Click on the links below to access all the ArticlePlus for this article.
Please note that ArticlePlus files may launch a viewer application outside of your web browser.
Spine, lung, and chest growth are interdependent. The coexistence of chest wall and spinal deformity is well documented, but their effect on lung growth is incompletely understood. Rib and chest wall deformity can be severe in early-onset scoliosis, but the focus of treatment has traditionally been on vertebral column deformity, with less attention paid to the chest wall deformation and its functional consequences. Adult respiratory compromise from early-onset scoliosis may be significant. Day et al have reported respiratory insufficiency in congenital scoliosis.1 Pehrsson et al2 brought attention to the association between infantile and juvenile onset scoliosis and increased mortality in the fourth and fifth decades. Goldberg et al3 noted diminished respiratory function with a mean forced vital capacity of only 40% in those patients with early-onset scoliosis requiring early fusion. A recent report4 of early-onset scoliosis patients undergoing fusion of five or more thoracic vertebrae before age 5 demonstrated reduced pulmonary function from 30% to 79% of predicted values, noting that earlier age at fusion and more thoracic levels fused correlated with diminished pulmonary volumes. Significant compromise of adult pulmonary function is unlikely following adolescent idiopathic scoliosis with pain and body image rather than pulmonary insufficiency predominating as functional outcome issues.5 This favorable respiratory outcome for adolescent idiopathic scoliosis6 may divert attention from the adverse pulmonary outcomes in early-onset spine deformity.
Thoracic Insufficiency Syndrome and the “Titanium Rib”
Traditional treatment of progressive early-onset scoliosis has assumed that control of chest deformity is best achieved by prevention of progressive spinal deformity. Treatments such as casting, orthoses, extensible spinal rods, hemi-epiphysiodesis, fusion, wedge resection, and vertebral stapling address spinal column deformity, with the implicit assumption that associated chest wall deformity will also be controlled. Some treatments for early-onset spinal deformity are based on arthrodesis, assuming that a shorter spinal column is preferable to worsened deformity. Campbell et al7,8 developed expansion thoracostomy and the “titanium rib” or vertical expandable prosthetic titanium rib (VEPTR) device to directly treat both spine and chest wall deformity during growth. Campbell and others also promoted a more comprehensive view of combined chest and spine deformities encountered in early-onset spinal disorders. The term by Campbell et al,7 “thoracic insufficiency syndrome” (TIS) is defined as the “inability of the thorax to support normal respiration and lung growth” and applies correctly to the combined chest wall and spinal deformity seen with early-onset spinal deformity of many types. Normal growth in volume of the lung and thorax continues through childhood and adolescence, but increases in the number of alveoli do not appear to continue beyond early childhood.
Respiratory insufficiency associated with early-onset scoliosis has several possible mechanical etiologies, including loss of thoracic volume, thoracic stiffness, chest wall muscle abnormalities, and inhibition of diaphragmatic movement. In congenital scoliosis, the thoracic spine may be shorter than normal9 and congenital chest wall abnormalities such as extensive rib fusions may be associated with a small hemi-thorax, resulting in diminished thoracic volume. In other types of progressive early childhood spinal deformity, a rotational chest wall deformity evolves with the convex side of the chest rotated posteriorly and the concave side rotated anteriorly, resulting in diminished pulmonary volume, particularly on the convex side of the scoliosis. Dubousset et al have described the “spinal penetration index”10 with loss of thoracic cross-sectional area in association with chest and spine rotation and thoracic lordosis. Either rib fusions or chest wall distortion result in thoracic stiffness and diminish the ability of the chest wall to expand and contract with respiration. Collapse of the thorax onto the pelvis seen with severe thoracolumbar deformity, particularly kyphosis, inhibits diaphragmatic excursion resulting in “secondary” thoracic insufficiency as described by Campbell et al.7
Expansion Thoracostomy and Implantation of VEPTR Devices
The major rationale of expansion thoracostomy and VEPTR is the treatment of TIS. Complementary goals include optimizing conditions for growth and function of the thorax and lungs, maximizing spine length and minimizing spine deformity and stiffness. The initial expansion thoracostomy seeks to restore more normal thoracic volumes and secondarily improve spinal contour. Subsequent lengthening procedures maintain thoracic volume relative to growth and allow continued spine growth. Except for attachment of rib-to-spine devices, no spinal procedure is performed in conjunction with expansion thoracostomy and VEPTR insertion. It is assumed that most patients will eventually require spinal fusion at or near maturity.
