SPINE
Early-onset scoliosis: clinical presentation, assessment and treatment options
Introduction Early-onset scoliosis (EOS) is defined by the Scoliosis Research Society (SRS) Growing Spine Study Group as ‘spine deformity that is present before 10 years of age’.1 It is an uncommon condition, which can be life limiting with a complex group of underlying diagnoses. As a result multiple different methods of treatment are used, in part due to the heterogenous nature of causes.
M Zaki B Choudhury Physiological importance
Athanasios I Tsirikos
EOS is clinically extremely significant due to the potentially devastating consequences on thoracic growth and lung development. Pehrsson et al.2 undertook a review of the natural history and reported early deaths and respiratory failure in patients with untreated scoliosis. Post-mortem studies have demonstrated not only small but also hypoplastic lungs. In the seminal work on spinal growth DiMeglio3 demonstrated the most rapid period of spinal growth occurs in the first 5 years of life. In this period the spine increases 50% of its length, but reaches 95% of adult canal diameter. Between 5 years and 10 years old spinal growth continues at a slower rate but this is a critical period of alveolar development, both in terms of number and functional complexity. Thoracic spine height averages 11 cm at birth, 18 cm at 5 years age and 22 cm at 10 years. Consider then that in the context of early fusion in early-onset scoliosis in a 5-year-old child this could lead to a 12.5 cm loss of spinal growth, leading to significantly curtailed pulmonary development.4 The cornerstone of treating this particular age group therefore is facilitation of lung development, thoracic growth and, if possible, preservation of movement. Spinal shape is regarded as a proxy marker of lung function with thoracic height contributing significantly to lung volume though in a non-linear way. The effects of spinal fusion prior to near maximal lung maturity at 10 years of age have been shown, with patients having a mean forced vital capacity (FVC) 41% of normal at maturity. This compared with 68% of normal FVC in patients having spinal fusion older than 10 years. It is worth noting however that the relationship is often not quite so simple, with interventions sometimes leading to loss of chest wall compliance and paradoxically less improvement in lung function. Campbell’s5 work focused on the analysis of lung function in the context of fused ribs and congenital scoliosis. This can create a three-dimensional (3D) thoracic deformity with adverse effects on thoracic growth and function termed ‘thoracic insufficiency syndrome’ (TIS), describing the inability of the thorax to support normal respiration or lung growth. This review shall be structured around the SRS consensus document but will seek to expand each aspect to provide a global overview. The sub-groupings of EOS are considered in Table 1.
David S Marks
Abstract Early-onset scoliosis (EOS) is an ‘umbrella’ term defining scoliosis presenting before the age of 10 years. It reflects a constellation of conditions, which are challenging to treat. EOS is subdivided aetiologically into: idiopathic, congenital, thoracogenic, neuromuscular and syndromic. Each group has unique issues to address. The cornerstone of treating EOS is the facilitation of optimal conditions for lung development, thoracic growth and spinal movement. As a proxy marker, thoracic height contributes significantly to lung volume. One must adopt a holistic approach with consideration of the impact of multiple treatment interventions in early life on overall development. Treatment is guided by the principles above. In certain patients, bracing or serial casting treatment can restore spinal parameters to normal with no late residual. In other cases, such as neuromuscular and syndromic, operative interventions are often necessary. Surgical options range from primary fusion, to growth sparing implants that are periodically extended to allow spinal lengthening and thereby thoracic volume increase. ‘Growing rods’ have evolved over timeesome require multiple surgeries, whilst others rely on a guided growth principle. A further recent development is the externally lengthened magnetic ‘growing rod’. This review addresses the underlying conditions, assessment and treatment of patients with EOS.
Keywords bracing; early-onset scoliosis; EOS; growing rods; serial casting; thoracic insufficiency
M Zaki B Choudhury BA(Hons) FRCS(Tr&Orth) DipSEM(UK&I) Spinal Fellow, Scottish National Spine Deformity Centre, Royal Hospital for Sick Children, Edinburgh, UK. Conflicts of interest: none declared. Athanasios I Tsirikos MD FRCS PhD Consultant Orthopaedic and Spine Surgeon, Scottish National Spine Deformity Centre, Royal Hospital for Sick Children, Edinburgh, UK. Conflicts of interest: educational/teaching commitments with DePuy/ Synthes Spine.
