Advances in nonfusion techniques for the treatment of scoliosis in children

Abstract

Scoliotic deformity in young children is a challenge for the spinal surgeon. Though traditional spinal correction and fusion techniques can improve these deformities, they inhibit growth of the spine. Nonfusion technologies are an effective approach to this problem. They not only correct the spinal deformity, but also allow the spine to keep growing and developing. These techniques include the growing rod, stapling, pedicle screw tethering, the vertical expandable prosthetic titanium rib (VEPTR), and multi‐vertebrae wedge osteotomy. This is a review of advances in nonfusion techniques for the treatment of scoliosis in children.

The management of scoliotic deformities in young children with skeletal immaturity is challenging, especially in those with early onset scoliosis (EOS) ¹ , ² , ³ , ⁴ , which presents earlier, progresses more rapidly and results in more serious deformity, creating great difficulties in the treatment of EOS. Though spinal correction and fusion techniques can improve scoliotic deformities, they inhibit growth of the spine, which affects the development of thoracic organs and can lead to the “crankshaft phenomenon” ⁵ . Furthermore, they may result in thoracic insufficiency syndrome (TIS) ⁶ . The question of how to preserve the growth potential of the spine has troubled spinal surgeons for a long time. Nonfusion instrumentation techniques are an effective way to solve this problem.

Currently, several nonfusion instrumentation techniques are used in the clinic, including the growing rod, stapling, pedicle screw tethering, the vertical expandable prosthetic titanium rib (VEPTR) and multi‐vertebrae wedge osteotomy. This article reviews advances in research on nonfusion instrumentation techniques for young scoliosis patients.

Growing rod

The growing rod is composed of two metal rods, a superior and an inferior rod, which are implanted on the same side of the spine. They are allowed to overlap and are connected by a growth connector. Both upper and lower rods are fixed to the vertebral body. This device can be lengthened periodically as required to correct the scoliosis deformity. In contrast to traditional spine fusion techniques, the growing rod preserves the capacity for spinal growth and development, while correcting the spinal deformity ⁷ , ⁸ . The growing rod technique can be used in all types of scoliosis, including idiopathic, congenital and neuromuscular scoliosis. Though no consistent indications have been established for it, most scholars believe that growing rod surgery must meet the following principles: (i) immaturity of the skeleton and a capacity for the spine to grow longitudinally; (ii) progression of the scoliosis and a Cobb's angle >50°; and (iii) the scoliosis is flexible or has become flexible following anterior release surgery ⁹ , ¹⁰ , ¹¹ . As for the skills required to perform growing rod surgery, it is necessary to minimize exposure of the vertebrae and avoid postoperative spinal fusion. With the exception of the need to expose the laminar segments for the upper and lower rods, the rest of the surgical procedure should be performed away from the vertebrae. The growing rods are implanted under the skin or superficial fascia and lengthened every 6–12 months postoperatively. The indications for lengthening are an increase in Cobb's angle of >15°or an increase in sitting height of >2 cm ¹² .

The single growing rod was first used to treat childhood scoliosis by Harrington ¹³ and achieved satisfactory results in correcting the scoliosis deformity, while at the same time preserving spinal growth. Subsequently, the single growing rod gradually evolved into the dual growing rod. Thompson et al. compared the effectiveness of single and dual rods in the management of severe spinal deformity in young children and concluded that the growing rod was effective in correcting spinal deformities and preserving spinal growth potential ¹⁴ . Moreover, the dual rod was stronger than the single rod and provided initial correction more effectively.

Mahar et al. studied the differences in stability between four of the foundation configurations used in the pediatric dual growing rod technique, namely hook‐hook with cross‐link, hook‐screw with cross‐link, screw‐screw with cross‐link and screw‐screw without cross‐link ¹⁵ . Their biomechanics experiments indicated that the screw‐screw with cross‐link configuration resulted in the greatest failure load and a cross‐link did not seem to enhance fixation. Sponseller et al. analyzed the outcomes and complications of fixing the growing rod to the pelvis in neuromuscular/syndromic scoliosis and found that pelvic fixation with the growing rod not only corrected pelvic obliquity, but also provided adequate fixation ¹⁶ . In addition, the incidence of complications was not higher than in patients without pelvic fixation.

