The Role of MRI in Children With Congenital Limb Deficiencies With Associated Scoliosis.

BACKGROUND: The association of scoliosis and congenital limb deficiency has been well described. However, the incidence of neural axis abnormalities in this population is not known. The ability to assess the neural axis by physical examination may be limited in patients with a limb deficiency. Although mobility of the spine is important for all children, it can be especially so in children with a limb deficiency. As spinal fusion in children with limb deficiency potentially has more functional impact, detecting reversible forms of scoliosis seems particularly important. METHODS: Retrospective review of children treated at 1 institution between 1990 and 2017 with both a diagnosis of a congenital limb deficiency, upper or lower, and scoliosis. Children were excluded if they had any neurological difference on history or physical examination, if they had sacral agenesis or spina bifida, or if their limb deficiency was related to trauma or early amniotic rupture sequence. RESULTS: Twenty-four children were identified, 11 with lower extremity deficiency, 14 with upper extremity deficiency with 1 having both. Fifteen children demonstrated neural axis abnormalities, 6 (40%) required neurosurgery. Five (45%) of 11 lower extremity deficiency children had MRI findings, 3 of these needing neurosurgery. Of the 14 upper extremity deficiency children, 10 had MRI changes, and 3 required neurosurgery. Eight children with congenital scoliosis, 5 had MRI findings, with 4 children requiring neurosurgery. The other 16 children had scoliosis without vertebral abnormalities, 10 had MRI findings, and 2 required neurosurgery. CONCLUSIONS: There is a high incidence of neural axis abnormalities (63%) in children with congenital limb deficiencies and scoliosis. A large portion of these require neurosurgical intervention. MRI should be considered soon after presentation in this population of children. LEVEL OF EVIDENCE: Level IV. DESIGN: Retrospective cohort.

Optimization of Casting in Early-onset Scoliosis.

BACKGROUND: Early-onset scoliosis is a spine deformity that presents before the age of 3 years. When compared with age-matched controls, children with the condition are known to be at risk for significant morbidity and mortality. Although many works support the use of casting for this condition, the key technical factors to optimize the outcomes of casting are not clear. This work was designed to evaluate the role of frequency of radiographic imaging and over the shoulder straps in the outcomes of casting. METHODS: Two surgeons at 2 centers followed nearly identical protocols for applying casts for early-onset scoliosis. At center A, the surgeon hoped to improve outcomes by obtaining radiographs after each cast and by reinforcing the cast with shoulder straps. At center B, the surgeon did not use shoulder straps and limited radiographs to once every 6 months. Children were included if they were 3 years or below of age and had a curve of ≥50 degrees at the time of the first cast and had a minimum of 3 years of follow-up. Center was used as a variable in a multivariable regression that also included: age at first cast, initial curve magnitude, and presence of a syrinx or genetic syndrome with the outcome of curve resolution. RESULTS: There were 40 children at center A, 9 of whom experienced resolution of their scoliosis. There were 36 children at center B, and 11 demonstrated scoliosis resolution. At center A, 2 of 10 children with a syrinx or genetic syndrome demonstrated curve resolution while 7 of 30 without these comorbidities did. At center B 3 of 10 children with a syrinx or genetic syndrome had curve resolution while 8 of 26 children without these comorbidities did. CONCLUSIONS: The children in group A demonstrated results very similar to the children of group B. Thus, the extra burden of shoulder straps and frequent radiographs are unnecessary. LEVEL OF EVIDENCE: Level III.

Results of Casting in Severe Curves in Infantile Scoliosis.

BACKGROUND: Previous work has demonstrated best results for casting in infantile scoliosis when the curves are small and the child begins casting under 2 years of age. This study examines if casting can delay the need for growth friendly instrumentation in severe curves (50 to 106 degrees) and how the comorbidities of syrinx or genetic syndromes affected outcomes. METHODS: All children undergoing casting for scoliosis at a single institution over an 8-year period were examined. Inclusion criteria included initial curve at first casting of ≥50 degrees, age ≤3 years at the start of casting, and a minimum follow-up of 3 years. Of 148 children undergoing casting during this period, 44 met our inclusion criteria. All children underwent magnetic resonance imaging. Ten children with a syrinx were identified. Ten children had known genetic syndromes (2 who also had a syrinx). The 26 children without these comorbidities were considered idiopathic. Curve magnitude ranged from 50 to 106 degrees. RESULTS: Nine of the 26 (35%) children in the children with idiopathic curves demonstrated resolution of their curves, while only 3 of the remaining 18 (17%) did. Of the children that did not have resolution of their curves, 14 were maintained over the entire follow-up period to within 15 degrees of their initial curve and 13 were improved 15 degrees or more. Only 5 children had an increase of 15 degrees or more over the follow-up period and 4 of these have undergone growth friendly instrumentation after a mean delay from initial cast of 71 months (range, 18 to 100 mo). CONCLUSION: This study demonstrates that even in severe curves, casting was effective in delaying instrumentation in all cases, and led to curve resolution of the curves in 12 of 44 children. LEVEL OF EVIDENCE: Level III-case control study.

