Journal Logo

Review Article

Use of All-pedicle-screw Constructs in the Treatment of Adolescent Idiopathic Scoliosis

Cuartas, Esteban MD; Rasouli, Alexandre MD; O'Brien, Michael MD; Shufflebarger, Harry L. MD

Author Information
Journal of the American Academy of Orthopaedic Surgeons: September 2009 - Volume 17 - Issue 9 - p 550-561
  • Free


Adolescent idiopathic scoliosis (AIS) is a common and potentially severe musculoskeletal disorder. It is characterized by spinal deformity in the coronal, sagittal, and axial planes (Figure 1). Most frequently, patients present with a right thoracic curve associated with thoracic hypokyphosis, along with rotation. Scoliosis is defined by a coronal curve of >10°. Using this measurement, epidemiologic studies show that 2% to 3% of the adolescent population is affected; however, <10% of these persons require treatment.1 Approximately 80% of AIS curves >20° occur in girls.1

Figure 1
Figure 1:
Anterior (A) and posterior (B) three-dimensional CT scans demonstrating coronal, sagittal, and rotational deformities present in patients with adolescent idiopathic scoliosis (AIS). Vertebral body wedging around the apex is evident. C, Axial view of a periapical vertebra in a patient with AIS, illustrating different pedicle morphologies and lung volumes, as well as vertebral rotation.

The etiology of AIS appears to be genetic,2 but the method of curve production is unknown. However, much is known about the natural history of AIS. Curve growth is related to curve size, peak growth rate, age at onset, sex, and curve type. Larger curve size (ie, >50°) has been associated with chronic back pain, disability, poor cosmesis, psychosocial issues, respiratory dysfunction, and increased mortality.1,3 Surgical indications for the patient with AIS include curve >50°, curve >40° to 45° in a skeletally immature patient, curve progression despite bracing, and deformity that is unacceptable to the patient.3

The King classification of AIS has several shortcomings. It describes only curves with a significant thoracic component, thereby omitting curves that are primarily thoracolumbar or lumbar. Additionally, it describes only the coronal dimension of curves, neglecting variations in sagittal alignment. Segmental instrumentation of King type II curves (ie, double curve, with the thoracic curve being more structural) has led to coronal decompensation in some patients. This outcome suggests that this is a nonhomogeneous group and points to the need for an improved classification. The King classification has been shown to have low interand intraobserver reliability.4,5

Lenke et al4 developed a two-dimensional classification of AIS based on coronal and sagittal radiographic views. This classification also takes into account curve type. The Lenke classification was also designed to assist in the selection of fusion levels using current methods of segmental instrumentation, much as the King classification was designed to aid in determining the extent of Harrington rod instrumentation.5,6 The Lenke classification has been shown to be more comprehensive, reliable, and reproducible than the King classification, allowing useful comparisons of different treatment methods and groups.4

The treatment of AIS has evolved in the past decade with the implementation of all-pedicle-screw constructs. The previous standard of treatment involved hook constructs. Use of all-pedicle-screw constructs was initially controversial because of safety concerns. However, the use of pedicle screws in the thoracic region has been shown to be safe and biomechanically advantageous. Specifically, this three-column method of fixation in patients with AIS has led to improved deformity correction and shorter fusions compared with hooks.7–9

The goals of surgical treatment in the patient with AIS are to obtain significant curve correction and to prevent progression or recurrence through the induction of spinal arthrodesis. Innovations in spinal instrumentation have led to improved correction of spinal deformity, lower nonunion rates, and early postoperative patient mobilization. Even with these improvements, attention to the basic tenets of spinal arthrodesis, including facetectomy, decortication, and bone grafting, is crucial in the attempt to attain satisfactory longterm results.

Clinical Safety of Thoracic Pedicle Screws

The shift from external to internal fixation in the management of AIS came in the 1960s with the introduction of Harrington distraction rods. Although this innovation led to better curve correction and fusion rates, such instrumentation was not adequate to control the sagittal plane. The advent of segmental instrumentation methods, including Luque wires and Cotrel-Dubousset hooks, made it possible to achieve sagittal control in addition to coronal correction. This instrumentation also allowed more rapid postoperative mobilization. Given its reliability, ease of placement, and relative safety, hook instrumentation quickly became the preferred treatment method in the 1980s. Thoracic pedicle screws appear to represent the next step in the evolution of posterior segmental instrumentation.

Pedicle screws were introduced in the 1950s by Boucher10 and popularized by Roy-Camille et al11 in the 1960s. Initial reports questioned the safety of pedicle screws in the thoracic spine.12,13 One cadaver study found a high percentage of pedicle wall violation and recommended the use of pedicle screws only in special circumstances.12 Specifically, the authors concluded that “pedicle-screw fixation is not of great benefit in the treatment of the deformed, rotated vertebrae seen in patients who have thoracic scoliosis.” However, several large series have established the relative safety of pedicle screws in the thoracic spine in a variety of settings, including trauma, infection, tumor, and deformity.7,9,14–28

Despite the significant rate of pedicle wall breaches documented on postoperative CT scans, the risk of serious injury appears to be small. For instance, although the risk of cortical breach has been reported to be as high as 43%,17 the risk of neurologic or vascular injury has been reported to be zero to 1.2%.15–17 Medial breach has been reported to occur with 1.7%21 to 14%17 of screws placed in the thoracic spine. This discrepancy between the rate of breach and the occurrence of injury led Gertzbein and Robbins15 to hypothesize that a 4-mm “safe zone,” composed of a 2-mm epidural space and a 2-mm subarachnoid space, exists around the pedicle wall. Belmont et al17 reported no neurologic deficits with 35 screws with medial breach of <4 mm. However, cadaver studies have shown that there is virtually no epidural space between the pedicles and the dura at the thoracic level.29,30 Thus, accurate placement is essential.