Materials and Methods
During the period 1999 to 2005, 31 patients underwent primary expansion thoracostomy and insertion of one to three VEPTR devices at Children’s Hospital Boston for TIS associated with fused ribs. Twenty-six patients had congenital vertebral anomalies (scoliosis and/or kyphosis) and fused ribs; two of them in association with multiple vertebral anomalies are best categorized as Jarcho-Levin syndrome (spondylo-costal or spondylo-thoracic dysplasia). Four patients had fused ribs with thoracogenic scoliosis: three with fused ribs following tracheo-esophageal fistula repair and one following excision and reconstruction of a chest wall malignancy during infancy. One patient had congenital chest wall deficiency. Four patients had prior thoracic spinal fusions at other institutions, which failed to stop the progression of the deformity (Table 1). All patients showed progressive spinal deformity, progressive chest deformity, or progressive hemi-thoracic constriction before surgery. Mean age ± SD at primary operation was 4.2 ± 3.6 years (range, 0.6–12.3). Mean follow-up ± SD was 2.6 ± 1.6 years (range, 0.5–5.4). Mean number of device lengthenings ± SD was 3.5 ± 2.6 (range, 0–10) per patient. Six patients have had device exchanges for growth (Figures 1 to 6).
Preoperative evaluation included an assessment by a pediatric orthopedic surgeon, pulmonologist, and thoracic surgeon, as well as evaluation of nutritional status, cardiac status with echocardiography, and pulmonary function tests when possible. If nutritional status was deemed inadequate, supplemental nutrition or percutaneous gastrostomy feedings were initiated in an attempt to thicken skin, muscle, and subcutaneous tissues. Spinal cord detethering was performed when needed at least 6 weeks before expansion thoracostomy and VEPTR insertion.
Imaging included chest and spine radiographs with calibrated scanogram ruler, spinal MRI, and CT scan of the chest and spine with three-dimensional image reconstruction to determine sites for thoracostomy and device attachment. Thoracic spine length was measured from T1–T12 on CT scans before surgery, after surgery and at last available CT. Growth of the thoracic spine in length after the initial procedure was measured by comparing first postoperative with last follow-up thoracic height. CT scans were acquired during quiet respiration in sedated patients and during a single breath-hold in patients who were able to suspend respiration. For measurement of lung volume, existing CT data sets were transferred to a Voxar three-dimensional imaging workstation (Barco three-dimensional imaging system, Edinburgh, UK). All three-dimensional lung volume reconstructions were performed by a pediatric radiologist experienced in three-dimensional postprocessing techniques. CT-derived lung volumes were created from the preoperative, postoperative, and the most recent CT examinations. Lung volumes of each lung were measured separately after removing central airways (trachea and mainstem bronchi), so as to report principally lung parenchymal volume. Lung volume measurements were performed on data sets obtained from either single- or multidetector CT, perhaps resulting in a small difference in three-dimensional lung volume measurements. We expect that any difference would not be significant since three-dimensional reconstruction parameters (HU window settings) and measurement tool (Voxar) were the same in all cases.
