a review of pediatric lumbar spine trauma...neurosurg focus / volume 37 / july 2014 pediatric lumbar...

Neurosurg Focus / Volume 37 / July 2014

Neurosurg Focus 37 (1):E6, 2014

1

©AANS, 2014

Pediatric spine fractures are usually the result of high-speed and impact injuries such as a motor vehicle accident or a fall from great height. Spine

fractures in children represent 1%–3% of all pediatric fractures.1,34 The incidence of pediatric spine injuries peaks in 2 age groups; children < 5 years old and children > 10 years old. There is a seasonal peak of pediatric spinal injuries from June to September, during summer break. There is another seasonal peak in the 2 weeks surrounding the Christmas holiday.

The mechanism of injury in the pediatric population varies with age. In young children ages 0–9 years, the pre-dominant cause of injury is falls and automobile-versus-pedestrian accidents (> 75%). In children ages 10–14 years, motor vehicle accidents (40%) are the major cause of lum-bar fractures, and falls and automobile-versus-pedestrian accidents are less prevalent. In children 15–17 years of age, motor vehicle and motorcycle accidents become the leading cause of spine injuries (> 70%), and there is also an increase in sports-related spine trauma.

The pediatric spinal column is different from the adult spine in many ways, and this can predispose infants and small children to flexion and extension injuries. They have proportionally larger heads compared with their

bodies and have underdeveloped neck musculature.26,30 They also have inherent ligamentous laxity, elasticity, and incomplete ossification.23,25,30,31 Their facet joints are small and more horizontally oriented, resulting in greater mobility and less stability.23–25,32 Because of these biome-chanical differences, younger children (0–8 years) tend to have fewer fractures and greater incidence of SCIWORA (spinal cord injury without radiographic abnormality). Hyperextension coupled with the hypermobility of the pe-diatric spine can result in momentary dislocation followed by spontaneous reduction, resulting in a damaged spinal cord but a radiographically normal–appearing vertebral column.20,24,26 Although SCIWORA has been reported to occur at rates as high as 20% in children, it falls dramati-cally to < 1% in adults. A more adult-like vertebral column (9–16 years) with sturdier osseoligamentous formation provides better protection of the spinal cord and therefore less severe spinal cord injury (SCI) in this age group com-pared with the younger age group. When subjected to high stress, the adult spine is more likely to suffer breakage of bones and frank rupture of ligaments in comparison with the pediatric spine, in which deformation and return to normal alignment is more common.

Lumbar Spine Radiographic Findings and Evaluation

Imaging studies are often used as an adjunct in evalu-ating and diagnosing injury in trauma. Findings can be misinterpreted as pathological in the setting of an imma-

A review of pediatric lumbar spine trauma

*Christina sayama, m.D., m.P.h.,1 tsulee Chen, m.D.,2 GreGory trost, m.D.,3 anD anDrew Jea, m.D.1

1Neuro-Spine Program, Division of Pediatric Neurosurgery, Texas Children’s Hospital, and Department of Neurosurgery, Baylor College of Medicine, Houston, Texas; 2Division of Pediatric Neurosurgery, Akron Children’s Hospital, Akron, Ohio; and 3Division of Spinal Surgery, Department of Neurological Surgery, University of Wisconsin, Madison, Wisconsin

Pediatric spine fractures constitute 1%–3% of all pediatric fractures. Anywhere from 20% to 60% of these frac-tures occur in the thoracic or lumbar spine, with the lumbar region being more affected in older children. Younger children tend to have a higher proportion of cervical injuries. The pediatric spine differs in many ways from the adult spine, which can lead to increased ligamentous injuries without bone fractures. The authors discuss and review pedi-atric lumbar trauma, specifically focusing on epidemiology, radiographic findings, types and mechanisms of lumbar spine injury, treatment, and outcomes.(http://thejns.org/doi/abs/10.3171/2014.5.FOCUS1490)

Key worDs • pediatric spine trauma • lumbar • fracture

Abbreviations used in this paper: ALL = anterior longitudinal lig-ament; AP = anteroposterior; PLL = posterior longitudinal ligament; rhBMP-2 = recombinant human bone morphogenetic protein–2; SCI = spinal cord injury; SCIWORA = SCI without radiographic abnor-mality; TLSO = thoracolumbosacral orthosis; VB = vertebral body.

* Drs. Sayama and Chen contributed equally to this work.

