32 Is There a Gold-Standard Surgical Option?
Defined as scoliotic spinal deformity of any etiology presenting before the age of 5 years, early onset scoliosis is one of the most challenging problems to treat in the world of pediatric orthopedic surgery. Dickson was the first to use the term early onset to reflect the establishment of scoliosis by 5 years of age. 1 Early onset scoliosis has a long list of possible etiologies. These include congenital, idiopathic, and neuromuscular conditions, as well as various syndromes. Nevertheless, the use of the term early onset scoliosis to describe these disorders collectively, regardless of the etiology, highlights an important concept: the age of the patient at onset is a crucial factor when treatment is considered because a deformed spine has major effects on the developing thorax and cardiopulmonary system. 2
Left untreated, early onset scoliosis has grave cardiopulmonary and skeletal consequences that were observed and documented as early as the middle of the last century. James et al, in 1959, reported that progressive early onset scoliosis “develops rapidly and relentlessly, causing the severest form of orthopaedic cripple with dreadful deformity, marked dwarfing and shortening of life.” 3 In 2003, Campbell et al used the term thoracic insufficiency syndrome to describe the compromised pulmonary development associated with severe early onset scoliosis. They defined thoracic insufficiency syndrome as the inability of the thorax to support normal lung growth and respiration. Disabling or life-threatening respiratory failure as a result of thoracic insufficiency syndrome is relatively common at, or before, late middle age. 4 , 5 , 6
The ultimate goal in the treatment of early onset scoliosis is to control the progression of deformity while allowing the development and growth of the spine, lungs, and thorax, thereby improving pulmonary function and providing a better quality of life. 2 , 7 Conservative options, such as casting, bracing, or a combination of both, are commonly employed as the first line of treatment. However, severe, relapsing, or progressing deformities may require early surgical intervention to protect the inherent growth potential for spinal height and lung development. For decades, a prevailing belief that a spine that was “short and straight” as a result of early fusion was preferable to one that was “longer but crooked” had been supported. 8 , 9 , 10 Therefore, the concept of early definitive fusion as a treatment was embraced. However, spinal fusion was not without its complications. The crankshaft phenomenon, described by Dubousset et al, and thoracic insufficiency syndrome, recognized by Campbell et al in young children who had undergone early spinal fusion for early onset scoliosis, have both prompted attempts to discover new modalities of treatment.
Skaggs classified fusionless surgical procedures into distraction-based, guided-growth, and tension-based systems. 7 Each of these strategies follows different principles, has different biomechanical characteristics, and has variable ability to correct deformity and maintain the correction while protecting the potential for growth. However, each strategy continues to evolve. Since the beginning of the last decade, we have witnessed a rapid expansion of the philosophy of treatment for early onset scoliosis. This has resulted in a better understanding and continued modifications of the widely accepted distraction-based systems (growth rods, vertical expandable prosthetic titanium rib [VEPTR]) as standard treatments. Additionally, there has been a revival of interest in the abandoned guided-growth systems following the emergence of newer techniques (modern Luque trolley, Shilla technique). Finally, the innovative tension-based systems (staples and tethers) also show exciting potential.
Despite the wide range of surgical treatment options for progressive early onset scoliosis, no solution has emerged as the single gold-standard option. The intention of this review is not to make treatment recommendations, but to document the options available to date and to examine the supporting evidence for each.
32.1 Methods
The keywords scoliosis and early onset were used in an electronic search of both PubMed and MEDLINE conducted on April 1, 2012. Inclusion criteria were that articles had been written in the English language and published after the year 2000 because since 2000, new surgical options have arisen while older methods have fallen out of favor.
The search for scoliosis yielded 5,764 titles, which were not all necessarily relevant to the objective of this particular review. When the term early onset was added as a keyword, 116 articles were found. The titles and abstracts of all 116 articles were examined to determine which were relevant to the quest for “surgical treatment methods.” Articles published in nursing journals, case reports, genetic reports, biomechanical studies, and reviews were then excluded. Articles on specific rare syndromes, such as neurofibromatosis and spinal muscular atrophy (SMA), as well as articles about scoliosis with a specific etiology (e.g., congenital or neuromuscular scoliosis), were also excluded from the review. A copy of the full article was obtained for each of the remaining titles. The references for all included articles were examined and checked for any relevant papers, and these papers were included.
A total of 19 articles were included in the final review. These were classified according to the modality of treatment they considered: (1) distraction-based, (2) guided-growth, or (3) tension-based systems. Thus, 13 papers were found to report distraction-based systems, two papers on guided-growth systems, and four papers on tension-based systems. The included articles in this review comprise case series as well as some experimental studies of new modalities. No randomized controlled studies were found.