Tibial Rotation Osteotomies in a Matched Cohort of Myelodysplasia and Cerebral Palsy Children.

BACKGROUND: The purpose of this study is to examine the frequency of complications in children with myelodysplasia (MD) undergoing tibial rotational osteotomies with a matched cohort of children with cerebral palsy (CP). It was postulated that because of the unique health issues facing children with MD more complications would be observed. METHODS: A retrospective chart review was performed to identify children with MD who underwent primary tibial rotational osteotomy between 1997 and 2012 and had a minimum 2-year follow-up. The 15 children thus identified were matched for age, body mass index, and functional ability with 15 children with CP. Outcome measures were complications that occurred within a year of osteotomy or hardware removal. Major complications were defined as nonunions or malunions, hardware failures, deep infections, fractures, and stage III or IV decubiti. Recurrence of rotational deformity requiring revision osteotomy at any time was also defined as a major complication. Minor wound problems healing within 6 weeks with only local care were considered minor complications. RESULTS: Fifteen children with MD, who underwent 21 tibial derotational osteotomies, were available for review with a mean 7-year follow-up. The 15 children with CP underwent 22 tibial derotational osteotomies with a mean of 6 years of follow-up. In each cohort there were 3 children classified as GMFCS I, 3 children as GMFCS II, 4 children as GMFCS III, and 5 as GMFCS IV. Three (20%) of the children with MD experienced major complications (1 infected nonunion and 2 children who experienced bilateral malunions requiring revisions). One child with a major complication was classified as GMFCS II and the other 2 as GMFCS IV. None of the children with CP experienced a major complication. CONCLUSIONS: The majority of children in both groups experienced good results, but children with MD have more frequent major complications. More frequent complications were seen in children with less functional ability. LEVEL OF EVIDENCE: Level III-prognostic study, case-control study.

Calcaneocuboid joint subluxation after calcaneal lengthening for planovalgus foot deformity in children with cerebral palsy.

BACKGROUND: Calcaneal lengthening is a common procedure for the treatment of symptomatic planovalgus deformity in children with cerebral palsy. Stabilization of the calcaneocuboid joint to prevent subluxation at the time of lengthening has been described. The purpose of this study was to evaluate the magnitude of calcaneocuboid joint subluxation and associated degenerative changes in patients with cerebral palsy who underwent calcaneal lengthening for planovalgus foot deformity with and without stabilization of the calcaneocuboid joint. METHODS: We conducted a retrospective review of children with cerebral palsy who underwent lateral column lengthening through the calcaneus. For the purposes of statistical analysis, the feet were divided into 2 groups: stabilized (those that received Steinmann pin stabilization at the time of lengthening) and nonstabilized (those feet that did not receive Steinmann pin stabilization). Initial, intraoperative, and most recent follow-up radiographs were reviewed for segmental foot analysis of planovalgus deformity, calcaneocuboid joint subluxation, and osteoarthritic changes. A minimum of 3-year follow-up was required. RESULTS: Sixty-one feet were included in this study; 28 feet in the stabilized group and 33 in the nonstabilized group. Radiographic assessment of segmental foot analysis demonstrated significant improvement with regard to planovalgus deformity (P<0.05, 5 measurements). Calcaneocuboid joint subluxation occurred in 24 feet in the stabilized group and 29 feet in the nonstabilized group (P=0.5269). At final follow-up, the magnitude of subluxation was not significantly different between the groups (P=0.076). There was no difference in the incidence of osteoarthritic changes at the calcaneocuboid joint between the groups (P=0.2856). CONCLUSIONS: Lateral column lengthening through the calcaneus, for planovalgus foot deformity, significantly improved the segmental alignment of the foot with respect to radiographic assessment. Stabilization of the calcaneocuboid joint at the time of lateral column lengthening through the calcaneus did not significantly reduce the incidence or magnitude of subluxation when compared with nonstabilized lengthening. In addition, stabilization did not have an effect on the development of radiographic osteoarthritic changes at the calcaneocuboid joint. LEVEL OF EVIDENCE: Level III, retrospective comparative study.

