Recent Advances in the Management of Early Onset Scoliosis




As the undesired results of early spinal fusion have become apparent, “growth-friendly” management methods for early onset scoliosis have been increasing during recent years. Current literature supports the use of repeated corrective cast applications as the initial management for most early onset progressive spinal deformities as either definitive treatment or as a temporizing measure. If casting is not an option or the deformity cannot be controlled via casting, one of the growth-friendly instrumentation techniques is chosen. Growth-friendly surgical methods and implants have been evolving as understanding of the disease improves.


Key points








  • Current literature supports the use of serial casting for most progressive early onset spinal deformities as either a definitive or a surgical delay treatment.



  • If and when surgery is required for early onset scoliosis treatment, the most commonly used system is growing rods, but this may change in the future with other viable options undergoing research, including magnetically controlled growing rods, nitinol staples, tethering, and the Shilla procedure.



  • Expansion thoracoplasty is generally recommended for patients with thoracic insufficiency syndrome.



  • Future goals include scoliosis treatment methods that involve less complications and to successfully treat patients with “fusionless” methods.






Introduction


Early onset scoliosis (EOS), as defined by Dickson, is severe spinal deformity of any cause affecting children less than 5 years of age. This definition is different from the standard definition of infantile scoliosis, taking into consideration the unique and profound respiratory consequences associated with moderate to severe curves in this age group. Spinal growth peaks by the age of 5; however, pulmonary development continues until the age of 8. A severe scoliotic spine in this young population impedes the multiplication of pulmonary alveoli, which are supposed to reach near adult numbers (300 million) by 8 years of age, thus restricting overall pulmonary development.


EOS curves can be broadly categorized based on their cause, including idiopathic, neuromuscular, syndromic, and congenital. Idiopathic scoliosis is a diagnosis of exclusion, defined as scoliosis that is not associated with any other abnormality. Neuromuscular scoliosis patients have innate abnormalities of the neuromuscular system. Common examples are cerebral palsy, spinal muscular atrophy, and muscular dystrophies. Syndromic scoliosis is a broad term that defines scoliotic curves associated with syndromes such as Marfan, Ehlers-Danlos, Pierre Robin, and velocardiofacial syndromes. Congenital scoliosis is classified either by a failure of vertebral formation or by segmentation. If left untreated, EOS curves can likely cause cosmetic disfigurement, rib-cage distortion, dyspnea, and cardiorespiratory failure in early adult life.


The previous standard of care for these children was early definitive anterior and posterior spinal fusion. Complications of this method, including crank shaft phenomenon and thoracic insufficiency, led to very modest results. Today “a short but straight spine” is no longer considered to be a good outcome for the treatment of EOS and providers have moved toward strategies for avoiding curve progression while still allowing for spine and thorax growth, and unrestricted multiplication of the pulmonary alveoli. The quest for “fusionless scoliosis correction” started to gain momentum in the previous decade. Nonsurgical and surgical treatment options continue to evolve as the understanding of the pathologic abnormality continues to improve and results of current treatment methods become available.




Introduction


Early onset scoliosis (EOS), as defined by Dickson, is severe spinal deformity of any cause affecting children less than 5 years of age. This definition is different from the standard definition of infantile scoliosis, taking into consideration the unique and profound respiratory consequences associated with moderate to severe curves in this age group. Spinal growth peaks by the age of 5; however, pulmonary development continues until the age of 8. A severe scoliotic spine in this young population impedes the multiplication of pulmonary alveoli, which are supposed to reach near adult numbers (300 million) by 8 years of age, thus restricting overall pulmonary development.


EOS curves can be broadly categorized based on their cause, including idiopathic, neuromuscular, syndromic, and congenital. Idiopathic scoliosis is a diagnosis of exclusion, defined as scoliosis that is not associated with any other abnormality. Neuromuscular scoliosis patients have innate abnormalities of the neuromuscular system. Common examples are cerebral palsy, spinal muscular atrophy, and muscular dystrophies. Syndromic scoliosis is a broad term that defines scoliotic curves associated with syndromes such as Marfan, Ehlers-Danlos, Pierre Robin, and velocardiofacial syndromes. Congenital scoliosis is classified either by a failure of vertebral formation or by segmentation. If left untreated, EOS curves can likely cause cosmetic disfigurement, rib-cage distortion, dyspnea, and cardiorespiratory failure in early adult life.