Unauthenticated | Downloaded 05/30/20 05:22 PM UTC

C. Sayama et al.

2 Neurosurg Focus / Volume 37 / July 2014

ture spine. Knowledge of the various stages of growth and the corresponding age of the patient are imperative for ac-curacy of diagnosis.

Vertebral formation occurs in 3 stages: membranous proliferation, chondrification, and ossification. Although ossification begins in the womb, it is completed decades after birth. There are 3 ossification centers in the verte-bra: the centrum of the vertebral body (VB), and right and left centers in the posterior arch (Fig. 1). At approximately 10 weeks of gestation, ossification of the centrum begins in the thoracolumbar region and proceeds in both direc-tions. Evidence for ossification of the neural arches has been documented as early as 9 weeks of gestation in the cervical regions and progresses in the cranial-to-caudal direction.32 The neurocentral synchondrosis delineates the junction where the centrum and both arch centers fuse.

During the mid- to late teenage years, secondary ossi-fication centers present in the transverse and spinous pro-cesses, and the ring apophyses fuse with the primary os-sification centers. During the midteenage years endplates begin to fuse, and this is completed at approximately 21–25 years of age (Fig. 2).

Radiographic imaging of the pediatric spine is lim-ited by the concern for radiation exposure to a maturing spine. Whereas an adult trauma evaluation may include a “pan scan” of the entire spine, evaluation of the pediatric patient relies heavily on a combination of the mechanism of injury and physical examination. Other factors such as “breath arrest” at the time of injury can be helpful in de-termining the likelihood of a VB fracture in the thoracic region.17

When imaging studies are obtained, there can be find-ings that may be misconstrued as pathological that in fact are normal in children. The neurocentral synchondrosis can still be visible at 3–6 years of age. Radiographically, this can appear as a groove at the corner of each VB. Until the endplates fully fuse in the mid-20s, these areas may be misinterpreted as VB fractures.

“Wedging” of the VBs is sometimes read as a com-pression fracture. Gaca et al.9 performed a retrospective radiographic review looking at the occurrence and degree of wedging of the VB from T-10 to L-3 based on anterior/posterior VB height measurements. Although the previ-ously reported ratio of 0.95 was used as the cutoff for concern about compression fracture,12 their study revealed that patients without spine injury can have an anterior/posterior ratio of as low as 0.893.15

All imaging modalities have their advantages and limitations. Plain radiographic films are useful in evalu-ating for gross evidence of misalignment both in the an-teroposterior (AP) and lateral views. Findings such as a trapezoidal shape to the VB, scoliosis, or offset spinous processes should raise the level of suspicion of acute inju-ry. Following plain radiographs, CT scanning has repeat-edly been shown to be superior in identifying acute bone injury. In the pediatric population, the concern for radia-tion exposure sometimes influences the practitioner to be judicious in obtaining a CT scan. Additionally, distraction and apophyseal injuries are difficult to determine on CT imaging alone. Magnetic resonance imaging studies are useful in determining spinal cord and soft-tissue injury (ligamentous, edema, and so on). Hemorrhage, edema,

and acute bone injury represented by marrow changes in a series of T1-weighted, T2-weighted, and STIR images can aid in delineating true injury from false positives and acute versus chronic findings. However, the MRI modality can show false-negative findings in detecting neural arch fractures.

Although imaging studies have their usefulness in evaluating the pediatric trauma patient, knowledge of the natural progression in growth of the immature spine is critical in correctly interpreting results. If the mechanism of injury and physical examination are suspicious for spine trauma, then AP and lateral plain radiographs are warranted. From there, a CT scan is helpful if a fracture is detected on radiographs or if the results of physical ex-amination do not match the imaging findings. An MRI scan is useful in the setting of neurological deficit.

Types of Lumbar Spine InjuriesSeveral classification systems exist to describe lum-

bar fractures systematically. The most comprehensive and accurate is probably the AO fracture classification. However, this system is cumbersome to apply in practice. Instead, most authors extrapolate from Denis’s 3-column model, originally described for fractures at the thoraco-lumbar junction. This model divides the spine into 3 col-umns: anterior (ventral half of the VB and intervertebral disc including the anterior longitudinal ligament [ALL]); middle (dorsal half of the VB and intervertebral disc in-cluding posterior longitudinal ligament [PLL]); and poste-rior (pedicles, lamina, facet joints, spinous processes, and interspinous and supraspinous ligaments). Depending on which combination of columns are injured, Denis labeled these fractures as compression, burst, seat belt, or frac-ture-dislocation. This classification system has thus been

Fig. 1. Axial CT of the spine obtained in a 5-month-old baby. Arrows delineate the synchondroses that separate the centrum, right, and left primary ossification centers.