Expansion thoracostomy with insertion of VEPTR devices as described by Campbell et al8 was performed with minor modifications, including extending the incision slightly more caudad, with deep incision in the latissimus dorsi more distal than the skin incision, and “delay” of the flap in one patient with multiple prior thoracotomy scars. When major chest wall anomalies were present, there were often major deficiencies of the normal chest wall musculature making creation of a satisfactory musculocutaneous flap difficult. Choice of number and location of expansion thoracostomies was dictated by the location of hemi-thoracic constriction, location of fused ribs, and secondarily by the location of spinal deformity or anomalies. Where possible, thoracostomies were planned to divide fused ribs (osteotomy in the potential interval between fused ribs), to not isolate groups of fewer than two or three ribs for fear of devitalizing the intervening rib segments, and to match thoracostomies to existing open disc spaces and intervals between fused spinal segments to maximize spinal correction. One to three thoracostomies were performed from the interval between rib heads to the costochondral junction anteriorly. When the pleura at the site of thoracostomy was disrupted sufficiently to allow herniation of lung or exceeded approximately 2 to 3 cm in width, reconstruction of the pleura with reabsorbable xenograft membrane was used. Intraoperative monitoring evolved to include upper extremity pulse oximetry, lower and upper extremity somatosensory evoked potentials, and transcranial motor-evoked potentials when possible. No other spinal procedure was performed in conjunction with the VEPTR procedure. All patients were initially managed in a pediatric intensive care unit and remained intubated for a range of 8 hours to 9 days before extubation. Skin overlying prominent devices was protected in the postoperative period by foam protectors.
Device Lengthenings and Exchanges.
Elective lengthening of all devices was planned at 4- to 6-month intervals, depending on growth rate, but occurred at more widely spaced intervals owing to intercurrent illnesses and difficulty in arranging travel to this institution. Lengthening was accomplished as described by Campbell with modification of the incision. Because the lengthening procedure is repetitive and the devices can be prominent beneath a relatively thin flap of muscle and skin, we later modified Campbell’s procedure to avoid full-thickness incisions over the devices during lengthening with a small skin incision in the line of the original incision and a separate deep muscular incision to approach the device, minimizing the risk of wound dehiscence and secondary infection.
Data are presented as mean ± SD. Paired t tests were used to determine statistical significance. Statistical tests were two-sided, and significance was determined at the 0.05 level using Microsoft Excel. The percent change in FVC and FEV1 in the visits following VEPTR placement was calculated as follows: [(first postprocedure value) − (last postprocedure value)]/(first postprocedure value). Growth of the thoracic spine after the initial VEPTR procedure was calculated as follows: (first postprocedure value) − (last follow-up value). The ratio between right and left lung volumes was compared with an arbitrary normal right to left ratio of 0.85 based on normative data of Gollogly et al.11 The study was approved by the institutional review board.
Growth in Thoracic Spine and Control of Scoliosis
The mean thoracic spine length increased at the time of the initial surgical procedure by 2.0 ± 1.7 cm (range,1.1–7.7) (Table 2). Mean increase in length (growth) of the thoracic spine after the initial procedure was 2.3 ± 1.6 cm (range, −0.3–5.4). Mean growth/y after the initial procedure was 1.2 ± 0.9 cm/year (range, −0.3–3.2). Patients with prior spinal fusions (n = 4) before their VEPTR procedures showed less thoracic growth per year (0.49 ± 0.5 cm/year) after the initial VEPTR procedure than those without prior fusions (n = 24) (1.3 ± 0.9 cm/year) (P = 0.05).
Control of progressive spinal deformity was accomplished in 30 of 31 patients (Table 2). In the main curve of patients with scoliosis (n = 29), the mean ± SD Cobb angle improved from 55 ± 16.4° (range, 30–92) before surgery to 39 ± 17° (range, 12–65) after surgery (P < 0.01). At last follow-up the mean ± SD Cobb angle was 43 ± 17° (range, 12–80), still improved compared with preoperative values (P < 0.01).