Prognosis When discussing prognosis there are two factors to consider: firstly the curve prognosis and secondly the overall morbidity and potential mortality to the patient (Table 2). As the diagnostic categories show this is a very diverse group of patients. As such in some of the diagnostic subgroups the incidence of significant cardiac and respiratory co-morbidities is much
David S Marks FRCS FRCS(Orth) Consultant Orthopaedic Spine Surgeon, Royal Orthopaedic Hospital and The Birmingham Children’s Hospital, Birmingham, UK. Conflicts of interest: consultancy agreement with DePuy/Synthes Spine; Speaker bureau: DePuy/Synthes Spine, Stryker Spine, Medtronic, Nuvasive. Royalties: DePuy/Synthes Spine.
ORTHOPAEDICS AND TRAUMA --:-
1
Ó 2017 Published by Elsevier Ltd.
Please cite this article in press as: Choudhury MZB, et al., Early-onset scoliosis: clinical presentation, assessment and treatment options, Orthopaedics and Trauma (2017), https://doi.org/10.1016/j.mporth.2017.09.006
SPINE
Diagnostic categories of early-onset scoliosis (EOS)1 Idiopathic C
C
C
Congenital
No apparent cause or related underlying aetiology Infantile idiopathic e a subgroup which develop scoliosis in infants and children less than 3 years old Juvenile idiopathic age 3 e10 years
C
C
C
Thoracogenic
Vertebrae develop incorrectly in uterus Sometimes associated with cardiac and renal abnormalities, e.g VATER, VACTERL Evaluation may include studies of heart and kidneys
C
C
C
C
Syndromic
Multiple congenital rib fusions as seen in spondylocostal or spondylothoracic dysostosis May have congenital vertebral anomalies May also be considered congenital scoliosis Changes in the chest wall following thoracic surgery which may function as a tether which promotes change in the shape of the spine
C
C C C
Syndromes, such as Marfan’s, Ehlers eDanlos and other connective tissue disorders Neurofibromatosis PradereWilli Numerous bone dysplasias may be associated with EOS
Neuromuscular C
C
C C C
Can develop in children with neuromuscular disorders Spinal muscular atrophy (SMA) Cerebral palsy Spina bifida Brain or spinal cord injury
Table 1
higher. Overall consideration must be given to the prognosis if the condition is left untreated as discussed above, but the fundamental assumption is that some form of treatment is better than observation alone. In patients with idiopathic scoliosis there are a number of features to suggest good prognosis if treated conservatively (i.e with casting or bracing). The early work identified that idiopathic EOS greater than 35 is likely to progress. In many children under 2 years old with infantile idiopathic curves less than 35 , scoliosis may resolve without treatment. Furthermore Mehta6 observed the rib vertebral angle difference (RVAD) as a predictor of curve progression in infantile idiopathic scoliosis (Figure 1). This measurement is performed by drawing a line perpendicular to the endplate of the most translated apical vertebra and a line down the midpoint of the concave and convex rib at this level. The angle created on the convexity is subtracted from the opposite concave angle to measure the RVAD. Mehta found
scoliosis resolution in 90% of children with RVAD less than 20 . Mehta also described the rib head phase as an adjunct to RVAD in determining curve progression. This is a rotational measure defined by the relationship of the rib head and vertebral body at the apex with overlapping of the vertebral body and the rib head predicting curve progression.
