Does Discontinuation of Aspirin Treatment Following Head Trauma Decrease the Incidence of Chronic Subdural Hematoma?

Scientific Background Chronic subdural hematoma (CSDH) is a well-described sequelae of minor head trauma in the elderly population. Known risk factors for the development of this condition are old age, previous bleed, intracranial hypotension, and anticoagulant or antiaggregant therapy prior to the trauma [1-3, 10].

Anti-aggregation therapy, including treatment with low-dose aspirin (LDA) is an established risk factor for the development of a CSDH. Among patients presenting with a chronic subdural hematoma, 16-76% have a history of antiaggregation therapy [6, 9, 13], however it remains unclear whether the increased risk is related to treatment prior to the trauma or to the continued treatment with LDA after the trauma, while the hematoma is supposedly developing.

This has implications for daily clinical practice. It is unclear whether continued antiaggregation therapy after a minor head trauma increases the risk for development of a CSDH, and therefore it is unclear whether cessation of such treatment after a minor head trauma is indicated.

We aim to assess whether continued aspirin treatment increases the risk of CSDH in mild head trauma patients 50 years and older who present with negative head CT.

Brief Review of the Literature The risk of bleeding complications associated with LDA prophylaxis has been extensively reported. In 2006, McQuaide and Laine [4] performed a systematic review and meta-analysis to assess the relative and absolute risk of clinically relevant adverse events in patients treated with LDA or clopidogrel for cardiovascular prophylaxis. They included randomized controlled trials comparing cardiovascular outcomes and adverse events in patients treated with placebo, LDA, and clopidogrel from 1966-2004 in their analysis. The authors found a very modest 0.3% (3 patients per thousand) increase in the absolute risk of intracranial bleeding associated with the use of aspirin, although CSDH in the wake of head trauma was not specifically discussed.

A year earlier, Serebruany et al [11] compared the risk of hemorrhage associated with < 100mg, 100-200mg, and > 200mg doses of aspirin in a meta-analysis of 31 studies involving 192,036 patients. They reported major bleeding events (intracranial bleeding; overt bleeding with decrease ≥ 5g/dl in hemoglobin or ≥ 15% in hematocrit) in 1.5% of patients on doses < 100 mg, 1.5% of patients at 100-200 mg, and 2.3% of patients on the highest dose. Incidence of CSDH was not reported.

Aspirin has been associated with an increased risk for postoperative intracranial hemorrhage [5, 7], and CSDH [13] However in spite of the strong evidence that aspirin treatment increases the risk of intracranial hemorrhage, there is a surprising dearth of research guiding the decision to continue or discontinue aspirin in patients who have sustained mild head trauma with no sign of injury on CT.

Tauber et al [14] recently assessed the incidence of secondary intracranial hemorrhage in a prospective study of 100 mild head trauma patients 65 years and older on LDA prophylaxis. Primary CT scan in all participants revealed no evidence of intracranial hemorrhage. Repeat CT, performed within 12-24 hours, revealed secondary intracranial hemorrhage leading to death in one patient, surgical intervention in one, and conservative management with an uneventful clinical course in two patients. The authors recommended repeat CT after 12-24 hours or 48-hour in-hospital observation for elderly patients on LDA prophylaxis.

In 2003, Spektor et al. investigated the role of LDA prophylaxis in the evolvement of acute traumatic findings after a mild-to-moderate head injury [12]. They evaluated 231 patients, comparing those on aspirin treatment to those who were not. Interestingly, the authors found no significant increase in the frequency of intracranial hemorrhage in patients receiving prophylactic LDA treatment, however they did not assess the incidence of delayed bleeding.

In 1992, Reymond et al reviewed the characteristics of 39 patients who developed chronic subdural hematoma in a population of 198 patients admitted after severe head trauma with various forms of intracranial bleeding on original CT [8]. They found that risk factors for intracranial hemorrhage were age, alcohol consumption, and anticoagulation or antiaggregation therapy. Five (13%) patients with chronic subdural hematoma had been treated with aspirin. The authors concluded that patients on prophylactic LDA treatment are at risk of developing chronic subdural hematoma.

Several other studies retrospectively evaluated the role of LDA treatment in the evolution of chronic subdural hematomas. The proportion of patients with chronic subdural hematoma who were treated by aspirin varied widely. O'Brien et al reported treatment w LDA in 93 of 123 (76%) patients with spontaneous hematomas [6]. On the other end of the scale, Rust et al. reported this treatment in only 18 of 81 patients (22%) [9] and Stroobandt in 16 of 100 (16%) [13].

Two studies reported no effect of aspirin treatment on the recurrence rate of chronic subdural hematomas after drainage. Torohashi et al [15] retrospectively reviewed 343 consecutive patients treated with surgical drainage for CSDH and subsequent recurrence. Patients were advised to discontinue anticoagulation and antiaggregation for one week after initial drainage. The authors found no statistically significant relationship between anticoagulation or antiaggregation therapy and recurrence of CSDH, however recurrence occurred with a shorter interval in patients treated with anticoagulation or antiaggregation therapy.

Research Objectives Assess whether continuation of low-dose aspirin therapy increases risk for development of chronic subdural hematoma in the 3-5 weeks following mild head trauma in patients 50 years and above.

Assess whether cessation of LDA therapy decreases the risk for development of chronic subdural hematoma in the 3-5 weeks following mild head injury in patients 50 years and above.