In a comparative study, Polly et al31 found that laminar hooks have canal intrusion equivalent to that found in pedicle screw breaches of 2 to 3 mm. Several surgeons consider a breach of <2 mm to be acceptable screw positioning. In their series involving 279 thoracic pedicle screws, Belmont et al17 found a 43% incidence of cortical breach, 68% of which were lateral. Ninety-nine percent of screws exhibited medial encroachment of <2 mm, and all were deemed acceptable.

Suk and colleagues7,16,32 pioneered the use of all-pedicle-screw constructs in AIS and found this technique to be safe and effective. In a review of 4,604 thoracic pedicle screws placed in 462 patients, 67 screw malpositions (1.5%) were reported in 48 patients (10.4%).16 The screw malpositions were inferior in 33, lateral in 18, superior in 12, and medial in 4. Four patients had neurologic complications, including three dural tears and one instance of transient paraparesis. However, there were no neurologic or visceral complications that adversely affected the long-term outcome.

More recently, it has been shown that thoracic pedicle screws can be safely placed even in cases of extreme deformity (ie, curve >90°).18 Kuklo et al19 reviewed 20 such patients with an average Cobb angle of 100.2° and found a 96.3% rate of accurate screw placement (ie, <2-mm breach) on postoperative CT scans. No neurologic complications were reported.

Most of these published series represent the experience of highly specialized spine surgeons working at high-volume centers. The hazards of pedicle screw fixation in the thoracic spine are real, and vascular, neurologic, and visceral injuries have been documented in several case reports.33–37

Pedicle Anatomy

Central to any discussion of the use of thoracic pedicle screws and their safety is a thorough understanding of normal pedicle anatomy13,32,38–41 (Figure 2). The morphometry of thoracic and lumbar pedicles has been well documented. The dimensions vary widely according to patient age, size of vertebra, and spinal level. For instance, pedicle width may range from 2.5 to 12 mm, and pedicle height, from 9.6 to 16.9 mm. However, some general trends have been observed. Thoracic pedicles are significantly smaller than their counterparts in the lumbar spine. Thoracic pedicles are ovoid in cross-section, with a pedicle height that is, on average, almost twice the width. Pedicle dimensions are generally smallest around T4-T6, slightly larger cranially, and more markedly enlarged caudally.

Figure 2
Figure 2:
A, Axial CT scan demonstrating normal adult thoracic pedicle anatomy and notations for determining transverse pedicle angle (a), pedicle width (b), and pedicle chord length (c). B, Sagittal CT scan demonstrating normal adult thoracic pedicle anatomy and notations for determining sagittal inclination angle (a) and pedicle height (b). These values vary according to thoracic level and individual. C, Axial CT scan of periapical morphology in a patient with adolescent idiopathic scoliosis. This artist-enhanced image shows the spinal cord (sc) position to be closer to the concavity, cerebrospinal fluid (CSF) within the dural sac, and the presence of epidural fat (*). Note that the pedicle is smaller on the concave side.

Zindrick et al39 documented the dimensions of 2,905 adult pedicles by taking measurements from spinal CT scans. Pedicle width varied from 5.0 to 10.0 mm (mean, 7.9 mm) at T1, 2.5 to 7.0mm (mean, 4.7 mm) at T4, and 3.0 to 11.0 mm (mean, 7.1 mm) at T12. Pedicle chord length (ie, cortex-tocortex distance along the axis of the pedicle) varied from 26 to 52 mm (mean, 36.9 mm) at T1, 33 to 49 mm (mean, 38.5 mm) at T4, and 23 to 61 mm (mean, 38.6 mm) at T12. Sagittal inclination varied from 5° to 18° in the thoracic spine. In the upper pedicles, the angle of entry was medially directed approximately 25°, progressing to nearly straightforward in the lower pedicles.

These trends determine screw trajectory as well as diameter and length. Preoperative imaging should be carefully studied to determine the feasibility of pedicle screw placement in a given patient. Some patients with AIS may have pedicle sizes and orientations markedly different from the dimensions noted by Zindrick et al.39

Cadaver studies have shown that the medial pedicle wall is two to three times thicker than the lateral wall and thus offers some measure of protection against breach.40 In a study of six fresh-frozen cadaver specimens, Misenhimer et al41 found that with increasing screw size, fracture occurred more frequently in the lateral wall than in the medial wall (72% versus 28%, respectively).

Several authors have studied pedicle morphology in AIS.20,42 In the patient with AIS, the pedicles on the concave side are typically smaller than those on the convex side. Pedicle angle in the coronal and sagittal planes appears to be unchanged, however. In the axial plane, the pedicles near the apex can have a windswept appearance. The dural sac is draped on and closer to the concaveside pedicles. These features can add to the difficulty of placing screws in periapical concave pedicles.

