When it comes to abdominal CT and MRI, radiologists are making good use of all available prior imaging before recommending additional exams, and they aren't ordering many unnecessary exams, according to a new study published in the Journal of the American College of Radiology.
The study findings contradict a generally held belief among policymakers that medical imaging exams -- particularly additional imaging intended to clarify indeterminate findings -- drive up healthcare costs and should be contained.
Researchers from NYU Langone Medical Center found that only 4% of recommendations for additional imaging (RAIs) for abdominal and pelvic CTs and MRs could have been avoided. The overall low percentage of avoidable additional exams demonstrates that, in general, radiologists effectively utilize prior imaging when crafting their interpretations and recommendations, according to lead author Dr. Ankur Doshi and colleagues (JACR, July 2015, Vol. 12:7, pp. 657-663).
Dr. Ankur Doshi from NYU Langone Medical Center."Overall, radiologists are very detail-oriented, and they make use of available priors," Doshi told AuntMinnie.com. "We weren't surprised that the percentage of avoidable RAIs was low -- but we were interested in finding out which prior studies would prove most useful."
Preventive priors
Prior imaging studies are helpful because they can often show the finding's long-term stability or additional features that demonstrate it is benign.
"Patients frequently have numerous prior studies available for review; each prior examination in a given modality provides an opportunity for potentially improved assessment of the indeterminate finding and avoidance of an unnecessary RAI," the group wrote.
For the study, Doshi and colleagues reviewed 1,006 recommendations for additional imaging from abdominal and pelvic CT and MRI reports; the exams came from inpatient, outpatient, and emergency department encounters. The group then reviewed reports and images from patients' prior imaging examinations, including those of all relevant body parts and modalities, to determine if the recommendation could have been avoided based on previous exams.
A recommendation was classified as avoidable if any prior exam provided any of the following information:
- The indeterminate finding had been stable for two years.
- The indeterminate finding had been definitively characterized in a prior exam.
- The recommended imaging exam had already been performed within the previous six months and had supplied the necessary clinical data.
The researchers found that 41 (4.1%) of the 1,006 RAIs could have been avoided. The key prior examination that established the recommendation as unnecessary was performed with a different modality in 53.7% of cases, and it was conducted on a different body area in 41.5% of cases.
The key finding was present on the prior images but not described in the report in nearly half of the 41 cases (46.3%). Not surprisingly, more prior exams were available in cases with avoidable recommendations than unavoidable ones.
| Key prior imaging exams, by modality type | ||
| Modality | Key prior exam for avoidable RAIs on CT | Key prior exam for avoidable RAIs on MR |
| CT | -- | 33.3% |
| MR | 28.1% | -- |
| Ultrasound | 18.8% | 11.1% |
| PET/CT | 3.1% | -- |
| X-ray | 3.1% | -- |
| Fluoroscopy | -- | 11.1% |
The researchers also found that the adrenal gland had a high rate of avoidable recommendations for additional imaging (18.2%), due to the failure of the radiologist to review prior chest CT exams.
"One of the interesting things that came out of this study is that prior chest imaging is really helpful for assessing indeterminate upper abdomen findings," Doshi said. "And it's better for radiologists to look at the images, rather than relying on the report."
PACS can help
Even though the percentage of avoidable recommendations was low, it's still important to further reduce their occurrence, according to Doshi and colleagues. PACS can help, but it must be customized. Scrolling through a large number of prior examinations can be burdensome if the PACS doesn't flag pertinent priors.
"Optimal PACS structures help highlight relevant studies and make images easily available for review," Doshi told AuntMinnie.com. "It's crucial to have strong information systems and IT support."
