Spinal muscular atrophy (SMA) is a progressive autosomal recessive neuromuscular disorder that typically manifests in infancy or early childhood. Hypotonia, symmetrical weakness, and muscle atrophy develop because of degenerative alterations in anterior horn cells resulting from deletions and/or mutations in the telomeric copy of the survival motor neuron gene on chromosome 5q (Table 1).1 Scoliosis is the most common spinal deformity in patients with SMA and is typically progressive in nature (5 to 15 degrees, annually),2–4 often resulting in severe deformity with detrimental effects on pulmonary function.5 Improved understanding of biology and advances in medical care have led to increased survival, improved function, and better quality of life in patients with SMA.6,7 Surgical correction and stabilization is generally recommended in patients with progressive deformity as bracing has been proven ineffective.4,8–10 However, the timing and method of surgical fixation remains controversial. Recent trends include increased use of growing rod constructs as early definitive fusion has been suggested to negatively impact lung development, respiratory function, and thoracic spine height—commonly referred to as thoracic insufficiency syndrome.11 The purpose of this study was to evaluate our 30-year experience of definitive posterior spinal fusion to treat scoliosis in SMA and determine the effects on pulmonary function, clinical complications, and radiographic outcomes to provide an historical baseline against which newer growth-preserving techniques can be measured.12,13
We retrospectively reviewed the demographics and clinical, radiographic and pulmonary function testing results of all children with SMA and associated scoliosis that underwent posterior spinal fusion between 1985 and 2013 after approval from the institutional review board. All were performed by the same surgeon (S.L.W.). A total of 17 patients (11 female) with muscle biopsy proven SMA underwent posterior spinal fusion during this time period. Of them, 16 patients had >2 years follow-up and all were included in the study (Table 1). Bracing was generally not effective and/or tolerated in this patient cohort. Although no formal bracing protocol was implemented in this group, a few patients did have short trials of soft bracing primarily for postural support rather than attempted curve correction or maintenance. All patients underwent posterior spinal fusion using the Luque rod constructs (including combination Harrington rod in 1 patient) extending to the pelvis. The mean age at time of surgery was 9.8±3.6 years (range, 6 to 18 y). The mean age at most recent follow-up was 19.4 years (range, 10 to 37 y), with a mean follow-up of 10.1 years (range, 3.1 to 26 y). All preoperative, initial postoperative, and final follow-up radiographs were evaluated by 2 independent observers (J.B.H. and S.L.W.) blinded to patient information. In attempt to minimize differences in magnification and subsequent measurement error, radiograph acquisition followed a strict protocol including documentation of the source-to-patient distance (standardized to 72 inches to minimize magnification) and full-size printing of radiographs. When comparing immediate postop to final follow-up radiographs, Luque rod size was considered to be stable and used as an internal sizer within each patient series. Therefore, after comparing measured rod length between images, the determined conversion factor was applied to all length measurements. Radiographic data included direct measures of the major curve (Cobb angle), coronal balance (horizontal distance from the center sacral line to the center of T1), pelvic obliquity (angle formed between a line from the spinous process of T1 to the sacrum and the perpendicular of a line crossing the iliac crests), T1-T12 height (distance from the T1 superior endplate to the T12 superior endplate), T1-S1 height (distance from the T1 superior endplate to the S1 superior endplate), and T1-rod length (distance from T1 superior endplate to cephalad end of rod). Radiographic estimates of rib collapse, thoracic cavity shape, and space-available-for-lung (T6:T12 width ratio, T6:T10 rib-vertebral-angle difference (RVAD) ratios, and lung height ratio) were also calculated.11–14 Preoperative and postoperative pulmonary function [including forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC)] was assessed in all patients (ambulatory and nonambulatory), who were able to comply with standard testing protocols.5 Standardization of lung function testing was performed using reference equations based on patient sex, age, height, weight, and/or arm length depending on ambulatory status.15,16
Normally distributed continuous measures are summarized by the mean and SD, and by the median when not normally distributed. The change in radiographic and pulmonary function measures from initial evaluation to final follow-up were analyzed with use of the paired t tests. Interobserver reliability was assessed by calculating the interclass correlation coefficient (ICC). Alpha was set at 0.05 for all pairwise comparisons.