Increase in Lung Volumes as Measured by CT
Eighteen of the 30 patients had a complete set of preoperative, postoperative, and most recent CT examinations of sufficient quality for three-dimensional lung volume measurement. (Table 3). Please explain what the asterisk and dagger signify in the footnotes to Table 3. Mean CT-derived total lung volume was 369 ± 279 cm3 (range, 32−1,254) before surgery and 394 ± 289 cm3 (range, 76–1,317) immediately after surgery (P = not significant). Compared with before surgery, immediate postoperative total lung volume increased in 15 patients but was diminished in 3 associated with postoperative atelectasis noted on CT. At last follow-up CT, there was a significant increase in total lung volume to a mean of 736 ± 462 cm3 (range, 266–1,840) compared with preoperative values (P < 0.01). In those patients with a unilateral VEPTR procedure with CT data (n = 17), lung volume on the side of the VEPTR procedure increased from 157 ± 124 cm3 (range, 10–496) before surgery to 168 ± 137 cm3 (range, 24–556) immediately after surgery (P = not significant) and to a mean of 326 ± 222 cm3 (range, 46–772) at last postoperative follow-up (P < 0.01). Percent increase in lung volume between preoperative and last follow-up for those patients with a unilateral VEPTR device (n = 17) was 219 ± 306% (range, 13–1,160) for the lung on the side of the VEPTR device, compared with 147 ± 176% (range, 24–731) increase for total lung volume (P = 0.05). The ratio of right to left lung volume was compared with a normal value of 0.85 in those patients with CT data. The mean absolute difference between observed right to left volume ratio and normal (0.85) improved from 0.46 ± 0.41 cm3 (range, 0.04–1.74) before surgery to 0.33 ± .53 cm3 (range, 0.02–2.29) but was not statistically significant.
Two patients with tracheotomy who were ventilator dependent are free of ventilator support at last follow-up. Three patients used supplemental part-time oxygen before surgery; two no longer require this. For patients with both preoperative and follow-up spirometry data (N = 8), the mean FVC percent predicted value before surgery was 72.9 ± 25.4 (range, 34–107) and the mean FEV1 percent predicted value was 71.6 ± 24.6 (range, 34–107). There was a significant decrease in FVC (P < 0.01) and FEV1 (P = 0.01) percent predicted values in the immediate postoperative period. At follow-up, mean FVC percent predicted value increased by 0.6% ± 16.6%, and FEV1 percent predictive value increased by 5.7% ± 13.5% as compared with preoperative values, but these increases were not statistically significant (Table 4).
Lengthening and Exchange Procedures
A total of 110 lengthening procedures were performed, which included six instances of device exchange for growth and three device conversions of rib-to-rib devices to rib-to-spine devices. The mean number of lengthenings per patient was 3.5 ± 2.6. One patient had both devices removed for deep infection; 1 patient had one of two devices removed because of deep infection, but the device was later reimplanted (Table 1).
Device Migration, Breakage
There were no instances of device breakage or device failure. Migration (complete or partial) of VEPTR devices through the bone at their anchor points was common (n = 8), and when necessary was treated at the time of an elective lengthening by reinsertion at the original anchor site where bone had usually reformed (Table 1). One instance of pelvic fixation required reinsertion because, although firmly embedded, it had become inaccessible for device exchange as it subsided into the pelvis.
No spinal cord injuries or lower extremity neurologic loss occurred. Two instances of brachial plexus palsy or acute thoracic outlet syndrome occurred. In 1 patient with severe thoracogenic scoliosis and iatrogenic rib fusions with severely scarred chest wall skin and muscle, a complete ipsilateral upper extremity motor and circulatory loss was noted at the close of the procedure. Shortening the devices afforded less correction and permitted easier flap closure with an immediate return of circulation to the arm. Complete return of motor function occurred over 6 months, with persistent Horner’s syndrome. In a second patient, after normal intraoperative monitoring and a normal strength examination after surgery, a mild distal ipsilateral hand weakness was recognized 2 weeks after surgery and resolved spontaneously over the subsequent 2 months.
Reoccurrence of Rib Fusions, Thoracic Scarring
Two patients showed re-fusion on CT between ribs in the area of previous rib osteotomy and expansion thoracostomy. Both underwent repeat separation of ribs, resection of bridging bone and thoracostomy at the time of device exchange or conversion. Improved spinal curve correction was noted in both after repeat thoracostomy. In every patient at the time of device exchange, dense soft tissue scarring was noted on the chest wall beneath the devices.
