Rare Post Traumatic Kyphoscoliosis of the Thoracolumbar Spine After Posterior Fusion for Adolescent Idiopathic Scoliosis: A Case Report and Review of the Literature

Abstract

Background:

Vertebral fracture after posterior arthrodesis and instrumentation for idiopathic scoliosis is a rare occurrence with limited reported cases in the literature.

Case Presentation:

A 16-year-old female patient surgically treated for adolescent idiopathic scoliosis with T2-L1 posterior spinal fusion was in a low-energy fall resulting in fracture of L1 and new kyphosis and scoliosis of the thoracolumbar spine at the distal aspect of the fusion. The fracture was initially managed conservatively, however pain persisted and thus she was indicated for extension of the fusion and correction of the post-traumatic kyphosis.

Conclusions:

Fractures after posterior spinal fusion for idiopathic scoliosis are rare and evidence for the appropriate management remains limited in the literature.

Level of Evidence: V

Introduction

Vertebral fracture after posterior arthrodesis and instrumentation for idiopathic scoliosis is rare; only a few cases have been reported in literature.^1-5 The authors present a case of traumatic fracture of the first lumbar vertebra (L1) at the distal aspect of the fusion construct. The L1 fracture resulted in a new kyphotic and scoliotic deformity initially treated conservatively however eventually necessitating surgical intervention.

Case Report

A 16-year-old Caucasian female was referred for evaluation for adolescent idiopathic scoliosis (AIS). She reported chronic back pain resistant to oral anti-inflammatory medication and physical therapy. Plain standing radiographs showed a 46 degrees right-going thoracic curve and 34 degrees left-going lumbar curvature; MRI was negative for intra-spinal pathology. She underwent T2 to L1 posterior spinal fusion with instrumentation (PSFI) (Figure 1). The operative and postoperative periods were uneventful. At her first follow-up she was doing well with regards to pain management and deformity correction. Forty days after the surgery, she fell down a flight of stairs in her home and landed on her left buttocks. She went to a local emergency department due to significant low back pain as well as left lateral flank paresthesias and pain. Spine X-rays obtained were initially interpreted as negative and she was released with pain medications. Two days after the fall, her pain persisted and she re-presented. Repeat plain radiographs showed a compression fracture of L1 with kyphotic deformity as well as left scoliotic deformity (40 degrees and 19 degrees respectively) of the thoracolumbar spine, without evidence of mechanical failure of the instrumentation (Figure 2). She was placed in a thoracolumbar brace acutely. At the first follow-up, 6 days after her injury, she stated that the pain was stable and constant in nature. She had pain when upright and significant night pain. She also continued to have left flank numbness. A comprehensive neurologic examination was normal. A CT scan was acquired, which showed a non-displaced L1 body fracture that extended about the left pedicle screw with an acute kyphosis at the distal aspect of the fusion construct without evidence of hardware fracture or failure (Figure 3). The fracture was considered stable, in the absence of hardware failure or neurologic deficit. At this time, conservative management was recommended, encompassing oral anti-inflammatory medications, cold therapy and full-time brace. After 3 months follow-up radiographs were stable, however the patient’s symptoms showed no improvement in symptoms. The patient complained of pain that limited her activities and sleep quality. For these reasons, she was indicated for partial hardware removal, correction of the thoracolumbar kyphosis and extension of the fusion to L4.

Figure 1.

Radiographs show the radiological status of the patient before the injury: a posterior spinal fusion from T2 to L1 with the hardware is in stable alignment without evidence of complications.

Figure 2.

Radiographs show vertebral height loss at L1, concerning for acute fracture. Interval increased 19 degrees thoracolumbar left scoliosis and 40 degrees thoracolumbar kyphosis.

Figure 3.

The CT images show the fracture line of the L1 body that start from the screw track.

The revision surgery was performed 10 months after the patient’s injury (approximately 11 months after index surgery). The spine was exposed T9 to L4. The rods were manually cut just above the T10 level. These rods and the pedical screws at L1 were removed. Two Ponte osteotomies were performed at levels T12-L1 and L1-L2 to achieve the correction of the kyphosis. Pedical screws were placed at L2-L4 and new rods were placed and connected to the old rods using an in-line connector at the right side and a side-to-side connector at the left side. The residual scoliotic and kyphotic deformities were corrected additionally with in situ rod benders. The operative and the postoperative hospital course were uneventful and she was discharged at the third postoperative day.

Radiographs acquired two and six months after surgery showed stable alignment without evidence of hardware complications. Her back pain improved by the second postoperative month and was completely resovled at six months after surgery. She was released to full activity without restrictions. At one-year follow-up, the patient was in excellent overall condition without any concerns of pain or functional limitation and radiographs demonstrated no further complications (Figure 4).