Percutaneous epiphysiodesis of the lower extremity: a comparison of single- versus double-portal techniques.

BACKGROUND: Percutaneous epiphysiodesis can be achieved using a single-portal or a double-portal technique. This study was performed to demonstrate any differences in outcomes, especially complications, between the 2 techniques. METHODS: This was a retrospective review of cases at a single institution from 1983 to 2002 that yielded 336 children, in which 63 qualified for the study with at least 3 years of clinical follow-up. A comparison was performed through clinic chart review and radiographic measurement outcomes, searching for patient satisfaction, surgical time, and complications. RESULTS: Minor complications included superficial infections, hematomas and effusions, whereas major complications included failure to arrest growth, partial arrest with angular deformity, fracture, and joint penetration. The single-portal group had an overall complication rate of 33.3%, with a major complication rate of 20% per patient. The double-portal group had a similar overall complication rate but only a 5.3% major complication rate per patient. There was no significant difference in patient demographics, operative times, or subjective complaints. CONCLUSIONS: Many methods of percutaneous epiphysiodesis exist in the literature that report low complication rates that are comparable with the complication rate of the original open procedure. Our study demonstrated a significantly higher rate of complications (both minor and major) compared with results previously reported. Moreover, the use of a single-portal approach increased the possibility of major complication by nearly 4-fold as compared with the use of a double-portal approach that avoids crossing the midline of the physis.

Cardiac screening in congenital spine disorders.

BACKGROUND: Children with congenital spine disorders are known to have a high prevalence of congenital heart disease. The purpose of this work is to examine the presentation of these cardiac defects. It was our hypothesis that the anomalies are clinically obvious; thus, screening with electrocardiogram (ECG) or echocardiogram for occult disease would prove unnecessary after orthopaedic referral. METHODS: The records and radiographs of 190 consecutive children known to have a congenital vertebral abnormality (excluding myelodysplasia and sacral agenesis) were retrospectively reviewed. Twenty-four children presented to the orthopaedist with a known cardiac anomaly. Of the remaining children, 64 underwent cardiac screening with echocardiogram, 27 underwent screening ECG and pediatric or cardiology evaluation, and 75 underwent evaluation by the orthopaedist only. Level of evidence was case series, level IV. RESULTS: Twenty-four children presented to the orthopaedist with an established diagnosis of congenital heart disease. One child, who underwent echocardiogram, was detected to have a bicuspid aortic valve. This prevalence is consistent with what would be expected in screening the general population with echocardiogram. None of the 27 children screened with ECG and pediatric or cardiology consultation, and none of the 75 who underwent evaluation only by the orthopaedist were detected to have a cardiac abnormality. Also, no additional cardiac abnormalities have been found with follow-up of 2 to 17 years. CONCLUSIONS: The presentation of congenital heart disease associated with congenital spine disorders is general clinically evident and made before the referral to the orthopaedist. Routine referral by the orthopaedist for ECG or echocardiogram is not supported by this work.

Function of children with myelodysplasia and lower extremity amputations.

This is a retrospective study of the functional status of children who underwent a lower extremity amputation for complications of myelodysplasia. With a computerized surgical database, 12 children with myelodysplasia who underwent an amputation at the Boyd level or above at a single children's referral hospital between 1983 and 2001 were identified. Four patients could not be contacted, but the remaining 8 patients were evaluated through chart review and interview to assess the impact of the amputation on their function. With a mean follow-up time of 9 years (range, 5-15 years), all 6 of the patients with a below-knee or Boyd amputation continued to ambulate using a prosthesis. Most patients occasionally reported having ulcers on their residual limb, but these cases were easily managed and did not result in amputation revisions.The only patient in this series with an above-knee amputation and the only patient with a knee disarticulation were exclusively wheelchair ambulators and no longer owned a prosthesis. This study supports the notion that children with myelodysplasia can have amputations and successfully wear a prosthesis to maintain their ambulation.