Biomechanics of Pedicle Screw Constructs

Theoretically, pedicle screw fixation provides better three-column fixation than does posterior columnonly fixation such as hooks or wiring. In fact, pedicle screws have been shown to be biomechanically superior to hooks in both laboratory and clinical settings.16,43 Liljenqvist et al20 found the pullout strength of pedicle screws to be 1.3 to 1.6 times that of pedicle hooks. Deviren et al44 studied various pedicle screw configurations in a cadaver model and found that increasing levels of iatrogenic destabilization required greater numbers of pedicle screws within the construct to achieve a similar level of stability throughout. Two pedicle screws at each level conferred the highest degree of stability. Torsional rigidity has been shown to increase an average of 26% with one crosslink and 44% with two.45

The most important variable for pullout strength is screw outer diameter. The wider the screw outer diameter, the greater the amount of pedicle wall purchase.41 Recommended screw size relative to pedicle size is based on patient age. In adults, screw width equal to 80% of the outer diameter of the pedicle is recommended. Because of the viscoelastic properties of younger bone, in adolescents, screw width may be ≤115% of the pedicle width.42

Cadaver studies have demonstrated that several factors are involved in determining the optimal degree of vertebral body penetration. One study found no difference in pullout strength between 50% and 100% vertebral body penetration at the lumbar level.46 However, with adequate (>70%) fill of the pedicle, increasing penetration of the vertebral body leads to increased pullout strength. This finding must be weighed against the risk of anterior penetration, which is heightened because of the anterior curvature of the vertebral body. To avoid this complication, ≤80% penetration should be obtained on the lateral view.46

Insertional technique affects pedicle insertional torque. In one study, undertapping by 1 mm resulted in a 93% increase in maximal insertional torque.47 From a lateral view, two screw trajectories are possible, one parallel to the end plates (ie, straightforward) and one in line with the pedicle axis (ie, anatomic). A straightforward trajectory has been shown to have a 39% higher maximal insertional torque than an anatomic trajectory.48 The in-out-in technique, in which the trajectory breaches the lateral pedicle wall and then penetrates the vertebral body farther down, has been shown to be feasible, but it is only about 75% as strong as the intrapedicular approach.49 These two techniques represent salvage options in the event of a breach.

Surgical Techniques for Placement of Thoracic Pedicle Screws

Despite the established safety record of thoracic pedicle screws, their placement remains a potentially dangerous undertaking. For best results, a thorough anatomic knowledge and strict adherence to a structured approach are required. Given the wide variations in pedicle morphology, a detailed analysis of each patient's preoperative imaging studies is essential.

Anatomic landmarks and fluoroscopic guidance are helpful in determining the starting points and proper screw trajectory. Fluoroscopy and neuromonitoring techniques are useful in confirming adequate screw placement. Kim et al21 described a surgical approach that involves relying on anatomic landmarks only (Figure 3, A). After exposure of the superior facet using an osteotome, the starting points are found according to their level and anatomic landmarks (transverse process, lamina, superior articular facet). The starting points are marked with a burr. A slightly curved, blunt, 2-mm probe (ie, gearshift) is then placed, with the curve initially facing laterally, so as to avoid medial breach. After reaching a depth of about 20 mm, the probe is redirected medially to obtain purchase in the vertebral body. The path is gently probed before and after tapping to evaluate for breaches, and measurements are taken to determine screw length. Kim et al21 recommend advancing 20 to 25 mm in the upper thoracic spine, 25 to 30 mm in the mid thoracic spine, and 30 to 35 mm in the lower thoracic spine. This method was found to be safe, with a reported 6.2% rate of cortical perforation with no neurologic injuries. One of the pitfalls of this method is its reliance on surgeon experience and detailed anatomic knowledge. Others have recommended using the inferolateral aspect of the facet as a constant landmark, with the starting point located approximately 2 mm medial to the junction of the transverse process and the lateral aspect of the superior facet.22,50,51

Figure 3
Figure 3:
A, Three-dimensional CT scan showing starting points (*) for thoracic pedicle screws. B, Fluoroscopy may be used to determine the starting point. To obtain a true PA view of each vertebral body, the image must be analyzed and several adjustments made to the C-arm in two planes. Left, The initial image is off in the cephalocaudad and mediolateral axes. Pedicle outlines are asymmetric, and the vertebral body outline is indistinct. Middle, With the cephalocaudad axis adjusted, the end plates become distinct. Right, Once the mediolateral axis is adjusted and a true PA view is obtained, the pedicles become distinct and symmetric with respect to the vertebral bodies. The starting point (x) is at the midpoint of the lateral outline of the pedicle.

Fluoroscopy may be used in addition to anatomic landmarks to ensure accurate screw placement.23,24 The key to this method is obtaining a true PA or AP view of each vertebra to be instrumented. This requires a skilled fluoroscopy operator because each level requires a different degree of rotation and inclination (Figure 3, B). On a true PA or AP view, the starting point that was determined according to anatomic landmarks should coincide with the middle of the lateral outline of the pedicle. As the probe or drill is advanced under fluoroscopic visualization, it is redirected as necessary to maintain the tip within the pedicle outline. Once a depth of 20 to 25 mm has been reached, it can be assumed that the tip of the probe has reached the vertebral body. At that point, the tip can safely be directed across the medial pedicle wall outline because it is anterior to the neural canal. The tip of the probe or screw should never cross the midline of the vertebral body; this would indicate a medial breach. Although it is cumbersome, fluoroscopic guidance increases the accuracy and safety of screw placement; thus, it is the preferred method of the senior author (H.L.S.). One drawback to this approach is the significant x-ray exposure to the patient and the surgical staff. Adequate protective equipment is recommended for the surgeon, including use of a thyroid shield and radioprotective gloves.52

