- Email: [email protected]
Scoliosis and the Impact in Neuromuscular Disease
Paediatric Respiratory Reviews 16 (2015) 35–42
Contents lists available at ScienceDirect
Paediatric Respiratory Reviews
Mini-symposium: Chest Wall Disease
Scoliosis and the Impact in Neuromuscular Disease Oscar Henry Mayer * Associate Professor of Clinical Pediatrics, Perelman School of Medicine at The University of Pennsylvania, Division of Pulmonary Medicine, The Children’s Hospital of Philadelphia, 3501 Civic Center Boulevard, Philadelphia, PA 19104
EDUCATIONAL AIMS Understand the differential pathophysiology of spinal muscular atrophy and Duchenne muscular dystrophy and how each can lead to scoliosis. Understand the considerations involved for when in the progression of scoliosis to intervene. Understand the options available for surgical intervention.
A R T I C L E I N F O
S U M M A R Y
Keywords: Scoliosis Spinal surgery Restrictive lung disease Spinal muscular atrophy Duchenne muscular dystrophy
Scoliosis can alter respiratory mechanics by changing the orientation of the muscles and joints of the respiratory system and in severe forms can put a patient at risk of severe respiratory morbidity or respiratory failure. However, perhaps the most important factor in determining the pulmonary morbidity in scoliosis is the balance between the ‘‘load’’ or altered respiratory mechanics and the ‘‘pump’’ or the respiratory muscle strength. Therefore, scoliosis in patients with neuromuscular disease will both lead to increased ‘‘load’’ and a weakened ‘‘pump’’, an exceptionally unfortunate combination. While progressive neuromuscular disease by its nature does not respond favorably to attempts to improve respiratory muscle strength, the natural approach of early proactive management of the ‘‘load’’ and in the case of scoliosis a variety of different strategies have been tried with variable short term and long term results. Figuring this out requires both an understanding of the underlying pathophysiology of a particular neuromuscular condition and the available options for and timing of surgical intervention. ß 2014 Elsevier Ltd. All rights reserved.
INTRODUCTION Patients with progressive hypotonic neuromuscular disease (NMDz) such as spinal muscular atrophy (SMA) and Duchenne muscular dystrophy (DMD) face a gradual decline with increasing respiratory morbidity progressive to the point of respiratory failure. There is a broad range in the rates of decline of the different types of NMDz, but the end point of respiratory failure is often the same. Unfortunately, what is also the same is the current absence of effective treatment options for each of the conditions to prevent the neuromuscular disease that drives this decline. Therefore, the historic and current focus to NMDz treatment is symptomatic and, to the extent possible, preventative therapy.
* Tel.: +215 590 3749; fax: +215 590 3500. E-mail address: [email protected]. http://dx.doi.org/10.1016/j.prrv.2014.10.013 1526-0542/ß 2014 Elsevier Ltd. All rights reserved.
While the focus of therapy in NMDz is in airway clearance and supporting mechanical ventilation using a variety of proven therapies, there has been a great deal of work over the years to support the respiratory system by treating the progressive scoliosis that occurs in NMDz. Scoliosis in patients with NMDz is different in many ways from that in patients with intact muscle function. First, the defect in patients with NMDz is based on weakness of the entire muscular component of the thorax, while in congenital scoliosis the scoliosis is typically caused by a significant skeletal defect or in idiopathic scoliosis by an asymmetry in the peri-spinal ligamentous and muscular support of the thorax. Second, because of the diffuse respiratory muscle weakness in NMDz there is less potential to resist the skeletal imbalance that occurs with progressive scoliosis, and as a result it can progress more rapidly. Third, the altered respiratory mechanics produced from scoliosis puts a burden on patients with NMDz that can more
36
O.H. Mayer / Paediatric Respiratory Reviews 16 (2015) 35–42
Figure 1. 18 year old female with SMA-2 and uncorrected scoliosis demonstrating severe unilateral rib cage collapse in the convex chest.
easily overwhelm their respiratory muscle function and lead to respiratory failure than in patients with scoliosis and intact muscle strength. This manuscript will explore the unique features of scoliosis in NMDz and the different ways that it develops in conditions with different pathophysiology, SMA and DMD. PATHOPHYSIOLOGY OF SCOLIOSIS IN NEUROMUSCULAR DISEASE In very general terms, spinal muscular atrophy causes more prominent weakness in the muscles of the chest wall than diaphragm, while in DMD the diaphragm is typically weaker with relative preservation of the chest wall muscles [1]. In SMA the
Figure 3. AP chest radiographs in a patient with SMA-2 with progressive scoliosis at a) 10 months of age; b) 3 years of age; c) 5 years of age; and d) 6.5 years of age.
Figure 2. 15 year old male with DMD and uncorrected scoliosis and upward diaphragm displacement.
O.H. Mayer / Paediatric Respiratory Reviews 16 (2015) 35–42
37
scoliosis develops early in the disease process [2], likely due to the substantial respiratory muscle weakness at diagnosis and in the case of SMA-2, patients are able to sit upright, but are nonambulatory [3]. In DMD the scoliosis occurs well in to the disease progression often not until early adolescence [4]. This coincides with the loss in ambulation and becoming wheelchair dependent usually around 11-13 years of age (Richard S. Finkel, MD unpublished data). The primary difference in the scoliosis in SMA compared to that in DMD is the presence of significant (usually unilateral) chest wall collapse in SMA with relatively well preserved rib positioning in DMD, but both can cause a significant scoliosis (Figures 1 and 2). Spinal Muscular Atrophy In SMA absent or very low production of the survival motor neuron (SMN) protein that is involved in maintaining neuronal integrity leads to progressive loss of efferent nerve signaling through the anterior horn cells of the spinal cord [5]. This causes severe peripheral muscle weakness and in the thorax prominent intercostal muscle weakness [5] and to scoliosis in between 60 and 95% of patients [6]. High chest wall compliance is the hallmark of SMA-1 in which patients are never able to sit upright without significant support and assistance [3]. In these patients there is often bilateral caudal rib rotation of the more superior ribs that produces a narrow superior chest and a typical triangular shaped chest with preservation of the lower chest contour due to the incompressible abdominal contents. In patients with SMA-1 who survive past infancy and are positioned in a semi-upright position because of their lack of truncal support [3] their thorax will bend in one direction and they will develop scoliosis [7]. In broad longitudinal reviews of patients with SMA-2, the onset of scoliosis has been reported as early as 4 years of age [2] and progressed at annual rates of between 58 and 158 per year [2]. With this progression in scoliosis comes a coincident decrease in forced vital capacity (FVC) with a loss of almost 5% predicted for every 108 increase in scoliosis curve [8]. Unlike in DMD, where there are often years of normal lung FVC before the onset of scoliosis and the respiratory decline, in SMA the trend in decline of FVC is not as well characterized since it begins before reliable testing can be performed between 5 and 6 years of age. This scoliosis happens most prominently in SMA-2 patients who can sit up without assistance but not stand. In patients with milder SMA-3 the scoliosis does not develop until after loss of ambulation much later in life [5]. Because patients with SMA-2 are able to sit upright in a wheelchair and often manipulate it with
Figure 4. Sagittal images of a patient with SMA-2 at a) 8 years of age and at b) 9 years of age demonstrating progressive kyphosis and narrowing in the sagittal plane. Figure 5. FVC as a percent of predicted as it changes with age in a cohort of 60 subjects with DMD between 5 and 24 years of age. The age represents mean data from 2 year epochs from 1 year younger and 1 year older than the age listed on the x-axis (Unpublished data, Complements of Richard S, Finkel, MD).
