Early-onset scoliosis (EOS) is comprised of a myriad of conditions, united by the documentation of scoliosis in young children. While authors do not agree on the older age limit at the time of diagnosis for EOS, with some reports including children 9 years or below at presentation, for the purpose of this article EOS will be defined as the onset of scoliosis at age 5 years or below.
EOS is not a diagnosis, but rather defined as an age of onset of spinal coronal plane deformity. As such, EOS includes spinal deformity resulting from congenital malformations, from neuromuscular conditions, from inherited bone dysplasias and syndromes, and in idiopathic cases with no underlying disorder. As EOS results from such a wide variety of etiologies, the natural history varies widely and in many cases is established by the child’s diagnosis which produces the spinal deformity. For example, the natural history of scoliosis occurring in infancy or early childhood due to spinal muscular atrophy is dependent on the form of spinal muscular atrophy rather than on the presence of scoliosis. Similarly, the natural history of patients who have bone dysplasias may be more dependent on the form of dysplasia than on the presence and magnitude of the coronal plane spinal deformity. It is of utmost importance for the treating physician to be familiar with the associated medical and surgical abnormalities associated with the various diagnoses in children with EOS.1
INFANTILE IDIOPATHIC SCOLIOSIS (IIS)
Patients diagnosed by 3 years of age or below with scoliosis without congenital vertebral malformations, neural axis abnormalities, and syndromes are labeled as having IIS. These children are considered to be otherwise normal, presenting with scoliosis that is either noted clinically or incidentally on chest or abdominal radiographs.
The natural history of IIS is linked to the likelihood of curve progression. Curve progression can be predicted based on several factors. Mehta2 was instrumental in establishing the natural history of IIS. Deformity progression is linked to the curve magnitude at presentation, being either ≤25 or ≥25 degrees. Smaller curves are more likely not only to be nonprogressive, but, unlike in adolescent idiopathic scoliosis, to actually resolve with age. In curves whose Cobb angles measure >25 degrees, the presence of overlap of the rib heads at the apex of the curve onto the vertebral bodies on an anteroposterior radiograph, labeled as the “phase” of the rib, is predictive. Those patients with phase 2 ribs which overlap the margins of the vertebrae are more likely to progress, and therefore merit treatment. Finally, in patients who do not have rib overlap (phase 1 ribs), quantification of the rib vertebral angle difference of Mehta can help sort patients who are unlikely to progress, with rib vertebral angle difference <20 degrees, from those who are likely to worsen.
The natural history of untreated progressive IIS is dismal, with a report by Scott and Morgan documenting progression of curves from 30 to 100 degrees. Moreover, 4 patients of 28 died before the age of 20 years of cardiorespiratory disease.3 Relentless curve progression, in the absence of treatment, results in increasing chest wall deformity. Rib rotation and curve progression produce restrictive pulmonary disease with worsening pulmonary function as documented by diminishing forced vital capacity (FVC) and total lung volume. Untreated, the spinal deformity produces chest wall rotation, which obliterates the space available for the lungs.
Pehrsson and colleagues studied mortality in idiopathic scoliosis, dividing the patients by age of onset. They found no significant increase in mortality in patients with adolescent-onset spinal deformity, the intermediate effect on lifespan in children diagnosed between the ages of 3 and 10 years, but significantly earlier age at death in 29 untreated children diagnosed before the age of 3 years with otherwise idiopathic scoliosis.4 The earlier the appearance of spinal deformity, the greater the disturbance of the chest with limited ability of the lungs to grow normally.
Because of the dismal natural history of untreated IIS, today’s spinal surgeon is tasked with identifying those patients likely to progress, and treating those children in a manner that facilitates pulmonary development and halts curve progression. The application of serial derotation casts in the manner of Mehta can halt or limit progression of scoliosis, and in some children, correct the curve in the long term.5 Mehta6 reported 2 groups of patients in which she applied serial derotation casts. In her first group, the children were very young, averaging 1 year and 7 months at treatment, and had smaller curves averaging 32 degrees. After the application of a mean of 5 casts, and then brace treatment, curves resolved and required no further treatment. In these children, the natural history of progressive cardiopulmonary compromise was altered with a favorable outcome. A second group of patients was slightly older (average age, 2.5 y) and had larger curves (average, 52 degrees) at the onset of serial casting. While the curves did not resolve, posterior spinal fusion was delayed until the age of above 10, and one may deduct that lifespan should be normal in these treated children.