Another method of ensuring proper pedicle screw placement and reducing the risk of medial breach involves performing a limited laminotomy to palpate the medial wall of the pedicle. This time-consuming approach is useful in salvage situations.53

Triggered electromyography is useful in confirming screw placement. This modality was initially described for evaluating the placement of lumbosacral pedicle screws.54,55 In the thoracic spine, electromyography is performed by applying current to the pedicle screw and recording signals at the rectus abdominis muscle (T7-T12). For higher levels, the signals are recorded at the intercostals. Shi et al56 found a 97.5% negative predictive value for pedicle breach with thresholds >11 mA. Impedance of <11 mA had a 50% positive predictive value for cortical breach on postoperative CT imaging. Potential pedicle breach should be assessed with fluoroscopic imaging and palpation of the pedicle walls with a ball-tip probe.

Adjunctive Posterior Procedures

Several adjunctive posterior procedures have been described that can function to release the posterior column and thus increase deformity correction. These include, with increasing levels of invasiveness and destabilization, facetectomy, removal of interspinous ligaments, thoracoplasty, Ponte osteotomy, and posterior vertebral column resection.

Facetectomy serves to expose the starting point for pedicle screws and prepare the bony surfaces for fusion, as well as to release the posterior facet joints. Similarly, thoracoplasty serves multiple purposes. It can be done to correct the cosmetic “rib hump” and to provide a source of autogenous bone graft. In patients with rigid spinal deformity, it can be done to release the spinal column from the thoracic cage. Some authors have found concave rib osteotomies to be of use in patients with rigid curvatures.57

The use of thoracoplasty is somewhat controversial. Kim et al25 found that in posterior spinal fusions, the use of thoracoplasty had a negative impact on percent predicted ventilatory values (ie, forced vital capacity [FVC], forced expiratory volume [FEV]). Percent predicted FVC and FEV were decreased by 10% and 8%, respectively, at 6.9 years (P < 0.001). A matched group of patients who did not undergo thoracoplasty had no change in percent predicted values. The clinical significance of this small decrease in percent predicted values is unknown. However, the authors suggested that thoracic cage violation should be avoided when possible.

Some authors have suggested that the increased rotational control provided by pedicle screws has reduced the need for thoracoplasty.58 However, Suk et al58 recently compared results with and without thoracoplasty in the setting of pedicle screw constructs. These authors found that thoracoplasty provided superior rib hump correction and patient satisfaction scores and no significant pulmonary compromise. Despite the controversy regarding its use, thoracoplasty remains a powerful tool for cosmetic correction, and it is an important component of the surgeon's armamentarium.

Ponte osteotomy was originally described for correction of thoracic kyphotic deformity. This technique also can serve to release the apical, more rigid portion of the curve. Ponte osteotomy consists of resection of the interspinous ligaments, ligamentum flavum, and facet joints and part of the lamina.59 The drawback to this procedure is the additional neurologic risk and blood loss sustained as a result of opening the neural canal. Because of these problems, some authors reserve this procedure for stiffer curves. Shufflebarger et al60 described a similar wide posterior release for use in the lumbar and thoracolumbar spine, specifically done to increase mobility and lordosis. This release led to improved coronal correction, from 64% to 76% (P < 0.005), and sagittal correction from 6° to 14° at 2-year follow-up.

Posterior vertebral column resection represents the extreme of posteriorly based releases. It is a technically demanding and lengthy procedure in which all or part of the anterior column is resected from a posterior approach to assist in deformity correction.16,26 This procedure has been used only in cases of severe and/or focal fixed deformities, which are not typical in AIS. Pedicle screws are used in these challenging cases because they provide stable fixation after the spinal column has been completely released, thus protecting the spinal cord from injury. Posterior vertebral column resection is exceptionally rare in the treatment of AIS. We therefore mention the procedure only to describe it as the end of the spectrum of posteriorly based spinal releases and because pedicle screw fixation plays a key role in this fixation technique.

These adjunctive release procedures allow increased biomechanical power with the pedicle screw constructs. Additionally, they allow the surgeon to obtain deformity correction through a single posterior approach, which is beneficial in terms of decreased morbidity and operating time.

Curve Correction

Once the pedicle screws are in place and posterior releases have been performed to the surgeon's satisfaction, the curve correction is undertaken. This is a challenging part of the procedure. Inadequate technique can lead to pedicle fracture, screw pullout, or postoperative decompensation.

A rod rotation maneuver is useful in the typical case of thoracic scoliosis combined with hypokyphosis. A rod is contoured to the desired sagittal profile, then placed into the concavity of the curve. Once the rod is engaged into all of the implants, a slow and careful 90° counterclockwise rotation is performed. This maneuver employs a multipurchase construct to gradually translate the coronal deformity into the desired sagittal profile. Once the rod rotation is complete, a second precontoured rod is placed on the remaining side.

Some systems include reduction screws with long tabs, which allow the rod to engage with the screws while still at a significant distance from the spine. With these reduction screws, the rod can be slowly brought to the screw head using the mechanical advantage of the set screw. The tabs can then be broken off or removed.

