Cost-Effectiveness of Universal Screening for Blunt Cerebrovascular Injury: A Markov Analysis

Abstract

Background: Blunt cerebrovascular injury (BCVI) is a significant cause of morbidity and mortality after blunt trauma. Numerous screening strategies exist, although which is used is institution- and physician-dependent. We sought to identify the most cost-effective screening strategy for BCVI, hypothesizing that universal screening would be optimal among the screening strategies studied.

Study design: A Markov decision analysis model was used to compare the following screening strategies for identification of BCVI: (1) no screening; (2) Denver criteria; (3) extended Denver criteria; (4) Memphis criteria; and (5) universal screening. The base-case scenario modeled 50-year-old patients with blunt traumatic injury excluding isolated extremity injures. Patients with BCVI detected on imaging were assumed to be treated with antithrombotic therapy, subsequently decreasing risk of stroke and mortality. One-way sensitivity analyses were performed on key model inputs. A single-year horizon was used with an incremental cost-effectiveness ratio threshold of $100,000 per quality-adjusted life-year.

Results: The most cost-effective screening strategy for patients with blunt trauma among the strategies analyzed was universal screening. This method resulted in the lowest stroke rate, mortality, and cost, and highest quality-adjusted life-year. An estimated 3,506 strokes would be prevented annually as compared with extended Denver criteria (incremental cost-effectiveness ratio of $71,949 for universal screening vs incremental cost-effectiveness ratio of $12,736 for extended Denver criteria per quality-adjusted life-year gained) if universal screening were implemented in the US. In 1-way sensitivity analyses, universal screening was the optimal strategy when the incidence of BCVI was greater than 6%.

Conclusions: This model suggests universal screening may be the cost-effective strategy for BCVI screening in blunt trauma for certain trauma centers. Trauma centers should develop institutional protocols that take into account individual BCVI rates.