There were significant improvements in all measurements of spinal deformity including major curve, coronal balance, and pelvic obliquity (Fig. 1) when comparing preoperative with final follow-up radiographs. There was a significant increase in T1-T12 height (2.3±2.4 cm; P<0.01), T1-S1 height (5.6±5.4 cm; P<0.001), and T1-rod length (2.0±1.6 cm; P<0.001) (Table 2). The space-available-for-the-lung ratio increased because of an increase in the lung height on the concave side and a stable lung height on the convex side (Table 3). The T6:T10 RVAD ratio did not change significantly over time. The ICC for length and angular measurements ranged from 0.86 to 0.99. Rib collapse worsened throughout the follow-up period (Fig. 1) in all but 1 patient.
A total of 10 patients were able to comply with pulmonary function testing (PFT) and their preoperative and postoperative data were reviewed. PFT measurements spanned from 9 years before surgery to 24 years after surgery. The mean number of FEV1 measurements per patient was 3.3 (range, 1 to 7) preoperatively and 4.4 (range, 1 to 7) postoperatively. The mean number of FVC measurements per patient was 3.3 (range, 1 to 7) preoperatively and 4.8 (range, 1 to 8) postoperatively. The mean annual rate of decline in FEV1 was 7.2% preoperatively (9.5% in type II, 4.3% in type III) compared with 2.3% postoperatively (4.8% in type II, 2.2% in type III). The mean annual rate of decline in FVC was 6.0% preoperatively (7.9% in type II, 3.6% in type III) compared with 2.9% postoperatively (5.8% in type II, 0.6% in type III) (Fig. 2). There was a small, acute decline in pulmonary function in the immediate postoperative testing results relative to the preoperative results. However, there was an overall decrease in the rate of decline in both FEV1 and FVC when comparing postoperative to preoperative rates.
The mean length of stay was 7.8±4.4 days. Complications included reintubation for low tidal volumes postoperatively (n=1), postoperative pneumonia (n=1), superficial wound breakdown requiring prolonged wound care (n=1), and superficial wound infection requiring operative irrigation and debridement 3 weeks postfusion (n=1). Upon return to the operating room with this patient the deep fascial layer was found to be intact without evidence of deep infection but operative cultures of the superficial tissues showed oxacillin-resistant Staphylococcus aureus so the patient was treated with 6 weeks of oral antibiotics and local wound care. There were no neurologic complications or complications related to instrumentation failure or symptomatic implants.
In the cohort reviewed here, definitive posterior spinal fusion for treatment of scoliosis associated with SMA was effective at controlling spinal curvature and pelvic obliquity without negatively impacting the space-available-for-lung ratio, trunk height, or pulmonary function as measured by pulmonary function testing at up to 26 years follow-up. In addition, children were not exposed to repeated hospitalizations, charges, and additional medical and surgical risks associated with repeated surgical encounters. Therefore, the authors recommend consideration for primary posterior spinal fusion as one of the treatment options for progressive scoliosis in patients with SMA and provide a baseline against which growth-preserving strategies can be measured.
Key radiographic findings observed at time of final follow-up included stable improvements in major curve, pelvic obliquity, and coronal balance. Clear demonstration of continued trunk growth and “growing off” of rods was observed in several patients (Fig. 3) until time of final follow-up or skeletal maturity. This is highlighted by the measured 2 cm mean increase in T1-rod length from immediate postoperative to final follow-up. Although a solid posterior fusion is believed to halt continued growth of the entire involved spinal column, continued growth may occur during the postoperative period as the fusion mass forms and matures as there is not segmental fixation with the Luque rod construct. Alternatively, progressive remodeling and lengthening of the symmetric posterior fusion mass may occur as the anterior column continues to grow until skeletal maturity.
