DOMINANT ASPIRIN THERAPY STRATEGY ELIMINATES ROLE FOR IMAGING SCREENING/WARFARIN THERAPY FOR STROKE PREVENTION IN MODEL OF ASYMPTOMATIC BLUNT NECK TRAUMA

Purpose:  

   Blunt neck trauma is associated with a small (1%) risk of major stroke consequent to cervical artery dissection. We model projected costs and outcomes of the most common clinical strategies to prevent stroke from dissection in asymptomatic blunt neck trauma (ABNT), in comparison with a novel treat-all-with-aspirin strategy. 

  Method(s):  

   Pretest probability of dissection from BNT, screening test characteristics, stroke rates, treatment complication rates, utilities, and direct costs were obtained from the literature and modeled over one year, using decision analysis software (TreeAge Pro 2016) for cost-effectiveness analysis.  Seven treatment strategies were assessed, four with various imaging tests and three without (treat-all-aspirin, treat-all-warfarin, and treat-none). One-way sensitivity analyses were performed on prior probability of dissection (Figure 2). A probabilistic Monte Carlo simulation was performed. Acceptability curves of the strategies were created (Figure 1) and the strategies were compared for effectiveness and incremental cost-effectiveness (Table 1).  A value of information analysis was also performed to determine the value of future research, given estimated US incidence of ABNT of 200,000.

  Result(s):  

   The treat-all-aspirin strategy dominates at any given probability of dissection; over a prior probability of dissection range of 0-15%, it is both more effective (0.9968 QALY) and less costly ($95.68) than any other strategy. It also dominates the other strategies given a willingness-to-pay threshold of $75,000.  The value of information analysis revealed an EVPI of $0.12 for cost and 0.0 QALYs for effectiveness, with a population EVPI of $24,000 for cost. 

  Conclusion(s):  

   Previously, we defined the cost-effective limits of imaging screening in a warfarin treatment paradigm to prevent strokes in BNT (poster, AUR 2014), with a role for CT angiography over the pretest probability range of 1.5-15%. The current analysis including the aspirin treatment strategy however would suggest treating all with aspirin to be a nearly perfect strategy for preventing stroke from dissection in asymptomatic blunt neck trauma, dominating all other strategies and eliminating any role for imaging screening.