Currently, Mehta derotation serial casting is an important part of the armamentarium of treatment options for not only IIS, but also syndromic and select congenital patients. Studies have shown a delay in the need to begin growth-friendly expansion surgeries, with a lower complication rate in patients in whom invasive surgery is delayed4 (Figs. 1A–C).
Congenital scoliosis is defined as spinal deformity resulting from the presence of ≥1 vertebral malformations. These malformations occur during embryogenesis at the fetal age of 6 weeks, a time period critical for the proper formation of the heart and kidneys as well. As many patients with congenital scoliosis have concomitant cardiac malformations, their prognosis in some children is directly related to the severity of their cardiac disease rather than their scoliosis.
McMaster and Ohtsuka7 published the seminal paper on progression in congenital scoliosis. Patients who have balanced congenital malformations, such as block vertebrae that are congenitally fused together, have little risk of curve progression, and their natural history is favorable. Those babies with unbalanced malformations, such as hemivertebrae contralateral to fused unilateral bars, have a very high likelihood of progression of scoliosis with growth. In the past, these children were treated with spinal fusion as soon as curve progression was documented on serial radiographs. After the publication of the crankshaft phenomenon by Dubousset,8 early anterior and posterior combined fusion became the treatment of choice for young children with progressive congenital scoliosis. It was believed that the obliteration of continued curve progression due to asymmetric growth by definitive fusion was necessary to reduce the likelihood of future severe spinal deformity and cardiopulmonary compromise. Not until 2002 were studies published negating this belief.
Goldberg and colleagues followed 43 patients who had undergone in situ fusion of congenital scoliosis (37 posterior only) to skeletal maturity. Despite successful fusion of the congenital segment, revision surgery was required in 26% due to progression of deformity.9
While complex congenital curves can result in severe deformity, simple segmented lumbar and lumbosacral hemivertebrae can produce progressive trunk shift due to worsening lumbosacral takeoff. These young children can be successfully treated with hemivertebral resection and limited (2 or 3 levels) spinal fusion. Recent reports of the results of resection of these hemivertebrae are positive, and the natural history of these treated limited fusions is likely benign following successful resection and fusion (Figs. 2A, B).10–12
More problematic are the complex multisegment congenital spinal deformities. These patients have limited growth potential in the abnormal vertebrae, which can result in shortening of the thoracic spine and therefore space available for the lungs. In addition, patients with thoracic congenital deformities frequently also have abnormalities in the ribs, with either rib fusions which limit intercostal movement during respiration, or missing ribs, resulting in paradoxical thoracic movement with diaphragmatic breathing. These children present the greatest difficulty to the pediatric spine surgeon.
Robert Campbell coined the term thoracic insufficiency syndrome in 2003, defined as the inability of the thorax to support normal respiration or growth.13 In his landmark paper, he proposed that thoracic insufficiency can result from progressive spinal deformity, from rib abnormalities, and/or from early fusion which limits the growth potential of the spine, and therefore the chest (Fig. 3).
THORACIC INSUFFICIENCY SYNDROME
Thoracic insufficiency syndrome is prevalent in patients with EOS because the presence or the fusion of the spinal deformity results in a lack of spinal growth. Dimeglio contributed greatly to the field of EOS by publishing studies on the growth of the thoracic spine.14 Thoracic growth is most rapid in the first 5 years of life, then decreases in velocity until the adolescent growth spurt. During the period of rapid early growth, the lungs also grow, with multiplication of alveoli occurring most rapidly until the age of 2. It has been proposed that the number of alveoli is established by the age of 8 years.15,16 Therefore, fusion of the spine before this age results in inadequate thoracic volume to support the growth of the lungs.