38
O.H. Mayer / Paediatric Respiratory Reviews 16 (2015) 35–42
their dominant hand, there has been some question about whether the scoliosis favors the side of the dominant hand. However, this has been explored and the results have been inconclusive with almost 2/3 of curves concave to the left [9], which is opposite of
what would be expected with the preponderance of righthandedness in the general population [9]. The typical progression has been a combination of either a broad C-shaped single scoliosis curve starting in the mid to upper thorax or the less common S-shaped curve with a curve in the thoracic region and a ‘‘compensatory’’ curve in the lumbar region [9]. The C-shaped curve occurs in 75-88% of curves [2], half of which are thoracolumbar [2]. There is substantial rib cage distortion (Figures 3a-d) with caudal rotation of the ribs in the convex chest to the point where they are approximately vertical and compression of the ribs in the concave chest occurs in the horizontal plane [10]. In addition to this thoracolumbar distortion is often kyphosis (Figure 4a & 4b) and occasionally pelvic obliquity with hip asymmetry and difficulty maintaining a comfortable sitting position [6]. The progression of pelvic obliquity and the degree of spinal rotation is closely related to the development of scoliosis [2]. Duchenne Muscular Dystrophy Because of the presumed inflammatory effect from myocyte death and apoptosis in the muscle of patients with DMD, there have been a number of studies exploring the proactive use of systemic steroids in preserving muscle function [11,12]. The results demonstrate that ambulation is preserved for approximately two years with an attenuation of the loss of forced vital capacity compared to those patients without steroid therapy [11,13]. This then connects to the development of scoliosis and in one single-center review, 83 of 85 patients not treated proactively with steroids developed scoliosis of at least 10 [14]. There is broad international consensus on the use of steroid therapy [15], but
Figure 6. Progressive scoliosis in a patient with DMD at a) 14 years of age, b) 16 years of age.
Figure 7. Patient with SMA-2 and kyphosis a) pre-operatively at 9 years of age with transverse CT Scan image of the chest below the carina and b) at 9.5 years of age post-operatively demonstrating chest narrowing in the sagittal plane, with the sagittal plan images represented in Figures 4a & b.
O.H. Mayer / Paediatric Respiratory Reviews 16 (2015) 35–42
uncertainty on the dosing that will minimize the most common side effects on bone mineralization and cardiac function [15]. Alternately there are newer non-steroidal agents under exploration that in early testing have had favorable results [12]; however, further review is beyond the scope of this manuscript. While there clearly is an association between decline in FVC as percent predicted (FVC%) and increase in scoliosis angle, the association is not significant [16], which suggests other factors in play such as progression of underlying disease and in particular loss of muscle strength. However, there have been a number of studies documenting the close correlation between loss of FVC% and decrease in muscle strength [17]. This speaks to the overall process of respiratory decline in DMD being multifactorial including both progression of underlying disease and made worse by mechanical changes such as development of scoliosis. There is a well-established 4-phase pattern of change in FVC%, in patients with DMD (Figure 5). The first phase is with FVC% solidly in the normal range and staying at a consistent value. The next phase is at around 11-13 years of age, where the FVC% begins to decline from its peak value and becomes lower in the normal range. The third phase is a more rapid decline FVC% that progresses until the FVC% becomes quite low, less than 20%. In the final phase the FVC declines significantly more slowly. In two separate cohorts of patients with DMD, the FVC% decreased at a rate of 8.0-8.3% per year before surgical intervention to a mean FVC% of 47% and a mean scoliosis angle of 38-458 before surgical intervention [18].
39
Similarly, three distinct patterns of scoliosis progression after loss of ambulation and with the initiation of wheelchair use have been described [14]. The first is a linear increase in scoliosis that begins just before or just after becoming wheelchair dependent [14]. The second is biphasic progression with a 1.5-3 year period of absent or slow growth after wheelchair dependency and then a steady linear increase [14]. The third is a rapidly collapsing curve in which the scoliosis develops rapidly over the first few months after wheelchair dependency [14]. From the time of loss of ambulation and the onset of wheelchair use and spending the majority of awake hours in the sitting position, there is a constant increase in scoliosis of about 68 per year [14]. Interestingly, the progression of scoliosis after wheelchair dependency is independent of age [14]. The progression in scoliosis continues unabated. As the scoliosis angle approaches 808, the rate of increase decreases substantially [19], likely due to a combination of resistance to further compression by the abdominal organs and the chest wall beginning to abut the pelvis. Without intervention this progression of scoliosis can continue until the rib cage abuts the pelvis or the curve increases to about 1208 [14]. Unlike in SMA in which scoliosis largely starts in the thorax (Figures 6a & 6b), in DMD, scoliosis starts in the lumbar region down to L5 and with progression causes sacral tilt and eventual pelvic obliquity and then progresses cephalad to include the thoracic spine [14]. In addition, the incidence of kyphosis was 62% and lumbar lordosis was 38% in a recently reported cohort of 60 patients with DMD [14]. As with SMA pelvic obliquity can also
Figure 8. a) 5 year old with SMA-1 and scoliosis with unilateral chest wall collapse; b) 6 year old with SMA-1 and scoliosis 14 months s/p growing rod placement (from Chandran S, et.al. Early treatment of scoliosis with growing rods in children with severe spinal muscular atrophy: a preliminary report. J Pediatr Orthop 2011;31(4):450–4.).
40
O.H. Mayer / Paediatric Respiratory Reviews 16 (2015) 35–42
be a major problem in DMD and can have a large impact on patient comfort and quality of life.