The previous standard of care for these children was early definitive anterior and posterior spinal fusion. Complications of this method, including crank shaft phenomenon and thoracic insufficiency, led to very modest results. Today “a short but straight spine” is no longer considered to be a good outcome for the treatment of EOS and providers have moved toward strategies for avoiding curve progression while still allowing for spine and thorax growth, and unrestricted multiplication of the pulmonary alveoli. The quest for “fusionless scoliosis correction” started to gain momentum in the previous decade. Nonsurgical and surgical treatment options continue to evolve as the understanding of the pathologic abnormality continues to improve and results of current treatment methods become available.




Current treatment and trends


Mehta developed a method to distinguish progressive from spontaneously resolving EOS, creating a means of early detection of curves that requirement treatment. This method uses the rib vertebral angle of difference (RVAD) between the apical vertebra and the corresponding concave and convex side ribs. The RVAD in resolving curves is typically less than 20° on initial radiograph and decreases on subsequent radiographs taken 2 or 3 months apart. The RVAD in progressive curves is typically greater than 20° on initial radiograph and either remains unchanged or increases on subsequent radiographs. In Mehta’s study, this pattern was noted in 83% of children with progressive scoliosis. Following this method, those patients who are diagnosed with progressive EOS should begin treatment immediately.


Nonoperative Treatment


Currently, the only widely accepted nonoperative management method that has shown effectiveness in the EOS population is serial casting. Bracing is often used as an adjunct to serial casting, but there is not enough evidence to support the use of bracing as a sole management method. A retrospective review by Smith and colleagues showed that bracing was the only treatment (compared with casting and expansion thoracoplasty [ET]) in their study that did not provide adequate curve control. In addition, the authors note that the patients for whom bracing treatment was effective had spontaneously resolving curves that may have improved despite bracing due to their significantly smaller RVAD and Cobb measurements. Another consideration in bracing treatment is brace-wear compliance. There is no study specifically on compliance in EOS patients, but compliance has been shown to be challenging in other patient populations; it may be inferred that the EOS population would encounter some of the same issues.


In contrast to bracing reports in the EOS population, casting has been thoroughly evaluated and has shown very good results. There are numerous earlier reports of casting present in orthopedic literature, but the serial corrective casting technique for EOS, as used today, was first described by Cotrel and Morel in 1964. Forty years passed until results from this technique were reported and gained acceptance as a definitive treatment method for certain deformities. As subsequent clinical results became available, serial casting became a well-accepted method for the management of EOS as both a definitive treatment and a surgical delay tactic.


Serial casting harnesses the patient’s own growth as a corrective force. An infant grows almost twice as fast during their first year, and as fast during their second year as adolescents do during their growth spurt. Mehta reports that harnessing the rigorous growth of infancy can straighten progressive curves that would otherwise develop into severe deformities. Historically, Risser casting was a popular method of scoliosis casting but, due to chest and abdominal expansion restriction and decreased chest wall compliance, it has since become nearly extinct in this capacity. Modern forms of casting are referred to as EDF casting: elongation, derotation, and flexion. Spinal deformities are 3-dimensional (3D) deformities, and EDF casts are able to address all dimensions. During the casting application, the patient is put in traction to “elongate” and gently derotate the spinal deformity; the corrected position is then held by the applied cast. Risser casts were only able to straighten the spine in one plane, allowing the deformity continued development. Mehta explains that the shape of an organ or part determines the direction of its continued growth; “if the direction remains constant, growth will simply perpetuate and enlarge its existing shape.” If a curved spine is left untreated, it will grow more curved, but with the external force of the cast, especially during a period of rapid growth, the spine will continue to grow in the corrected position and the final shape can be altered. In addition, EDF casts use a large “mushroom”-shaped cutout on the front of the cast to allow for proper chest expansion, as well as a smaller cutout on the back over the concavity to allow for rib-cage expansion, which better aligns the spine and corrects rotation ( Fig. 1 ).