Pediatric lumbar spine trauma

3

extrapolated to the thoracic spine, the lumbar spine, and the pediatric spine. Treatment approaches were then sug-gested by the type of fracture.

Compression FracturesCompression fractures are the most frequently seen

fractures in the lumbar spine and usually occur as a result of an axial compression load in flexion. This is a stable injury in which the posterior column is intact. Spinal cord injury has not been documented in this type of fracture, and spinal canal compromise does not occur. Kyphotic angulation can occur because there is more reduction in height of the VB anteriorly. A long-term increase in de-formity can occur when there is an associated kyphosis of > 30%.

The patient in Case 1 was a 16-year-old girl who fell from a horse and was thrown over, landing on her but-tocks. She presented to the emergency department with back pain and left wrist pain. She was found to have an L-2 compression fracture in addition to a left wrist frac-ture. She was neurologically intact. A CT scan of the lumbar spine showed an L-2 compression fracture with approximately 19° of loss of height anteriorly. She was treated conservatively in a brace and did well with con-servative management, with improved back pain (Fig. 3).

Burst FracturesBurst fractures occur with axial load without flexion,

and affect the anterior and middle columns. This type of fracture makes up 15%–20% of all major VB fractures.

The VB is partially or completely comminuted and there is disruption of the posterior wall. There are often frag-ments that break into the spinal canal and there may be (SCI).

The patient in Case 2 was a 15-year-old male unre-strained passenger involved in a motor vehicle accident who presented with back pain and bilateral ankle pain. On CT imaging his lumbar spine demonstrated an L-1 burst fracture with approximately 19° of kyphotic defor-mity and disruption of the PLL with retropulsion of VB fragments posteriorly (Fig. 4A). An MRI study was per-formed, which demonstrated an associated contusion of the conus medullaris, disruption of the ALL and PLL, and no ongoing spinal cord compression (Fig. 4B). The patient was noted to have decreased strength in his bilateral lower extremities; his physical examination was limited due to pain but he was able to wiggle his toes bilaterally. The patient did have bowel and bladder dysfunction as well as some associated dysesthetic pain. Conservative treatment in a brace was attempted because there was no ongoing compression of the spinal cord on MRI or worsening of findings on his neurological examination; however, the pa-tient had intractable back pain and increased kyphosis on upright lumbar spine radiographs (26° in brace; Fig. 4C). Therefore, the decision was made to perform posterior spinal stabilization and fusion from T-11 to L-3 (Fig. 4D). Postoperatively the patient did well, and after 2 weeks in the hospital was transferred to an inpatient rehabilitation center. He has been seen in a 3-month follow-up visit and at that time he was ambulatory, with a left orthotic foot brace and a walker.

Fig. 2. Axial CT scans demonstrating the primary (medium blue arrows, dot) and secondary (light blue lines, dots) ossification centers of the L-4 vertebra as they change over a lifespan from an infant to an adult. There are 5 secondary ossification centers: 1 at the tip of the spinous process, 1 at the tip of each transverse process, and 2 ring epiphyses. yo = years old.


C. Sayama et al.


Anterior and Posterior Element Injuries (3-Column Injuries)

This type of injury pattern is inherently unstable and is characterized by failure of all 3 columns. These account for approximately 20% of all spinal injuries and are asso-ciated with the highest likelihood of SCI. Approximately 75% of patients with fracture-dislocation injuries have some neurological deficit and 50% of these have complete SCIs.

There are different types of injury patterns in 3-col-umn injuries: those with distraction and those with rota-tion. The first (with distraction) is a flexion-distraction type injury where the flexion force vector is acting around an anteriorly placed axis of rotation and the posterior ele-ments are distracted. This is commonly seen in association with motor vehicle–associated seat belt injuries. Seat belt injuries are often associated with intraabdominal trauma, and neurological injury is infrequent. The second type of injury (with rotation) usually occurs with falls from great heights or with heavy objects falling onto a body with a bent torso. On CT scans, this is best appreciated on axial views where one can see rotation of the superior and in-ferior VBs. This type of injury has a high propensity for

neurological sequelae. There may be disruption of the fac-ets as well as narrowing of the spinal canal. It is the most unstable and severe of all the injury mechanisms.