Progression of rib collapse continued despite primary spinal fusion. This was observed throughout the chest but was most marked in the upper thorax, resulting in progressive development of a bell-shaped thoracic cavity (Fig. 1). Proximal compensatory scoliosis requiring routine observation, cervicothoracic bracing, or extension of the fusion has been previously reported in patients with SMA, and is believed to be the result of poor muscle balance throughout the trunk.12 Although we observed new proximal compensatory scoliosis in this sample, none of our patients required proximal extension of the fusion construct because the wheelchair bound patients remained well-balanced clinically and ambulatory patients maintained their functional level and showed no negative clinical sequela. Although attempts were made to minimize measurement error, radiographic outcome results are limited by slight differences in beam orientation and distance from patient, patient rotation and position, and processing technique (eg, digital vs. film).
Pulmonary function in the perioperative period poses a significant concern, as further respiratory decompensation is commonly associated with surgery and anesthesia in SMA patients with already limited pulmonary function. The subsequent effects on the intercostal musculature lead to rib collapse with an eventual bell-shaped, triangular thoracic cavity with reduced lung volume which McElroy et al first attempted to quantify radiographically with the T6:T10 RVAD ratio and the T6:T12 thoracic width ratio.12,17 The results of McElroy et al12 and Lenhart et al13 demonstrated the inability of growing rods to halt progression of rib collapse. Our results demonstrate a statistically nonsignificant decrease in the T6:T10 RVAD ratio from preoperative to final follow-up; however, this ratio is maintained only as a result of the progressive, yet similar, decline in both the T6 and T10 RVAD measurements. We observed a profound narrowing of the upper thoracic region with relatively less narrowing of the lower thoracic region. The end result was a progressive collapse of all the ribs with overall decrease in the T6:T12 width ratio at time of final follow-up, resulting in a bell-shaped thoracic cavity, consistent with the results seen after treatment with growing rods.12,13 Although not a true measure of thoracic cavity shape, T1-T12 height, T1-S1 height, and chest width measures have been shown to predict pulmonary function in patients with early onset scoliosis and were therefore selected as appropriate measures in the current study.18
Extensive efforts have been made to maintain lung volume and thoracic height and width in patients with SMA undergoing spinal fusion.19 Proponents of growing rods raise concerns regarding early definitive fusion because of the potential negative effects on lung development, respiratory function, and thoracic spine height—commonly referred to as thoracic insufficiency syndrome.11 Although lung development and maturation continues into adulthood, the maximal number of both alveoli and respiratory passages are generally reached by age 8, with nearly 90% of total alveoli formed by 48 months of age.20 In addition, Fujak et al19 showed that pulmonary function in children with SMA is already markedly reduced by the age of 4 to 6 years relative to age-matched norms. Uncertainty remains with respect to the minimal thoracic spine length required for adequate pulmonary function in adulthood. Although not explicitly recommended by Karol et al21 in their initial report, many authors have adopted the suggestion that a thoracic spine length of ≤18 cm may result in restriction in pulmonary function with absolute conviction. Recommendation that a thoracic height threshold of 18 cm should be exceeded in effort to achieve adequate pulmonary function has generally been adopted across the literature.18,22–25 However, lung function in patients with SMA is multifactorial and failure to achieve a defined minimum thoracic height should not independently negate consideration for definitive spinal fusion. Rather, pulmonary function should be monitored preoperatively and aid in decisions regarding surgical timing and technique. The average thoracic spine length at the time of surgery in our patient cohort was 19.3 cm. This increased an average of 2.3 cm to measure 21.6 cm at time of final follow-up. The fragile pulmonary system and high rate of perioperative pulmonary complications associated with growing rod treatment are common arguments against repeated surgical exposures for rod lengthening. However, the effect of progressive rib collapse on pulmonary function is not clear. In a study evaluating pulmonary function and radiographic outcomes after early fusion in patients with congenital scoliosis, Bowen et al26 found a decreased FVC in all patients, but no difference between early fusion and no fusion groups at 10 years follow-up despite radiographic differences. To test the relationship between radiographic findings and pulmonary function, we correlated the pulmonary capacity using preoperative and postoperative FVC and FEV1 with radiographic changes from preoperative to final follow-up. Although limited by patient number and incomplete preoperative and postoperative serial testing on all patients, these results confirm prior findings that surgical intervention may cause a small acute decline in pulmonary function in the immediate postoperative period, but an overall decrease in the rate of decline in pulmonary function over time.3,27–29 Prior studies evaluating use of growth-preserving instrumentation have attributed the modest slowing of percentage FVC decline to the use of these techniques13,30,31; however, further clarity is needed as similar findings are reported here after definitive primary fusion. A gradual decline is the expected natural history,19 so it comes as no surprise that surgical treatment that does not attenuate the progressive rib collapse does not significantly alter measured pulmonary function either.