Assessment of EOS A detailed history and examination should be undertaken at initial clinical presentation. It is possible that what is labelled as ‘EOS’ is actually a missed congenital scoliosis. A detailed birth history can reveal complications suggestive of periods of difficult or obstructed labour, and thus potential for a diagnosis of cerebral palsy, albeit perhaps the more subtle end of the spectrum. Similarly a thorough perinatal history can reveal issues with growth and intrauterine development suggestive of VATER/
Factors affecting prognosis Idiopathic
Congenital
Thoracogenic
Syndromic
Neuromuscular
Note: consider role of MRI in curve >20 or rapid progression e higher incidence of syrinx and Chiari malformation
Vertebrae develop incorrectly in utero Sometimes associated with cardiac and renal abnormalities, e.g VATER, VACTERL Evaluation may include studies of heart and kidneys
Can be considered either: ‘cause’ e as in the case of thoracic insufficiency syndrome, with rib malformations and issues arising from ventilatory inadequacy, or: ‘effect’ e Changes following thoracic/cardiac surgery in early infancy which may function as an external tether. Note also potential ongoing thoracic/cardiac issues
Severity of connective tissue disorders Neurofibromatosis with associated complications: optic glioma, dural ectasia, soft tissue neurofibromata, bone quality PradereWilli: behavioural issues and compliance with certain forms of treatment may limit their effectiveness
Related to severity of underlying condition, can have bearing on ability to recover from anaesthesia. Can also limit expected postoperative function
Table 2
ORTHOPAEDICS AND TRAUMA --:-
2
Ó 2017 Published by Elsevier Ltd.
Please cite this article in press as: Choudhury MZB, et al., Early-onset scoliosis: clinical presentation, assessment and treatment options, Orthopaedics and Trauma (2017), https://doi.org/10.1016/j.mporth.2017.09.006
SPINE
a
vertebrae themselves, or the ribs. There are limitations to any modality whereby image data needs to be processed e the algorithms used can sometimes generate false-negatives or falsepositives when producing reconstructed views. CT is useful for identifying potential problems with instrumentation placement (e.g. narrow or non-existent pedicles) and can make some quantitative assessment of bone quality. However there is significant radiation exposure with CT scans, particularly if a highly detailed assessment of bony structures is needed and in young patients this has a potential cumulative effect. MRI negates the risk of radiation exposure at the cost of loss of detail in bony assessment. Detailed soft tissue studies are possible along with evaluation of potential neural axis anomaly. However MRI is time consuming and requires the patient to remain immobile. In children compliance is often an issue and therefore studies are frequently undertaken under general anaesthetic, which carries its own inherent risks. Therefore it is reasonable to consider whole spine or indeed whole neural axis imaging if surgical intervention is considered in EOS as approximately 20% of patients have an underlying intraspinal anomaly, such as Chiari malformation, syringomyelia or diastematomyelia. Alternatively, if treatment is not needed in the early years then MRI can be delayed to after the age of 6 years when many children can tolerate the investigation without anaesthesia.
b
Vertebral body Convex rib
Concave rib
Figure 1 Diagram demonstrating calculation of the rib vertebral angle difference: convex (a) minus concave (b).
VACTERL syndromes. TIS should be considered as it may have a profound impact on the patient’s quality and length of life. Initial identification of the curve can be purely incidental, either clinically or on a chest X-ray taken for unrelated reasons. It is important to identify any potential progression which has been observed by the parents.
Radiographic studies Standard whole spine postero-anterior and lateral radiographs are appropriate investigations for the assessment of EOS. Consideration should be given to weight-loaded upright views, although in some patients with syndromic or neuromuscular conditions a sitting or supported view may be the only possibility. In such cases suspended or traction views can give an idea of curve flexibility and an indication as to the contribution of trunk muscle control on posture. Cobb angle is universally measured to give an indicator of the degree of spinal deformity. However, particularly in the case of EOS the degree of rotation and thoracic deformity cannot be easily identified by looking at Cobb angle alone and this is where Mehta’s rib head phases become more informative. With regards to thoracic deformity both a structural and volumetric/functional assessment is necessary. T1eT12 and T1 eS1 height are recorded in the case of the growing thorax, as once T1eT12 height is 22 cm or greater Karol and Johnston7 have shown that lung function would be satisfactory. Formal spirometry and respiratory review are appropriate particularly in patients where functional compromise is evident from the clinical history or surgery is considered. Skeletal maturity can be