Impact of the Spring Distraction System on scoliotic spine morphology: an MRI- and CT-based analysis
Summary
Study Design
Prospective cohort study
Summary of Background Information
Early Onset Scoliosis (EOS) is a three-dimensional (3D) deformation of the spine and trunk below the
age of 10, which can lead to pulmonary insufficiency and even death. At UMC Utrecht, the Spring
Distraction System (SDS) was developed to provide dynamic and continuous reduction of the scoliotic
spine while allowing for near-physiological spinal growth. This study investigated morphometric
differences in the intervertebral discs (IVDs) and vertebral bodies (VBs) before and at an average of 1.5
years after SDS implantation surgery in a prospective study cohort. For this analysis, MRI imaging and
CT-based imaging was used to assess the 3D shape of individual IVDs and VBs, as well as of the relative
position of the nucleus pulposus (NP) in the annulus fibrosus.
Methods
3D morphology of peri-apical levels of ten EOS patients were compared before surgery and at least 1-
year after surgery using MRI- and CT-based imaging. Peri-apical VBs and IVDs were semiautomatically segmented to determine the lengths of the anterior, posterior, convex and concave aspects
of these structures, as well as their rotation, while taking their 3D-orientation into account. Additionally,
NP morphology was analyzed using MRI-based IVD reconstructions.
Results
Primary thoracic curves showed a height increase in every segment, except for the IVD convexity, with
a mean gain of 7.6 ± 3.1 mm. This increase was mainly seen in the VB (7.0 ± 5.2 mm), compared to the
IVD (0.6 ± 3.1 mm) Relative changes were most prevalent along both the VB and IVD concavity
showing 17% and 13% height increase respectively, compared to 11% and -6% in the convexity.
Reduction of the deformity was mainly seen in the IVD where concave/convex length ratios normalized
from 0.65 to 0.80, translating to 5.9 mm of coronal wedging reduction. Anterior/posterior ratios
normalized from 1.25 to 1.21, indicating less lordosis. The VB also showed wedging reduction, but to a
much lesser extent. Here, concave/convex ratios normalized from 0.85 to 0.89, indicating a 1.8 mm
coronal correction, however anterior/posterior ratios increased from 0.97 to 1.00.
Both VB and IVD rotation appeared stable pre- versus post-surgery showing a positive correlation
between rotational severity and (peri-)apical level.
The NP center of volume showed a pre-surgery mean distance to the IVD center of volume of 4.4 ± 1.1
mm. This distance decreased to 2.6 ± 1.0 mm post-surgery.
Conclusion
The reduction of scoliotic spines in EOS patients using the SDS results in a transition towards a more
physiological shape, affecting both IVD and VB morphology. Most correction of the deformity takes
place in the IVD while the spine continues to grow in the VBs. A trend is shown where this bony growth
is modulated, resulting in decreased coronal wedging. Additionally, a trend is observed where the NP
takes on a more centralized position in the IVD.