Fig. 1


Mehta casting. ( A ) Pretreatment supine radiograph of 37° curve in an 8-month-old girl. ( B ) Radiograph of patient in applied cast. ( C ) Radiograph of 13° curve in same patient at 37 months from initial cast. ( D ) Clinical photo of patient in cast (front). ( E ) Clinical photo of patient in cast (back).


For those patients whose scoliosis is completely or “definitively” corrected via serial casting, success is correlated with treatment beginning during the first 2 years of rapid growth (ie, before the age of 2). In Mehta’s study, all curves in patients (with syndromic or idiopathic scoliosis) who began casting between 15 and 21 months old with Cobb angles between 27° and 35° resolved; conversely, curves in patients who began their treatment between 27 and 34 months old with Cobb angles between 47° and 53° did not resolve. Further evidence has shown that idiopathic curves up to 60° can resolve if treatment begins before the age of 2. In the definitive treatment group, casting is usually discontinued when radiographs show restoration of rib-cage symmetry, derotation of apical vertebra, and complete, or nearly complete, correction of the curve. These patients are generally prescribed a bracing regimen for 6 months to a year to ensure that the spine stabilizes in its new position. If the spine continues to show complete correction, brace wear may be ceased at this time and the patient should be followed clinically at intervals of 6 months and then annually until skeletal maturity.


For those patients whose curves cannot be definitively treated with serial casting, this treatment can still serve as a means to delay surgical intervention and its associated complications, including wound-healing problems, infection, premature fusion, implant failure, and many others. Furthermore, the concept of “diminished returns,” or the progressive decrease in spinal length gained at each lengthening surgery, is of great concern in these patients. This phenomenon is presumably related to spinal autofusion and, according to Sankar and colleagues, is especially notable after the seventh lengthening procedure. Their study found that the average distraction gained at the initial lengthening surgery was 1.04 cm and decreased to 0.41 cm by the seventh lengthening. Following the typical 6-month lengthening schedule, these effects are seen in just 3.5 years. A patient beginning treatment at 4 years of age would experience these effects by age of 7, before full lung development. In a study of serial casting outcomes conducted in 2012 by Fletcher and colleagues, surgery was delayed in 15 of 29 patients by 39 ± 25 months (the equivalent of nearly 7 growing rod lengthenings) from the time of their first cast. The other 14 patients had not required surgery at publication. Overall, 72% of the patients had avoided surgery at an average of 5.5 years after starting casting treatment. These patients had both congenital and neuromuscular diagnoses with Cobb angles greater than 50° and did not begin their casting treatment until an average age of 4.4. Surgical intervention may even be delayed until final fusion is indicated, bypassing any need for growth-friendly surgical techniques.


Although casting prevents most complications associated with surgery, it does come with its own set of complications. Most are minor skin problems; however, a case report of lateral subclavian vein thrombosis in a patient with a Cotrel cast suggests that casting treatment is not free of more severe complications. This serious potential complication may be avoided by careful and generous trimming of the cast at the groin and axillary regions. Another concern is that the application of the rigid and molded cast over the chest wall may impair ventilation. A recent study investigating the change in pulmonary pressures during and after the application of scoliosis casts in children with EOS reported increased peak inspiratory pressures during the application of the cast, with a decrease (but still higher than the baseline) after casting windows were cut out. They reported difficulty in maintaining ventilation in two procedures, one hypotensive episode, one case of hypoxemia after casting application and breathing difficulties in one patient. In addition to medical complications, there is concern that casts and braces restrict a child’s interaction with their environment and with others, which may, in turn, create psychological stigma and impede social development.