Pulmonary function following early thoracic fusion has been studied in both IIS and congenital scoliosis patients. Day et al17 measured pulmonary function in 36 patients with a variety of congenital scoliosis deformities, finding FVC measured only 68% of normal expected values if children had been fused. Goldberg and colleagues published the pulmonary function results of patients with IIS following early fusion. FVC measured only 41% of predicted normal values if patients were fused before age 10, compared with 68% if fused at an older age.18
Our center published a longer-term study of the pulmonary function of patients with EOS following definitive spinal fusion in 2008.19 In total, 28 patients returned for pulmonary function test testing at an average of 11 years following anterior and posterior spinal fusion. The majority of the patients had congenital scoliosis, but there were smaller numbers who had neurofibromatosis, syndromes, or infantile idiopathic curves. All patients had been fused before the age of 9 years, averaging 3.3 years at the time of surgery. Results were sobering. FVC averaged 58% predicted, but was as low as 22% predicted in some patients. A severe restrictive pattern of pulmonary disease, defined as an FVC<50% normal predicted value, was present in 43% of the tested population. FVC was statistically correlated with the percent of the thoracic spine that had been fused, and was strongly related to the length of the thoracic spine from T1 to T12 on the anterior-posterior radiograph. Four patients already had thoracic insufficiency syndrome, and used respiratory assistance such as supplemental oxygen or bilevel positive airway pressure at night. Interestingly, patients who achieved 22 cm of thoracic height measured between T1 and T12 had near normal pulmonary function. This has led many EOS surgeons to adopt 22 cm as a goal in thoracic growth.
Although there are isolated case reports of maintained long-term pulmonary function following early spinal fusion,20 other centers have published series with similar unfavorable results as ours. Emans et al21 noted that FVC averaged 62% predicted in their early fusion patients. Bowen et al22 and Vitale et al23 each reported reduced FVC in studies of EOS patients with shorter postoperative follow-up.22,23
Pulmonary volumes normally decline during adulthood, beginning in the mid-fourth decade. Pehrsson et al24 stated that when FVC measures <43% predicted, severe pulmonary compromise will occur as the lung volume deteriorates in adulthood. It may be presumed that EOS patients who have diminished pulmonary volumes at the end of growth will have a high mortality rate due to respiratory failure as further lung volume is lost during the normal aging process. Our center is studying the natural history of respiratory decline in the patients who were tested in the 2008 study, finding an average of 20% loss of FVC over the last 10-year period in these compromised patients.
There is great interest currently in the concept of continued growth of the spine and chest while treating the spinal deformity in EOS patients. The VEPTR device (vertically expandible prosthetic titanium rib) allows fixation on the ribs, the ribs and spine, or the ribs and pelvis, and through repetitive surgical expansions, spinal and thoracic growth is enabled.25,26 Similarly, dual submuscular spine to spine growing rods have been successful in allowing continued spinal growth as a result of multiple surgical rod lengthenings. Unfortunately, these techniques carry a high complication rate, ranging from infection, loss of fixation, inadvertent fusion, and progressive kyphosis. Bess and the Growing Spine Study Group published that there is a 24% complication rate for each time a growing rod construct is surgically lengthened, and a 13% decrease in complications for each additional year of age at the time of initial growing rod implantation.27 Rod implantation below age 7 years, increasing kyphosis, and more severe major curve magnitude have been shown to correlate with a higher rate of complications overall.28 Such surgical difficulties, as well as the potential harmful effect of repetitive anesthesias, has led to the adoption of the magnetically expandible growing rod. Problems with loss of fixation and failure of the implants in some cases continue, however29 (Figs. 4A, B). In addition, due to the frailty of many children with EOS and their serious comorbidities, mortality rates of up to 18% have been published even with growth-friendly surgery.30
