Underreporting of spinal epidural lipomatosis: A retrospective analysis of lumbosacral MRI examinations from different radiological settings.

PURPOSE: The purpose of this study was to assess the rate with which radiologists reported spinal epidural lipomatosis (SEL) when interpreting lumbosacral magnetic resonance imaging (MRI) examination.

MATERIALS AND

METHODS

A total of 450 lumbosacral MRI examinations obtained in 450 patients were included (199 men, 251 women; mean age, 56.7 ± 13.5 [SD] years; age range: 18-91 years).  Three senior radiologists assessed and classified independently SEL on MRI according to the Borré grading system (Grade 1 to Grade 3). Depiction of SEL on MRI reports (i. e., reporting rate) and association with patients' symptoms were verified.

RESULTS

SEL was found in 75/450 patients (prevalence = 16.7%), and classified as grade-1 (mild) in 49/75 (65.3%) patients, grade-2 (moderate) in 24/75 (32%), and grade-3 (severe) in 2/75 (2.7%). SEL was diagnosed on MRI report in 6/75 (8%) patients. SEL prevalence based on MRI reports was 1.3% (6/450), significantly lower than its actual prevalence based on MRI examinations (P < 0.0001). The reporting rate was 0% in grade-1 (0/19), 10.2% in grade-2 (5/49) and 50.0% in grade-3 (1/2), and variable on the radiologist subspecialty (10.0% among musculoskeletal radiologists, 11.1% among neuroradiologists, and 3.7% among generalists). SEL was considered as the only cause of symptoms in 7/75 patients (9.3%) and a concurrent cause in 9/75 (12%).

CONCLUSION

SEL reporting rate is extremely low, leading to an important underestimation of disease prevalence. SEL diagnosis and grading should be refined to improve reports quality and subsequently patient care.