INTERVENTIONS Unlike in DMD when scoliosis typically begins soon before puberty and progresses to the point of needing surgical intervention in mid-adolescence [4], the scoliosis in SMA occurs in early childhood and progresses to the point of needing surgical intervention in early puberty and often well in advance of the pubertal growth spurt [20]. There is an inherent conflict between spinal stabilization to correct the mechanical defect and improve comfort and the alternative of preserving growth as much as possible. This conflict brings up the potential risk that too early spinal fusion and growth cessation would produce a significant enough restrictive respiratory defect to prevent the patient from ventilating adequately to meet his/her metabolic demands. That is a natural concern in patients with full exertional capacity, however, in patients with NMDz that concern is harder to reconcile and remains largely unresolved. As a result chest wall bracing is often used as a temporizing, but never curative, procedure, which and has been shown to preserve a comfortable sitting position [8]. In the case of SMA there is the potential to make the caudal rib rotation worse with a near 50% increase in the rate of scoliosis development per year [2] due to outward pressure on the chest wall. This rate of progression is most rapid after 10 years of age [2]. Bracing can have a negative impact on vital capacity [6] and can decrease both the respiratory system compliance and the tidal ventilation [21]. For these reasons, spinal stabilization using a variety of different techniques has been the standard approach for treating scoliosis in patients with NMDz and scoliosis. Over the last 40 years the approaches have changed substantially [22]. Early on the preferred technique was spinal fusion using Harrington rods, which stabilized the spine cephalad and caudally, altered the normal curve of the spine and required post-operative bracing for as much as a month [22]. The Luque-Galveston technique then became popular and had the advantage of maintaining a normal spinal contour by attaching a custom curved using sub-laminar wires at each vertebra, but required long operating time and relatively high blood loss [22]. More recent modifications have been using pedicle screws or laminar hooks to make attachments more stable, to minimize the number of fixation points and to reduce surgical time [22]. The decision to extend the caudal attachment point into the sacrum is largely related to the presence of significant pelvic obliquity [23]. Surgical intervention in patients with NMDz poses a variety of risks. The non-ambulant patients with NMDz (effectively all patients with NMDz and scoliosis) are at risk of having bone demineralization that can be worsened by poor nutritional status [22]. Chronic respiratory failure in both SMA and DMD and cardiomyopathy in patients with DMD can also complicate both the operative and preoperative course [22]. As a result, patients with NMDz have an increased post-operative length of stay (10.3 vs. 7.7 days) and an increased cost of hospitalization compared to patients with non-NMDz scoliosis [24]. However, a good portion of this risk can be mitigated by proactive assessment and preoperative management by the appropriate specialists.
However, in a minority of patients there is the potential for loss of the spinal curve correction over time [20]. In patients that have only a posterior fusion occasionally ‘‘crank shaft’’ deformity can cause a hyperlordosis with spinal growth due to spine growth being greater in the anterior spine compared to the posterior spine [25]. The loss of correction is less in patients on whom a long segment fusion was performed [20]. Kyphosis can be successfully corrected surgically, but in doing so the sagittal diameter of the chest can become substantially
Spinal Muscular Atrophy In SMA spinal fusion has been demonstrated to substantially improve the scoliosis curve [8] and the pelvic obliquity [25]. In addition while the FVC% does not improve after spine fusion [8], the decline in FVC% is substantially slower [26].
Figure 9. a) Pre- and b) post-operative AP chest radiographs in a patient with SMA-2 after bilateral VEPTR insertion.
O.H. Mayer / Paediatric Respiratory Reviews 16 (2015) 35–42
narrow (Figures 4a, 4b & 7a, 7b), likely because of the intercostal muscle weakness. The significance of this change has not been well described. Because of the interest in intervening as early as possible in the development of scoliosis in SMA to minimize respiratory morbidity and maximize comfort, growth sparing procedures have been used using telescoping rods [25]; growing rods [27] (Figure 8a & b); and the vertical expandable prosthetic titanium rib (VEPTR) device (Robert Campbell, MD personal communication) [28] (Figure 9a & b). One additional problem that impacts patients with SMA is that the rib cage collapse, usually unilateral (Figures 3a-3d), that occurs coincident with the scoliosis is exceptionality difficult to correct, or prevent, with spinal fusion (Figures 10a & b) [27]. Even when there is correction, the chest wall asymmetry often recurs within 4 1/2 years of surgery when spine based support such as growing rods (Figure 8b) are used and expanded with growth [10]. Because the chest wall correction in patients with juvenile scoliosis remains stable using this approach [10] the progression in chest wall disease is very likely due to the underlying chest wall weakness in subjects with SMA and speaks to a need to support a new approach that better supports the chest wall. In being able to be used
Figure 10. a) 13 year old female with SMA-2 who is s/p spine fusion and has significant bilateral rib cage collapse; b) the same female at 21 years old, with return of scoliosis and worsening of the bilateral rib cage collapse.
41
laterally directly on the ribs to support the rib cage and spine together, the VEPTR has potential to address both issues (Figures 9a & b), but there are no long-term data in the literature yet to demonstrate the long-term efficacy. Duchenne Muscular Dystrophy One of the hypotheses for progression of disease and loss of function in DMD is inflammation due to sarcomere apoptosis and exposure of intact sarcomeres to cytoplasmic components [11]. Based on this, varying preparations of corticosteroids have been used and have been shown by some to alter the progression of FVC decline, presumably by delaying the loss of ambulation [11]. The impact on the progression of scoliosis is controversial [13]. In DMD most patients will have scoliosis surgery between 12 and 16 years and usually a little later than in patients with SMA [4]. In a study comparing patients who had or didn’t have surgical repair of scoliosis, the cohort who did have scoliosis surgery had larger scoliosis angles and substantially lower FVC% than the cohort of with less severe disease and without significant scoliosis [29]. Interestingly, in the less severe non-surgical cohort, the higher FVC% was maintained, in spite of loss of FVC% as a result of the progression of DMD [29]. Therefore, FVC% declines in all patients with DMD, but is substantially lower in patients with DMD and scoliosis. The improvement in scoliosis and pelvic obliquity is usually quite good [30] (Figures 6b & 11). However, the improvement in FVC% bears no relation to the improvement in scoliosis angle [16], and in fact FVC% still gets worse, but at a lower rate than what is seen in patients with significant scoliosis who opted not have scoliosis surgery [18]. Pelvic obliquity is also an issue and as in SMA can substantially impact quality of life and comfort in a sitting position. In a review
Figure 11. Surgical correction of scoliosis and pelvic obliquity in a patient with DMD at 16.5 years compared to preoperative image in Figure 6B.