The patient in Case 3 was a 9-year-old girl who was a restrained passenger in a motor vehicle accident in which she had no loss of consciousness. On presentation to the emergency department she complained of back, neck, and abdominal pain. She was neurologically intact with full strength and sensation in her bilateral lower extremities. A CT scan of her lumbar spine demonstrated an L-1 seat belt–type fracture with anterior VB wedging and a frac-ture through the bilateral facets and the T-12 spinous pro-cess (Fig. 5A). The patient had layering fluid in her abdo-men and, due to concern for bowel injury, she was initially taken to the operating room by pediatric surgery for an exploratory laparotomy and small-bowel resection. She was kept on strict spinal precautions with no alteration in findings on her neurological examination. Lumbar spine MRI studies demonstrated mild compression of L-1 with STIR signal extending through both pedicle and pars with disruption of the interspinous ligaments and the ligamen-tum flavum (Fig. 5B). Because this injury was unstable, the patient was taken to the operating room for a T10–L3 posterior spinal fusion. Postoperatively she did well and was discharged home on postoperative Day 5. On 6-week follow-up the patient was doing well and thoracolumbar radiographs were obtained, which showed intact hardware and stable alignment (Fig. 5C).

Apophyseal Ring FracturesThe apophyseal ring is attached to the anulus fibro-

sus; it ossifies between 4 and 6 years of age and fuses at 18 years of age. There is an osteocartilaginous portion that exists between the VB and apophyseal ring that is rela-tively weak and susceptible to repeated stresses. Apophy-seal ring fractures often result from lifting a heavy object, but may also occur from falls or twisting injuries. Many patients with apophyseal ring fractures describe a “pop” at the time of injury, which is then followed by radiculopa-thy. Diagnosis is confirmed with MRI and/or CT studies to pinpoint the exact location and configuration of the le-sion. In the pediatric spine the growth zone in the VBs is found in the endplates. Disruption of this zone can easily occur with moderate shearing forces. This fracture is typi-cally seen in the adolescent or young adult and presents just like a herniated disc with a radiculopathy.33 Yen et al. reported 4 adolescent cases and noted a strong associa-tion of adolescent apophyseal ring fractures with a slipped capital femoral epiphysis and overweight/obesity.35

The patient in Case 4 was a 13-year-old girl who fell on her tailbone while playing volleyball and developed acute onset of back pain with radicular pain down her left posterior extremity to the dorsum of her left foot. A CT scan was done, which demonstrated a posterior detached fragment at the superior S-1 endplate and a defect in the bone adjoining the fracture site (Fig. 6A and B). Lumbar spine MRI demonstrated a left eccentric disc herniation with impingement on the exiting left S-1 nerve root (Fig. 6C and D). Her clinical picture and radiographic imaging was consistent with an apophyseal ring fracture and as-sociated L5–S1 disc herniation causing an S-1 radiculopa-

Fig. 3. Case 1. Sagittal CT scan of the lumbar spine showing an L-2 compression fracture with 19° loss of height anteriorly.




5

thy. Five months of conservative management, with rest, NSAIDs, steroids, and physical therapy, was attempted; however, she continued to have debilitating activity-lim-iting pain. Therefore, surgery was recommended and she underwent a minimally invasive left L5–S1 microdiscec-tomy. She did well postoperatively and on 3-month follow-up was pain free.

Treatment OptionsDifferent classification schemes exist for thoracic and

lumbar spine fractures. The Thoracolumbar Injury Clas-sification and Severity Score has commonly been applied in the adult population with trauma to help determine non-operative versus operative management of thoracolumbar

Fig. 4. Case 2. A: Sagittal lumbar spine CT scan showing an L-1 burst fracture with 19° of kyphosis. B: Axial and lateral MRI studies of the lumbar spine demonstrating the L-1 burst fracture with moderate narrowing of the canal and a contusion of the conus medullaris, without signs of ongoing compression. C: Lateral thoracolumbar radiograph obtained with the patient in a TLSO brace showing increased kyphotic deformity. D: AP and lateral thoracolumbar spine radiographs showing the postopera-tive fusion construct with improvement in kyphotic deformity at L-1.