Deep wound infection occurred in 2 patients and is presumed to have occurred in a third One patient with thoracogenic scoliosis and rib fusions and multiple prior incisions had a history of prior chest wall infections. Direct trauma and partial dehiscence of thin skin with no muscle over a prominent device may have contributed to a deep postoperative infection noted 3 weeks after surgery. Both devices were removed and could not be reimplanted because of inadequate soft tissue coverage. Spinal fusion was eventually performed. In another patient, adjacent eczema at the time of device lengthening appeared to contribute to infection, which followed a routine device lengthening. One of two devices was removed and was eventually be reimplanted. In a third patient with chest wall hypoplasia, toxic-shock syndrome occurred 2 weeks after surgery. It is presumed that a deep infection occurred, although no bacteria were isolated and no site of infection demonstrated.
Two instances of rib fracture, one at the time of exchange and lengthening and one at the time of device insertion, occurred. In 1 patient, the upper device attachment fractured when a preexisting rib-to-rib device was converted to a rib-to-spine device and tensioned and was revised by reinsertion in a new more caudad rib anchor point. In another patient with severe preoperative deformity due to multiple fused ribs and chest wall deficiency, fracture of the upper rib attachments occurred at the time of the initial procedure. Correction was reduced and the ribs allowed to heal. Attempts at revision after the ribs had healed failed to achieve adequate fixation at the upper points of attachment, and this patient is presently managed with both the original VEPTR devices and extensible spinal rods.
Expansion thoracostomy and VEPTR insertion directly address TIS-associated chest wall deformity and hemi-thoracic constriction while indirectly controlling spinal deformity. In this group of complex patients, there was an overall improvement in lung volume and control of spinal deformity while permitting continued spinal growth. However, at the time of follow-up, all patients can still be considered to have TIS, in so far as they do not have normal lung function nor normal lung growth. TIS is difficult to quantify, but we believe that in this series, no treatment would have resulted in worsening of the TIS, while treatment has in most instances stabilized or improved, but not eliminated TIS. Comparison with natural history and the evaluation of the results of opening wedge thoracostomy and insertion of VEPTR devices in this group of patients are made difficult by the variability in location and extent of rib fusions, age at treatment, underlying inhibition of growth by congenital spinal abnormalities or prior spinal fusions, and difficulty in assessing pulmonary function in small children. In this series, few patients have reached maturity, and none has advanced into adulthood, making a definitive statement of the effectiveness of treatment of thoracic insufficiency impossible.
Growth of the Thoracic Spine
Fused ribs are a common accompaniment of congenital scoliosis, and several reports of arthrodesis-based treatment of congenital scoliosis include patients with fused ribs but do not specify the severity of rib fusion.12,13 McMaster et al felt that fused ribs did not contribute materially to the tendency of curves to progress, feeling instead that the growth differential created by hemivertebrae and tethering effect of bony bars predominated as the cause of curve progression.14,15 Our series includes 4 patients and Campbell’s includes 5 patients who had fused ribs in whom spinal arthrodesis alone did not stop the progression of spinal deformity, arguing that the tethering effect of fused ribs can be a powerful deforming force. Control of progressive deformity was achieved in all 4 of our patients with arthrodesis after expansion thoracostomy and VEPTR insertion. Because no spinal arthrodesis is performed and the VEPTR device contacts the spine only at the point of attachment of rib-to spine devices, there is no direct interference with spinal growth by expansion thoracostomy and VEPTR insertion. In this series and that of Campbell et al,16 average thoracic spine growth per year was significantly less in those patients who had undergone arthrodesis before expansion thoracostomy and VEPTR procedures, than in those who had not had a prior spinal fusion. In our series, spinal deformity was also adequately controlled in those patients without prior arthrodesis, arguing that in patients with fused ribs and combined spine and chest wall deformity, expansion thoracostomy, and VEPTR alone can control both spine and chest deformity during growth, and early arthrodesis is unnecessary. Mean growth of the thoracic spine after the initial procedure in our series was 1.2 ± 0.9 cm/year, compared with the prediction of DiMeglio and Bonnel17 for normal children of 1.4 cm/year from age birth to 5 years and 0.6 cm/year from age 5 to 10 years. Early expansion thoracostomy and VEPTR insertion appears to permit continued thoracic spine growth while controlling deformity. However, it is assumed that at or near maturity spinal arthrodesis will be needed for most patients.