No crankshaft phenomenon or curve progression was observed at final follow-up in this series, in contrast to prior studies reporting rates of up to 36%.32 This is likely the result of routinely fusing from the upper thoracic spine (T2-T4) to the pelvis. We did not observe loss of fixation. Two patients required readmission within 90-days postoperatively. One developed a superficial wound infection requiring operative irrigation and debridement with repeat wound closure and oral antibiotics 3 weeks after the index surgery. The second was readmitted with pneumonia 4 weeks after discharge. No ambulatory patients lost the ability to walk. One patient died at the age of 28, 20 years after spinal fusion and 6 months after final follow-up, from complications related to respiratory failure. In general, perioperative complications occurred less frequently than that reported with growing rod instrumentation.12,27
Definitive posterior spinal fusion for treatment of progressive scoliosis associated with SMA is an effective treatment without negatively impacting pulmonary function at up to 26 years follow-up, without exposing children to the medical and surgical risks of repeated surgical encounters. Magnetically expandable growing rods (MAGEC; NuVasive Inc., San Diego, CA) do not require repeated operative lengthening procedures and may be of additional benefit to patients with SMA, but should be considered with cautious optimism until proven clinically effective in this select group of high-risk patients. Primary spinal fusion as definitive treatment for progressive scoliosis in children with SMA is a proven, safe, and effective option for children with SMA and should be considered when counseling patients and families.
1. Mercuri E, Bertini E, Iannaccone ST. Childhood spinal muscular atrophy
: controversies and challenges. Lancet Neurol. 2012;11:443–452.
2. Riddick MF, Winter RB, Lutter LD. Spinal deformity in patients with spinal muscular atrophy
3. Granata C, Merlini L, Magni E, et al. Spinal muscular atrophy
: natural history and orthopaedic treatment of scoliosis
4. Rodillo E, Marini ML, Heckmatt JZ, et al. Scoliosis
in spinal muscular atrophy
: review of 63 cases. J Child Neurol. 1989;4:118–123.
5. Robinson D, Galasko CS, Delaney C, et al. Scoliosis
and lung function in spinal muscular atrophy
. Eur Spine
6. Wang CH, Finkel RS, Bertini ES, et al. Consensus statement for standard of care in spinal muscle atrophy. J Child Neurol. 2007;22:1027–1049.
7. Oskoui M, Levy G, Garland CJ, et al. The changing natural history of spinal muscular atrophy
type 1. Neurology. 2007;69:1931–1936.
8. Aprin H, Bowen JR, MacEwen GD, et al. Spine
fusion in patients with spinal muscular atrophy
. J Bone Joint Surg Am. 1982;64:1179–1187.
9. Merlini L, Granata C, Bonfiglioli S, et al. Scoliosis
in spinal muscular atrophy
: natural history and management. Dev Med Child Neurol. 1989;31:501–508.
10. Hensinger RN, MacEwen GD. Spinal deformity associated with heritable neurological conditions: spinal muscular atrophy
, Friedreich’s ataxia, familial dysautonomia, and Charcot-Marie-Tooth disease. J Bone Joint Surg Am. 1976;58:13–24.