C. Sayama et al.


fractures.16 It has subsequently been studied and applied to thoracic fractures and/or lumbar fractures alone. A score is calculated based on injury mechanism, neurological sta-tus, and integrity of the posterior ligamentous complex; a score of ≤ 3 suggests nonoperative treatment, a score of 4 suggests nonoperative or operative treatment, and a score of ≥ 5 suggests operative treatment.15 The Thoracolum-bar Injury Classification and Severity Score has recently been studied retrospectively in the pediatric population, and Garber et al. report excellent validity of this scoring system in their pediatric cohort (unpublished data).

As a basic rule of thumb, many stable fractures can be treated conservatively, whereas unstable fractures require surgical stabilization. There are many different braces that can be used to obtain thoracic and lumbar immobilization for conservative management. The thoracolumbosacral orthosis (TLSO) or the Jewett brace (used if hyperexten-sion is desired) both can be used to achieve this result. For upper thoracic spine fractures between T-1 and T-4, a SOMI (sterno-occipito-mandibular immobilizer) brace is used in conjunction with a TLSO brace for treatment. Duration of bracing therapy depends on the injury and neurosurgeon, but is typically at least 3 months.

Surgical stabilization for unstable lumbar fractures can often be managed via a posterior approach. Adoles-cents can often be stabilized with adult-type instrumen-tation; the pediatric spine reaches adult maturity at ap-proximately 9 years of age. Younger children, however, have smaller pedicles, and thus pedicle screw placement may be challenging. They also have smaller spinal canals, so placement of sublaminar hooks is not safe. Stereotactic guidance has helped with placement of instrumentation in these cases. Also, rhBMP-2 has been used in young pa-tients to promote bone fusion because they routinely have a very limited amount of bone autograft available. Although the FDA has only approved the use of rhBMP-2–soaked absorbable collagen sponges for use in adult anterior in-terbody lumbar fusion, “off-label” applications have been

reported.4,17,18 We previously reported the early safety and efficacy of rhBMP-2 use for posterior spinal fusions in the pediatric population (mean follow-up of 11 months).8

Lumbar compression fractures are treated with bed rest, with gradual resumption of activities as tolerated when the kyphotic wedge of the vertebra is < 10°. Howev-er, when the kyphotic deformity is > 10° in the immature spine, immobilization in hyperextension is recommended for 2 months, followed by bracing for ≥ 1 year. Surgery with posterior and/or anterior stabilization is recommend-ed in adolescents who are near skeletal maturity with > 30° of kyphosis,5,28 or if the patient has had a prior lami-nectomy. Stabilization and correction of the deformity can be accomplished with pedicle screw constructs, a univer-sal segmental fixation system, or the Harrington rod dis-traction system.

Burst fractures may be treated conservatively when there is no neurological injury. Conservative treatment usually consists of hyperextension casting for 2–3 months and bracing for an additional 6–12 months. Surgical in-tervention is relatively indicated with > 50% loss of body height, > 50% canal compromise from retropulsed bone fragments, or > 30° of kyphosis. Absolute indications for surgical decompression and/or fusion include neurologi-cal deficits or progressive kyphosis.2,5 In the presence of multiple nerve root palsies or SCI, anterior decompression must be considered.

Unstable injuries, such as the 3-column fractures mentioned above, are most often operative injuries. The only situation that can be treated conservatively is when there is fracture-dislocation with minimal displacement and the fracture line goes through the bone (a “bony chance fracture”). In this case, conservative management consists of case immobilization for 8–10 weeks followed by bracing thereafter.34 Surgical stabilization is indicated with progressive neurological deficit or displacement > 17° of kyphosis or when the predominant injury is dis-coligamentous instead of bony. Three-column injury with

Fig. 5. Case 3. T12–L1 Seat belt injury. A: Sagittal CT scan of the lumbar spine showing L-1 anterior wedging with distraction and fracture (red arrow) through the posterior elements. B: Sagittal MRI study of the lumbar spine demonstrating STIR signal abnormality (arrow with dashed line) with disruption of the posterior ligamentous complex. C: Postoperative AP and lateral thoracolumbar radiographs demonstrating a stable construct on 6-week follow-up.




7

rotation is the most unstable and is uniformly treated with surgery. The procedure consists of posterior instrumenta-tion and fusion 1 or 2 levels above and below the level of injury.11,19,21

Conservative management for apophyseal ring frac-tures consists of rest, analgesics, physical activity modifi-cation, and physical therapy; however, it is rarely success-ful and surgical excision of the limbus fragment is usually required. Surgery is done via a posterior approach to re-move the offending fragments with or without an inter-laminal laminectomy.13,35 Typically these are not unstable injuries, and thus fusion is not necessary and outcomes are usually favorable.