42
O.H. Mayer / Paediatric Respiratory Reviews 16 (2015) 35–42
of 50 patients with DMD and scoliosis who had scoliosis surgery, it was recommended to have pelvic fixation in all cases except if the fusion is done early, soon after wheel chair dependence, if there is a small curve or if there is negligible pelvic obliquity [30]. While some recommend early scoliosis surgery with curves between 108 and 308 [18] to minimize the loss of FVC%, which is effectively unrecoverable, nothing is known about the potential impact of early spine fusion in early adolescence or earlier on spine height or FVC. However, it would be reasonable to draw upon some of the recent experience in SMA, in which scoliosis surgery is often performed in pre-adolescence using growth sparing procedures such as telescoping rods [25], growing rods [10] and the VEPTR (Robert Campbell, MD personal communication) [28]. However, as of yet no data are available to definitively address this issue. CONCLUSION In its most basic sense respiratory success or failure is determined by the balance between the respiratory ‘‘pump’’, including the respiratory muscles and spinal and spinal and chest wall support, and the respiratory ‘‘load’’ including anything that can alter respiratory mechanics, such as sources of airway obstruction (e.g. obstructive sleep apnea), alterations in respiratory compliance or the resistance to respiratory motion. The load is clearly impacted by progressive scoliosis and with the ‘‘pump’’ progressively worsening because of the underlying NMDz, the only true therapeutic option is minimizing the respiratory ‘‘load’’. Understanding this and the options available for addressing scoliosis in NMDz and the optimal time to do so is critical in caring for patients with NMDz in a comprehensive and complete way. FUTURE DIRECTIONS FOR RESEARCH Attempts should be made to better define the long-term risks and benefits of long-term systemic corticosteroids and other new non-steroidal therapies in DMD in terms of respiratory function, scoliosis and survival. The impact of newer spinal surgical interventions such as VEPTR on long term decline in FVC in neuromuscular disease need to be performed to further optimize health and survival.
References [1] Gozal D. Pulmonary manifestations of neuromuscular disease with special reference to Duchenne muscular dystrophy and spinal muscular atrophy. Pediatr Pulmonol 2000. [2] Granata C, et al. Spinal muscular atrophy: natural history and orthopaedic treatment of scoliosis. Spine 1989;760–2. [3] Lunn MR, Wang CH. Spinal muscular atrophy. The Lancet 2008;2120–33. [4] Roberto R, et al. The Natural History of Cardiac and Pulmonary Function Decline in Patients With Duchenne Muscular Dystrophy. Spine 2011;E1009–17.
[5] Kaufmann P, et al. Observational study of spinal muscular atrophy type 2 and 3: functional outcomes over 1 year. Arch Neurol 2011;779–86. [6] Mesfin, A., P.D. Sponseller, and A.I. Leet, Spinal muscular atrophy: manifestations and management., J Am Acad Orthop Surg. 2012, American Academy of Orthopaedic Surgeons. p. 393–401. [7] Tsirikos AI, Baker ADL. Spinal muscular atrophy: Classification, aetiology, and treatment of spinal deformity in children and adolescents. Current Orthopaedics 2006;430–45. [8] Robinson D, et al. Scoliosis and lung function in spinal muscular atrophy. Eur Spine J 1995;268–73. [9] Fujak, A. et al., Natural course of scoliosis in proximal spinal muscular atrophy type II and IIIa: descriptive clinical study with retrospective data collection of 126 patients., in BMC Musculoskelet Disord. 2013, BioMed Central Ltd. p. 283. [10] McElroy MJ, et al. Growing Rods for Scoliosis in Spinal Muscular Atrophy. Spine 2011;1305–11. [11] Biggar WD, et al. Deflazacort treatment of Duchenne muscular dystrophy. The Journal of Pediatrics 2001;45–50. [12] Buyse G.M. et al., Effects of glucocorticoids and idebenone on respiratory function in patients with duchenne muscular dystrophy, in Pediatr Pulmonol 2012, Wiley Subscription Services, Inc., A Wiley Company. p. 912–920. [13] Houde S, et al. Deflazacort Use in Duchenne Muscular Dystrophy: An 8-Year Follow-Up. Pediatric Neurology 2008;200–6. [14] Shapiro F, et al. Progression of spinal deformity in wheelchair-dependent patients with Duchenne muscular dystrophy who are not treated with steroids: coronal plane (scoliosis) and sagittal plane (kyphosis, lordosis) deformity. Bone Joint J 2014;100–5. [15] Bushby K, et al. Report on the 124th ENMC International Workshop. Treatment of Duchenne muscular dystrophy; defining the gold standards of management in the use of corticosteroids 2–4 April 2004, Naarden, The Netherlands. Neuromuscular Disorders 2004;526–34. [16] !!! INVALID CITATION !!!. [17] Nicot F, et al. Respiratory muscle testing: a valuable tool for children with neuromuscular disorders. American journal of respiratory and critical care medicine 2006;67–74. [18] Velasco MV, et al. Posterior spinal fusion for scoliosis in duchenne muscular dystrophy diminishes the rate of respiratory decline. Spine 2007;459–65. [19] Oda T, et al. Longitudinal study of spinal deformity in Duchenne muscular dystrophy. Journal of Pediatric Orthopaedics 1993;478–88. [20] Zebala LP, et al. Minimum 5-year radiographic results of long scoliosis fusion in juvenile spinal muscular atrophy patients: major curve progression after instrumented fusion. J Pediatr Orthop 2011;480–8. [21] Tangsrud SE, et al. Lung function measurements in young children with spinal muscle atrophy; a cross sectional survey on the effect of position and bracing. Archives of Disease in Childhood 2001;521–4. [22] Mehta JS, Gibson MJ. The treatment of neuromuscular scoliosis. Current Orthopaedics 2003;17(4):313–21. [23] Alexander WM, et al. The effect of posterior spinal fusion on respiratory function in Duchenne muscular dystrophy. Eur Spine J 2013;22(2):411–6. [24] Barsdorf, A.I., D.M. Sproule, and P. Kaufmann, Scoliosis surgery in children with neuromuscular disease: findings from the US National Inpatient Sample, 1997 to 2003., in Arch. Neurol. 2010, American Medical Association. p. 231–235. [25] Fujak A, et al. Operative treatment of scoliosis in proximal spinal muscular atrophy: results of 41 patients. Arch Orthop Trauma Surg 2012;1697–706. [26] Chng SY, et al. Pulmonary function and scoliosis in children with spinal muscular atrophy types II and III. J Paediatr Child Health 2003;673–6. [27] Chandran S, et al. Early treatment of scoliosis with growing rods in children with severe spinal muscular atrophy: a preliminary report. J Pediatr Orthop 2011;450–4. [28] Tobert DG, Vitale MG. Strategies for treating scoliosis in children with spinal muscular atrophy. Am J Orthop 2013;E99–103. [29] Van Opstal N, et al. The effect of Luque-Galveston fusion on curve, respiratory function and quality of life in Duchenne muscular dystrophy. Acta Orthop Belg 2011;659–65. [30] Sengupta DK, et al. Pelvic or lumbar fixation for the surgical management of scoliosis in duchenne muscular dystrophy. Spine 2002;2072–9.
