Growing Rod Instrumentation for Early-Onset Scoliosis

 

 

 

DEFINITION

Early-onset scoliosis (EOS) is defined by the diagnosis of scoliosis at or before the age of 5 years. The many etiologies of EOS include the following:

Congenital vertebral or spinal anomalies: for example, vertebral bars, hemivertebrae Neuromuscular diseases: for example, cerebral palsy, spinal dysraphism, muscular dystrophy Syndromes associated with scoliosis: for example, neurofibromatosis

Idiopathic causes

Progressive and severe curves can be associated with deformity, thoracic insufficiency, restrictive pulmonary disease, pulmonary hypertension, cardiac disease, and increased mortality.

 

 

ANATOMY

 

Two periods of increased growth velocity are associated with increased incidence of curve progression. T1 to S1 growth velocity is greatest from birth until the age of 5 years (more than 2 cm per year), and by the age of 5 years, two-thirds of the final sitting height is achieved. Growth rate slows between ages 5 and 10 years (1 cm

per year) before increasing again during puberty and the adolescent growth spurt (1 to 2 cm per year).5, 14

 

The increased spinal growth during the first years of life is paralleled by an increase in thoracic and lung dimensions. Thoracic volume at birth is about 5% of the adult volume; by 5 years of age, it equals 30% of adult volume. A slower rate of thoracic growth occurs from 5 to 10 years of age, by which time it has reached 50% of the adult volume. The final 50% of adult volume is achieved during the adolescent growth spurt from 10 to 15 years of age.

 

PATHOGENESIS

 

The pathogenesis of EOS depends on its etiology.

 

 

Vertebral anomalies cause scoliosis by an imbalance in bone growth, secondary to either an increase in growth on a side associated with a hemivertebrae or growth retardation on the side associated with a vertebral bar.

 

In neuromuscular and central nervous disorders, an imbalance in muscular forces is pathogenic, likely following the Heuter-Volkmann principle that the physeal growth rate is related to the forces it is exposed to, with compression inhibiting growth and tension promoting it.

 

The etiology and pathogenesis of infantile idiopathic scoliosis (IIS) (0 to 3 years of age) is, by definition, unknown, but there is likely a component of genetic susceptibility. The external factors resulting in scoliosis are not yet clearly delineated but may include intrauterine molding as well as infant positioning. The etiology of IIS most likely differs from that of adolescent idiopathic scoliosis (AIS).

NATURAL HISTORY

 

The natural history of EOS also depends on the etiology.

 

 

The natural history of EOS due to IIS is favorable when compared with late-onset scoliosis (LOS). Spontaneous resolution occurs in a large number of patients. Progression of congenital curves depends on the type of anomaly and growth potential.

 

EOS due to neuromuscular etiologies usually follows the natural history of said neuromuscular disease, in addition to specific problems associated with progressive curves in this age group.

 

Regardless of the etiology, progression of scoliosis during the first 5 years of life adversely affects growth as well as pulmonary function.

 

 

A history of EOS is associated with a higher risk of cardiopulmonary decompensation in middle-aged patients, which can lead to disabling and even fatal respiratory failure.

 

PATIENT HISTORY AND PHYSICAL FINDINGS

 

Evaluation of the patient with EOS includes a complete history, including the family history, prenatal history, birth history, and developmental history.

 

 

IIS has been associated with breech presentation and, in boys, with premature birth.

 

Physical examinations includes observation of gait (if patient is ambulatory), respiration, truncal and pelvic balance in the coronal and sagittal planes, cutaneous lesions, and any prominence on Adams forward bending test.

 

Any deficits in motor, sensory, or reflex function, including abdominal reflexes, may indicate central nervous system pathology and should be thoroughly evaluated with advanced diagnostic studies.

 

Flexibility of the curve can be assessed either by the manual application of traction through the cervical spine or by applying a three-point bending force at the apex of the curve.