In conclusion, nonoperative treatment of EOS patients is available in the form of serial casting. It may serve as definitive treatment (generally for idiopathic and syndromic curves that are treated before the age of 2) or as a means to delay surgical intervention. Because of the significant differences in outcomes that can be caused by a delay in casting treatment of even a few months, Mehta suggests confirming the EOS diagnosis by measuring the RVAD on 2 successive radiographs as previously discussed. If progressive scoliosis is diagnosed, treatment should begin immediately. Mehta found that delays in treatment were often caused by physicians who monitored curve progression over too long a period. At times, physicians stated the monitoring process was extended while they were investigating other anomalies; Mehta recommends treating the scoliosis in conjunction with these investigations, which often occur in EOS patients.


Operative Treatment


Surgical indications for EOS patients may still greatly vary between providers, but a survey of the Growing Spine Study Group found that surgery was often recommended when curves progressed to 60° in patients younger than 8 to 10 years of age. Historically, definitive fusion was the only reliable surgical treatment for these patients, but outcomes were less than desirable with continued deformity, low rates of circumferential fusion, and poor pulmonary function and cosmesis. In light of these poor outcomes, there has been a transition to growth-friendly surgery. This transition has also been encouraged by considerable advances in growth-friendly instrumentation and techniques, which are resulting in better outcomes. Recently, Skaggs and colleagues proposed a classification of these techniques to create common terminology and to facilitate comparative studies. According to this classification, currently available techniques fall into 3 categories: distraction-based, compression-based, and guided growth systems.




Distraction-based systems


Growing Rods


Currently, growing rods are probably the most commonly used surgical management method for EOS in North America. This technique has existed for nearly 2 decades, promoted first by Paul Harrington, who began with a single growing rod construct. This single growing rod has been replaced with a dual rod technique due to evidence of better correction, increased stability, and a greater proportion of expected spinal growth. Although the technique has changed, the premise has remained the same: provide distraction between the ends of a scoliotic curve by anchoring instrumentation to the spine proximally and to the spine or pelvis distally, avoiding instrumentation in the intervening spinal segment ( Fig. 2 ). The rods are composed of 2 sections joined by a tandem connector. As the spine grows, the distraction needs to be “reset” by surgically lengthening the rods (typically every 6 months). The goal is to provide distraction through the early adolescent growth period until sufficient vertebral column growth is achieved. Generally, at the termination of growing rod treatment, they are replaced with a permanent rod system during a definitive fusion.




Fig. 2


Growing rods. ( A ) Preoperative posterior anterior (PA) radiograph of an 8-year-old boy. ( B ) Preoperative lateral (LAT) radiograph. ( C ) Postoperative PA radiograph of same patient 30 months from initial surgery. ( D ) Postoperative LAT radiograph.


It has been established by numerous studies that repeated growing rod distractions provide maintenance of deformity correction and promote spinal growth, possibly even greater than normal growth rates. The Hueter-Volkmann law states that spinal distraction may promote growth of individual vertebral bodies. This theory is supported by the findings of a recent study that found there was considerably more growth per vertebra within the instrumented zone compared with the vertebra outside the instrumented zone. Other studies have concluded that patients lengthened at intervals of less than 6 months have a higher annual growth rate than those lengthened between 9 and 20 months (1.8 cm and 1.0 cm, respectively).


In a study of general growing rod outcomes in 23 patients with idiopathic, congenital, and syndromic scoliosis, Akbarnia reported a curve reduction from an average of 82° preoperatively to 36° at final follow-up and an average of almost 9 cm in growth from T1-S1. Patients underwent an average of 6.3 lengthening procedures and had a 48% complication rate. Spinal elongation from T1-S1 averaged 10.7 cm after final fusion. In addition, Elsebai and colleagues were also able to confirm that growing rods are able to improve the space available for lung (SAL). Overall advantages of distraction-based systems are regeneration of disks by relieving surrounding pressure and restoration in the decrease of disk height and vascular channel volume in the endplates caused by compression.