Akbarnia et al. reviewed children who had completed dual growing rod treatment, and who had a three to eleven years follow‐up ¹⁷ . They found that patients whose rods had been lengthened at ≤6 months (range, 5.5–6.7 months) had a higher annual growth rate and significantly greater scoliosis correction than those who had been lengthened less frequently (range, 9–20 months). Sponseller et al. applied the growing rod technique in infantile scoliosis with Marfan syndrome and concluded that extensible spinal growing rods were an effective solution to this syndrome, not only correcting the deformity, but also minimizing disproportion of the trunk ¹⁸ . Sankar et al. reviewed a multicenter database of 782 growing rod surgeries performed in 252 patients ¹⁹ . Five hundred and sixty‐nine of the 782 cases were performed with neural monitoring. They found that the rate of nerve injury was very low, only 0.1% patients sustaining a temporary nerve injury and no neurologic events occurring in 361 lengthenings in patients with no previous neurologic events. Therefore, the authors raised the question as to whether intraoperative neuromonitoring is necessary for simple lengthenings in children who had uneventful primary implantations.

McCarthy et al. recently devised a “Shilla” growth guidance system that does not require repeated surgical lengthening ²⁰ . The Shilla system guides growth at the ends of the dual rods, the apex of the curve being corrected, fused, and fixed to them. Growth occurs through extraperiosteally implanted pedicle screws that slide along the rods at either end of the construct. They implanted immature goats with the dual rod system, evaluated plain radiographs 6 months later and found that all of the goat spines grew with the implants in place and that growth had occurred at both the thoracic and lumbar ends of the rods for a total average of 48 mm.

Though the growing rod is effective in the correction of children scoliosis, the complications that accompany it cannot be ignored. Qiu et al. investigated the clinical outcomes of the growing rod for the correction of childhood scoliosis with an average 54 months follow‐up ²¹ . They found that curve response to correction tended to decline with consecutive lengthening and that there was a high rate of complications during the lengthening procedure. Li et al. retrospectively studied 11 scoliotic children who had undergone dual growing rod surgery and found that 45.5% of them had complications, including hook displacement, pedicle screw loosening and broken rod ²² . Therefore, the authors advised that scoliotic patients undergoing dual growing rod treatment requirej strict and regular follow‐up.

Stapling

Staple treats scoliosis by blocking growth of the convex side of the vertebrae. Many factors can affect the longitudinal growth of vertebrae, but the most important one is stress ²³ . The Hueter‐Volkmann law demonstrates that the more stress a growth plate receives, the slower it grows ²⁴ . The stapling technique was firstly used to treat scoliosis in the 1950s. However, the indications were not uniform. In 2003, Betz et al. reviewed scoliotic cases treated by stapling, and proposed that the indications for stapling surgery be as follows: (i) progressive scoliosis and the patient's age >9 years; (ii) Risser's sign ≤II degrees; and (iii) a Cobb's angle of 20°–45° ²⁵ . They also stated that a kyphosis Cobb's angle of >40°, insufficient pulmonary function and allergy to metal are contraindications to this surgery.

Zhang et al. implanted shape memory alloy (SMA) staples into the lateral aspect of the thoracic vertebrae of goats and observed the effects on spinal growth ²⁶ . They found that SMA staples inhibited spinal growth and changed the spinal growth orientation, and that greater compression resulted in larger curves. Song et al. also found that staples can significantly alter the growth rates of the two sides of vertebrae in scoliosis, the growth rate of the concavity exceeding that of the convexity ²⁷ . Betz et al. reported the results of vertebral body stapling with a minimum 2‐year follow‐up in patients with idiopathic scoliosis and found a success rate of 87% for all lumbar curves and 79% for thoracic curves <35 degrees ²⁸ . No instances of staple dislodgement or neurovascular injury occurred.

Stücker studied scoliotic patients with a minimum follow‐up of 2 years who had undergone staple correction surgeries ²⁹ . He found that curves of more than 35 degrees at the time of surgery showed signs of progression, while curves of less than 35 degrees did not. The author therefore suggested that careful patient selection for stapling is indicated for curves of <35 degrees.

Currently, there is little data on long term follow‐up of stapling. The indications for stapling are not clear and it has not been widely accepted by spinal surgeons. Most of the studies have been focused on animal experiments ³⁰ , ³¹ , ³² . Thus, a large number of cases and long term follow‐up are needed to clarify the effectiveness of stapling in childhood scoliosis.

VEPTR

VEPTR was first used to treat TIS. TIS, a condition which was first described by Campbell et al., is a syndrome in which the thoracic cage cannot support normal respiration and pulmonary development ³³ . Congenital or acquired deformities of the thoracic cage, spine or ribs reduce the volume of the thoracic cavity and make the thoracic wall rigid, which can eventually lead to respiratory insufficiency. In VEPTR, which can expand the ribs and thoracic cage and promote the growth and development of the ribs, a titanium rib distractor is placed on the concave side of the spinal curve, indirectly correcting thoracic scoliosis in children. Because scoliosis in children is often associated with TIS, VEPTR can be used to correct the deformity of spine and chest while preserving the capacity for growth of the spine and lungs ³⁴ , ³⁵ , ³⁶ .