Growth and Function of the Chest
Expansion thoracostomy with repetitive lengthening increased the size and volume of the treated hemithorax in the Campbell et al series.18 In this series, we document increased lung volumes by CT, greater in the hemithorax treated by VEPTR (mean, 219%) than for total lung volume (mean, 147%). However, the volume, contour, and function of the expanded hemithorax are not normal after VEPTR treatment. Normal thoracic function is complex and requires a stable, adequate thoracic volume and the ability to move the volume with respiration. Expansion thoracostomy and VEPTR appear to create adequate stable volume, larger than before surgery, but the chest wall on the treated side is still stiffer than normal. Diaphragmatic function can also be helped by expansion thoracostomy and VEPTR, either by displacing the lower ribs with their diaphragmatic attachment caudad, or by relief of “secondary” thoracic insufficiency when overall spinal alignment is improved and the chest displaced cephalad away from the pelvis, allowing easier diaphragmatic excursion. Our limited spirometry data and CT lung volumes suggest that VEPTR worsens thoracic function immediately after surgery but improves function over time. In the immediate postoperative period, percent predicted pulmonary function worsened in those patients old enough to have preoperative spirometry; however, by last follow-up, pulmonary functions were improved slightly in the group of patients old enough for preoperative spirometry. Three patients demonstrated diminished lung volumes on CT immediately after surgery associated with atelectasis. However, at last follow-up, lung volumes were more symmetric and total lung volume larger. Ventilator dependence and need for supplemental oxygen diminished in nearly all affected patients. Only long-term follow-up of these patients will determine the effect of expansion thoracostomy and VEPTR on pulmonary status and life expectancy.
Complications were common in this series of patients, similar to historical series of extensible rods without fusion.19–21 Device migrations were frequent and may be inevitable because of continued distraction pressure of the device and movement of anchor points. Two deep infections occurred early in this series, both in patients with multiple prior procedures and deficient soft tissue coverage for the VEPTR devices. Following these cases, we modified the original procedure to facilitate closure of the soft tissue flap and have avoided full-thickness incisions over the most prominent part of the devices wherever possible.
One profound brachial plexus palsy was presumably due to an acute thoracic outlet syndrome created when the upper thorax was moved cephalad with expansion thoracostomy and the scapula and clavicle were drawn caudad to provide coverage of the newly expanded thorax and devices. Following this initial instance of brachial plexus palsy, institution of intraoperative upper extremity monitoring led to two instances in which closure after initial, acute expansion resulted in altered upper extremity somatosensory evoked potential signals. The amount of correction was reduced and no postoperative deficit noted. The second, mild instance of brachial plexus palsy was noted 2 weeks after surgery and occurred despite normal intraoperative monitoring and a normal postoperative strength examination.
Indications for VEPTR: Changed Attitudes Toward Early Childhood Scoliosis
Stimulated by awareness of respiratory insufficiency in adults after early childhood scoliosis, Campbell’s description of TIS, and the availability of VEPTR and expansible rods, treatment options for early childhood scoliosis now include consideration of growth-preserving procedures. Our experience with VEPTR for TIS associated with congenital or acquired fused ribs has been satisfactory in that spine and chest deformity were controlled, lung volume improved, and growth continued in the thoracic spine. Surgical intervention with VEPTR can be considered early in growth, before deformity is severe, since spinal growth will continue with treatment. Not all progressive congenital scoliosis and not all fused ribs should be treated with VEPTR. We have used several criteria for trying to decide between VEPTR and more conventional treatments such as fusion, hemi-epiphysiodesis, hemivertebra excision, and extensible spinal rods. If the fused ribs are extensive, then expansion thoracostomy with VEPTR is the logical choice. If conventional treatment requires an extensive thoracic spinal arthrodesis at an early age, then VEPTR may be the logical choice if fused ribs or constricted hemi-thorax are present or extensible rods may be preferable if the chest wall deformity is less severe. If the deformity in a young child can be controlled by a localized fusion-based procedure such as hemivertebra excision or limited in situ arthrodesis, or if the child is approaching maturity, conventional treatments may be preferred.