Contents lists available at ScienceDirect
Paediatric Respiratory Reviews
Mini-symposium: Chest Wall Disease
Scoliosis and the Impact in Neuromuscular Disease Oscar Henry Mayer * Associate Professor of Clinical Pediatrics, Perelman School of Medicine at The University of Pennsylvania, Division of Pulmonary Medicine, The Children’s Hospital of Philadelphia, 3501 Civic Center Boulevard, Philadelphia, PA 19104
EDUCATIONAL AIMS Understand the differential pathophysiology of spinal muscular atrophy and Duchenne muscular dystrophy and how each can lead to scoliosis. Understand the considerations involved for when in the progression of scoliosis to intervene. Understand the options available for surgical intervention.
A R T I C L E I N F O
S U M M A R Y
Keywords: Scoliosis Spinal surgery Restrictive lung disease Spinal muscular atrophy Duchenne muscular dystrophy
Scoliosis can alter respiratory mechanics by changing the orientation of the muscles and joints of the respiratory system and in severe forms can put a patient at risk of severe respiratory morbidity or respiratory failure. However, perhaps the most important factor in determining the pulmonary morbidity in scoliosis is the balance between the ‘‘load’’ or altered respiratory mechanics and the ‘‘pump’’ or the respiratory muscle strength. Therefore, scoliosis in patients with neuromuscular disease will both lead to increased ‘‘load’’ and a weakened ‘‘pump’’, an exceptionally unfortunate combination. While progressive neuromuscular disease by its nature does not respond favorably to attempts to improve respiratory muscle strength, the natural approach of early proactive management of the ‘‘load’’ and in the case of scoliosis a variety of different strategies have been tried with variable short term and long term results. Figuring this out requires both an understanding of the underlying pathophysiology of a particular neuromuscular condition and the available options for and timing of surgical intervention. ß 2014 Elsevier Ltd. All rights reserved.
INTRODUCTION Patients with progressive hypotonic neuromuscular disease (NMDz) such as spinal muscular atrophy (SMA) and Duchenne muscular dystrophy (DMD) face a gradual decline with increasing respiratory morbidity progressive to the point of respiratory failure. There is a broad range in the rates of decline of the different types of NMDz, but the end point of respiratory failure is often the same. Unfortunately, what is also the same is the current absence of effective treatment options for each of the conditions to prevent the neuromuscular disease that drives this decline. Therefore, the historic and current focus to NMDz treatment is symptomatic and, to the extent possible, preventative therapy.
* Tel.: +215 590 3749; fax: +215 590 3500. E-mail address: [email protected]. http://dx.doi.org/10.1016/j.prrv.2014.10.013 1526-0542/ß 2014 Elsevier Ltd. All rights reserved.
While the focus of therapy in NMDz is in airway clearance and supporting mechanical ventilation using a variety of proven therapies, there has been a great deal of work over the years to support the respiratory system by treating the progressive scoliosis that occurs in NMDz. Scoliosis in patients with NMDz is different in many ways from that in patients with intact muscle function. First, the defect in patients with NMDz is based on weakness of the entire muscular component of the thorax, while in congenital scoliosis the scoliosis is typically caused by a significant skeletal defect or in idiopathic scoliosis by an asymmetry in the peri-spinal ligamentous and muscular support of the thorax. Second, because of the diffuse respiratory muscle weakness in NMDz there is less potential to resist the skeletal imbalance that occurs with progressive scoliosis, and as a result it can progress more rapidly. Third, the altered respiratory mechanics produced from scoliosis puts a burden on patients with NMDz that can more
36
O.H. Mayer / Paediatric Respiratory Reviews 16 (2015) 35–42
Figure 1. 18 year old female with SMA-2 and uncorrected scoliosis demonstrating severe unilateral rib cage collapse in the convex chest.
easily overwhelm their respiratory muscle function and lead to respiratory failure than in patients with scoliosis and intact muscle strength. This manuscript will explore the unique features of scoliosis in NMDz and the different ways that it develops in conditions with different pathophysiology, SMA and DMD. PATHOPHYSIOLOGY OF SCOLIOSIS IN NEUROMUSCULAR DISEASE In very general terms, spinal muscular atrophy causes more prominent weakness in the muscles of the chest wall than diaphragm, while in DMD the diaphragm is typically weaker with relative preservation of the chest wall muscles [1]. In SMA the
Figure 3. AP chest radiographs in a patient with SMA-2 with progressive scoliosis at a) 10 months of age; b) 3 years of age; c) 5 years of age; and d) 6.5 years of age.
Figure 2. 15 year old male with DMD and uncorrected scoliosis and upward diaphragm displacement.
O.H. Mayer / Paediatric Respiratory Reviews 16 (2015) 35–42
37
scoliosis develops early in the disease process [2], likely due to the substantial respiratory muscle weakness at diagnosis and in the case of SMA-2, patients are able to sit upright, but are nonambulatory [3]. In DMD the scoliosis occurs well in to the disease progression often not until early adolescence [4]. This coincides with the loss in ambulation and becoming wheelchair dependent usually around 11-13 years of age (Richard S. Finkel, MD unpublished data). The primary difference in the scoliosis in SMA compared to that in DMD is the presence of significant (usually unilateral) chest wall collapse in SMA with relatively well preserved rib positioning in DMD, but both can cause a significant scoliosis (Figures 1 and 2). Spinal Muscular Atrophy In SMA absent or very low production of the survival motor neuron (SMN) protein that is involved in maintaining neuronal integrity leads to progressive loss of efferent nerve signaling through the anterior horn cells of the spinal cord [5]. This causes severe peripheral muscle weakness and in the thorax prominent intercostal muscle weakness [5] and to scoliosis in between 60 and 95% of patients [6]. High chest wall compliance is the hallmark of SMA-1 in which patients are never able to sit upright without significant support and assistance [3]. In these patients there is often bilateral caudal rib rotation of the more superior ribs that produces a narrow superior chest and a typical triangular shaped chest with preservation of the lower chest contour due to the incompressible abdominal contents. In patients with SMA-1 who survive past infancy and are positioned in a semi-upright position because of their lack of truncal support [3] their thorax will bend in one direction and they will develop scoliosis [7]. In broad longitudinal reviews of patients with SMA-2, the onset of scoliosis has been reported as early as 4 years of age [2] and progressed at annual rates of between 58 and 158 per year [2]. With this progression in scoliosis comes a coincident decrease in forced vital capacity (FVC) with a loss of almost 5% predicted for every 108 increase in scoliosis curve [8]. Unlike in DMD, where there are often years of normal lung FVC before the onset of scoliosis and the respiratory decline, in SMA the trend in decline of FVC is not as well characterized since it begins before reliable testing can be performed between 5 and 6 years of age. This scoliosis happens most prominently in SMA-2 patients who can sit up without assistance but not stand. In patients with milder SMA-3 the scoliosis does not develop until after loss of ambulation much later in life [5]. Because patients with SMA-2 are able to sit upright in a wheelchair and often manipulate it with
Figure 4. Sagittal images of a patient with SMA-2 at a) 8 years of age and at b) 9 years of age demonstrating progressive kyphosis and narrowing in the sagittal plane. Figure 5. FVC as a percent of predicted as it changes with age in a cohort of 60 subjects with DMD between 5 and 24 years of age. The age represents mean data from 2 year epochs from 1 year younger and 1 year older than the age listed on the x-axis (Unpublished data, Complements of Richard S, Finkel, MD).