Early studies are now being published on the pulmonary outcome of patients with EOS who are treated with growth-sparing instrumentation. Glotzbecker et al31 found that growth of the thoracic spine is realized with repetitive expansions at 6.5-year follow-up. Motoyama et al32 noted that in short-term follow-up, the FVC of 10 children treated with expandible instrumentation without fusion increases, but the percent predicted does not improve up to 3 years despite treatment. Johnston et al recently published on the pulmonary results of growing rod treatment of a heterogenous group of 12 patients with EOS, finding that FVC averaged 55% predicted at final follow-up.33
While some patients do thrive with growth-friendly surgical treatment, there remains a compromised group of patients who still have a poor prognosis for pulmonary longevity in adulthood. Future studies must continue to focus on stratifying results based on the specific diagnoses of groups of patients with EOS. Vitale’s classification system is a valuable tool in comparing similar patient groups with EOS.34 One should expect that successful treatment that encourages growth of the spine and chest should lead to favorable outcomes in patients with early-onset idiopathic scoliosis. Patients with progressive neuromuscular conditions, such as the congenital muscular dystrophies and (until recently) spinal muscular atrophy, may realize spinal growth with expandible instrumentation, but the worsening muscle weakness may negate the positive effects of the growth-friendly implants. Patients with bone dysplasias that limit skeletal growth may experience improvement in spinal deformity, but have intermediate results compared with their normal-growing peers. The intimate relationship between the vertebral alignment and growth and the 3-dimensional chest and rib volume must continue to be at the forefront of outcomes research in this difficult patient population.
1. Campbell RM. Spine deformities in rare congenital syndromes: clinical issues. Spine. 2009;34:1815–1827.
2. Mehta MH. The rib-vertebra angle in the early diagnosis between resolving and progressive infantile scoliosis. J Bone Joint Surg Br. 1972;54:230–243.
3. Scott JC, Morgan TH. The natural history and prognosis of infantile idiopathic scoliosis. J Bone Joint Surg Br. 1955;37:400–413.
4. Pehrsson K, Larsson S, Oden A, et al. Long-term follow-up of patients with untreated scoliosis: a study of mortality, causes of death, and symptoms. Spine. 1992;17:1091–1096.
5. Fletcher ND, McClung A, Rathjen KE, et al. Serial casting as a delay tactic in the treatment of moderate-to-severe early-onset scoliosis. J Pediatr Orthop. 2012;32:664–671.
6. Mehta MH. Growth as a corrective force in the early treatment of progressive infantile scoliosis. J Bone Joint Surg Br. 2005;87:1237–1247.
7. McMaster MJ, Ohtsuka K. The natural history of congenital scoliosis. J Bone Joint Surg Am. 1982;64:1128–1147.
8. Dubousset J, Herring JA, Shufflebarger H. The crankshaft phenomenon. J Pediatr Orthop. 1989;9:541–550.
9. Goldberg CJ, Moore DP, Fogarty EE, et al. Long-term results from in situ fusion for congenital vertebral deformity. Spine. 2002;27:619–628.
10. Chang DG, Kim JH, Ha KY, et al. Posterior hemivertebra resection and short segment fusion with pedicle screw fixation for congenital scoliosis in children younger than 10 years: greater than 7-year follow-up. Spine. 2015;15:E484–E491.
11. Yaszay B, O’Brien M, Shufflebarger HL, et al. Efficacy of hemivertebra resection for congenital scoliosis: a multicenter retrospective comparison of three surgical techniques. Spine. 2011;36:2052–2060.
12. Bollini G, Docquier PL, Viehweger E, et al. Lumbar hemivertebra resection. J Bone Joint Surg Am. 2006;88:1043–1052.
13. Campbell RM, Smith MD, Mayes TC, et al. The characteristics of thoracic insufficiency syndrome associated with fused ribs and congenital scoliosis. J Bone Joint Surg Am. 2003;85:399–408.
14. Dimeglio A, Bonnel F. Le rachis en croissance. Paris: Springer; 1990.
15. Burri PHMcDonald JA. Structural aspects of prenatal and postnatal development and growth of the lung. Lung Growth and Development. New York, NY: Dekker; 1997:1–35.
