Introduction
Epidural lipomatosis has been identified as a spinal stenosis etiology since 1975.
With longer life expectancy, more chronic diseases and cancers are being treated with steroids. Also, given the obesity epidemic, lipomatosis from obesity has been recognized increasingly in stenosis. Epidural lipomatosis is associated with protease inhibitor treatment of HIV.
There is debate whether epidural steroid injections are a cause of epidural lipomatosis or a treatment (emulsifying fat).
Rustom felt epidural lipomatosis was 55% from chronic steroids, 25% obesity, 17% idiopathic and 3% Cushing’s disease. Significant risk occurred with a mean daily dose of 30-100 mg prednisone and mean duration 5-11 years. Their youngest patient was a child age 6.
Epidural lipomatosis diagnostic criteria is a fat layer >7mm or >40% cross sectional area of the spinal canal. Lumbar spine is the most frequent site, but cases have been described in the thoracic and cervical areas.
Borre performed a retrospective study of 2528 patients with lumbar stenosis. There were 1095 men, 1433 women, mean age 47. The percentage of cross-sectional area occupied by fat was measured on axial MRI at the S1 upper end plate. Lipomatosis >40% of the spinal canal was found in 41% of patients.
Methods
Despite our increased awareness, 41% of Borre’s patients having epidural lipomatosis seemed relatively high. Our co-author RH has kept a Denver registry of >12,000 injections for the spine, SI joints and hips. The registry was queried for (a) number of patients with diagnosis of spinal stenosis in the thoracic and lumbar spine and (b) number with a secondary diagnosis of epidural lipomatosis.
Results
In the Denver registry, 1758 patients had a primary spinal stenosis diagnosis, 31 had a secondary diagnosis of epidural lipomatosis, an incidence of 1.8% (vs. 41% per Borre). Denver average age 61.6 (vs. 47 Borre).
Lipomatosis etiology differed. Denver (vs. Rustom) idiopathic 52% (17%), obesity 22% (25%), previous multiple ESI 13% (not reported), exogenous steroids 10% (25%), HIV 3% (not reported), Cushings 0% (3%).
All 31 Denver patients were treated by interlaminar epidural steroid injections. Symptoms improved in 22/31=71% and surgery avoided. Surgery was performed in 9 patients. Eight were candidates for microsurgical decompression using McCulloch’s bilateral decompression via a unilateral approach developed for degenerate stenosis. Only 1 patient required multilevel lumbar laminectomy.
Discussion
Epidural steroid injections appear to be a treatment and a cause of epidural lipomatosis.
Patient population may impact incidence. Denver average age was 61.6 (vs. 47 Borre), thus, the Denver registry likely had more patients with degenerative stenosis. Also, the Colorado population is the leanest in the US.
For surgical planning, the key is whether lipomatosis is a continuous mass narrowing the entire canal i.e. tube type stenosis vs. segmental hourglass distribution (typical of degenerate stenosis). Pathoanatomy dictates midline laminectomy for adequate decompression (tube stenosis) vs. segmental microsurgical decompression (hourglass stenosis).
Surgical tip: the fat consistency in epidural lipomatosis is generally more dense, organized and adherent to the dura than storage fat. Epidural lipomatosis generally requires direct removal, not flushing.