 

Examination techniques unique for EOS include the thumb excursion test for thoracic expansion and sitting height measurement.

 

IMAGING AND OTHER DIAGNOSTIC STUDIES

 

All patients should have full-length standing anteroposterior (AP) and lateral radiographs (FIG 1) covering the cervical spine to the pelvis, including the entire thorax. For patients who are unable to stand, supine radiographs encompassing the same area should be taken.

 

 

The cervical spine, lumbosacral spine, pelvis, and hips all may need to be studied to elicit whether or not developmental hip

 

P.802

dysplasia or other vertebral anomalies are contributing to the scoliosis.

 

 

 

FIG 1 • A,B. Preoperative AP and lateral radiographs, respectively, of a 5-year-old boy with severe EOS and associated kyphosis in the setting of a congenital diaphragmatic hernia, which progressed to 88 degrees, apex left, despite attempts at bracing. C,D. Preoperative AP and lateral radiographs of a 3-year-old girl with severe EOS that progressed to 76 degrees, apex right, despite attempts at bracing.

 

 

Bolster bending, side-bending, or traction radiographs are necessary to help delineate the degree of flexibility of the curves.

 

The Cobb angle is used to assess initial curve severity and is followed over successive visits to evaluate for curve progression.

 

Spinal height is obtained by measuring the distance from the top of T1 to the top of S1 on the AP view of the spine.

 

Coronal balance is measured by the distance from the center of C7 to a line drawn up from S1.

 

The sagittal balance is measured from the posterior cranial corner of S1 to a line drawn down from the center of C7.

 

All of these measurements should be recorded and compared on successive visits to document any change in curve magnitude or growth of the spine.

 

The rib-vertebral angle difference (RVAD) of Mehta (FIG 2), first described in 1972, measures the amount of rotation at the apex vertebra and has some prognostic value.10

 

 

 

FIG 2 • The RVAD measures the angle of a line drawn perpendicular to the apical thoracic vertebra endplate and a line drawn down the center of the concave and convex ribs. The difference is calculated by subtracting the convex from the concave angle.

 

 

P.803

 

The angles formed by a line perpendicular to the vertebra and a line drawn down the center of the rib is compared between the convex and concave side. If the difference calculated by subtracting the convex angle to the concave angle is 20 degrees or less, there is an 85% to 90% chance the curve will resolve; when there is a difference of 21 degrees or more, it will likely progress.

 

The phase of the rib head is determined by whether or not the head of the convex rib overlaps the vertebral body.

 

 

 

If there is no overlap (phase 1), then the RVAD is calculated as previously mentioned. If there is overlap (phase 2), the risk of progression is high, regardless of RVAD.

 

The space available for the lung (SAL) is calculated by taking the ratio of the distance from the apex of the most cephalad rib to the highest point of the hemidiaphragm of the concave side divided by the convex side.

 

 

A lower SAL points toward a poorer prognosis for lung function.

 

Magnetic resonance imaging (MRI) is recommended to evaluate for spinal cord anomalies in children with rapidly progressive spinal deformities, clinical findings concerning for spinal cord anomalies, or preoperative patients. Before embarking on a repetitive distraction technique such as growing rods, it is particularly

valuable to know whether there is a tethered spinal cord.

 

 

MRI is also used to measure lung volume and assess thoracic architecture when thoracic insufficiency is an issue.

 

In severe congenital deformity, the ribs may spiral around the vertebrae, causing the thoracic volume on

one side to be severely diminished while the other is larger, creating what Campbell4 calls a “windswept thorax.”

 

Computed tomography (CT) scanning is not routinely used, particularly in an era where we are increasingly concerned about high levels of medical radiation in very young patients.

 

DIFFERENTIAL DIAGNOSIS

Congenital vertebral or spinal anomalies Vertebral bars

Hemivertebrae

Syrinx Tethered cord

Neuromuscular diseases Cerebral palsy Myelodysplasia

Muscular dystrophy

Syndromes associated with scoliosis Beel syndrome

Trisomies

IIS

 

 

NONOPERATIVE MANAGEMENT

 

Nonoperative treatment for EOS is indicated in curves that are not expected to progress.