Assess whether cessation of LDA therapy is associated with an increase in clinically significant adverse events, including myocardial infarction, transient ischemic attack, or vascular or thrombo-embolic events in the 3-5 weeks following mild head trauma in patients 50 years and above.

Establish a study protocol that may serve as the basis of a larger and more definitive multi-institutional study.

We hypothesize that continuation of LDA will not increase the risk of CSDH in the 3-5 week follow-up period after minor head trauma.

We hypothesize that discontinuation of LDA for 3-5 weeks will not be associated with an increase in the number of in clinically significant cardiovascular events.

Expected Significance There is currently no consensus regarding the continuation or cessation of prophylactic aspirin treatment following mild head trauma in patients with no immediate evidence of hemorrhagic injury. The evidence-based knowledge that will be gained in this study, and from larger multi-institutional studies will have significant clinical implications.

Methods All patients seen in Hadassah-Hebrew University Medical Center's Department of Emergency Medicine after head trauma are examined by the neurosurgeon on call. A complete history is taken, and patients undergo a thorough physical and neurological examination. Patients reporting to the Emergency Department after mild trauma who are treated with LDA at the time of injury are routinely referred for immediate non-contrast head CT. Blood tests to assess prothrombin time, partial thromboplastin time, INR, and platelet count are performed. A detailed history is taken to ascertain the reason for aspirin therapy. Patients with head CT that is negative for traumatic findings, and with normal neurologic examination or no change from neurological baseline before trauma, are discharged from the emergency department. The policy regarding continuation or discontinuation of LDA after mild head trauma is not well established.

We propose to offer patients aged 50 and above who suffer mild head trauma, who are on LDA therapy (75-100 mg) at time of trauma, who have normal neurological examination or no change from neurological baseline before trauma, no evidence of intracranial hemorrhage on initial CT, and who meet other inclusion and exclusion criteria, participation in a randomized trial to assess the effects of aspirin continuation versus aspirin cessation.

Intake- and follow-up examinations will be completed according to standard forms that will be used by all physicians involved in the care of participants in the study (Appendix 1-2). Informed consent will be obtained (Appendix 3). Patients will be randomized. An attending neuroradiologist will read the admission CT. Patients with a finding of intracranial hemorrhage, or possible intracranial hemorrhage will be excluded from the study and followed appropriately by a physician from the Department of Neurosurgery.

Two weeks after injury, a study nurse will contact patients by telephone, to schedule follow-up CT and clinic evaluation, with follow-up reminders as needed. Follow-up clinical / neurological examination and head CT will be performed 3-5 weeks after injury. Achieving follow-up CT 3-5 weeks after the injury is routine in the patients' population included in this study.

At final assessment participants will be evaluated for clinical or imaging evidence of CSDH as well as other intracranial hemorrhage, or cardiac, thrombotic, or embolic condition. In the case of a clinically significant adverse cardiovascular or cerebrovascular event, the IRB will be notified, and the patient will be treated as appropriate for the clinical situation. At study completion, LDA prophylaxis will be prescribed for individual patients as clinically indicated, with cardiology or other consultation as appropriate.

Informed Consent The attending neurosurgeon will fully explain the basis for the study and its clinical importance, as well as the risks involved with the continuation or cessation of LDA prophylaxis, and invite the eligible patient to participate in the study. Eligible patients who are fully conscious and who agree to participate in the study, or their legally authorized representatives, will be asked to sign an informed consent form (Appendix 3). Patients with no legal guardian who suffer from dementia or any other concurrent illness which prevents them from giving informed consent, will be excluded from the study.

Randomization Randomization of all participants will be performed remotely via an online program.

Patients in arm 1 will be requested to continue their aspirin therapy according to previous instructions; those in arm 2 will be requested to cease aspirin therapy for the initial 3-5 weeks following their mild head trauma.

Primary Endpoint

• Occurrence of CSDH Secondary Endpoints
• Intervention for surgical evacuation of CSDH
• Occurrence of other clinically significant intracranial hemorrhage
• Intervention for surgical evacuation of other intracranial hemorrhage
• Occurrence of clinically significant cerebrovascular, cardiovascular, thrombolic, or embolic event
• Intervention for treatment of cerebrovascular, cardiovascular, thrombolic, or embolic event

The possible relationship between LDA prophylaxis and a variety of other demographic and clinical variables will also be assessed, including patient age and gender, coagulation profile, mechanism of injury, loss of consciousness, and time elapse from traumatic head injury to any clinically significant sequelae. We will assess the relationship between aspirin cessation and incidence of clinically significant events in patients with ischemic heart disease.

Statistical Analysis Since the incidence of CSDH in patients on LDA is has not been studied directly, it is not currently possible to estimate the number of patients required to achieve statistical power for this research. Study enrollment will continue through the first year, with an estimated 100 participants in this period. Based on preliminary findings at that point, we expect to determine the total required study population.

The two-proportions test will be used to assess whether there is a statistically significant difference in the rate of CSDH in patients randomized to cessation of LDA therapy and those randomized to continuation of LDA. Relationships between the explanatory the dependent variables will be explored with classical parametric and nonparametric statistical methods, including multivariate analysis, logistic regression, the two proportions test, and the independence test. Several measures of association/correlation between pairs of variables will be analyzed as well.

With the pilot data from 100 patients, our statistician will be consulted to perform a power analysis to assess the required sample size to have a greater than 80% chance of detecting a significant difference between the two groups at significance level of p < 0.05.