Another method consists of placing a contoured rod into the concave side of the curve. After all screws are engaged, a controlled in situ correction with in situ rod benders is performed. Several passes are necessary. Additional correction may be obtained by using compression-distraction maneuvers to adjust coronal and sagittal alignment.21

In 1999, Lee et al27 described a maneuver involving direct vertebral rotation (DVR) with pedicle screws (Figure 4). This technique specifically addresses the rotational component of the curve with the placement of derotating posts over the periapical screws bilaterally. Pressure is applied to the screws placed on the convex side and to the thoracic prominence, thus derotating the vertebra. The vertebrae are then locked into the derotated position by placing the set screws through the posts. The authors cautioned that when a compensatory lumbar curve crosses the midline (ie, center sacral vertebral line), the lowermost one or two screws should be rotated in the opposite direction to decrease the amount of lumbar rotation. This method resulted in a 42.5% correction of the apical vertebral rotation, as measured on CT. Additional unexpected benefits of the DVR technique included improved coronal correction in the thoracic curve (79.6% with DVR versus 68.9% with simple rod derotation) as well as in the noninstrumented compensatory lumbar curve (80.5% in the DVR group versus 62.2% in the simple rod derotation group).

Figure 4
Figure 4:
A, Illustration demonstrating the axial view of the direct vertebral rotation technique. Long tubes acting as levers (L) are attached to the pedicle screws (P). The levers are moved simultaneously (left) to achieve the desired amount of vertebral body rotation (right). Photographs of a spine model before (B) and after (C) direct vertebral rotation.

As mentioned previously, the increased rotational correction afforded by DVR has obviated the need for thoracoplasty in some centers.58


Correction of coronal plane deformity is considerably better with posterior pedicle screw constructs than with hook constructs (Figure 5). In 1995, Suk et al7 compared hook constructs with screw constructs. They reported major curve correction of 72% with segmental screws compared with 55% with hooks. Loss of correction was also less with screws (1% versus 6%, respectively). Correction in the second, compensatory, curve was 70% with segmental pedicle screws and 57% with hooks. In 2004, Kim et al8 reviewed 52 patients treated with either segmental pedicle screws or hooks and reported similar results. At 2-year follow-up, 76% average major curve correction was achieved in patients treated with screws compared with 50% correction with hooks. Blood loss and surgical time were not significantly different. Sagittal correction and maintenance of correction were also higher in this series. Kim et al8 also documented a significant improvement in pulmonary function in patients treated with pedicle screws rather than hooks (P < 0.05).

Figure 5
Figure 5:
PA (A) and lateral (B) preoperative radiographs of a 13-year-old girl with Lenke type 6C adolescent idiopathic scoliosis and hypokyphosis. PA (C) and lateral (D) radiographs following treatment with a pedicle screw construct, showing successful treatment of the coronal deformity (83% lumbar Cobb angle correction and 77% thoracic Cobb angle correction) and an improved sagittal profile (thoracic kyphosis increased from 0° to 14°). Also note the thoracoplasty on the right side in panel C.

The effect of pedicle screw constructs on sagittal plane correction is not as clear. Patients with AIS often display hypokyphosis in the thoracic region. Ideally, surgery would address this. Kim and colleagues8,28 reported a decrease in thoracic kyphosis in patients treated with pedicle screw constructs. However, Suk et al61 found better improvement of hypokyphosis with pedicle screws than with hooks. Recently, Clement et al62 reported consistently improved kyphosis in hypokyphotic patients treated with pedicle screw constructs. These authors also found sagittal correction to be technique-dependent. In their study, a two-rod simultaneous translation technique was more effective than a cantilever technique. Improved sagittal correction may be surgeon-, positioning-, or techniquedependent. Further research is needed on the effect of instrumentation and fixation type on hypokyphosis.

The Lenke classification is useful in determining which curves should be fused. The exact determination of the upper and lower instrumented vertebra in each curve type is a more controversial issue that is beyond the scope of this article.9 However, several authors have reported that the improved corrective power of pedicle screws has allowed for slightly shorter posterior fusions. Kim et al8 found that, compared with hooks, pedicle constructs saved an average of 0.8 levels from the distal vertebra. Suk et al9 reviewed their experience with single thoracic curves and found that, instead of ending the fusion at the stable vertebra (ie, the vertebra most closely bisected by the center sacral vertebral line), the fusion could be stopped at the neutrally rotated vertebra (N) or one level above it (N-1). Although these differences may seem small, sparing one level in the lower lumbar spine may have important long-term benefits with regard to adjacent segment degeneration.

Another major benefit of pedicle screw constructs was described by Luhmann et al.18 In a review of curves between 70° and 100° in patients with AIS, the authors found coronal correction to be equivalent regardless whether anterior release and posterior fusion or posterioronly fusion with screws was performed. Similar results were found with sagittal correction. This study seems to indicate that an anterior release with its concomitant morbidity can be successfully avoided in some patients with severe curves. However, it is important to note that some surgeons still prefer to perform anterior releases, which can be done thoracoscopically, in curves between 70° and 100°.

There is experimental63 as well as clinical16,64,65 evidence to suggest that all-pedicle-screw constructs may be stiff enough to prevent the crankshaft phenomenon (ie, postoperative worsening of the curvature resulting from continued anterior spinal growth after posterior fusion).66 Traditionally, this complication has been prevented by performing surgery with dual anterior and posterior approaches in at-risk patients. However, a posterior-only approach greatly diminishes morbidity in these patients.16,63–65

Some authors have promoted the use of hybrid constructs, with hooks at the upper portions and pedicle screws at the bottom of the instrumentation.28 There are few published comparisons of hybrid constructs versus all-hook constructs. However, there is some indication that hybrid constructs provide intermediate results in terms of curve correction. Kim et al28 found significantly better average major curve correction with all-pedicle-screw constructs than with hybrid constructs (70% versus 56%, respectively; P = 0.001).