 

 

Patients with a curve of less than 25 degrees and RVAD less than 20 degrees may be followed with serial radiographs every 4 to 6 months to document any progression.

 

Active treatment is warranted in the following:

 

 

Progression greater than 10 degrees

 

Phase 2 rib-vertebral relationship, RVAD greater than 20 degrees, or a Cobb angle greater than 25 degrees in any skeletally immature patient

 

Nonoperative treatment generally starts with casting or bracing.

 

 

Brace treatment should be abandoned in favor of surgical management when unacceptable curve magnitude or progression is seen.

 

SURGICAL MANAGEMENT

 

Surgical treatment of EOS attempts to stop progression of the scoliosis, allowing improvements in growth of the spine, thorax, and lungs.

 

 

Surgery is recommended for progressive curves with a Cobb angle greater than 45 degrees. The age of the patient helps to determine the type of surgery needed.

 

Adolescents and more skeletally mature patients may do well with spine fusions, which stabilize the spine but also stop growth.

 

Younger patients with substantial growth potential suffer from the “crankshaft” phenomenon if fusion is performed early in life from an isolated posterior approach. They suffer from severe growth retardation in height and thoracic volume if fusion is performed using a combined anterior and posterior technique.

 

The growing rod technique for EOS was developed to correct spinal deformity while allowing spinal growth to continue or even enhancing that growth.

 

Preoperative Planning

 

Careful evaluation of radiographic studies allows planning of surgical levels. Typically, the cranial level of the construct includes T2 and extends two or three levels caudal to the end vertebra of the curve.

 

Medical and subspecialty consultations should be obtained before operation if the patient has any history of medical comorbidities.

 

 

Pulmonary function tests may be obtained in children who are able to cooperate if thoracic insufficiency is suspected.

 

Positioning

 

The patient is placed under general anesthesia on the stretcher and then placed on the operating room table in the prone position on two longitudinal chest rolls or tightly rolled blankets.

 

Neurologic monitoring is used during the procedure for neurologically intact patients. Leads should be placed before prone positioning, as should a Foley catheter.

 

Care must be taken to be sure all bony prominences and compressible nerves are well padded.

 

Approach

 

The growing rod technique is performed posteriorly through either a single long midline incision or two smaller incisions cranially and caudally.

 

 

P.804

 

TECHNIQUES

• Placing the Foundation Anchors

Unlike traditional exposures used for thoracolumbar fusions, placement of growing rods begins by only exposing the two anchor clusters, first distally for pedicle screw placement then proximally for hook and sublaminar cable placement.

After radiographically localizing the exposure sites for the two sets of anchor clusters, the distal anchor site is exposed subperiosteally (TECH FIG 1A,B).

Pedicle screws are generally placed bilaterally at two or three adjacent vertebra (TECH FIG 1C), optimal placement of the screws is confirmed radiographically, and the wound is packed and attention turned to

 

placing the proximal anchors.

 

 

 

TECH FIG 1 • A. A single skin incision may be used with subperiosteal exposure of the cranial and caudal foundation sites. The rods and tandem connectors are placed under the fascia in a bed of paraspinal muscle. Pedicle hooks have been used as anchors for the cranial foundation and pedicle screws for the caudal foundation. B. The lateral view shows the straight tandem connector placed in the thoracolumbar region. The trajectory of the pedicle screws can also be seen and varies by patient. C. Close-up of the cranial foundation shows two transverse process hooks and two pedicle hooks spanning two levels in the thoracic region. D. Close-up of the caudal foundation shows four pedicle screws spanning two levels in the lumbar spine. Hooks may also be used for this foundation.