Length changes in tibial osteotomy with angular correction.

Twelve boys, aged 11-17 years, who underwent percutaneus proximal tibial osteotomy with acute angular correction and application of external fixator for unilateral Blount's disease were retrospectively reviewed. Preoperative radiographs were compared with radiographs at healing to evaluate changes in tibial length and overall limb length. Angular correction increased overall limb length by a mean of 1.4 cm (range -0.4 cm to 3.2 cm). This increase was a mean 0.7 cm less than was predicted by adding the preoperative tibial and femoral lengths. This failure to achieve the predicted limb length occurs due to shortening in the tibia and should be considered when planning an osteotomy.

Scoliotic curve patterns in patients with Chiari I malformation and/or syringomyelia.

STUDY DESIGN: A retrospective radiographic review was performed on 41 patients with scoliosis associated with a Chiari I malformation and/or syringomyelia. OBJECTIVES: To characterize curve patterns and curve features in this population and possibly refine the radiographic indications for magnetic resonance imaging in patients with a normal history and physical examination. SUMMARY OF BACKGROUND DATA: A subset of patients with "idiopathic" scoliosis may have an underlying neurologic abnormality. The radiographic indications for magnetic resonance imaging in asymptomatic patients with a normal clinical examination are not well defined. METHODS: Data were collected from standing posteroanterior and lateral radiographs. The curve pattern and specific curve features were recorded and compared with historic controls. Thoracic kyphosis and total lumbar lordosis were also measured. RESULTS: Fifty-one percent of patients were male. Ten curve patterns were identified, and, based on our criteria, approximately 50% of patients had an "atypical" pattern (left thoracic, double thoracic, triple, long right thoracic). A subset of those with "typical" patterns (right thoracic, right thoracic/left lumbar) had atypical features including a superior or inferior shift of the apex and/or the upper or lower end vertebrae. The mean kyphosis (T3-T12) was 41.8 degrees. CONCLUSIONS: Although the decision to obtain magnetic resonance imaging in a patient with scoliosis should be based on both clinical and radiographic criteria, we suggest that a heightened index of suspicion is warranted with certain curve patterns (left thoracic, double thoracic, triple, and a long right thoracic curve with end vertebra caudal to T12), and with a high or low apex and/or end vertebra, especially in males and patients with a normal to hyperkyphotic thoracic spine.

Rehabilitation after femoral osteotomy in cerebral palsy.

This is a study of the time required to return to preoperative functional levels after proximal femoral osteotomy in children with cerebral palsy. Seventy-one consecutive children who underwent proximal femoral osteotomy to treat an unstable hip secondary to cerebral palsy are retrospectively reviewed. All children returned to their preoperative ambulatory function within 30 months of the procedure. Children who were community or household ambulators returned to their preoperative function at a mean of 7 months after osteotomy, while wheelchair and therapeutic ambulators required a mean of 10 months. Children who had regular visits with a licensed therapist tended to return to function more quickly than those who had exercises preformed by their parents or care-takers after instruction by a therapist. The authors conclude that families should be advised that rehabilitation after osteotomy requires on average 7-10 months, but times up to 30 months are possible.

Spine behavior caudal to instrumentation in King II and IV curves.

This retrospective review of children surgically treated for King Type II or IV curvature of the spine required a minimum lumbar Cobb angle of 40 degrees and a minimum lumbar inclination (the angle formed between a line through the spinous processes of the three most caudal lumbar vertebrae and a line perpendicular to the floor) of 10 degrees. Twenty children had combined anterior thoracolumbar and posterior instrumentations whereas 20 had only posterior instrumentation. Children who had combined surgery had significantly better corrections of their lumbar Cobb angles. They had a mean correction of 43.3 degrees compared with 26.7 degrees in children with posterior instrumentation only. These superior corrections of the lumbar Cobb angles did not result in significantly better improvements in the lumbar inclinations. Patients who had the combined procedures had a mean improvement of 10.1 degrees, whereas patients who had posterior instrumentation only had a mean improvement of 8.0 degrees in lumbar inclination. Instead of having superior corrections of the lumbar inclinations, the combined surgeries resulted in a significant worsening of the angle between the end plates of the last instrumented vertebra and the next most caudal end plate. In patients who had combined surgery this angle averaged 8.4 degrees, whereas in patients who had posterior instrumentation only this angle averaged 4.1 degrees.