One of the arguments against the use of all-pedicle-screw constructs is the higher implant cost. Kim et al8 found average implant cost with screws (average number of fixation points, 17.1; $14,200 US) to be significantly higher than with hooks (average number of fixation points, 11.8; $9,228 US). Despite these higher costs, they found no statistically significant differences in SRS-24 scores at 2-year follow-up. One potential cost benefit of pedicle screw constructs is the lower rate of loss of fixation.16 In cases of severe deformity, the higher implant cost is offset by the avoidance of dual anterior and posterior approaches. The impact of improved corrections and shorter fusions on long-term outcomes is unknown. The theoretic benefits with this technique include decreased adjacent segment degeneration and improved health-related outcome measures. Long-term studies are needed to prove those projected outcomes.


All-pedicle-screw constructs are popular in the management of AIS, and their clinical safety has been demonstrated. Accurate placement of pedicle screws requires meticulous attention to technique. All-pedicle-screw constructs have been associated with improved correction in all three planes of deformity. In patients with severe deformity, such constructs can obviate the need for anterior approaches. The mid- and long-term impact of these improved corrections is unknown. Short-term, patient-based outcome measures indicate results similar to those achieved with hook constructs. Evidence suggests that pulmonary function may be improved with all-pedicle-screw constructs. It is thought that these constructs may offer improved outcomes in the long term, decreased adjacent segment degeneration, and a decreased need for revision surgery. It is also likely that more severe deformities benefit proportionally more from pedicle screw instrumentation.

Implant costs are higher with allpedicle-screw constructs than with hooks. No cost-effectiveness analyses are available.67 A recent evidencebased review found that, to date, all studies favoring the use of pedicle screw constructs in AIS have been level III evidence, that is, based on expert opinion, case series, and biomechanical studies.67 Despite the lack of level I evidence, there is a growing consensus that the increased deformity correction obtained with pedicle screw constructs is advantageous.


Evidence-based Medicine: Levels of evidence are listed in the table of contents. In this article, no level I or II studies are cited. Level III studies include references 7, 8, 25, 28, 32, 56–58, 61, and 62. References 14–24, 26, 27, 52, 55, 63, and 66 are level IV studies. References 1, 3–6, 9, and 59 are level V studies.

Citation numbers printed in bold type indicate references published within the past 5 years.