32.2 Review
Although an algorithm for managing early onset scoliosis has been proposed by Gillingham et al, 4 no rigid indications for the different surgical interventions currently exist. Numerous options are available and at the disposal of the treating surgeon. Vitale et al investigated variability in decision making with regard to the treatment of early onset scoliosis among a group of 13 experienced pediatric spinal surgeons. 11 They reported a wide variation in choice of construct type, number of constructs, and level of instrumentation. This review describes the systems based on three surgical principles (distraction, guided growth, and tension), with particular attention given to reporting recent advances and their updated outcomes.
32.3 Distraction-Based Systems
32.3.1 Growing Rods
The concept of using growing rods in the management of early onset scoliosis has been around since the 1960s. Harrington was the first to report on the technique in 1962. 12 He connected a single rod to the spine with a single hook proximally and a single hook distally and performed periodic lengthening. Since then, several modifications have been introduced to improve outcomes, and up until 2004, several authors reported on their results, including Moe et al, 13 Klemme et al, 14 Blakemore et al, 15 and Mineiro and Weinstein. 16 Their articles described the same principle of active distraction with a single rod. They all reported variable but high complication rates. It was not until 2005 that Akbarnia et al published their first report on the dual growing rod technique. 17 Since the introduction of this concept (dual rods instead of a single rod), focus has shifted away from single rods, and multiple reports and experiments have been published. Because the aim of this review is to examine current practice, we considered the earlier reports on single growing rods to be beyond its scope and excluded them. Seven articles on dual growing rods were found in our search, and these are the subject of this part of the review (Table 32.1).
Author (reference) | Year | No. | Intervention | Follow-up (range) | Deformity correction, % | Complications, No. (%) |
Akbarnia et al 17 | 2005 | 23 | DGRs | 4.7 yrs (2–9) | 54% | 11 (48%) |
Akbarnia et al 18 | 2008 | 13 (lengthening <6 months [7] and >6 months [6]) | DGRs | 3–11 yrs | 64% | 6 (46%) |
Sankar et al 19 | 2011 | 38 | DGRs | 3.3. yrs (2–7) | 52% | |
Bess et al 20 | 2010 | 69 | DGRs | 53.8 months (24–126) | 48.7% | 38 (55%) |
Schroerlucke et al 21 | 2012 | 90 (26 K–, 35 N, 29 K+) | DGRs (64) SGRs (26) | >2 yrs | 12 in K– (46%) 12 in N (34%) 18 in K+ (62%) | |
Sponseller et al 22 | 2009 | 36 (severe deformity extending to pelvis) | DGRs (30) SGRs (6) All fixed to pelvis distally | 40 months (± 20) | 44% | |
Yang et al 23 | 2011 | 327 | DGRs (206) SGRs (121) | Only rod fractures (11% in DGRs, 26% in SGRs) | ||
Abbreviations: DGRs, dual growing rods; K–, hypokyphosis; N, normal sagittal balance; K+, hyperkyphosis; SGRs, single growing rods. | ||||||
In 2005, Akbarnia et al published their first retrospective review of their case series, reporting on the technique and its early results. 17 They used dual growing rods to treat 23 children with early onset scoliosis of various etiologies between 1993 and 2001, with a minimum of 2 years of follow-up (average, 4.7 years). They described the technique of connecting two upper and two lower rods to the upper and lower regions of the spine, respectively, with two or more vertebral levels used at each end as foundations. The upper and lower rods were then connected with a tandem connector at the thoracolumbar junction. All instrumentation was done through a limited subperiosteal exposure. Distraction was applied, and the tandem connector was tightened. This procedure was followed by serial lengthening procedures at an average of 7.4-month intervals. Akbarnia et al reported 56% improvement in the Cobb angle (from an average of 82 degrees preoperatively to 36 degrees at the latest follow-up). There were 13 complications in 11 patients (48%) during the treatment period.
In 2008, Akbarnia et al reported on their series of 13 children with noncongenital early onset scoliosis who had completed dual growing rod treatment before final fusion. 18 They found even better results after the final fusion (64% improvement in the Cobb angle and 45% increase in the T1-S1 length). In this report, they examined the effect of frequency of lengthening and found that patients who underwent lengthening at intervals of less than 6 months had a higher annual growth rate (1.84 vs. 1.02 cm) and even a significantly greater correction of scoliosis (from 89 to 20 degrees, or 78%) than those who underwent lengthening less frequently (48%). More recently, Sankar et al 19 reported the overall results of their study, which were comparable with those in previous reports by Akbarnia et al (Cobb angle correction from 74 to 35 degrees and T1-S1 growth of 1.74 cm per year). They also noticed a “law of diminishing returns” in T1-S1 growth with repeated lengthening procedures (1.04 cm after the first lengthening but only 4 mm after the seventh) as a result of autofusion. As in previous studies, correction of the Cobb angle was achieved predominantly following the initial instrumentation. These findings are significant in a number of ways. They warn us not to expect too much distraction intraoperatively in a child who has undergone multiple lengthening procedures, and so to avoid excessive distraction and subsequent instrumentation failure. Also, they suggest delaying the initial surgery and stopping treatment after fewer lengthening procedures to prevent early autofusion and reduce the risk for complications from unnecessary surgeries.