11. Campbell RM Jr, Smith MD, Mayes TC, et al. The characteristics of thoracic insufficiency syndrome associated with fused ribs and congenital scoliosis
. J Bone Joint Surg Am. 2003;85A:399–408.
12. McElroy MJ, Shaner AC, Crawford TO, et al. Growing rods for scoliosis
in spinal muscular atrophy
(Phila Pa 1976). 2011;36:1305–1311.
13. Lenhart RL, Youlo S, Schroth MK, et al. Radiographic and respiratory effects of growing rods in children with spinal muscular atrophy
. J Pediatr Orthop. 2016. [Epub ahead of print].
14. Campbell RM Jr, Smith MD, Mayes TC, et al. The effect of opening wedge thoracostomy on thoracic insufficiency syndrome associated with fused ribs and congenital scoliosis
. J Bone Joint Surg Am. 2004;86:1659–1674.
15. Miller MR, Crapo R, Hankinson J, et al. General considerations for lung function testing. Eur Respir J. 2005;26:153–161.
16. Pellegrino R, Viegi G, Brusasco V, et al. Interpretative strategies for lung function tests. Eur Respir J. 2005;26:948–968.
17. Mehta MH. The rib-vertebral angle in the early diagnosis between resolving and progressive infantile scoliosis
. J Bone Joint Surg Br. 1972;52:230–243.
18. Glotzbecker M, Johnston C, Miller P, et al. Is there a relationship between thoracic dimensions and pulmonary function in early-onset scoliosis
19. Fujak A, Raab W, Schuh A, et al. Natural course of scoliosis
in proximal spinal muscular atrophy
type II and IIa: descriptive clinical study with retrospective data collection of 126 patients. BMC Musculoskelet Disord. 2013;14:283.
20. Dunnill MS. Postnatal growth of the lung. Thorax. 1962;17:329–333.
21. Karol LA, Johnston C, Mladenov K, et al. Pulmonary function following early thoracic fusion in non-neuromuscular scoliosis
. J Bone Joint Surg Am. 2008;906:1272–1281.
22. Karol LA. Early definitive fusion in young children: what we have learned. Clin Orthop Relat Res. 2011;469:1323–1329.
23. Cook Christopher. Pediatric orthopedics. Pediatr Clin North Am. 2014;61(number 6):1232–1233.
24. Canavese F, Dimeglio A. Normal and abnormal spine
and thoracic cage development. World J Orthop. 2013;4:167–174.
25. Yazici Muharrem. Non-Idiopathic Spine
Deformity in Young Children. New York, NY: Springer; 2011:6–11.
26. Bowen RE, Scaduto AA, Banuelos S. Does early thoracic fusion exacerbate preexisting restrictive lung disease in congenital scoliosis
patients? J Pediatr Orthop. 2008;28:506–511.
27. Fujak A, Raab W, Schuh A, et al. Operative treatment of scoliosis
in proximal spinal muscular atrophy
: results of 41 patients. Arch Orthop Trauma Surg. 2012;132:1697–1706.
28. Evans GA, Drennan JC, Russman BS. Functional classification and orthopaedic management of spinal muscular atrophy
. J Bone Joint Surg Br. 1981;63B:516–522.
29. Piasecki JO, Mahinpour S, Lovine DB. Longterm followup of spinal fusion in spinal muscular atrophy
. Clin Orthop Relat Res. 1986;207:44–54.
30. Dede O, Motoyama EK, Yang CI, et al. Pulmonary and radiographic outcomes of VEPTR (vertical expandable prosthetic titanium rib) treatment in early-onset scoliosis
. J Bone Joint Surg Am. 2014;96:1295–1302.
31. Yoon WW, Sedra F, Shah S, et al. Improvement of pulmonary function in children with early-onset scoliosis
using magnetic growth rods. Spine
32. Zebala LP, Bridwell KH, Baldus C, 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;31:480–488.