 

 

The proximal anchor site is exposed subperiosteally. Generally, we use bilateral transverse process hooks at the upper instrumented vertebrae and bilateral pedicle hooks at the level below, protected with bilateral sublaminar cables at the level of the pedicle hook (TECH FIG 1D). In the past, pedicle screws were used for proximal rod fixation, but recent literature has revealed the potential for significant

complications.12

 

 

P.805

• Placing the Rods and Rod-to-Rod Connectors

   

  The next step is to place the concave rod. The fascia on the concave side is divided, and a bed is

  created in the paraspinous muscle for the rod. To avoid unwanted autofusion, care is taken to avoid exposure of the posterior elements of the spine (as is done in a normal spine exposure).

   

  Either of two types of connectors may be used: a tandem connector, which houses the cranial and caudal

  rods inside a rectangular box so the ends meet end-to-end, or side-to-side connectors, which allow the rods to overlap.

   

  Tandem connectors are used most commonly. Because these connectors are straight and cannot be contoured, the rods are measured so that they meet inside the tandem connector in the relatively straight thoracolumbar region.

   

  The ends of the rods that fit inside the connector also must be straight. If any contouring is necessary in the region where the cranial and caudal rods meet, closed dual connectors must be used, with an overlap of 2 to 4 inches to allow future lengthening.

   

  After the concave rod is placed, the spine is manually corrected to its maximum amount, then the concave rod is tightened. For the initial implant, it is best to correct the spine and let the rod hold that correction passively rather than try to drive the correction by distracting the rod at a time when the anchor sites are not yet fused.

   

  Next, the convex rod is placed. Generally, it must be contoured with increased kyphosis.

   

  After the rods are placed in the hooks or screws of each foundation, transverse connectors are often placed between the two cranial rods and the two caudal rods, preferably between the points of fixation on each foundation.

   

  If distraction is required, the caudal set screw is tightened, a distractor is implemented in the slot of the tandem connector between the two rods, and the cranial set screw is tightened (TECH FIG 2A).

   

  Similarly, a rod clamp can be used to distract against if a closed dual connector, or even a tandem connector, is used (TECH FIG 2B).

   

  The surgical area is then irrigated, followed by a limited arthrodesis, decorticating and applying autograft bone or other graft extenders between the vertebrae making up each foundation.

   

  Before final closure, AP and lateral radiographs are taken to confirm alignment and proper position of the implants (TECH FIG 2C-F).

   

  The wound is then closed in standard fashion.

   

   

   

  TECH FIG 2 • A. Lengthening can be performed by inserting the distractor between the rods through the slot of the tandem connector. One set screw is loosened, distraction is performed, and the set screw secured. B. Alternatively, a rod clamp can be placed on the rod a few centimeters from the connector and the distractor placed between the rod clamp and the end of the connector. The set screw nearest the rod

  clamp is then loosened, the distractor employed, and the screw retightened. C,D. AP and lateral radiographs, respectively, after the dual growing rod procedure using tandem connectors was performed on the patient shown in FIG 1A,B. (continued)

   

   

  P.806

   

   

   

  TECH FIG 2 • (continued) E,F. AP and lateral radiographs, respectively, after the dual growing rod procedure using side-to-side connectors was performed on the patient in FIG 1C,D. (A,B: From Bagheri R, Akbarnia BA. Pediatric ISOLA (DePuy Spine) instrumentation. In: Kim DH, Vaccaro AR, Fessler RG, eds.

  Spinal Instrumentation: Surgical Techniques. New York: Thieme, 2005:640,642.)

• Lengthening and Exchange

   

  Lengthening of the dual rod construct may be performed as either an inpatient or outpatient procedure with neural monitoring for patients with normal neurologic function.

   

  The connector is located through palpation or fluoroscopy, and a small incision is made over that area where the lengthening is planned.

   

  After dissection of the connector is performed, lengthening similar to that performed during the index procedure is carried out by loosening the set screw (mostly cranial), distracting between the two rods, and then tightening the set screw again.

