Brendan J. McCullough MD, PhD, Presenter: Nothing to Disclose

Germaine Richard Johnson MD, Abstract Co-Author: Nothing to Disclose

Brook I. Martin MPH, Abstract Co-Author: Nothing to Disclose

Jeffrey Gil Jarvik MD, MPH, Abstract Co-Author: Consultant, HealthHelp Founder, PhysioSonics, Inc Stockholder, PhysioSonics, Inc Stockholder, Nevro Imaging, Inc Research Advisory Board, General Electric Company

To examine in patients with uncomplicated low back pain or radiculopathy the association between epidemiological information added to lumbar spine MRI reports and subsequent imaging and treatment by primary care providers. Our hypothesis was that including epidemiological information would decrease subsequent imaging and treatment.

Since 2005, a brief statement referencing the prevalence of lumbar spine MRI findings in asymptomatic individuals (Jarvik et al. Spine 2002) was arbitrarily added to lumbar spine MRI reports. We queried the radiology information system for patients of primary care physicians who had lumbar spine MRIs for uncomplicated low back pain or radiculopathy. Our primary outcome of interest was subsequent cross-sectional imaging of the lumbar spine within one year. Secondary outcomes were narcotics prescriptions, injection treatments, and surgery. Comparisons were performed using the chi-squared test and multivariate logistic regression.

We identified 71 patients with lumbar spine MRI reports containing the statement and 167 controls without. Baseline characteristics of both groups were similar, although the control group had a non-significant greater proportion of patients with “moderate or severe” degenerative findings than the statement group (55% versus 44%, P = 0.11). Re-imaging with MR or CT was seven times more common in the control group (7% versus 1%, P = 0.12), although this was not statistically significant. Narcotics prescriptions following receipt of the MRI report for low back pain were three times higher in the control group (22% versus 7%, P = 0.01 adjusted for severity of MRI findings). Similarly, the control group was borderline significantly more likely to receive physical therapy (36% versus 24%, P = 0.07). The rates of lumbosacral steroid injections and surgery were similar between groups.

Epidemiological information in lumbar spine MRI reports is associated with significantly decreased rates of narcotics prescriptions by primary care providers and may also lower rates of subsequent high cost imaging.

Management of low back pain by primary care providers may be influenced by a simple statement as to the prevalence of MRI findings in asymptomatic individuals.

McCullough, B, Johnson, G, Martin, B, Jarvik, J, Lumbar Imaging and Reporting Epidemiology (LIRE) Study: Does Epidemiological Data in Reports Affect Clinical Management?.  Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL. http://archive.rsna.org/2010/9013313.html