Figure 4.

Twelve month post-operative radiographs show distal extension of posterior spinal instrumentation with hardware extending from T2 to L4, removal of the L1 pedicle screws and the correction of the thoracolumbar kyphosis and scoliosis.

Discussion

Fractures of the thoracolumbar spine represent around 90% of traumatic spine fractures in adults.⁶ Most of these fractures do not manifest neurological complications, and thus can be treated conservatively with brace and rest. In these cases, persistent pain and kyphotic deformity are the most common indications for surgical correction.^7-8

However, thoracolumbar fractures after PSFI for AIS are very rare. In this case, the traumatic fracture was initially stable and the patient was neurologically intact; thus, conservative management was initially trialed. The persistence of pain and the presence of a kyphotic deformity rendered surgical intervention a reasonable option for this patient’s symptoms after 10 months of conservative management. Due to the length of intervening time between injury and surgical intervention, two Ponte osteotomies were required to successfully correct the rigid kyphotic deformity. The instrumentation below the fracture from L2 to L4 was considered necessary to maintain the reduction of the kyphosis and the scoliosis.

Only few similar cases have been reported in literature,^1-5 and each differs from our case. To the best of our knowledge, this is the first report in literature of a vertebral body hyper-flexion fracture that involved the distal-most vertebra of a construct with pedicle screws.

In 2014, Pirris and Kimes reported a case of a 12-year old patient with PSFI for AIS treated with Harrington’s rodding technique. At the age of 43 years, she developed adjacent-segment disease that was treated with extension of the PSFI from L3 to ileum. After 8 months, the patient fell and reported a compression fracture of L2, which was within the fusion mass cranial to the new pedicle screw construct. After 2 weeks of observation a balloon kyphoplasty was performed because of pain and kyphosis. In this case report, the procedure was unsuccessful and she developed pseudarthrosis and relapse of the kyphosis.¹

Neyt and Weinstein, in 1999, reported a fracture-dislocation three years after a patient underwent PSFI with Cotrel-Dobousset instrumentation for idiopathic scoliosis. The patient was involved in a car accident and L2 was injured, which was the first free vertebra below the prior arthrodesis. In that case, the fracture was determined to be unstable with risk of neurologic compromise. The authors extended the arthrodesis to L4 with two cross-linked side-to-side rods by a posterior approach.²

Bago et al., in 1998, reported a case of a 30-year-old woman who underwent anterior and posterior fusion for AIS with Cotrel-Dobousset instrumentation. She was involved in a car accident two years after her index surgery, sustaining a wedge fracture of T11 with deformity of the rod and loss of correction. Because the fracture was stable and within a solid fusion mass, conservative treatment was successful.³

King and Bredford, in 1980, reported on a fracture-dislocation of T11 in a patient involved in motor-vehicle collision two years after PSFI with Harrington’s rod. They chose an operative approach because the angulation of the rod interfered with stable closed reduction.⁵

Other similar reports in literature refer to older or a geriatric population. Various authors reported fractures in fused segments with or without minor trauma,^9-10 after the removal of the implant,¹¹ or in patients with other comorbidities like idiopathic skeletal hyperostosis.¹²

The literature posits that these fractures are unusual because the fusion mass and instrumentation protect fused spinal segments. This hypothesis was namely coined by by King and Bago.^3,5 No current reports in the literature involve segments with pedicle screw instrumentation. Pedicle screws are biomechanically superior to previously ubiquitously utilized hooks, providing better pull-out strength, higher stiffness and stability to the construct.¹³ After immobilization and arthrodesis of a long spinal segment, there is an increased mobility between the upper and the lower adjacent segments.^14-15 We hypothesize that, in this case, fracture occurred at the end vertebra of the construct for three reasons: first, there is greater mobility of the caudal adjacent segment at the junction of the rigid thoracic spine and the mobile lumbar spine; second, the injury occurred only 40 days after the surgery and there was not yet a solid fusion mass; and third, the fracture occurred through the weakened bone of the screw track because of time course and relatively recent screw insertion. The presence of the pedicle screw construct gave enough stability such that the fracture did not grossly change during 10 months of observation.

Conclusion

We present an unusual case of vertebral fracture through a pedicle screw track 40 days after AIS surgery. Thoracolumbar fractures in patients who are status post PSFI A for AIS are very rare and mainly the consequence of high-energy injuries. Secondary kyphosis with persistence of pain should be considered an indication for extension of the fusion and to restore normal sagittal balance.