16. Thurlbeck WM. Postnatal human lung growth. Thorax. 1982;37:564–571.
17. Day GA, Upadhyay SS, Ho EK, et al. Pulmonary functions in congenital scoliosis. Spine. 1994;19:1027–1031.
18. Goldberg CJ, Gillic I, Connaughton O, et al. Respiratory function and cosmesis at maturity in infantile-onset scoliosis. Spine. 2003;28:2397–2406.
19. Karol LA, Johnston C, Mladenov K, et al. Pulmonary function
following early thoracic fusion in non-neuromuscular scoliosis. J Bone Joint Surg Am. 2008;90:1272–1281.
20. Winter RB, Lonstein JE. Ultra-long-term follow-up of pediatric spinal deformity problems: 23 patients with a mean follow-up of 51 years. J Orthop Sci. 2009;14:132–137.
21. Emans J, Kassab F, Caubet JF, et al. Earlier and more extensive thoracic fusion is associated with diminished pulmonary function
: outcome after spinal fusion of four or more thoracic spinal segments before age five. (paper presentation no. 101). Scoliosis Research Society Annual Meeting, Buenos Aires, Argentina. September 6–9, 2004.
22. Bowen RE, Scaduto AA, Baneulos S. Does early thoracic fusion exacerbate preexisting restrictive lung disease in congenital scoliosis patients? J Pediatr Orthop. 2008;28:506–511.
23. Vitale MG, Matsumoto H, Bye MR, et al. A retrospective cohort study of pulmonary function
, radiographic, measures, and quality of life in children with congenital scoliosis: an evaluation of patient outcomes after early spinal fusion. Spine. 2008;33:1242–1249.
24. Pehrsson K, Nachemson A, Olofson J, et al. Respiratory failure in scoliosis and other thoracic deformities. A survey of patients with home oxygen or ventilator therapy in Sweden. Spine. 1992;17:714–718.
25. Campbell RM, Hell-Vocke AK. Growth of the thoracic spine in congenital scoliosis after expansion thoracoplasty. J Bone Joint Surg Am. 2003;85:409–420.
26. Emans JB, Caubet JF, Ordonez CL, et al. The treatment of spine and chest wall deformities with fused ribs by expansion thoracostomy and insertion of verticle expandable prosthetic titanium rib: growth of thoracic spine and improvement of lung volumes. Spine. 2005;30 (suppl 17):558–568.
27. Bess S, Akbarnia BA, Thompson GH, et al. Complications of growing-rod treatment for early-onset scoliosis: analysis of one hundred and forty patients. J Bone Joint Surg Am. 2010;92:2533–2543.
28. Upasani VV, Parvaresh KC, Pawalek JB, et al. Age at initiation and deformity magnitude influence complication rates of surgical treatment with traditional growing rods in early-onset scoliosis. Spine Deform. 2016;4:344–350.
29. Akbarnia BA, Pawalek JB, Cheung KM, et al. Traditional growing rods versus magnetically controlled growing rods for the surgical treatment of early-onset scoliosois: a case-matched 2-year study. Spine Deform. 2014;2:493–497.
30. Phillips JH, Knapp DR, Herrera-Soto J. Mortality and morbidity in early-onset scoliosis surgery. Spine. 2013;38:324–327.
31. Glotzbecker MP, Gold M, Miller P, et al. Distraction-based treatment maintains predicted thoracic dimensions in early-onset scoliosis. Spine Deform. 2014;2:203–207.
32. Motoyama EK, Deeney VF, Fine GF, et al. Effects on lung function of multiple expansion thoracoplasty in children with thoracic insufficiency syndrome: a longitudinal study. Spine. 2006;31:284–290.
33. Johnston CE, Tran DP, McClung A. Functional and radiographic outcomes following growth-sparing management of early-onset scoliosis. J Bone Joint Surg Am. 2017;99:1036–1042.
34. Williams BA, Matsumoto H, McCalla DJ, et al. Development and initial validation of the classification of early-onset scoliosis (C-EOS). J Bone Joint Surg Am. 2014;20:1359–1361.