Appropriately Ordered Brain Mris Based on the American College of Radiology Appropriateness Criteria Often Did Not Lead to a Change in Management

Background:

The United States has the highest expenditure per capita and fastest growth when it comes to health care. Imaging is one of the fastest growing services in medicine reaching approximately $100 billion annually. There have been criticisms that imaging studies are often over utilized and inappropriate which results in unnecessary costs and waste. Our study wished to address whether brain MRIs done at our institution were appropriate based on the ACR(American College of Radiology) criteria and its impact on management. 

Methods:

We reviewed brain MRIs done at Temple University Hospital during the period of January 2012 through May 2012 (n=256) to assess how appropriate they were based on the ACR appropriateness guidelines, their findings and its impact on management.  The ACR stratified brain MRIs into three subcategories: “Usually Not Appropriate”, “May Be Appropriate”, “Usually appropriate”. These subcategories were made by the ACR in efforts to ascertain the necessity of a brain MRI based on the clinical findings and history present. 

Results:

Of the 256 MRI brains we analyzed, 86% were categorized as “usually appropriate”, 14% were “maybe appropriate” and none were categorized as “usually not appropriate”. Only 31% of the “usually appropriate” brain MRI studies had positive findings and 3% of the “maybe appropriate” brain MRIs were positive. Of note, 41% of the brain MRIs that were categorized as usually appropriate lead to a change in management. We evaluated all the references (n=340) regarding brain MRIs that were used in the ACR appropriateness criteria based on the ACR study quality index. Only 4% of the references used were based on well designed studies that accounted for common biases and 83 % of studies had either important study limitations or was considered not useful as primary evidence. 

Conclusions:

Most brain MRIs done at Temple University Hospital during the study period were considered usually appropriate based on the ACR appropriateness criteria. But only a third of these MRIs showed positive findings and less than half of the brain MRIs lead to a change in management. A possible explanation for this could be that the ACR appropriateness criteria for Brain MRIs are based on studies with important limitations and rarely looked at clinical outcomes as their endpoint. High quality studies that focus on the impact imaging tests have on patient’s outcome are needed, to help further reduce unnecessary imaging tests from being done.