assessed using the Risser sign, observation of the status of the tri-radiate cartilage, or a Sanders digital index assessment. It is worth noting that skeletal maturity can diverge from chronological age, particularly in the case of skeletal dysplasia, syndromic conditions and patients with severe underlying medical illnesses. The radiological measures of maturity, coupled with physiological factors such as menarche or Tanner eWhitehouse staging can also give an idea of prognosis with respect to progression and can guide decisions such as timing of surgery. With respect to volumetric and structural analysis 3D imaging such as CT or more recently low-dose tomography can be used to calculate actual lung volumes and gain a detailed anatomical perspective, including structural anomalies, either in the
ORTHOPAEDICS AND TRAUMA --:-
Treatment goals in EOS The treatment options are guided by the following principles: minimize spinal deformity and maximize thoracic volume and function over the patient’s life minimize the extent of final spinal fusion, maximize motion of chest and spine minimize complications, procedures, hospitalizations and burden for the family and consider overall development of the child. In the ‘truly’ idiopathic scoliosis nonoperative treatment including bracing and casting may suffice to restore spinal parameters to normal with no residual issues going forward. In other patients, such as those with neuromuscular and syndromic conditions, operative treatment is usually required. Options range from short segment fusions to growing rod technology, whereby implants are placed that are periodically extended to allow spinal lengthening and thoracic volume increase. Growing rods act as an internal brace and have evolved over time to encompass different techniques e some require multiple surgeries, whilst others rely on a guided growth principle. A further development is magnetic lengthening systems, in which externally applied magnetic fields can periodically achieve lengthening.
Casting or bracing Plaster techniques have been used in scoliosis since the 19th century for correction of spinal deformities with traction or lateral pressures and application of cast. Hibbs described the use of casting in the pre- and postoperative spinal fusion technique which he described prior to the advent of instrumentation.
3
Ó 2017 Published by Elsevier Ltd.
Please cite this article in press as: Choudhury MZB, et al., Early-onset scoliosis: clinical presentation, assessment and treatment options, Orthopaedics and Trauma (2017), https://doi.org/10.1016/j.mporth.2017.09.006
SPINE
modified using manual moulding of the rib hump instead of straps. Success of casting in EOS is correlated with treatment within the first 2 years of life with a mild scoliosis. Mehta showed complete resolution in patients with syndromic or idiopathic scoliosis who began casting at 15e21 months of age with curves 27e35 . In patients starting treatment at 27e34 months of age with curves 47e53 complete resolution was not achieved, though curve progression was better controlled. In these patients casting can probably delay surgical intervention. Fletcher et al.12 showed surgery could be delayed in nearly 50% of patients by 14 e64 months from the time of initial casting, equivalent to potentially seven growing rod lengthening procedures. The remaining 14 patients did not require surgery. Overall in both congenital and neuromuscular patients at average follow-up 5.5 years, 72% of patients were deemed not to require surgery. This group had curves >50 and did not begin their casting treatment until an average age of 4.4 years. Johnston et al.13 compared growing rod versus cast treatment and showed that casting is an acceptable alternative for children with EOS. Comparison of two age- and curve-matched cohorts of 27 patients showed better correction in the growing rod group but with 23 complications (85%). In the casting group, the magnitude of deformity was maintained rather than decreased but with no major complications. The study included children who were older than the age group in which cast treatment is most successful. Proven prevention of progression is very important as it delays surgical intervention, More recently, there has been a resurgence in casting as a method of delaying the need for commencing surgical treatment to avoid later complications which have become apparent from long-term follow-up studies of ‘growing rod’ techniques.
Localizer casts 8
Risser described application of a corrective cast coupled with spinal fusion. He detailed the use of a specialized table during cast application for curve correction. The patient is supported by a series of longitudinal straps with pads under the buttocks and shoulders. Traction is applied via a pulley system. A ‘localizer’ or adjustable solid arch applies corrective forces in order to aid correction during casting. A window is cut on the cast across the spine to allow for surgical fusion. Postoperatively the cast is changed once at 7e10 days and then every 3e4 days until bony fusion has been achieved. Interestingly, Risser did not report the technique of casting alone in very young patients.