OutcomesA majority of lumbar spine fractures in children and

younger adolescents are minor, stable, and do not pre sent with neurological deficit. One must approach pediatric spine fractures in a systematic way, taking into consid-eration immediate stability, the need for neurological de-compression, and late stability and propensity for healing. Because children have more laxity of the spine, physicians must keep a high suspicion for SCIWORA and therefore order and proceed with more imaging (such as MRI) when clinically indicated. Previous reports estimate that anywhere between 7.5% and 30% of patients with thora-columbar fractures require operative intervention.6,15,23,29

Erfani et al. have described the results of thoracic and lumbar instrumentation for pediatric traumatic spine fractures over a 10-year period. Of 102 individuals, L-1 was the most frequently injured vertebra (31 patients), followed by L-2 (19 patients), L-3 (18 patients), T-12 (16 patients), and L-4 (10 patients), and the remainder of verte-bral levels had < 5 patients each. Twenty patients required surgical stabilization; however, only 15 patients partici-pated in follow-up (mean 49 months). Eight (53.3%) of the 15 patients had lumbar spine fractures, and all underwent posterior spinal instrumentation and fusion alone. Three patients with lower thoracic or thoracolumbar fractures underwent an anterior followed by a posterior approach (those with delay in diagnosis). Four patients with SCI (2 with cauda equina and 2 with incomplete SCIs) had return of baseline neurological function. They reported no post-operative complications and all patients were functionally independent at follow-up.7

Dogan et al. also reviewed a series of 89 pediatric patients with traumatic thoracolumbar fractures, 23 of whom underwent surgery. Of these fractures, 19.2% were at the thoracolumbar junction and 29.8% were somewhere between L-2 and L-5. Of the 23 patients who underwent operation, 14 (60.9%) had lumbar spine fractures. The mean follow-up was 17.2 months, and all patients had a stable fixation, visible callus formation, and maintenance of alignment.6 Twenty-nine patients with mild lumbar or

Fig. 6. Case 4. Apophyseal ring fracture with L5–S1 disc herniation. Preoperative AP (A) and lateral (B) lumbar spine CT scans demonstrat-ing a posterior detached fragment at the superior S-1 endplate and a defect in the bone adjoining the fracture site eccentric to the left near the left S-1 nerve root (red arrows). Preoperative AP (C) and lateral (D) lumbar spine MRI studies demonstrating a left eccentric disc herniation with impingement on the exiting left S-1 nerve root.


C. Sayama et al.


sacral injuries were treated successfully with bed rest alone and gradual resumption of activities. There have been several previous studies that have reported success-ful fusion associated with normal thoracolumbar growth and good alignment after both anterior and posterior ap-proaches for pediatric lumbar trauma.3,6,7,10,22,27

Many pediatric fractures are managed conservatively, and children have been noted to have a high propensity for healing. Karlsson et al. observed and reported on the abil-ity of the pediatric spine to remodel after traumatic frac-ture in the thoracolumbar spine.14 The authors followed a 24-patient observational cohort of children with thoracic or lumbar fractures for up to 47 years (range 27–47 years, mean 33 years). Twenty-one patients had simple compres-sion fractures, and all fractures were treated conserva-tively. There were 17 thoracic fractures and 7 lumbar frac-tures. The most notable result from this study is that they found a significant improvement in the ratio of the anterior height/posterior height (from 0.75 at time of injury to 0.87 at follow-up) in patients < 13 years old. This improvement in wedging of the VB demonstrates the potential of VB remodeling during growth when a fracture is encountered in a younger patient. This is a stark contrast to adults, in whom we usually see increased posttraumatic kyphosis.

In terms of quality of life measures, Erfani et al. re-ported a 33.3% incidence of occasional back pain in the long term.7 They also reported near complete improvement in all 4 patients with neurological injury in their series. In another study by Moller et al., 5 (21.7%) of 23 patients had postoperative pain at long-term follow-up. They also re-ported predominantly favorable long-term outcomes with no or minor neurological deficits.22 Dogan et al. noted that all neurologically intact patients remained so at follow-up and that 75% of the patients with incomplete SCIs recov-ered completely.6

ConclusionsPediatric lumbar spine fractures constitute approxi-

mately 1%–3% of all pediatric fractures. Because of the differences in the pediatric spine, the incidence of SCI-WORA is much higher than in adults. Conservative versus surgical management of different types of fractures de-pends on the stability of the fracture, the neurological sta-tus of the patient, and his/her injury-healing ability. When surgical intervention is indicated it is safe and effective in the pediatric population, and outcomes in terms of fusion and neurological status are good.