Growing rod complications include high rates of implant breakage and pull-out and wound complications. As previously discussed, there appears to be a phenomenon of “diminishing returns” with multiple lengthenings. Diminishing returns with multiple lengthenings is thought to be due to increasing spine stiffness, but whether it represents autofusion or ankylosis of the instrumented segment is unclear. Numerous studies report a complication rate of 40% to 60%, with chances increasing with each additional surgical procedure by as much as 24%. Although there is a high rate of complications, most are addressed during planned lengthening surgeries and the occurrence of major complications is rare. Regarding neurologic safety, growing rod placement and lengthening procedures are extremely safe with an overall neuromonitoring change incidence of 0.9%.


Magnetically Controlled Growing Rod System


There have been major advances toward the clinical use of a magnetically controlled growing rod (MCGR) in the past few years. The implant includes a telescopic actuator that contains a small internal magnet; otherwise, the construct is the same as traditional growing rods. The internal magnet is rotated by an electrically powered remote control, which is placed externally over the spine to lengthen or shorten the rod. Lengthenings can be performed in an outpatient setting without anesthesia or analgesia. Because of the inverse relationship between growth rate and lengthening intervals and the relative ease of lengthenings, 3-month lengthening intervals are recommended to optimize spinal growth.


Dannawi and colleagues released findings on the largest series of MCGR use in humans. Thirty-four patients, diagnosed with EOS of any cause, received either a single or a dual MCGR (mean age at implant was 8). Their goal was to distract the rod faster than the rate of predicted growth to allow for better curve correction. At follow-up, they reported a mean of 4.8 distractions (minimum of 3), a mean of 66 days to first distraction, and a mean of 87 days between distractions. Both groups showed significant reduction in mean Cobb angles from preoperative to immediate postoperative; a significant difference between mean preoperative, immediate postoperative, and final distance between T1 and S1; a significant increase in the mean distance between T1 and S1 from preoperative to immediate postoperative; and an overall significant difference between the mean preoperative, immediate postoperative, and final Cobb angles. Although both constructs showed significant results, the dual rod improved the Cobb angle significantly better than the single rod.


Like growing rods, the MCGR construct also comes with complications. In Dannawi’s study, one patient in each group had a superficial wound infection and a broken rod requiring revision. Two single rods lost distraction, which was rectified by subsequent lengthening. However, there were no spontaneous fusions or neurologic deficits and, in comparison with traditional growing rods, these patients experienced less wound infections and broken rods. It should be noted that this study had a shorter follow-up period than most growing rods studies. In addition, it is likely that stiffness, spontaneous fusions, and “diminished returns” will also be observed with this technique, but avoidance of multiple surgeries is a colossal advantage over conventional growing rods. The numerous complications associated with surgical lengthenings, including wound infection, rod breakage, autofusion, anchor failure, and implant prominence, should be greatly reduced. There is no evidence that the electromagnetic field causes any persistent or major side effect with repeated distraction. Although these results are promising, longer follow-up is necessary to clarify the effectiveness and the complication rates of this exciting device. MCGR was approved for use in the United States in February 2014.


ET, Vertical, Expandable, Prosthetic Titanium Rib, and Hybrid Rib to Spine Devices


In contrast to growing rods, which focus on the spinal deformity, Expansion Thoracoplasty (ET) focuses on the thoracic deformity. Typical constructs use a proximal rib anchor (hook or a rib cradle) and connect to either another rib, to the lumbar spine, or to the posterior iliac crest. Vertical, expandable, prosthetic titanium rib (VEPTR) is an instrument specifically developed for this purpose and is the most commonly used construct for ET. As with growing rods, the construct is surgically lengthened at 6-month intervals ( Fig. 3 ). The deformed thoracic segment is not exposed and there is no intervention in the spinal column. The expansion of the chest indirectly provides correction of the thoracic spinal deformity.


Feb 23, 2017 | Posted by in ORTHOPEDIC | Comments Off on Recent Advances in the Management of Early Onset Scoliosis

Full access? Get Clinical Tree

Get Clinical Tree app for offline access