Predictors of Rod Length Gain and Sagittal Alignment Change After Magnetically Controlled Growing Rod Lengthening

Abstract

Study design: This was a retrospective cohort study.

Objective: There are preoperative characteristics that predict rod lengthening of magnetically controlled growing rods (MCGR). The lengthening of MCGR will lead to increased kyphosis.

Summary of background data: The amount of length gained by MCGR is variable, and predictors are lacking. Similarly, sagittal plane changes over the MCGR lengthening period have not been investigated.

Materials and methods: Patients with MCGR and a minimum 2-year follow-up were identified and those with previous spine surgery were excluded. Preimplantation, postimplantation, and last follow-up postlengthening radiographs were examined. Multiple linear regression analyses were used for rod length gain predictors.

Results: Fifty-six patients with early-onset scoliosis met inclusion criteria: nine idiopathic, five congenital, 14 neuromuscular, 20 syndromic, and eight skeletal dysplasia patients. No difference was seen between subtypes of early-onset scoliosis for rod length gain ( P =0.62). Shorter preoperative T1-T12 height and higher curve correction rate after implantation were significant predictors for rod length gain ( P <0.001). Preoperative major curve magnitude and kyphosis were not significant predictors. Flattening of the spine around the actuator and compensative increase in T1-T5 kyphosis were seen secondarily after MCGR implantation. Maximum sagittal kyphosis, T1-T5 kyphosis, and T5-T12 kyphosis did not change during lengthening. Lumbar lordosis significantly decreased postimplantation (first erect) but then increased during lengthening. Pelvic incidence and sacral slope both increased during lengthening, but this may be age related.

Conclusion: Patients who gained the most rod length with MCGR were those with an initially shorter T1-T12 height and better initial curve correction at implantation, likely representing that implants work best in small patients with flexible curves. Diagnosis, preoperative curve magnitude, and thoracic kyphosis were not related to rod length gain. Thoracic kyphosis did not deteriorate over the lengthening phase.

Level of evidence: Level III-retrospective cohort study.