38
O.H. Mayer / Paediatric Respiratory Reviews 16 (2015) 35–42
their dominant hand, there has been some question about whether the scoliosis favors the side of the dominant hand. However, this has been explored and the results have been inconclusive with almost 2/3 of curves concave to the left [9], which is opposite of
what would be expected with the preponderance of righthandedness in the general population [9]. The typical progression has been a combination of either a broad C-shaped single scoliosis curve starting in the mid to upper thorax or the less common S-shaped curve with a curve in the thoracic region and a ‘‘compensatory’’ curve in the lumbar region [9]. The C-shaped curve occurs in 75-88% of curves [2], half of which are thoracolumbar [2]. There is substantial rib cage distortion (Figures 3a-d) with caudal rotation of the ribs in the convex chest to the point where they are approximately vertical and compression of the ribs in the concave chest occurs in the horizontal plane [10]. In addition to this thoracolumbar distortion is often kyphosis (Figure 4a & 4b) and occasionally pelvic obliquity with hip asymmetry and difficulty maintaining a comfortable sitting position [6]. The progression of pelvic obliquity and the degree of spinal rotation is closely related to the development of scoliosis [2]. Duchenne Muscular Dystrophy Because of the presumed inflammatory effect from myocyte death and apoptosis in the muscle of patients with DMD, there have been a number of studies exploring the proactive use of systemic steroids in preserving muscle function [11,12]. The results demonstrate that ambulation is preserved for approximately two years with an attenuation of the loss of forced vital capacity compared to those patients without steroid therapy [11,13]. This then connects to the development of scoliosis and in one single-center review, 83 of 85 patients not treated proactively with steroids developed scoliosis of at least 10 [14]. There is broad international consensus on the use of steroid therapy [15], but
Figure 6. Progressive scoliosis in a patient with DMD at a) 14 years of age, b) 16 years of age.
Figure 7. Patient with SMA-2 and kyphosis a) pre-operatively at 9 years of age with transverse CT Scan image of the chest below the carina and b) at 9.5 years of age post-operatively demonstrating chest narrowing in the sagittal plane, with the sagittal plan images represented in Figures 4a & b.
O.H. Mayer / Paediatric Respiratory Reviews 16 (2015) 35–42
uncertainty on the dosing that will minimize the most common side effects on bone mineralization and cardiac function [15]. Alternately there are newer non-steroidal agents under exploration that in early testing have had favorable results [12]; however, further review is beyond the scope of this manuscript. While there clearly is an association between decline in FVC as percent predicted (FVC%) and increase in scoliosis angle, the association is not significant [16], which suggests other factors in play such as progression of underlying disease and in particular loss of muscle strength. However, there have been a number of studies documenting the close correlation between loss of FVC% and decrease in muscle strength [17]. This speaks to the overall process of respiratory decline in DMD being multifactorial including both progression of underlying disease and made worse by mechanical changes such as development of scoliosis. There is a well-established 4-phase pattern of change in FVC%, in patients with DMD (Figure 5). The first phase is with FVC% solidly in the normal range and staying at a consistent value. The next phase is at around 11-13 years of age, where the FVC% begins to decline from its peak value and becomes lower in the normal range. The third phase is a more rapid decline FVC% that progresses until the FVC% becomes quite low, less than 20%. In the final phase the FVC declines significantly more slowly. In two separate cohorts of patients with DMD, the FVC% decreased at a rate of 8.0-8.3% per year before surgical intervention to a mean FVC% of 47% and a mean scoliosis angle of 38-458 before surgical intervention [18].
39
Similarly, three distinct patterns of scoliosis progression after loss of ambulation and with the initiation of wheelchair use have been described [14]. The first is a linear increase in scoliosis that begins just before or just after becoming wheelchair dependent [14]. The second is biphasic progression with a 1.5-3 year period of absent or slow growth after wheelchair dependency and then a steady linear increase [14]. The third is a rapidly collapsing curve in which the scoliosis develops rapidly over the first few months after wheelchair dependency [14]. From the time of loss of ambulation and the onset of wheelchair use and spending the majority of awake hours in the sitting position, there is a constant increase in scoliosis of about 68 per year [14]. Interestingly, the progression of scoliosis after wheelchair dependency is independent of age [14]. The progression in scoliosis continues unabated. As the scoliosis angle approaches 808, the rate of increase decreases substantially [19], likely due to a combination of resistance to further compression by the abdominal organs and the chest wall beginning to abut the pelvis. Without intervention this progression of scoliosis can continue until the rib cage abuts the pelvis or the curve increases to about 1208 [14]. Unlike in SMA in which scoliosis largely starts in the thorax (Figures 6a & 6b), in DMD, scoliosis starts in the lumbar region down to L5 and with progression causes sacral tilt and eventual pelvic obliquity and then progresses cephalad to include the thoracic spine [14]. In addition, the incidence of kyphosis was 62% and lumbar lordosis was 38% in a recently reported cohort of 60 patients with DMD [14]. As with SMA pelvic obliquity can also
Figure 8. a) 5 year old with SMA-1 and scoliosis with unilateral chest wall collapse; b) 6 year old with SMA-1 and scoliosis 14 months s/p growing rod placement (from Chandran S, et.al. Early treatment of scoliosis with growing rods in children with severe spinal muscular atrophy: a preliminary report. J Pediatr Orthop 2011;31(4):450–4.).
40
O.H. Mayer / Paediatric Respiratory Reviews 16 (2015) 35–42
be a major problem in DMD and can have a large impact on patient comfort and quality of life.
