mortality benefit nor statistically significant difference in hematoma expansion between the treatment and control groups was found.
The Myth of Delayed Bleed
Despite the increased risk of bleeding in elderly patients on antiplatelet agents with blunt head trauma, the vast majority of CT scans will turn out to be negative. Little evidence supports routine repeat CT scans to detect delayed bleeding in this population. (J Am Coll Radiol 2018;15:319; Neurol Clin Pract 2017;7:296; http://bit.ly/2NysliM; J Emerg Med 2014;46:410; West J Emerg Med 2015;16:43; http://bit.ly/2NukiDt; J Trauma Acute Care Surg 2015;79:310; Emerg Med J 2010;27:537.) Some studies suggest observing these patients for up to 48 hours or repeat CT scanning at 24 hours, but these are costly in time and resources and likely confer no benefit.
The presence of delayed bleed in the anticoagulated patient has been found to be relatively uncommon. The authors of one review concluded that the prevalence of delayed bleeding after a normal initial CT scan ranged from 0.6 to six percent. (J Emerg Med 2014;46:410.) They pointed out that this is a surrogate outcome; it does not speak to any patient-oriented endpoint such as death, morbidity, mortality, or need for operative intervention. When they looked at patient-centered outcomes, they found a range of zero to 1.1 percent of patients who died or had a neurosurgical intervention. This review included studies in which patients were taking warfarin as well as antiplatelet medications, and was not a pure antiplatelet agent population. The 2012 paper with more than 1,000 patients found that the delayed bleed rate was 0.6 percent for those on warfarin and zero for those on clopidogrel. (Ann Emerg Med 2012;59:460; http://bit.ly/2OEWMjG.)
A better, cheaper, and likely equally efficacious strategy may be to provide detailed instructions to patients and family about the warning signs of progressive head injury and concussive symptoms and encourage prompt return should worrisome symptoms develop.
Clinical care should be evidence-based and clinicians informed of these data that guide their day-to-day practice. Blunt head trauma in this population will continue to put strain on our ED throughput and resources owing to varying practice patterns and fear of missing a delayed bleed. Data suggest that many of these fears are unfounded and rooted in dogma. Prospective clinical trials are needed to definitively change practice, but the current evidence should serve as a foundation from which to build your clinical practice for now.Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.