Elongation, de-rotation and lateral flexion (EDF) casting Cotrel and Morel9 described the use of serial casting in 75 patients with idiopathic scoliosis. Their technique differs from Risser’s in the way de-rotational force is applied. Risser’s localizer cast uses traction and postero-laterally directed forces over the rib prominence, the inference of which is that Risser also employed de-rotation of the spine via the thoracic cage. Cotrel and Morel applied the corrective force through straps instead of a localizer, termed ‘de-rotation’. In the description of their EDF method they detail the corrective forces, achieving elongation of the spine through pelvic straps and halter head traction, similar to Risser. De-rotation is achieved with straps around the rib prominence, pulled perpendicular to the ribs. Lateral flexion is achieved through lateral straps, adjusting the pelvic and halter traction accordingly to achieve clinical correction. Cotrel suggested that spinal growth may be controlled with casting to prevent deformity progression. In patients with juvenile scoliosis casting was an adjunct to posterior fusion. Mehta classified EOS into resolving and progressive types, defined by using the RVAD and the rib stage. Mehta further delineated the progressive type as being either benign or malignant according to the rib phase and rotational profile. Mehta and Morel10 reported the use of casting as sole treatment in EOS. Their group of 21 patients with progressive benign infantile scoliosis had casting at early age. Six of the patients were skeletally mature with curves maintained at greater than 20 . The remaining skeletally immature patients had well-controlled curves. They noted that lateral bending or wedge-type casts correct thoracic and thoracolumbar curves better and distractiontype casts are more effective at correcting thoracic and lumbar combined curves.
Role of bracing during cast treatment The Milwaukee and Boston braces have been successfully used in scoliosis treatment but the evidence for bracing in EOS as anything but an adjunct is limited. It has been observed that bracing was effective in curves that may perhaps have spontaneously resolved, and that bracing alone was not sufficient to provide adequate curve control. Brace wear compliance is problematic in some patient groups; very few studies on compliance include EOS patients.
Surgical treatment Surgical treatment is a difficult undertaking in this diverse group and although initial results are typically very encouraging, ultimate outcomes are often poor. There are a number of considerations but chief amongst these is the aim to control the deformity whilst allowing continued spinal growth to minimize the constitutional impact to the child and reduce the need for repetitive surgery. As such there are a number of potential options but unfortunately there is no ‘ideal’ treatment solution to address such aggressive deformities. The first consideration is the distinction between ‘growth-friendly’1 and fusion techniques.14 Growth-friendly implants are divided into three categories: distraction-based, guided growth and compressionbased.
EDF modification (according to Mehta) A well-applied cast in scoliosis provides a constant corrective force in the growing spine. In a small and fast-growing child braces are not practical as the rate of growth often outstrips how quickly the brace can be refashioned. The downside is that casting has to be undertaken under general anaesthetic. It is well established that the most rapid period of spinal growth is within the first 2 years of life. Serial casting aims to utilize this fast period of growth as the ‘corrective force’. Mehta11 reported a prospective study of 136 patients with progressive infantile scoliosis on cast treatment. Cotrel and Morel’s method was
ORTHOPAEDICS AND TRAUMA --:-
4
Ó 2017 Published by Elsevier Ltd.
Please cite this article in press as: Choudhury MZB, et al., Early-onset scoliosis: clinical presentation, assessment and treatment options, Orthopaedics and Trauma (2017), https://doi.org/10.1016/j.mporth.2017.09.006
SPINE
lengthening, in theory increasing lung volume. The VEPTR attaches between proximal and distal ribs, spine or pelvis to control the rib cage. As the ribs and spine form a coupled lever arm there is also potential to control scoliosis via the rib deformity using a similar philosophy to de-rotation applied in casting whereby external forces to the chest wall rotate the spine. However, as observed in the growing rod systems, there are reports of implant prominence and failure leading to skin breakdown and infection, decreased lung compliance and proximal junctional kyphosis.
Growing spine techniques A. Distraction-based implants These are commonly used devices in EOS. Four types can be considered: traditional growing rods (TGR) vertical expandable prosthetic titanium rib (VEPTR) hybrid systems magnetically controlled growing rods (MCGR). All rely on two fixed points proximally and distally with a connection between which allows elongation as the child’s spine grows.