   

  Lengthening is generally performed every 6 months initially and in children with long constructs and a flexible spine. Over time, the interval is usually lengthened to 8 to 12 months as diminishing returns are noted.

   

  Once further distraction is no longer achievable, final correction and arthrodesis are performed.

• Changing the Connector or Rod

 

Exchange of the tandem connector or the rod may be needed if the amount of lengthening exceeds the initial length of the tandem connector.

 

In such a case, both set screws should be loosened and the tandem connector slid cephalad until full

clearance of the caudal rod is achieved.

The connector can then be removed off the cranial rod, replaced by a longer connector, and slid onto the caudal rod again.

Connectors longer than 70 mm are rarely used to minimize the adverse effect on sagittal balance.

If the needed length exceeds the longest connector or if the longest connector is too long, it is necessary to fashion new rods and remove the old ones.

This entails exposing and removing the tandem connectors, exposing the foundation, removing the rods, and replacing them with longer rods, creating a construct similar to the initial procedure.

Replacement of the cephalad rods is most common.

 

 

Exposure

• Avoid subperiosteal dissection anywhere except the foundation to avoid

premature fusion.

Implants

• Perform a careful radiographic examination or use image-guided navigation if

  pedicle screws are desired.

• Use proper rod contouring to correct both coronal and sagittal deformity. Perhaps the most common pitfall is to attempt to overcorrect a kyphotic deformity. Over time, overcorrection leads to frequent anchor failure, especially at the proximal anchors.

• Tandem connectors are straight and should be placed in the thoracolumbar region, which also is straight.

Lengthening ▪ Do not be too aggressive with lengthenings, especially at the index procedure

and first lengthening, to avoid implant issues.

Indications

• Growing rods may not be indicated in very stiff curves, poor bone quality, older

children with limited growth potential, or children too young to allow internal fixation.

PEARLS AND PITFALLS

 

 

 

 

POSTOPERATIVE CARE

P.807

 

Patients are braced postoperatively with a thoracolumbosacral orthosis for up to 6 months to facilitate fusion of the foundations. Rehabilitation proceeds according to the patient's tolerance and ability.

 

OUTCOMES

The available literature on growing rods for EOS shows it to be a safe and effective method for correcting spinal curves while preserving growth.1, 2, 6, 13, 15

 

 

In a retrospective series of 23 patients with EOS of any etiology, Akbarnia and colleagues2 demonstrated an improvement in the mean curve from 82 degrees preoperatively to 38 degrees after the initial implantation procedure. Patients in this study saw a mean T1-S1 length increase of 1.2 cm per year and an increase in lung space ratio from 0.87 to 1.

A more recent series by Akbarnia and coworkers1 showed Cobb angle improvement from 81.0 degrees to 35.8 degrees after the initial procedure and to 27.7 degrees after final fusion.

Complications, while similar in frequency to other growth preserving spinal procedures,11 are frequent1, 2, 15 and must be anticipated and thoughtfully managed. Complication risk worsens with increased number of surgical procedures and younger age at first surgery.3 However, more frequent lengthenings are also associated with improved curve correction and T1-S1 growth.1

In a study of 140 growing rod patients, Bess and coworkers3 found that 81 (58%) experienced at least one complication. Improved complication profiles were associated with dual rods as opposed to single ones and in submuscular rod placement versus subcutaneous placement.

Another study by Watanabe and colleagues16 of 88 patients with EOS and growing rods found that 50 patients had complications (57%). Complications occurred in 119 of 538 procedures and included 86 implant-related failures (72%), 19 infections (16%), 3 neurologic impairments (3%), and 11 other complications. The most frequent implant-related failure was dislodgement (71%), with 95% of the dislodgements occurring at the proximal foundation.