The indications for VEPTR surgery are not yet consistent. Campbell et al. have proposed that the indications should be as follows: (i) age between six months and skeletal maturity; (ii) progressive TIS; (iii) a greater than 10% reduction in the height of the hemithorax on the concave side of the curve compared with the height of the contralateral hemithorax; and (iv) three or more anomalous vertebral bodies with three or more fused ribs at the apex of the deformity ³⁷ . Emans et al. have suggested that the indications should be: (i) unilateral or bilateral restrictive thoracic deformity or TIS which has resulted from rib fusion or defects involving thoracic and spinal deformity; (ii) direct treatment of restrictive thoracic deformity, increasing the chest volume and correcting spinal deformity while simultaneously allowing continuous growth of the thoracic spine; and (iii) treatment of thoracic and spinal deformity related to rib fusion/defect and TIS in young children ³⁸ .

Campbell et al. prospectively studied VEPTR surgery for treatment of cervical tilt and head/truncal decompensation in children with TIS associated with congenital thoracic scoliosis ³⁹ . They found that the primary thoracic scoliosis and space available for the lung improved, the cervical tilt stabilized, and head and truncal decompensation improved. Caubet et al. found that hemoglobin concentrations increased after VEPTR surgery ⁴⁰ . Therefore, the authors believed that VEPTR surgery could significantly improve the pulmonary function of EOS patients and advised using the hemoglobin concentration as a method of evaluating the outcome of surgery. Motoyama et al. retrospectively studied 24 patients with moderate‐to‐severe restrictive lung defects and TIS who had undergone VEPTR surgery ⁴¹ . After a median of 3.2 years follow‐up, the patients' forced vital capacity was found to have increased by an average of 11.1%/year.

Skaggs et al. conducted a prospective multicenter study to evaluate the risk of neurologic injury during surgical procedures using VEPTR and to determine the efficacy of intraoperative spinal cord neuromonitoring ⁴² . The authors concluded that intraoperative neuromonitoring was effective in VEPTR procedures. However, it may not be necessary for children with no history of neurologic deficit.

Hasler et al. compared the correction effectiveness of VEPTR and the growing rod, and considered that the complication rate was lower and control of the sagittal plane and pelvic obliquity as good, but correction of the coronal plane deformity was less than with the growing rod ⁴³ . Samdani et al. applied VEPTR to older children (>10 years old) with complex spinal deformities and achieved good results ⁴⁴ . They therefore believe that VEPTR is a reasonable alternative to potentially risky vertebral osteotomy for correcting deformities in selected patients.

Pedicle screw‐tether

The principles for treating childhood scoliosis by pedicle screw‐tether are the same as for stapling. This technique also corrects scoliotic deformity by blocking the growth of the convex side of the vertebrae. The pedicle screws are implanted on the convex side of the curve and connected by tethers. Compressed by the tethers, the growth of the convex side of the vertebrae is inhibited, but the concave side is not affected. Thus, this technique can correct scoliotic deformity, without limiting the growth and development of the spine.

Newton et al. used the screw‐tether technique in bovine and porcine models and found that anterolateral tethering of the spine created kyphosis and scoliosis in the rapidly growing models ⁴⁵ . However, the total lateral bending motion returned to previous levels after removal of the tether. They also found coronal vertebral body wedging and disc wedging in all the tethered vertebrae. However, magnetic resonance images revealed no evidence of disc degeneration ⁴⁶ . Thus the authors believe that the method of screw‐tether provides a possible treatment for the correction of spine deformities without arthrodesis in patients who are skeletally immature.

Braun et al. reported a study on the bone anchor‐ligament tethers technique in goat scoliosis models ⁴⁷ . They found that the flexible ligament tethers attached to bone anchors demonstrated greater efficacy and integrity than the more rigid shape memory alloy staples. Later, Braun et al. compared the 3‐D effect of bone anchor‐ligament tethers and shape memory alloy staples on an experimental idiopathic‐type scoliosis ⁴⁸ . They found the ligament tether attached to a bone anchor provided modest correction of scoliosis in the coronal plane, but not in the sagittal or transverse plane, and the deformity correction effect was better than that achieved with staples.