3. Hybrid system: this comprises a traditional growing rod attached to the ribs, either via conventional hooks or via VEPTR rib cradles producing a rib-to-spine fixation. There are also fixation options to attach the VEPTR to the pelvis.
1. Traditional growing rods: this is a well described technique, combining the concept of a small fusion block proximal and distal to secure anchor points using pedicle screw or hook constructs. Using rods passed extraperiosteally to avoid triggering fusion but sub-fascial to avoid prominence the fusion areas are linked with rods which are connected with a cross-link used to lengthen the spine periodically (usually every 6 months). This necessitates repeat general anaesthetics, potential wound problems and the scar to be re-opened each time. Over the years different configurations have been proposed but a dual rod system is considered optimal due to biomechanical and physiological reasons.15 Growing rods have transformed the care of these patients but are not without problems. In very young, small size or malnourished children implant prominence can cause skin breakdown, muscular pain and cosmetic problems. Bess et al.16 reviewed 910 growing rod procedures and reported 20% complication rates. In addition to wound infection, biomechanical issues arise which can limit the effectiveness of treatment. Sankar et al.17 described the ‘law of diminishing returns’ and showed that following seven lengthenings spinal compliance reduces due to auto-fusion. After this procedure on average each lengthening allows less than 8 mm distraction. Considering these limitations, children who have growing rods placed at 3e4 years of age may potentially hit the biomechanical limits of lengthening within 2.5e3.5 years, much earlier than full alveolar development has been achieved. It is, therefore, prudent to postpone growing rod application until as late as possible, especially as complications decrease by 13% with each year that surgery is delayed. Given the increased stiffness coupled with multiple surgeries and potential for fretting of the implants there is a relatively high incidence of rod breakage requiring revision surgery. Another observation with all posteriorly based distraction systems is the preponderance toward causing sagittal imbalance and generating kyphosis whilst distracting, in particular the risk of proximal junctional kyphosis which is extremely difficult to address and often results in the need for proximal fusion extension to the detriment of respiratory function and spinal movement.
4. Magnetically controlled growth rods: This is a new class of implants which relies on constructs with distal spinal anchors and proximal rib or spinal anchors in the same way as TGR systems. They are connected by telescoping rods which are magnetically controlled and allow lengthening without need for surgery (Figure 2). The lengthenings can be performed in the outpatient setting and are done either with X-ray control or ultrasound guidance to assess the amount of distraction. Preliminary studies are divided into two camps. Nnadi’s group18 report promising early results. There have been concerns raised about the mechanical complications of MCGR technology.19 A number of centres have identified early mechanical failure leading to loss of ability to lengthen, though these have been with the earlier generation of magnets which have been recently modified. Similarly there have been concerns about the development of metallosis in the surrounding soft tissues with metal particles released in the blood circulation with as yet unknown impact on long-term health. Complications seen in TGR techniques with screw pullout and instrumentation failure can also occur. Despite these concerns all studies, whether favourable or not, report scoliosis improvement around 20 degrees. There is need for further work to determine the precise role of MCGR technology in the management of EOS. 5. Guided growth techniques: in guided growth techniques a different philosophy is used in that implants are deployed to allow the deformity to correct as the spine grows. The original technique used segmental Luque sublaminar wires secured around straight rods that corrected scoliosis with growth. However this technique can lead to spontaneous fusion limiting spinal development. McCarthy developed the Shilla technique, in which pedicle screws are placed with minimal dissection in the hope of avoiding spontaneous fusion (Figure 3). The apex of the curve is corrected and fused (initially this was recommended by 360 anterior/posterior fusion though current recommendation is for the surgery to be performed all posteriorly). Rods can slide across the proximal and distal anchor points guiding growth and allowing 3D deformity correction. These techniques reduce the need for multiple surgeries. McCarthy’s study of Shilla technique compared to growing rods demonstrated fewer operations in the Shilla group (2.8 versus 7.4). However Shilla resulted in less spinal growth and less
2. Vertical expandable prosthetic titanium rib (VEPTRÒ): the effect of scoliosis on thoracic function leading to TIS has already been discussed. Campbell5 developed an expansion thoracoplasty device in skeletally immature patients. The VEPTR was developed to allow stabilization of the hemithorax gradual
ORTHOPAEDICS AND TRAUMA --:-
5
Ó 2017 Published by Elsevier Ltd.