Disclosure

Dr. Trost is a consultant for Medtronic.Author contributions to the study and manuscript preparation

include the following. Conception and design: Jea. Acquisition of data: Sayama, Chen, Jea. Analysis and interpretation of data: Sayama, Chen, Jea. Drafting the article: Sayama, Chen, Jea. Critically revising the article: all authors. Reviewed submitted version of manuscript: Sayama, Trost, Jea. Approved the final version of the manuscript on behalf of all authors: Sayama.

References

1. Akbarnia BA: Pediatric spine fractures. Orthop Clin North Am 30:521–536, x, 1999

2. Andreychik DA, Alander DH, Senica KM, Stauffer ES: Burst fractures of the second through fifth lumbar vertebrae. Clini-cal and radiographic results. J Bone Joint Surg Am 78:1156–1166, 1996

3. Black BE, O’Brien E, Sponseller PD: Thoracic and lumbar spine injuries in children: different than in adults. Contemp Orthop 29:253–260, 1994

4. Carlisle E, Fischgrund JS: Bone morphogenetic proteins for spinal fusion. Spine J 5 (6 Suppl):240S–249S, 2005

5. Crawford AH: Operative treatment of spine fractures in chil-dren. Orthop Clin North Am 21:325–339, 1990

6. Dogan S, Safavi-Abbasi S, Theodore N, Chang SW, Horn EM, Mariwalla NR, et al: Thoracolumbar and sacral spinal injuries in children and adolescents: a review of 89 cases. J Neurosurg 106 (6 Suppl):426–433, 2007

7. Erfani MA, Pourabbas B, Nouraie H, Vadiee I, Vosoughi AR: Results of fusion and instrumentation of thoracic and lumbar vertebral fractures in children: a prospective ten-year study. Musculoskelet Surg [epub ahead of print], 2014

8. Fahim DK, Whitehead WE, Curry DJ, Dauser RC, Luerssen TG, Jea A: Routine use of recombinant human bone morpho-genetic protein-2 in posterior fusions of the pediatric spine: safety profile and efficacy in the early postoperative period. Neu rosurgery 67:1195–1204, 2010

9. Gaca AM, Barnhart HX, Bisset GS III: Evaluation of wedging of lower thoracic and upper lumbar vertebral bodies in the pe-diatric population. AJR Am J Roentgenol 194:516–520, 2010

10. Gambardella G, Coman TC, Zaccone C, Mannino M, Ciurea AV: Posterolateral approach in the treatment of unstable ver-tebral body fractures of the thoracic-lumbar junction with in-complete spinal cord injury in the paediatric age group. Childs Nerv Syst 19:35–41, 2003

11. Gumley G, Taylor TK, Ryan MD: Distraction fractures of the lumbar spine. J Bone Joint Surg Br 64:520–525, 1982

12. Hegenbarth R, Ebel KD: Roentgen findings in fractures of the vertebral column in childhood examination of 35 patients and its results. Pediatr Radiol 5:34–39, 1976

13. Ishihara H, Matsui H, Hirano N, Tsuji H: Lumbar intervertebral disc herniation in children less than 16 years of age. Long-term follow-up study of surgically managed cases. Spine (Phila Pa 1976) 22:2044–2049, 1997

14. Karlsson MK, Moller A, Hasserius R, Besjakov J, Karlsson C, Ohlin A: A modeling capacity of vertebral fractures ex-ists during growth: an up-to-47-year follow-up. Spine (Phila Pa 1976) 28:2087–2092, 2003 [Erratum in Spine (Phila Pa 1976) 30:2237, 2005 and Spine (Phila Pa 1976) 31:72, 2006]

15. Kraus R, Stahl JP, Heiss C, Horas U, Dongowski N, Schnettler R: [Fractures of the thoracic and lumbar spine in children and adolescents.] Unfallchirurg 116:435–441, 2013 (Ger)

16. Lee JY, Vaccaro AR, Lim MR, Oner FC, Hulbert RJ, Hed-lund R, et al: Thoracolumbar injury classification and sever-ity score: a new paradigm for the treatment of thoracolumbar spine trauma. J Orthop Sci 10:671–675, 2005

