Early MRI Versus CT Scan for Evaluation of Cerebrovascular Events in a Community Hospital: A Cost Minimization...

Diffusion-weighted MRI is shown to be equally effective, as a CT scan, in diagnosing ischemic and hemorrhagic strokes. Would it be cost-effective to perform an early MRI instead of a CT head?

A retrospective chart review was conducted between October 1, 2015, through October 1, 2017, for patients admitted for possible cerebrovascular accident (CVA). Inclusion criteria were age >/= 18 years and symptoms suggestive of a stroke. Exclusion criteria were pregnancy and age <18 years. We obtained information regarding patients' length of hospital stay, imaging modalities performed, and the related cost. We performed a cost analysis by calculating the total duration and cost of hospitalization, and cost for each investigation.

The study included 828 patients who underwent CT head without contrast initially. A total of 634 (76.5%) patients got MRI brain without contrast, 261 (31.5%) had MRI brain with and without contrast, 406 (49%) had magnetic resonance angiography (MRA) head without contrast, 60 patients (7.2%) had MRA neck without contrast, 272 (32.8%) had MRA neck with and without contrast, and 1 patient (0.1%) had MRA head with and without contrast. The hospital duration for all patients was 1,797 days. The average duration per patient was 1.9364 days. The total health care cost for all patients was $25,383,983. Average per patient hospitalization cost was $25,383,983/828 = $30656.98. Average per day cost for all patients would be $25383983/1.93days = $13,152,322.8. Combined costs of all MRIs performed on all patients = $1,413,014. If MRI brain with and without contrast was considered as an initial modality, a total of $335,340 can be saved on the diagnostic imaging. Ultimately, it can also help reduce the hospitalization duration.

Early MRI (in appropriately selected patients) can reduce the length of hospitalization and cut some health care costs. However, more studies are required to develop appropriate patient selection criteria.