Please cite this article in press as: Choudhury MZB, et al., Early-onset scoliosis: clinical presentation, assessment and treatment options, Orthopaedics and Trauma (2017), https://doi.org/10.1016/j.mporth.2017.09.006
SPINE
Figure 2 Radiographs of an 8-year-old child with infantile idiopathic thoracic and lumbar scoliosis. He underwent placement of magnetically controlled growing rods with consecutive lengthenings which preserved spinal growth and delayed the definitive posterior spinal fusion for 3 years.
Figure 3 Radiographs of a 7-year-old patient with syndromic kyphoscoliosis. He underwent a Shilla procedure which has delayed progression of the deformity in both planes over 2 years follow-up and allowed growth to occur across the proximal and distal anchor points.
overcorrection in which an opposite direction curve developed. There are reports of instrumentation failure and it is unclear what should be done with the implant at the end of spinal growth.
scoliosis correction with higher complication rates. Therefore, its use has become limited to selected cases. B. Compression-based implants These techniques are focused on the convexity of the curve and are in an experimental phase with limited US Federal Drug Administration (FDA) and European CE Mark approval. The rationale is similar to that of a hemi-epiphysiodesis in the long bones aiming to stop spinal growth on the convex side of the curve allowing remaining concave growth to gradually correct the deformity. The main methods which have been used include more historical such as anterior vertebral stapling but there is a new resurgence of interest with anterior vertebral body tethering (AVBT). As long-term results are in progress it is difficult to judge the indications and utility. Studies at present are limited to case series which have demonstrated curve correction with growth in skeletally immature patients with adolescent idiopathic scoliosis.20 In terms of problems observed there have been cases of
ORTHOPAEDICS AND TRAUMA --:-
Fusion Both in terms of the final operation of a growing rod lengthening programme, a ‘bail out’ for failed bracing and across short segments to control a congenital deformity, performing a spinal fusion has its issues in EOS. Fusion surgery should be delayed for as long as possible to avoid pulmonary complications due to the negative effect of a shortened thoracic spine on lung development and maturation. Fusion surgery in the growing spine usually requires a combined anterior and posterior approach to be effective. If an isolated posterior fusion is performed, remaining anterior vertebral growth in skeletally immature patients can result in worsening rotational deformity described as ‘crankshaft
6
Ó 2017 Published by Elsevier Ltd.
Please cite this article in press as: Choudhury MZB, et al., Early-onset scoliosis: clinical presentation, assessment and treatment options, Orthopaedics and Trauma (2017), https://doi.org/10.1016/j.mporth.2017.09.006
SPINE
phenomenon’. Long-term results of early posterior fusion in EOS demonstrated that almost 40% of patients underwent reoperation due to deformity progression.
8 Risser JC. Scoliosis treated by cast correction and spine fusion. Clin Orthop Relat Res 1976; 116: 86e94. 9 Cotrel Y, Morel G. The elongation-derotation-flexion technic in the correction of scoliosis. Rev Chir Orthop Reparatrice Appar Mot 1964; 50: 59e75. 10 Mehta M, Morel G. The non-operative treatment of infantile idiopathic scoliosis. In: Zorab P, Siegler D, eds. Sixth symposium on scoliosis; September 17e18, 1979. London: Academic Press, 1980; 71e84. 11 Mehta MH. Growth as a corrective force in the early treatment of progressive infantile scoliosis. J Bone Jt Surg (Br) 2005; 87: 1237e47. 12 Fletcher ND, McClung A, Rathjen KE, Denning JR, Browne R, Johnston 3rd CE. Serial casting as a delay tactic in the treatment of moderate-to-severe early-onset scoliosis. J Pediatr Orthop 2012; 37: 664e71. 