INTERVENTIONS Unlike in DMD when scoliosis typically begins soon before puberty and progresses to the point of needing surgical intervention in mid-adolescence [4], the scoliosis in SMA occurs in early childhood and progresses to the point of needing surgical intervention in early puberty and often well in advance of the pubertal growth spurt [20]. There is an inherent conflict between spinal stabilization to correct the mechanical defect and improve comfort and the alternative of preserving growth as much as possible. This conflict brings up the potential risk that too early spinal fusion and growth cessation would produce a significant enough restrictive respiratory defect to prevent the patient from ventilating adequately to meet his/her metabolic demands. That is a natural concern in patients with full exertional capacity, however, in patients with NMDz that concern is harder to reconcile and remains largely unresolved. As a result chest wall bracing is often used as a temporizing, but never curative, procedure, which and has been shown to preserve a comfortable sitting position [8]. In the case of SMA there is the potential to make the caudal rib rotation worse with a near 50% increase in the rate of scoliosis development per year [2] due to outward pressure on the chest wall. This rate of progression is most rapid after 10 years of age [2]. Bracing can have a negative impact on vital capacity [6] and can decrease both the respiratory system compliance and the tidal ventilation [21]. For these reasons, spinal stabilization using a variety of different techniques has been the standard approach for treating scoliosis in patients with NMDz and scoliosis. Over the last 40 years the approaches have changed substantially [22]. Early on the preferred technique was spinal fusion using Harrington rods, which stabilized the spine cephalad and caudally, altered the normal curve of the spine and required post-operative bracing for as much as a month [22]. The Luque-Galveston technique then became popular and had the advantage of maintaining a normal spinal contour by attaching a custom curved using sub-laminar wires at each vertebra, but required long operating time and relatively high blood loss [22]. More recent modifications have been using pedicle screws or laminar hooks to make attachments more stable, to minimize the number of fixation points and to reduce surgical time [22]. The decision to extend the caudal attachment point into the sacrum is largely related to the presence of significant pelvic obliquity [23]. Surgical intervention in patients with NMDz poses a variety of risks. The non-ambulant patients with NMDz (effectively all patients with NMDz and scoliosis) are at risk of having bone demineralization that can be worsened by poor nutritional status [22]. Chronic respiratory failure in both SMA and DMD and cardiomyopathy in patients with DMD can also complicate both the operative and preoperative course [22]. As a result, patients with NMDz have an increased post-operative length of stay (10.3 vs. 7.7 days) and an increased cost of hospitalization compared to patients with non-NMDz scoliosis [24]. However, a good portion of this risk can be mitigated by proactive assessment and preoperative management by the appropriate specialists.
However, in a minority of patients there is the potential for loss of the spinal curve correction over time [20]. In patients that have only a posterior fusion occasionally ‘‘crank shaft’’ deformity can cause a hyperlordosis with spinal growth due to spine growth being greater in the anterior spine compared to the posterior spine [25]. The loss of correction is less in patients on whom a long segment fusion was performed [20]. Kyphosis can be successfully corrected surgically, but in doing so the sagittal diameter of the chest can become substantially
Spinal Muscular Atrophy In SMA spinal fusion has been demonstrated to substantially improve the scoliosis curve [8] and the pelvic obliquity [25]. In addition while the FVC% does not improve after spine fusion [8], the decline in FVC% is substantially slower [26].
Figure 9. a) Pre- and b) post-operative AP chest radiographs in a patient with SMA-2 after bilateral VEPTR insertion.
O.H. Mayer / Paediatric Respiratory Reviews 16 (2015) 35–42
narrow (Figures 4a, 4b & 7a, 7b), likely because of the intercostal muscle weakness. The significance of this change has not been well described. Because of the interest in intervening as early as possible in the development of scoliosis in SMA to minimize respiratory morbidity and maximize comfort, growth sparing procedures have been used using telescoping rods [25]; growing rods [27] (Figure 8a & b); and the vertical expandable prosthetic titanium rib (VEPTR) device (Robert Campbell, MD personal communication) [28] (Figure 9a & b). One additional problem that impacts patients with SMA is that the rib cage collapse, usually unilateral (Figures 3a-3d), that occurs coincident with the scoliosis is exceptionality difficult to correct, or prevent, with spinal fusion (Figures 10a & b) [27]. Even when there is correction, the chest wall asymmetry often recurs within 4 1/2 years of surgery when spine based support such as growing rods (Figure 8b) are used and expanded with growth [10]. Because the chest wall correction in patients with juvenile scoliosis remains stable using this approach [10] the progression in chest wall disease is very likely due to the underlying chest wall weakness in subjects with SMA and speaks to a need to support a new approach that better supports the chest wall. In being able to be used
Figure 10. a) 13 year old female with SMA-2 who is s/p spine fusion and has significant bilateral rib cage collapse; b) the same female at 21 years old, with return of scoliosis and worsening of the bilateral rib cage collapse.
41
laterally directly on the ribs to support the rib cage and spine together, the VEPTR has potential to address both issues (Figures 9a & b), but there are no long-term data in the literature yet to demonstrate the long-term efficacy. Duchenne Muscular Dystrophy One of the hypotheses for progression of disease and loss of function in DMD is inflammation due to sarcomere apoptosis and exposure of intact sarcomeres to cytoplasmic components [11]. Based on this, varying preparations of corticosteroids have been used and have been shown by some to alter the progression of FVC decline, presumably by delaying the loss of ambulation [11]. The impact on the progression of scoliosis is controversial [13]. In DMD most patients will have scoliosis surgery between 12 and 16 years and usually a little later than in patients with SMA [4]. In a study comparing patients who had or didn’t have surgical repair of scoliosis, the cohort who did have scoliosis surgery had larger scoliosis angles and substantially lower FVC% than the cohort of with less severe disease and without significant scoliosis [29]. Interestingly, in the less severe non-surgical cohort, the higher FVC% was maintained, in spite of loss of FVC% as a result of the progression of DMD [29]. Therefore, FVC% declines in all patients with DMD, but is substantially lower in patients with DMD and scoliosis. The improvement in scoliosis and pelvic obliquity is usually quite good [30] (Figures 6b & 11). However, the improvement in FVC% bears no relation to the improvement in scoliosis angle [16], and in fact FVC% still gets worse, but at a lower rate than what is seen in patients with significant scoliosis who opted not have scoliosis surgery [18]. Pelvic obliquity is also an issue and as in SMA can substantially impact quality of life and comfort in a sitting position. In a review
Figure 11. Surgical correction of scoliosis and pelvic obliquity in a patient with DMD at 16.5 years compared to preoperative image in Figure 6B.