Based on this series of 31 patients, we recommend expansion thoracostomy and VEPTR for treatment of TIS with fused ribs in growing children. The initial procedure commits the child to multiple subsequent procedures over many years, a drawback that should limit indications to those complex and extensive spinal deformities that cannot be effectively controlled by more conventional means.
- Chest and spine deformity associated with fused ribs in children may result in unilateral or bilateral constriction of the thorax and thoracic insufficiency syndrome.
- Expansion thoracostomy and insertion of VEPTR directly treats chest wall constriction, improves lung volume, and indirectly corrects spinal deformity, while allowing thoracic spine growth.
- Expansion thoracostomy and VEPTR may be the preferred treatment for combined chest wall and spinal deformity associated with fused ribs and thoracic insufficiency syndrome in growing children.
1.Day GA, Upadhyay SS, Ho EK, et al. Pulmonary functions in congenital scoliosis. Spine
2.Pehrsson K, Larsson S, Oden A, et al. Long-term follow-up of patients with untreated scoliosis: a study of mortality, causes of death, and symptoms. Spine
3.Goldberg CJ, Gillic I, Connaughton O, et al. Respiratory function and cosmesis at maturity in infantile-onset scoliosis. Spine
4.Emans J, Kassab F, Caubet JF, et al. Earlier and more extensive thoracic fusion is associated with diminished pulmonary function: outcome after spinal fusion of 4 or more thoracic spinal segments before age 5 [paper presentation no. 101]. Scoliosis Research Society Annual Meeting, Buenos Aires, Argentina, September 6–9, 2004.
5.Weinstein SL, Dolan LA, Spratt KF, et al. Health and function of patients with untreated idiopathic scoliosis: a 50-year natural history study. JAMA
6.Pehrsson K, Danielsson A, Nachemson A. Pulmonary function in adolescent idiopathic scoliosis: a 25 year follow up after surgery or start of brace treatment. Thorax
7.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
8.Campbell RM Jr, Smith MD, Hell-Vocke AK. Expansion thoracoplasty: the surgical technique of opening-wedge thoracostomy [surgical technique]. J Bone Joint Surg Am
9.Goldberg CJ, Moore DP, Fogarty EE, et al. The natural history of early-onset scoliosis. Stud Health Technol Informatics
10.Dubousset J, Wicart P, Pomero V, et al. Spinal penetration index: new three-dimensional quantified reference for lordoscoliosis and other spinal deformities. J Orthop Sci
11.Gollogly S, Smith JT, White SK, et al. The volume of lung parenchyma as a function of age: a review of 1050 normal CT scans of the chest with three-dimensional volumetric reconstruction of the pulmonary system. Spine
12.Winter RB, Moe JH. The results of spinal arthrodesis for congenital spinal deformity in patients younger than five years old. J Bone Joint Surg Am
13.Winter RB, Moe JH, Lonstein JE. Posterior spinal arthrodesis for congenital scoliosis: an analysis of the cases of two hundred and ninety patients, five to nineteen years old. J Bone Joint Surg Am
14.McMaster MJ, Ohtsuka K. The natural history of congenital scoliosis: a study of two hundred and fifty-one patients. J Bone Joint Surg Am
15.McMaster MJ, Singh H. Natural history of congenital kyphosis and kyphoscoliosis: a study of one hundred and twelve patients. J Bone Joint Surg Am
16.Campbell RM Jr, Hell-Vocke AK. Growth of the thoracic spine in congenital scoliosis after expansion thoracoplasty. J Bone Joint Surg Am
17.DiMeglio A, Bonnel F. Le Rachis en Croissance
. Paris: Springer, 1990.
18.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
19.Acaroglu E, Yazici M, Alanay A, et al. Three-dimensional evolution of scoliotic curve during instrumentation without fusion in young children. J Pediatr Orthop
20.Klemme WR, Denis F, Winter RB, et al. Spinal instrumentation without fusion for progressive scoliosis in young children. J Pediatr Orthop
21.Mineiro J, Weinstein SL. Subcutaneous rodding for progressive spinal curvatures: early results. J Pediatr Orthop