17. Leroux J, Vivier PH, Ould Slimane M, Foulongne E, Abu-Ama ra S, Lechevallier J, et al: Early diagnosis of thoracolum-bar spine fractures in children. A prospective study. Orthop Trau matol Surg Res 99:60–65, 2013

18. Lu DC, Sun PP: Bone morphogenetic protein for salvage fusion in an infant with Down syndrome and craniovertebral instabil-ity. Case report. J Neurosurg 106 (6 Suppl):480–483, 2007

19. Magerl F, Aebi M, Gertzbein SD, Harms J, Nazarian S: A com-prehensive classification of thoracic and lumbar injuries. Eur Spine J 3:184–201, 1994

20. Marar BC: Hyperextension injuries of the cervical spine. The pathogenesis of damage to the spinal cord. J Bone Joint Surg Am 56:1655–1662, 1974

21. McCormack T, Karaikovic E, Gaines RW: The load sharing classification of spine fractures. Spine (Phila Pa 1976) 19: 1741–1744, 1994




9

22. Moller A, Hasserius R, Besjakov J, Ohlin A, Karlsson M: Ver-tebral fractures in late adolescence: a 27 to 47-year follow-up. Eur Spine J 15:1247–1254, 2006

23. Mortazavi MM, Dogan S, Civelek E, Tubbs RS, Theodore N, Rekate HL, et al: Pediatric multilevel spine injuries: an institu-tional experience. Childs Nerv Syst 27:1095–1100, 2011

24. Pang D, Pollack IF: Spinal cord injury without radiographic abnormality in children—the SCIWORA syndrome. J Trau-ma 29:654–664, 1989

25. Pang D, Wilberger JE Jr: Spinal cord injury without radio-graphic abnormalities in children. J Neurosurg 57:114–129, 1982

26. Pang D, Zovickian JG: Vertebral column and spinal cord in-juries in children, in Winn HR (ed): Youmans Neurologi-cal Surgery, ed 6. Philadelphia: Elsevier Saunders, 2011, pp 2293–2332

27. Parisini P, Di Silvestre M, Greggi T: Treatment of spinal frac-tures in children and adolescents: long-term results in 44 pa-tients. Spine (Phila Pa 1976) 27:1989–1994, 2002

28. Pouliquen JC, Kassis B, Glorion C, Langlais J: Vertebral growth after thoracic or lumbar fracture of the spine in chil-dren. J Pediatr Orthop 17:115–120, 1997

29. Puisto V, Kääriäinen S, Impinen A, Parkkila T, Vartiainen E, Jalanko T, et al: Incidence of spinal and spinal cord injuries and their surgical treatment in children and adolescents: a pop-ulation-based study. Spine (Phila Pa 1976) 35:104–107, 2010

30. Reddy SP, Junewick JJ, Backstrom JW: Distribution of spinal

fractures in children: does age, mechanism of injury, or gender play a significant role? Pediatr Radiol 33:776–781, 2003

31. Roche C, Carty H: Spinal trauma in children. Pediatr Radiol 31:677–700, 2001

32. Skórzewska A, Grzymisławska M, Bruska M, Lupicka J, Woźniak W: Ossification of the vertebral column in human foetuses: histological and computed tomography studies. Folia Morphol (Warsz) 72:230–238, 2013

33. Swischuk LE, Swischuk PN, John SD: Wedging of C-3 in in-fants and children: usually a normal finding and not a fracture. Radiology 188:523–526, 1993

34. Vialle LR, Vialle E: Pediatric spine injuries. Injury 36 (Suppl 2):B104–B112, 2005

35. Yen CH, Chan SK, Ho YF, Mak KH: Posterior lumbar apophy-seal ring fractures in adolescents: a report of four cases. J Or-thop Surg (Hong Kong) 17:85–89, 2009

Manuscript submitted March 12, 2014.Accepted May 19, 2014.Please include this information when citing this paper: DOI:

10.3171/2014.5.FOCUS1490. Address correspondence to: Christina Sayama, M.D., M.P.H.,

Division of Pediatric Neurosurgery, Department of Neurosurgery, Texas Children’s Hospital, Baylor College of Medicine, 6621 Fan-nin St., CCC 1230.01, Houston, TX 77030. email: [email protected].


a review of pediatric lumbar spine trauma...neurosurg focus / volume 37 / july 2014 pediatric lumbar...

Documents