42
O.H. Mayer / Paediatric Respiratory Reviews 16 (2015) 35–42
of 50 patients with DMD and scoliosis who had scoliosis surgery, it was recommended to have pelvic fixation in all cases except if the fusion is done early, soon after wheel chair dependence, if there is a small curve or if there is negligible pelvic obliquity [30]. While some recommend early scoliosis surgery with curves between 108 and 308 [18] to minimize the loss of FVC%, which is effectively unrecoverable, nothing is known about the potential impact of early spine fusion in early adolescence or earlier on spine height or FVC. However, it would be reasonable to draw upon some of the recent experience in SMA, in which scoliosis surgery is often performed in pre-adolescence using growth sparing procedures such as telescoping rods [25], growing rods [10] and the VEPTR (Robert Campbell, MD personal communication) [28]. However, as of yet no data are available to definitively address this issue. CONCLUSION In its most basic sense respiratory success or failure is determined by the balance between the respiratory ‘‘pump’’, including the respiratory muscles and spinal and spinal and chest wall support, and the respiratory ‘‘load’’ including anything that can alter respiratory mechanics, such as sources of airway obstruction (e.g. obstructive sleep apnea), alterations in respiratory compliance or the resistance to respiratory motion. The load is clearly impacted by progressive scoliosis and with the ‘‘pump’’ progressively worsening because of the underlying NMDz, the only true therapeutic option is minimizing the respiratory ‘‘load’’. Understanding this and the options available for addressing scoliosis in NMDz and the optimal time to do so is critical in caring for patients with NMDz in a comprehensive and complete way. FUTURE DIRECTIONS FOR RESEARCH Attempts should be made to better define the long-term risks and benefits of long-term systemic corticosteroids and other new non-steroidal therapies in DMD in terms of respiratory function, scoliosis and survival. The impact of newer spinal surgical interventions such as VEPTR on long term decline in FVC in neuromuscular disease need to be performed to further optimize health and survival.
References [1] Gozal D. Pulmonary manifestations of neuromuscular disease with special reference to Duchenne muscular dystrophy and spinal muscular atrophy. Pediatr Pulmonol 2000. [2] Granata C, et al. Spinal muscular atrophy: natural history and orthopaedic treatment of scoliosis. Spine 1989;760–2. [3] Lunn MR, Wang CH. Spinal muscular atrophy. The Lancet 2008;2120–33. [4] Roberto R, et al. The Natural History of Cardiac and Pulmonary Function Decline in Patients With Duchenne Muscular Dystrophy. Spine 2011;E1009–17.
[5] Kaufmann P, et al. Observational study of spinal muscular atrophy type 2 and 3: functional outcomes over 1 year. Arch Neurol 2011;779–86. [6] Mesfin, A., P.D. Sponseller, and A.I. Leet, Spinal muscular atrophy: manifestations and management., J Am Acad Orthop Surg. 2012, American Academy of Orthopaedic Surgeons. p. 393–401. [7] Tsirikos AI, Baker ADL. Spinal muscular atrophy: Classification, aetiology, and treatment of spinal deformity in children and adolescents. Current Orthopaedics 2006;430–45. [8] Robinson D, et al. Scoliosis and lung function in spinal muscular atrophy. Eur Spine J 1995;268–73. [9] Fujak, A. et al., Natural course of scoliosis in proximal spinal muscular atrophy type II and IIIa: descriptive clinical study with retrospective data collection of 126 patients., in BMC Musculoskelet Disord. 2013, BioMed Central Ltd. p. 283. [10] McElroy MJ, et al. Growing Rods for Scoliosis in Spinal Muscular Atrophy. Spine 2011;1305–11. [11] Biggar WD, et al. Deflazacort treatment of Duchenne muscular dystrophy. The Journal of Pediatrics 2001;45–50. [12] Buyse G.M. et al., Effects of glucocorticoids and idebenone on respiratory function in patients with duchenne muscular dystrophy, in Pediatr Pulmonol 2012, Wiley Subscription Services, Inc., A Wiley Company. p. 912–920. [13] Houde S, et al. Deflazacort Use in Duchenne Muscular Dystrophy: An 8-Year Follow-Up. Pediatric Neurology 2008;200–6. [14] Shapiro F, et al. Progression of spinal deformity in wheelchair-dependent patients with Duchenne muscular dystrophy who are not treated with steroids: coronal plane (scoliosis) and sagittal plane (kyphosis, lordosis) deformity. Bone Joint J 2014;100–5. [15] Bushby K, et al. Report on the 124th ENMC International Workshop. Treatment of Duchenne muscular dystrophy; defining the gold standards of management in the use of corticosteroids 2–4 April 2004, Naarden, The Netherlands. Neuromuscular Disorders 2004;526–34. [16] !!! INVALID CITATION !!!. [17] Nicot F, et al. Respiratory muscle testing: a valuable tool for children with neuromuscular disorders. American journal of respiratory and critical care medicine 2006;67–74. [18] Velasco MV, et al. Posterior spinal fusion for scoliosis in duchenne muscular dystrophy diminishes the rate of respiratory decline. Spine 2007;459–65. [19] Oda T, et al. Longitudinal study of spinal deformity in Duchenne muscular dystrophy. Journal of Pediatric Orthopaedics 1993;478–88. [20] Zebala LP, et al. Minimum 5-year radiographic results of long scoliosis fusion in juvenile spinal muscular atrophy patients: major curve progression after instrumented fusion. J Pediatr Orthop 2011;480–8. [21] Tangsrud SE, et al. Lung function measurements in young children with spinal muscle atrophy; a cross sectional survey on the effect of position and bracing. Archives of Disease in Childhood 2001;521–4. [22] Mehta JS, Gibson MJ. The treatment of neuromuscular scoliosis. Current Orthopaedics 2003;17(4):313–21. [23] Alexander WM, et al. The effect of posterior spinal fusion on respiratory function in Duchenne muscular dystrophy. Eur Spine J 2013;22(2):411–6. [24] Barsdorf, A.I., D.M. Sproule, and P. Kaufmann, Scoliosis surgery in children with neuromuscular disease: findings from the US National Inpatient Sample, 1997 to 2003., in Arch. Neurol. 2010, American Medical Association. p. 231–235. [25] Fujak A, et al. Operative treatment of scoliosis in proximal spinal muscular atrophy: results of 41 patients. Arch Orthop Trauma Surg 2012;1697–706. [26] Chng SY, et al. Pulmonary function and scoliosis in children with spinal muscular atrophy types II and III. J Paediatr Child Health 2003;673–6. [27] Chandran S, et al. Early treatment of scoliosis with growing rods in children with severe spinal muscular atrophy: a preliminary report. J Pediatr Orthop 2011;450–4. [28] Tobert DG, Vitale MG. Strategies for treating scoliosis in children with spinal muscular atrophy. Am J Orthop 2013;E99–103. [29] Van Opstal N, et al. The effect of Luque-Galveston fusion on curve, respiratory function and quality of life in Duchenne muscular dystrophy. Acta Orthop Belg 2011;659–65. [30] Sengupta DK, et al. Pelvic or lumbar fixation for the surgical management of scoliosis in duchenne muscular dystrophy. Spine 2002;2072–9.