Risk factors for radiologic adjacent segment degeneration after lumbar fusion

Abstract

Adjacent segment degeneration (ASD) is one of the major complications after lumbar fusion. Several studies have evaluated the risk factors of ASD. Although the paraspinal muscles play an important role in spine stability, no study has assessed the relationship between paraspinal muscle atrophy and the incidence of ASD after lumbar fusion. The purpose of this clinical study was to verify the known risk factors of ASD such as body mass index (BMI), pre-operative adjacent facet joint degeneration, disc degeneration, and to investigate the relationship between paraspinal muscle atropy and ASD.

To calculate the appropriate sample size for the study, we performed a pre-study analysis of the paraspinal muscle cross-sectional area (CSA), and estimated that at least 35 cases would be needed for each group. Among the 510 patients who underwent posterior lumbar fusion for degenerative lumbar disease between January 2009 and October 2009, a total of 50 patients with ASD after surgery were selected. Another group of 50 matched patients with degenerative lumbar disease without ASD after spinal fusion were selected as the control group. Each patient in the ASD group was matched with a control patient according to age, sex, fusion level, and follow-up period.

The risk factors considered were higher BMI, pre-operative adjacent segment disc and facet degeneration, and pre-operative paraspinal muscle atrophy and fatty degeneration. The radiographic data were compared between the ASD and control groups, to determine the predictive factors of adjacent segment degeneration after posterior lumbar fusion by using logistic regression analysis.

Multivariate logistic regression analysis indicated that higher BMI (OR: 1.353, p = 0.008), preoperative facet degeneration on computed tomography (CT) examination (OR: 3.075, p = 0.011), disc degeneration on magnetic resonance imaging (MRI) (OR: 2.783, p = 0.003), fatty degeneration (OR: 1.080, p=0.044), and a smaller relative cross sectional area (CSA) of the paraspinal muscle preoperatively (OR: 0.083, p = 0.003) were significant factors for predicting the development of ASD.

In conclusion, the occurrence of radiologic ASD is most likely multifactorial, and is associated with a higher BMI, pre-existing facet and disc degeneration on preoperative examination, and a smaller pre-operative relative CSA of the paraspinal muscle on MRI.