1. Lonstein JE: Scoliosis: Surgical versus nonsurgical treatment. Clin Orthop Relat Res 2006;443:248-259.
2. Ogilvie JW, Braun J, Argyle V, Nelson L, Meade M, Ward K: The search for idiopathic scoliosis genes. Spine 2006;31: 679-681.
3. Bridwell KH: Surgical treatment of idiopathic adolescent scoliosis. Spine 1999;24:2607-2616.
4. Lenke LG, Betz RR, Harms J, et al: Adolescent idiopathic scoliosis: A new classification to determine extent of spinal arthrodesis. J Bone Joint Surg Am 2001;83:1169-1181.
5. Lenke LG, Edwards CC II, Bridwell KH: The Lenke classification of adolescent idiopathic scoliosis: How it organizes curve patterns as a template to perform selective fusions of the spine. Spine 2003; 28:S199-S207.
6. Puno RM, An KC, Puno RL, Jacob A, Chung SS: Treatment recommendations for idiopathic scoliosis: An assessment of the Lenke classification. Spine 2003;28: 2102-2115.
7. Suk SI, Lee CK, Kim WJ, Chung YJ, Park YB: Segmental pedicle screw fixation in the treatment of thoracic idiopathic scoliosis. Spine 1995;20:1399-1405.
8. Kim YJ, Lenke LG, Cho SK, Bridwell KH, Sides B, Blanke K: Comparative analysis of pedicle screw versus hook instrumentation in posterior spinal fusion of adolescent idiopathic scoliosis. Spine 2004;29:2040-2048.
9. Suk SI, Lee SM, Chung ER, Kim JH, Kim WJ, Sohn HM: Determination of distal fusion level with segmental pedicle screw fixation in single thoracic idiopathic scoliosis. Spine 2003;28:484-491.
10. Boucher HH: A method of spinal fusion. J Bone Joint Surg Br 1959;41:248-259.
11. Roy-Camille R, Saillant G, Mazel C: Internal fixation of the lumbar spine with pedicle screw plating. Clin Orthop Relat Res 1986;203:7-17.
12. Vaccaro AR, Rizzolo SJ, Balderston RA, et al: Placement of pedicle screws in the thoracic spine: Part II. An anatomic and radiographic assessment. J Bone Joint Surg Am 1995;77:1200-1206.
13. Vaccaro AR, Rizzolo SJ, Allardyce TJ, et al: Placement of pedicle screws in the thoracic spine: Part I. Morphometric analysis of the thoracic vertebrae. J Bone Joint Surg Am 1995;77:1193-1199.
14. Lonstein JE, Denis F, Perra JH, Pinto MR, Smith MD, Winter RB: Complications associated with pedicle screws. J Bone Joint Surg Am 1999;81: 1519-1528.
15. Gertzbein SD, Robbins SE: Accuracy of pedicular screw placement in vivo. Spine 1990;15:11-14.
16. Suk SI, Kim WJ, Lee SM, Kim JH, Chung ER: Thoracic pedicle screw fixation in spinal deformities: Are they really safe? Spine 2001;26:2049-2057.
17. Belmont PJ Jr, Klemme WR, Dhawan A, Polly DW Jr: In vivo accuracy of thoracic pedicle screws. Spine 2001;26:2340-2346.
18. Luhmann SJ, Lenke LG, Kim YJ, Bridwell KH, Schootman M: Thoracic adolescent idiopathic scoliosis curves between 70 degrees and 100 degrees: Is anterior release necessary? Spine 2005; 30:2061-2067.
19. Kuklo TR, Lenke LG, O'Brien MF, Lehman RA Jr, Polly DW Jr, Schroeder TM: Accuracy and efficacy of thoracic pedicle screws in curves more than 90 degrees. Spine 2005;30:222-226.
20. Liljenqvist UR, Halm HF, Link TM: Pedicle screw instrumentation of the thoracic spine in idiopathic scoliosis. Spine 1997;22:2239-2245.
21. Kim YJ, Lenke LG, Bridwell KH, Cho YS, Riew KD: Free hand pedicle screw placement in the thoracic spine: Is it safe? Spine 2004;29:333-342.
22. Elliott MJ, Slakey CJ: Thoracic pedicle screw placement: Analysis using anatomical landmarks without image guidance. J Pediatr Orthop 2007;27: 582-586.
23. Carbone JJ, Tortolani PJ, Quartararo LG: Fluoroscopically assisted pedicle screw fixation for thoracic and thoracolumbar injuries: Technique and short-term complications. Spine 2003; 28:91-97.
24. Lee CS, Kim MJ, Ahn YJ, Kim YT, Jeong KI, Lee DH: Thoracic pedicle screw insertion in scoliosis using posteroanterior C-arm rotation method. J Spinal Disord Tech 2007;20:66-71.
25. Kim YJ, Lenke LG, Bridwell KH, Kim KL, Steger-May K: Pulmonary function in adolescent idiopathic scoliosis relative to the surgical procedure. J Bone Joint Surg Am 2005;87:1534-1541.
26. Suk SI, Kim JH, Kim WJ, Lee SM, Chung ER, Nah KH: Posterior vertebral column resection for severe spinal deformities. Spine 2002;27:2374-2382.
27. Lee SM, Suk SI, Chung ER: Direct vertebral rotation: A new technique of three-dimensional deformity correction with segmental pedicle screw fixation in adolescent idiopathic scoliosis. Spine 2004;29:343-349.
28. Kim YJ, Lenke LG, Kim J, et al: Comparative analysis of pedicle screw versus hybrid instrumentation in posterior spinal fusion of adolescent idiopathic scoliosis. Spine 2006;31:291-298.
29. Ebraheim NA, Jabaly G, Xu R, Yeasting RA: Anatomic relations of the thoracic pedicle to the adjacent neural structures. Spine 1997;22:1553-1556.
30. Ugur HC, Attar A, Uz A, Tekdemir I, Egemen N, Genç Y: Thoracic pedicle: Surgical anatomic evaluation and relations. J Spinal Disord 2001;14:39-45.
31. Polly DW Jr, Potter BK, Kuklo T, Young S, Johnson C, Klemme WR: Volumetric spinal canal intrusion: A comparison between thoracic pedicle screws and thoracic hooks. Spine 2004;29:63-69.
32. Suk SI, Lee CK, Min HJ, Cho KH, Oh JH: Comparison of Cotrel-Dubousset pedicle screws and hooks in the treatment of idiopathic scoliosis. Int Orthop 1994;18:341-346.
33. Minor ME, Morrissey NJ, Peress R, et al: Endovascular treatment of an iatrogenic thoracic aortic injury after spinal instrumentation: Case report. J Vasc Surg 2004;39:893-896.
34. Choi JB, Han JO, Jeong JW: False aneurysm of the thoracic aorta associated with an aorto-chest wall fistula after spinal instrumentation. J Trauma 2001;50:140-143.
35. Papin P, Arlet V, Marchesi D, Rosenblatt B, Aebi M: Unusual presentation of spinal cord compression related to misplaced pedicle screws in thoracic scoliosis. Eur Spine J 1999;8:156-159.
