Obesity and spinal epidural lipomatosis in cauda equina syndrome.

BACKGROUND CONTEXT

Although lumbar disc herniations are common, only a small portion of these herniations lead to cauda equina syndrome (CES), which is an uncommon but debilitating disorder. Why some patients with herniation develop CES, when most do not, remains unknown. Preexisting subclinical epidural lipomatosis may limit canal space such that an otherwise benign herniation causes CES.

PURPOSE

This study determines whether patients with an acute disc herniation and CES have a greater body mass index (BMI) and greater quantity of epidural fat compared with control subjects with non-CES symptomatic lumbar herniated discs.

STUDY DESIGN/SETTING: A retrospective case-control series at a university-based level-1 trauma center was carried out.

PATIENT SAMPLE

There were 33 CES and 66 control subjects identified from a prospectively maintained database of patients who underwent surgical management for a lumbar disc herniation between 2007 and 2012. Each CES case had two non-CES control patients matched by gender and age within 5 years except 5 CES cases that matched only one non-CES control.

OUTCOME MEASURES

The outcome measures included weight, height, age, gender, and BMI. Radiographic outcome measures included the proportion of lumbar spinal canal occupied by fat and herniated disc on preoperative magnetic resonance imaging.

METHODS

Patient charts and preoperative radiographs were retrospectively reviewed. For each patient, a blinded reviewer determined the proportion of lumbar spinal canal occupied by fat, and the maximal proportion of the canal occupied by herniated material at the involved level. Patient demographics and radiographic measures were compared between CES and control groups using chi-square or Student t tests. A second blinded reviewer re-assessed a series of radiographs, and the intraobserver variability was determined by Spearman correlation. Logistic regression was used to model the preoperative factors associated with having an acute disc herniation and CES.

RESULTS

The CES cases had higher BMI (31.8 kg/m, 95% confidence interval [CI] 29.5-34.0 vs. 28.1 kg/m, 95% CI 26.7-29.5 in controls; p=.007), focally narrower canals (14.6 mm, 95% CI 13.8-15.3 mm vs. 16.4 mm, 95% CI 15.4-17.3 mm in controls; p=.003), and a greater percentage of spinal canal occupied by epidural fat (31.3%, 95% CI 26.1%-36.6% vs. 21.9%, 95% CI 18.7%-25.1% in controls; p=.003) and herniated disc material (54.5%, 95% CI 46.9%-62.0% vs. 34.4%, 95% CI 30.3%-38.5% in controls; p<.0001). Logistic regression confirmed canal width at the involved level, BMI, amount of canal occupied disc, and proportion of canal occupied by fat as independent predictors of having an acute disc herniation and CES.

CONCLUSIONS

Obesity is a risk factor for CES from disc herniation. The CES cases also had a greater amount of herniated material, focally narrower canal, and larger epidural fat deposits. The latter may be the mechanism linking obesity with CES.