13 Johnston CE, McClung AM, Thompson GH, Poe-Kochert C, Sanders JO. Comparison of growing rod instrumentation versus serial cast treatment for early-onset scoliosis. Spine Deform 2013; 1: 339e42. 14 Skaggs DL, Akbarnia BA, Flynn JM, Myung KS, Sponseller PD, Vitale MG, Chest Wall and Spine Deformity Study Group; Growing Spine Study Group; Pediatric Orthopaedic Society of North America; Scoliosis Research Society Growing Spine Study Committee. A classification of growth friendly spine implants. J Pediatr Orthop 2014; 34: 260e74. 15 Akbarnia BA, Marks DS, Boachie-Adjei O, et al. Dual growing rod technique for the treatment of progressive early-onset scoliosis: a multicenter study. Spine (Phila PA 1976) 2005; 30(suppl 17): S46e57. 16 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 Jt Surg Am 2010; 92: 2533e43. 17 Sankar WN, Skaggs DL, Yazici M, et al. Lengthening of dual growing rods and the law of diminishing returns. Spine (Phila PA 1976) 2011; 36: 806e9. 18 Thompson W, Thakar C, Rolton DJ, Wilson-MacDonald J, Nnadi C. The use of magnetically-controlled growing rods to treat children with early-onset scoliosis: early radiological results in 19 children. Bone Jt J 2016; 98-B: 1240e7. https://doi.org/10.1302/ 0301-620X.98B9.37545. 19 Teoh KH, Winson DM, James SH, et al. Magnetic controlled growing rods for early-onset scoliosis: a 4-year follow-up. Spine J 2016; 16(suppl 4): S34e9. https://doi.org/10.1016/j.spinee.2015. 12.098. Epub 2016 Feb 1. 20 Samdani AF, Ames RJ, Kimball JS, et al. Anterior vertebral body tethering for immature adolescent idiopathic scoliosis: one-year results on the first 32 patients. Eur Spine J 2015; 24: 1533e9.
Conclusion It is important to recognize that EOS does not represent a single condition but instead a number of underlying aetiologies presenting with early-onset and often highly aggressive deformity affecting all planes of the spine associated with marked distortion of rib cage and lung development. Treatment is challenging and can produce life-threatening complications. This aims to allow the child to grow, develop respiratory function and minimize impact on quality of life and in the more severe cases prolong lifespan. We recommend the use of conservative measures such as casting and bracing in order to control the deformity and delay scoliosis surgery. Growth preservation techniques have their own limitations and complications and should be used when non-operative treatment has failed in very young children. Spinal fusion should be postponed as much as possible at least until alveolar growth and significant chest development has been completed close to puberty. A
REFERENCES 1 Skaggs DL, Guillaume T, El-Hawary R, Emans J, Mendelow M, Smith J, SRS Growing Spine Committee. Early onset scoliosis consensus statement, SRS Growing Spine Committee. Spine Deform 2015; 3. A1eA8, 105e204. 2 Pehrsson K, Nachemson A, Olofson J, Strom K, Larsson S. Respiratory failure in scoliosis and other thoracic deformities: a survey of patients with home oxygen or ventilator therapy in Sweden. Spine (Phila PA 1976) 1992; 17: 714e8. 3 DiMeglio A. Growth of the spine before age 5 years. J Pediatr Orthop B 1992; 1: 102e7. 4 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 (Phila PA 1976) 2008; 33: 1242e9. 5 Campbell Jr RM. VEPTR past experience and the future of VEPTR principles. Eur Spine J 2013; 22(suppl 2): 106e17. 6 Mehta MH. The rib-vertebra angle in the early diagnosis between resolving and progressive infantile scoliosis. J Bone Jt Surg (Br) 1972; 54: 230e43. 7 Karol LA, Johnston C, Mladenov K, Schochet P, Walters P, Browne RH. Pulmonary function following early thoracic fusion in non-neuromuscular scoliosis. J Bone Jt Surg (Am) 2008; 90: 1272e81. https://doi.org/10.2106/JBJS.G.00184.
ORTHOPAEDICS AND TRAUMA --:-
7
Ó 2017 Published by Elsevier Ltd.
Please cite this article in press as: Choudhury MZB, et al., Early-onset scoliosis: clinical presentation, assessment and treatment options, Orthopaedics and Trauma (2017), https://doi.org/10.1016/j.mporth.2017.09.006