36. Heini P, Schöll E, Wyler D, Eggli S: Fatal cardiac tamponade associated with posterior spinal instrumentation: A case report. Spine 1998;23:2226-2230.
37. Yalçin S, Güven O: Reversible anterior cord syndrome due to penetration of the spinal canal by pedicular screws. Paraplegia 1995;33:423-425.
38. Panjabi MM, Takata K, Goel V, et al: Thoracic human vertebrae: Quantitative three-dimensional anatomy. Spine 1991; 16:888-901.
39. Zindrick MR, Wiltse LL, Doornik A, et al: Analysis of the morphometric characteristics of the thoracic and lumbar pedicles. Spine 1987;12:160-166.
40. Kothe R, O'Holleran JD, Liu W, Panjabi MM: Internal architecture of the thoracic pedicle: An anatomic study. Spine 1996;21:264-270.
41. Misenhimer GR, Peek RD, Wiltse LL, Rothman SL, Widell EH Jr: Anatomic analysis of pedicle cortical and cancellous diameter as related to screw size. Spine 1989;14:367-372.
42. O'Brien MF, Lenke LG, Mardjetko S, et al: Pedicle morphology in thoracic adolescent idiopathic scoliosis: Is pedicle fixation an anatomically viable technique? Spine 2000;25:2285-2293.
43. Hackenberg L, Link T, Liljenqvist U: Axial and tangential fixation strength of pedicle screws versus hooks in the thoracic spine in relation to bone mineral density. Spine 2002;27:937-942.
44. Deviren V, Acaroglu E, Lee J, et al: Pedicle screw fixation of the thoracic spine: An in vitro biomechanical study on different configurations. Spine 2005; 30:2530-2537.
45. Dick JC, Zdeblick TA, Bartel BD, Kunz DN: Mechanical evaluation of cross-link designs in rigid pedicle screw systems. Spine 1997;22:370-375.
46. Krag MH, Beynnon BD, Pope MH, et al: An internal fixator for posterior application to short segments of the thoracic, lumbar, or lumbosacral spine: Design and testing. Clin Orthop Relat Res 1986;203:75-98.
47. Kuklo TR, Lehman RA Jr: Effect of various tapping diameters on insertion of thoracic pedicle screws: A biomechanical analysis. Spine 2003;28:2066-2071.
48. Lehman RA Jr, Polly DW Jr, Kuklo TR, Cunningham B, Kirk KL, Belmont PJ Jr: Straight-forward versus anatomic trajectory technique of thoracic pedicle screw fixation: A biomechanical analysis. Spine 2003;28:2058-2065.
49. White KK, Oka R, Mahar AT, Lowry A, Garfin SR: Pullout strength of thoracic pedicle screw instrumentation: Comparison of the transpedicular and extrapedicular techniques. Spine 2006; 31:E355-E358.
50. Kuklo TR, Polly DW Jr: Surgical anatomy of the thoracic pedicle. Seminars in Spine Surgery 2002;14:3-7.
51. Cinotti G, Gumina S, Ripani M, Postacchini F: Pedicle instrumentation in the thoracic spine: A morphometric and cadaveric study for placement of screws. Spine 1999;24:114-119.
52. Ul Haque M, Shufflebarger HL, O'Brien M, Macagno A: Radiation exposure during pedicle screw placement in adolescent idiopathic scoliosis: Is fluoroscopy safe? Spine 2006;31:2516-2520.
53. Xu R, Ebraheim NA, Ou Y, Yeasting RA: Anatomic considerations of pedicle screw placement in the thoracic spine: Roy-Camille technique versus openlamina technique. Spine 1998;23:1065-1068.
54. Calancie B, Lebwohl N, Madsen P, Klose KJ: Intraoperative evoked EMG monitoring in an animal model: A new technique for evaluating pedicle screw placement. Spine 1992;17:1229-1235.
55. Calancie B, Madsen P, Lebwohl N: Stimulus-evoked EMG monitoring during transpedicular lumbosacral spine instrumentation: Initial clinical results. Spine 1994;19:2780-2786.
56. Shi YB, Binette M, Martin WH, Pearson JM, Hart RA: Electrical stimulation for intraoperative evaluation of thoracic pedicle screw placement. Spine 2003;28: 595-601.
57. Mann DC, Nash CL Jr, Wilham MR, Brown RH: Evaluation of the role of concave rib osteotomies in the correction of thoracic scoliosis. Spine 1989;14:491-495.
58. Suk SI, Kim JH, Kim SS, Lee JJ, Han YT: Thoracoplasty in thoracic adolescent idiopathic scoliosis. Spine 2008;33:1061-1067.
59. Geck MJ, Macagno A, Ponte A, Shufflebarger HL: The Ponte procedure: Posterior only treatment of Scheuermann's kyphosis using segmental posterior shortening and pedicle screw instrumentation. J Spinal Disord Tech 2007;20:586-593.
60. Shufflebarger HL, Geck MJ, Clark CE: The posterior approach for lumbar and thoracolumbar adolescent idiopathic scoliosis: Posterior shortening and pedicle screws. Spine 2004;29:269-276.
61. Suk SI, Kim WJ, Kim JH, Lee SM: Restoration of thoracic kyphosis in the hypokyphotic spine: A comparison between multiple-hook and segmental pedicle screw fixation in adolescent idiopathic scoliosis. J Spinal Disord 1999;12:489-495.
62. Clement JL, Chau E, Kimkpe C, Vallade MJ: Restoration of thoracic kyphosis by posterior instrumentation in adolescent idiopathic scoliosis: Comparative radiographic analysis of two methods of reduction. Spine 2008;33:1579-1587.
63. Kioschos HC, Asher MA, Lark RG, Harner EJ: Overpowering the crankshaft mechanism: The effect of posterior spinal fusion with and without stiff transpedicular fixation on anterior spinal column growth in immature canines. Spine 1996;21:1168-1173.
64. Burton DC, Asher MA, Lai SM: Scoliosis correction maintenance in skeletally immature patients with idiopathic scoliosis: Is anterior fusion really necessary? Spine 2000;25:61-68.
65. Sarlak AY, Atmaca H, Buluç L, Tosun B, Musaoğlu R: Juvenile idiopathic scoliosis treated with posterior arthrodesis and segmental pedicle screw instrumentation before the age of 9 years: A 5-year follow-up. Scoliosis 2009;4:1.
66. Dubousset J, Herring JA, Shufflebarger H: The crankshaft phenomenon. J Pediatr Orthop 1989;9:541-550.
67. Mulpuri K, Perdios A, Reilly CW: Evidence-based medicine analysis of all pedicle screw constructs in adolescent idiopathic scoliosis. Spine 2007;32(19 suppl):S109-S114.
© 2009 by American Academy of Orthopaedic Surgeons