Magnetically controlled growing rods (MCGR) are commonly implanted for the treatment of early-onset scoliosis. While most authors report favorable short-term results, little is known about long-term deformity correction. This prospective cohort study assesses spinal deformity control in a homogeneous spinal muscular atrophy (SMA) patient group treated with MCGR implants, a standardized lengthening protocol and a minimum follow-up of four years. 17 SMA patients with progressive scoliosis were treated with MCGR implanted parallel to the spine with rib-to-pelvis fixation. Radiologic measurements were performed before and after MCGR implantation and during external lengthening procedures. These included measurements of the scoliotic curve, kyphosis, lordosis, pelvic obliquity and the spinal length. Additional clinical data of the complications were also analyzed. 17 children (mean age 7.4 years) were surgically treated and underwent a total of 376 lengthenings. Complication rates were 3.5% in respect to all interventions or 41% of the patients had complications during 3.5% of the lengthening sessions. The initial implantation significantly reduced the main scoliotic curve by 59%, with the correction remaining constant throughout the follow-up. Pelvic obliquity was also significantly and permanently corrected by 72%, whereas kyphosis and lordosis were not influenced. The spinal length could be significantly increased mostly during the first year of treatment. Bilateral implantation of MCGRs for correction of spinal deformity in children with SMA showed no decrease of the lengthening potential during a four-year follow-up. Therefore, the previously described ‘law of diminishing returns’ could not be applied to this patient population.Level of Evidence/Clinical relevance: Therapeutic Level IV.