Variable but high complication rates following dual growing rod surgery have been reported, including wound problems (superficial and deep infections); implant problems (fracture, loss of fixation, and implant prominence); alignment complications (kyphosis and curve progression); and general complications. 20 Because the Growing Spine Study Group (GSSG) is a multicenter international organization that houses a registry of patients with early onset scoliosis, it is not surprising that they have published the largest and most inclusive evaluations of complications. Bess et al in 2010 reported on 140 patients treated with growing rods (single and dual) over 18 years. 20 Of these, 69 were treated with dual growing rods, and they are a subject of interest in this review. Of the 69 children, 38 (55%) had 83 complications (average of 1.2 complications per patient) requiring 32 unplanned procedures. Two factors were found to increase the risk for a complication: the number of surgical procedures (24% increased risk with each additional procedure beyond the index surgery) and younger age at the time of index surgery (13% reduction in risk for each year of increase in age). Two points are worth noting here. First, in some of the children, the growing rods were implanted subcutaneously, and these children had the highest rate of complications. The subcutaneous placement of rods is not recommended anymore, and the complication rate might have been less if all rods had been implanted submuscularly. Secondly, as highlighted by Karol, 24 only 14 children had undergone final fusion at the time of the report. Therefore, the reported risk rate is likely to increase as the remaining children undergo further lengthening procedures.
The GSSG continued to investigate the outcome of growing rod surgery. Schroerlucke et al studied the effect of preoperative thoracic kyphosis on the complication rate. 21 They reported a 3.1 times greater risk for a complication in children with kyphosis of more than 40 degrees than in children with normal kyphosis. They also found an increased risk for implant-related complications (especially rod fractures) in the group with hyperkyphosis. Even though this increase was not statistically significant, it surely is of clinical importance. Yang et al 23 reported a 15% rate of rod fractures (31 fractures in 206 children treated with dual growing rods). In their study of 327 children treated with growing rods (single and dual), they identified ambulation, syndromic diagnosis, single rods, stainless steel rods, small-diameter rods, and small tandem connectors as risk factors for rod fractures. Interestingly, preoperative hyperkyphosis was not found to be a risk factor in the analysis of Yang et al (contradictory to the findings of Schroerlucke et al).
Growing rods have proved to be a very effective, versatile, and reproducible treatment for early onset scoliosis. However, a number of questions remained unanswered. Can we avoid the high risk for complications associated with repeated lengthening surgeries? Can growing rods be used in severe early onset scoliosis extending to the pelvis? Can we somehow avoid proximal thoracic fusion and preserve motion? And apart from their known effect on the growing spine, do growing rods influence the thoracic geometry as well? Investigators undertook a quest to find answers to these questions.
In 2010, Sabourin et al developed a special imaging system to create a three-dimensional reconstruction of the spine and rib cage. 25 They evaluated seven children who had undergone growing rod treatment and found that growing rods corrected not only spinal deformity but also chest wall deformity (rib orientation), thoracic axial rotation, and thoracic asymmetry. Sponseller et al reviewed the results of growing rods anchored distally to the pelvis in 36 children. 22 The indications for such a procedure were either severe early onset scoliosis extending to the pelvis or a lack of alternative anchor points in the lumbar spine. Once again, growing rods (especially dual rods) lived up to their expectations. The children showed substantial improvement in coronal and sagittal balance as well as pelvic obliquity and spinal growth. The rods were well tolerated without any increase in the rate of complications.
Repeated lengthening operations remain a big obstacle to reducing the rate of complications in growing rod surgery. However, very exciting reports have emerged recently indicating the beginning of a new era in the development of growing rods. A noninvasive, remotely distractible, magnetically controlled growing rod system has been developed. 26 Akbarnia et al first evaluated its safety and efficacy in an animal model, and the results were very encouraging. 27 Currently, the magnetically controlled growing rod system is being applied in multiple centers around the world, and preliminary results are promising. 26 This technique allows noninvasive lengthening of the rods with precise incremental distraction. Two systems currently exist: the MAGEC rod and the Phenix rod. The MAGEC Remote Control Spinal Deformity System (Ellipse Technologies, Irvine, California) is supplied sterile and can be implanted as a single rod or as dual rods. If dual rods are implanted, each can be individually adjusted with an external remote controller. Rod lengthening is done as an outpatient procedure, usually on a monthly basis. The Phenix rod, on the other hand, is used as a single rod. It is delivered clean but nonsterile. The rod is custom-made to specifications determined by the child’s spinal deformity. Parents are given a magnetic device that they hold over the child’s back and rotate to extend the rod. They turn the magnet once a day to lengthen the rod by 0.2 mm.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
