CORR Insights®: Is Growth-friendly Surgical Treatment Superior to One-stage Posterior Spinal Fusion in 9- to 11-year-old Children with Congenital Scoliosis?

Where Are We Now?

Growing rods have provided major advancements for skeletally immature patients with severe spinal deformities. Historically, fusion surgery in very young patients with scoliosis has resulted in a short truncal height and thoracic insufficiency [4]. Growing rods are indicated in patients who still have a considerable amount of remaining growth to allow spinal growth, sufficient thoracic height, and normal lung development. Typically, the goal for T1 to T12 height at skeletal maturity is set at 20 cm to 25 cm, but this varies based on the child’s underlying disease condition and growth potential [3]. After growth is complete, surgery is performed to remove the growing rods and fuse the spine.

Traditional growing rods require twice-yearly operations to mechanically lengthen the rods. These have been largely superseded by magnetically controlled growing rods that can be lengthened in clinic and that require fewer scheduled surgical procedures; nonetheless, 30% to 40% of patients still undergo one or more unscheduled reoperations with this approach [1, 2]. Further, because the points of fixation are limited so as to span the spine and allow for growth, the ability of growing rods to correct a spinal deformity is less than that of fusion, which includes powerful segmental instrumentation. At a minimum, patients with growing rods undergo two operations: one to place the rods and a second to revise the construct to fusion surgery. This final fusion procedure is fraught with complications and is among the most difficult scoliosis procedures to perform. One study found that 20% of these procedures are complicated by unplanned reoperations [7]. The vertebrae spanned by the growing rods are susceptible to stress shielding, with resulting osteopenia that can lead to poor-quality bone fixation. In addition, autofusion may occur during treatment, resulting in a rigid spine [1]. Further, after years of distraction, adjacent deformities such as proximal junctional kyphosis develop that can call for longer fusions, occasionally in the cervical spine. Because of autofusion and osteopenia, from a technical standpoint, the final fusion surgery after growing rod treatment holds greater challenges and fewer rewards than one-stage fusion in a patient with no previous spine surgery.

Thus, the decision to delay surgery and treat a spinal deformity with a single operation versus embark on growing rod treatment remains critically important. This topic has been addressed in 9- to 11-year-old patients with idiopathic scoliosis but has not been fully explored in patients with congenital spinal deformities [6]. In this paper, Xu et al. [9] provided a detailed comparison of growing rod treatment and one-stage posterior spinal fusion for a select population of 9- to 11-year-old children with long segments of congenital deformity. Similar to findings in patients with idiopathic scoliosis, the authors found that one-stage posterior spinal fusion resulted in superior correction to and fewer complications than growing rod treatment, with similar health-related quality of life scores at the final follow-up visit. This is important because this patient population can be spared the growing rod experience, which entails multiple surgical interventions over time with a deleterious impact on healthcare cost, patient quality of life, school and work schedules, and reoperation rates. Xu et al. [9] suggest that patients with multilevel congenital scoliosis will have a satisfactory result with observation followed by a single-stage intervention, which is clearly easier on the patient and families than growing rods with multiple surgeries over time. This paper evaluated a variety of growing spine devices, whereas in many practices today, the standard of care has become magnetically controlled growing rods that eliminate surgery for routine rod lengthening.

Where Do We Need To Go?

For the question of definitive fusion versus growing rod placement, research to date has focused on the transitional population of patients aged 9 to 11 years [6]. One-stage fusion is considered very acceptable for children older than 10 years, so this is a relatively modest question. A more important question is, could a similar strategy (of delaying spinal surgery and avoiding growing rods at the cost of treating a more-severe deformity with a single intervention) be used in an even younger age group? Further, with the advent of nonfusion strategies, it will be important to determine which early-onset scoliosis conditions can be treated with anterior spinal instrumentation and growth modulation, rather than growing rods. A further challenge is how to introduce new technology safely into treatment of pediatric spine deformities with consistent indications to allow for rapid assembling of data and early assessment of safety and efficacy.

How Do We Get There?

To answer the question of whether definitive fixation or growing rod placement is appropriate in younger children, additional research could address whether patients younger than 9 to 11 years might benefit from single-stage treatment. Health-related quality of life, Early Onset Scoliosis Questionnaire-24 and Scoliosis Research Society-22 scores, and curve correction and spinal height at maturity could be compared between matched cohorts of 6- to 9-year-old children with early-onset scoliosis treated with growing rods and those treated with a program of bracing, halo traction, and definitive fusion. Although magnetically controlled growing rods reduce the number of planned operative interventions, they still require four clinic visits per year to perform rod lengthening. Surprisingly, health-related quality of life scores for patients with magnetically controlled growing rods have not shown an improvement compared with those reported by patients with traditional growing rods [8]. A single high-intensity period of orthopaedic treatment (halo traction and fusion) may be less burdensome on families than years of medium-intensity interventions with frequent lengthening procedures in clinic and periodic operations to exchange rods and perform final fusion.

Studies will also need to determine whether early-onset scoliosis can be successfully treated with anterior spinal instrumentation and growth modulation, rather than growing rods. Anterior vertebral body tethering is now approved by the FDA for treating idiopathic scoliosis. These procedures present a novel avenue to foster growth without fusion and preserve some degree of spinal motion over the instrumented vertebrae [5]. Although current instrumentation is not amenable to the treatment of very young patients, it holds much promise for preserving normal spine motion and growth. The anterior thoracoscopic approach is technically challenging, and its long-term effects on pulmonary function, particularly in a young patient, are unknown. However, anterior surgery could be used to treat flexible spine deformities without kyphosis and is well-suited for 9- to 11-year-old children. Similar further advancements in posterior treatment strategies that will allow for instrumentation with an elastic modulus that is more similar to bone may occur for congenital and kyphotic deformities. Much of the growing rod problems (implant pull-out, loss of fixation, junctional kyphosis, autofusion, and stress shielding) are because of the mismatch of placing stiff rods in young, active, flexible patients.

Although the pace of innovation is rapid in early-onset scoliosis, it is important to continue to critically evaluate the application of new technology. Each treatment strategy is likely suited for a specific group of patients, and wide application of a single new technology across multiple diverse conditions is likely to be met with failure. Thus, a careful prospective study of each device according to standardized criteria and outcomes, followed by rapid dissemination of knowledge, is necessary to prevent surgical centers from repeating the same mistakes. Although a single center can initiate a prospective study, multicenter studies with agreed-upon treatment indications should achieve practice convergence and consistent indications. The paper by Xu et al. [9] should compel each of us to reexamine our use of new technology and conclude that sometimes previously established techniques are preferable to the newest approach.