Also of concern is the psychosocial toll of repeat operative procedures, and physicians treating these patients should be vigilant for possible adverse psychological outcomes in this population.8, 9

Final fusion is usually performed in late childhood or early adolescence and involves a similar number of levels as the growing rod instrumentation, achieves some additional correction in most cases, and has a

similar complication rate to other spinal fusion procedures.7

 

COMPLICATIONS

Wound breakdown Infection

Junctional kyphosis Crankshaft phenomenon Curve progression Implant failure

Patients with more frequent lengthenings have fewer implant problems but more wound problems, whereas patients with less frequent lengthenings have more implant problems and fewer wound complications. Implant complications often can be treated during scheduled lengthenings, but wound infections should be treated urgently.

 

 

ACKNOWLEDGMENTS

 

We would like to acknowledge Victor Hsu and Behrooz Akbarnia for their work in writing the previous edition of

this chapter.

 

REFERENCES

1. Akbarnia BA, Breakwell LM, Marks DS, et al. Dual growing rod technique followed for three to eleven years until final fusion: the effect of frequency of lengthening. Spine 2008;33(9):984-990.

    

    

2. Akbarnia BA, Marks DS, Boachie-Adjei O, et al. Dual growing rod technique for the treatment of progressive early-onset scoliosis: a multicenter study. Spine 2005;30(17 suppl):S46-S57.

    

    

3. Bess S, Akbarnia BA, Thompson GH, et al. Complications of growing rod treatment for early-onset scoliosis: analysis of one hundred and forty patients. J Bone Joint Surg Am 2010;92(15):2533-2543.

    

    

4. Campell 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 2003;85-A(3):399-408.

    

    

5. DiMeglio A. Growth of the spine before age 5 years. J Pediatr Orthop B 1993;1:102-107.

    

    

6. Elsebai HB, Yazici M, Thompson GH, et al. Safety and efficacy of growing rod techniques for pediatric congenital spinal deformities. J Pediatr Orthop 2011;31(1):1-5.

    

    

7. Flynn JM, Matsumoto H, Torres F, et al. Psychological dysfunction in children who require repetitive surgery for early onset scoliosis. J Pediatr Orthop 2012;32(6):594-599.

    

    

8. Flynn JM, Tomlinson LA, Pawelek J, et al. Growing-rod graduates: lessons learned from ninety-nine patients who completed lengthening. J Bone Joint Surg Am 2013;95(19):1745-1750.

    

    

9. Matsumoto H, Williams BA, Corona J, et al. Psychosocial effects of repetitive surgeries in children with early-onset scoliosis: are we putting them at risk? J Pediatr Orthop 2014;34(2):172-178.

    

    

10. Mehta MH. The rib-vertebra angle in the early diagnosis between resolving and progressive infantile scoliosis. J Bone Joint Surg Br 1972;54:230-243.

     

     

11. Sankar WN, Acevedo DC, Skaggs DL. Comparison of complications among growing spinal implants. Spine 2010;35(23):2091-2906.

     

     

12. Skaggs KF, Brasher AE, Johnston CE, et al. Upper thoracic pedicle screw loss of fixation causing spinal cord injury: a review of the literature and multicenter case series. J Pediatr Orthop 2013;33(1):75-79.

     

     

13. Thomspson GH, Akbarnia BA, Kostial P, et al. Comparison of single and dual growing rod techniques followed through definitive surgery: a preliminary study. Spine 2005;30:2039-2044.

     

     

14. Tis JE, Karlin LI, Akbarnia BA, et al. Early onset scoliosis: modern treatment and results. J Pediatr Orthop 2012;32:647-657.

     

     

15. Wang S, Zhang J, Qiu G, et al. Dual growing rods technique for congenital scoliosis: more than 2 years outcomes: preliminary results of a single center. Spine 2012;37(26):E1639-E1644.

     

     

16. Watanabe K, Uno K, Suzuki T, et al. Risk factors for complications associated with growing-rod surgery for early-onset scoliosis. Spine 2013;38(8):E464-E468.