Magnetically Controlled Devices Parallel to the Spine in Children with Spinal Muscular Atrophy

Abstract

Background: Children with severe spinal deformity frequently are managed with growth-friendly implants. After initial surgery, externally controlled magnetic rods allow spinal deformity correction during growth without further surgical intervention. The ability to lengthen the spine without additional surgical procedures is especially beneficial in high-risk children, such as those with spinal muscular atrophy (SMA). The purpose of the present study was to assess the level of control of spinal deformity in a homogeneous group of patients with SMA who were managed with magnetically controlled implants for 2 years.

Methods: This prospective, nonrandomized study included 21 non-ambulatory children with type-II SMA and progressive scoliosis who were managed bilaterally with a magnetically controlled implant that was inserted parallel to the spine with use of rib-to-pelvis hook fixation. Radiographic measurements of scoliotic curves, kyphosis, lordosis, pelvic obliquity, and spinal length were performed before and after implantation of the magnetically controlled device and during external lengthening. The mean duration of follow-up was 2 years.

Results: The mean main curve of patients without prior vertical expandable prosthetic titanium rib (VEPTR) treatment decreased from 70° before implantation of the magnetically controlled device to 30° after implantation of the device. Correction was maintained during the follow-up period, with a mean curve of 31° at the time of the latest follow-up at 2.2 years. Pelvic obliquity was surgically corrected by 76% (from 17° to 4°) and remained stable during follow-up. Thoracic kyphosis could not be corrected within the follow-up period. Spinal length of children without prior spinal surgery increased by >50 mm immediately after device implantation and steadily increased at a rate of 13.5 mm/yr over the course of treatment. During treatment, 4 general complications occurred and 6 lengthening procedures failed, with 3 patients requiring surgical revision.

Conclusions: Bilateral implantation of an externally controlled magnetic rod with rib-to-pelvis fixation represents a safe and efficient method to control spinal deformity in children with SMA, achieving sufficient and stable curve correction as well as increased spinal length. The complication rate was lower than those that have been described for VEPTR and other growing rod instrumentation strategies.

Level of evidence: Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.

Fig. 1

Posteroanterior (Figs. 1-A, 1-B, and 1-C) and lateral (Figs. 1-D, 1-E, and 1-F) radiographs of the spine of a 7-year-old girl with SMA and spinal deformity. The main curve was corrected from 58° (Fig. 1-A) before implantation of the magnetically controlled device to 12° (Fig. 1-B) after implantation. This result was maintained over the course of the 2-year follow-up (Fig. 1-C). In the sagittal plane, kyphosis in the thoracolumbar junction (Fig. 1-D) was initially corrected (Fig. 1-E), but increasing thoracic kyphosis occurred over the course of treatment (Fig. 1-F).

Fig. 2

Chart showing the development of the main curve angle before and after each intervention. The values are given as the mean, and the I-bars indicate the standard deviation. The first n value shown in parentheses indicates the number of radiographs analyzed before each intervention, and the second n value indicates the number of radiographs analyzed after each intervention. *P < 0.05. ***P < 0.001.

Fig. 3

Chart showing the age-dependent development of the main curve angle after each intervention. The values are given as the mean, and the I-bars indicate the standard deviation. Patients are grouped according to their age (<6 years of age and 6 to 10 years of age). The n values indicate the numbers of patients in each age group. Two children (>10 years of age) were excluded from the analysis. *P < 0.05.

Fig. 4

Chart showing the development of the pelvic obliquity before and after each intervention. The values are given as the mean, and the I-bars indicate the standard deviation. The first n value shown in parentheses indicates the number of radiographs analyzed before each intervention, and the second n value indicates the number of radiographs analyzed after each intervention. *P < 0.05. ***P < 0.001.

Fig. 5

Chart showing the development of age-adapted thoracic kyphosis before and after each intervention. The values are given as the mean, and the I-bars indicate the standard deviation. The first n value shown in parentheses indicates the number of radiographs analyzed before each intervention, and the second n value indicates the number of radiographs analyzed after each intervention.

Fig. 6

Chart showing the development of age-adapted lumbar lordosis before and after each intervention. The values are given as the mean, and the I-bars indicate the standard deviation. The first n value shown in parentheses indicates the number of radiographs analyzed before each intervention, and the second n value indicates the number of radiographs analyzed after each intervention.

Fig. 7

Chart showing the development of spinal length before and after each intervention. The values are given as the mean, and the I-bars indicate the standard deviation. The first n value shown in parentheses indicates the number of radiographs analyzed before each intervention, and the second n value indicates the number of radiographs analyzed after each intervention. ***P < 0.001.