Nowadays, studies on pedicle screw‐tether are still few in number. Most of have been animal experiments and the clinical studies are scarce. However, it has introduced a new idea for the treatment of children scoliosis. By selecting different tethers and anchors, an absorbable nonfusion technique may be developed, which would avoid multiple lengthening surgeries and eventual fusion.

Multi‐vertebral wedge osteotomy

Vertebral wedge osteotomy technique corrects scoliosis deformity by wedge osteotomy of the convex vertebrae of the curve, combined with internal and/or external fixation to maintain the effectiveness and force line. The implants and/or brace are removed when the bone has healed. There is no bone graft fusion during the surgery (except for vertebral osteotomy on the convex side), and the spine retains growth potential and activity. Therefore, it corrects scoliosis deformity without affecting the longitudinal growth of the spine.

Maruyama et al. used the multi‐vertebral wedge osteotomy technique to treat the deformities of 20 children with idiopathic scoliosis ⁴⁹ . A specially‐designed internal fixation system was implanted to maintain effective correction and the line. Twelve weeks postoperatively, the implant was removed. With an average of 8.9 years follow‐up, no patient had neurologic complications. The average Cobb angle of 64.0° before surgery was corrected to 48.2°after surgery, and differences in postoperative pulmonary function tests were not statistically significant. In conclusion, the authors thought that a fusionless, multi‐vertebral wedge osteotomy technique could safely correct the deformity of scoliosis.

Guille et al. analyzed the feasibility, safety, and utility of vertebral wedge osteotomy for nonfusion treatment of paralytic scoliosis in immature patients ⁵⁰ . They believed that the procedure was feasible and safe for the treatment of paralytic scoliosis, with maintenance of spinal mobility and no nonunions. However, long‐term follow‐up (to skeletal maturity) is needed to confirm the efficacy of the technique. McCarthy et al. treated 14 children with scoliosis secondary to spinal cord injury or myelodysplasia with a nonfusion vertebral wedge osteotomy technique ⁵¹ . Their indications for surgery included: (i) diagnosis of progressive thoracolumbar or lumbar paralytic scoliosis secondary to myelodysplasia or spinal cord injury (SCI); (ii) diagnosis of SCI or myelodysplasia (at L3 or above); and (iii) Cobb's angle >35°. After surgery, all patients were placed in a thoracolumbosacral orthosis which was removed when skeletal maturity had been attained. Spinal function scores had improved on 2 years follow‐up. In conclusion, the authors believed that vertebral wedge osteotomy is potentially an effective treatment option for immature paralytic scoliosis.

Though multi‐vertebral wedge osteotomy surgery is effective in correcting scoliosis, the incidence of complications, including neurologic injury, wound infection and pseudoarthrosis, is high ⁵⁰ . It is also unable to correct the kyphosis deformity. The patients had to wear a brace until the bone had healed, which is very inconvenient for them.

Conclusions

So far, the treatment of childhood scoliosis continues to be a problem. Though orthosis is sometimes effective in correcting the deformity, it does not stop progression of scoliosis. Spinal fusion surgery can significantly correct scoliosis deformity, but it inhibits spinal growth. Nonfusion instrumentation techniques offer an alternative and effective way of treating childhood scoliosis, preserving the capacity of the spine to grow and delaying eventual spinal fusion.

However, each nonfusion technique has its advantages and disadvantages. Growing rods are effective in correcting scoliosis deformity while retaining spinal and pulmonary growth potential, but the incidence of postoperative complications is high. Sometimes, spontaneous vertebral fusion occurs in the fixation zone. The patient has to undergo multiple operations and wear a brace after the surgery, which is inconvenient. Staples are effective in correcting scoliosis when the Cobb's angle is not large. They fix the vertebrae by crossing the disc and imposing great stress. Therefore, the staples easily become loose, dislodge and break. VEPTR is useful for expanding the volume of the chest cavity and increasing the pulmonary capacity of children with congenital scoliosis, but there is a high incidence of pulmonary complications. The pedicle screw‐tether technique can modulate spinal growth, which may provide a new nonfusion method for the treatment of scoliosis. However, most published studies are of animal experiments. The safety, indications and long term effectiveness are still unknown. Multi‐vertebral wedge osteotomy surgery corrects scoliosis deformity by a purely anterior approach, but it does not correct kyphosis deformity. In addition, the patient needs to wear a brace after the surgery, which is inconvenient.

In conclusion, nonfusion techniques need further improvement to achieve better correction and to reduce the pain associated with repeated lengthening surgeries and a high incidence of complications.

Disclosure

The authors did not receive any outside funding or grants in support of the research for, or preparation of, this work.