13 Selection of Fusion Levels

10.1055/b-0034-82167

13 Selection of Fusion Levels

Mulconrey, Daniel S. and Lenke/, Lawrence G.]

The selection of fusion levels for the surgical treatment of adolescent idiopathic scoliosis (AIS) has been debated since the inception of this surgical procedure. This debate began before the introduction of instrumentation for scoliosis 1–5 and has intensified during the modern era of segmental spinal fixation. After the introduction of posterior instrumentation, Harrington addressed the concept of the stable zone to identify the distal extent for a spinal fusion. Harrington defined the stable zone as the area between two parallel vertical lines running through the lumbosacral joint, and recommended that the end-vertebra of a spinal fusion be within that stable zone.6,7 Moe introduced the practice of evaluating curve flexibility and vertebral rotation to select fusion levels; thus initiating the concept of giving flexible curves the ability to correct spontaneously while performing a selective fusion of the more rigid curvature.4,8 Later, King and colleagues9 categorized patients with AIS and created a classification system for it. Included in King’s evaluation of patients with AIS was a description of the center sacral vertical line (CSVL), a vertical line that bisects the sacrum and is perpendicular to the level iliac crests. The CSVL falls in the middle of Harrington’s stable zone6,7 and bisects the stable vertebra ( Fig. 13.1 ).

Lenke et al10 further subdivided and analyzed patients with AIS to develop a comprehensive classification system. The Lenke classification included a two-dimensional (2D) evaluation of the scoliotic curvature with sagittal-plane emphasis, specific objective criteria for increasing interobserver reliability, and the ability to provide a template for spinal-fusion surgery.11 The fundamentals of this classification system will be the focus of the remaining portion of this chapter. This classification system, as outlined in Chapter 9, provides an algorithm for assisting in the selection of spinal fusion levels. The selection of fusion levels is a complex decision-making process that focuses on both clinical and radiographic examination.

History and Physical Examination

The history and physical examination is a critical facets in the process of spinal-fusion surgery. The patient’s skeletal maturity, family history of scoliosis, medical comorbidities, activity level, and reported self-image influence the management of scoliosis. The clinical deformity (shoulder balance, trunk shift, thoracic and lumbar prominence) in addition to the radiographic deformity will determine the selection of fusion levels. The difficulty in fusion surgery is in predicting how the correction and fusion of the patient’s scoliosis will affect this clinical deformity while maintaining coronal and sagittal balance.

**Fig. 13.1** Illustration of the center sacral vertical line (CSVL). The figure has been modified to include the stable vertebra (SV) and Harrington’s stable zone (*shaded area*). (From the Spinal Deformity Study Group Radiographic Measurement Manual, page 47, published by Medtronic, Inc., 2004.)

Radiographic Evaluation

Radiographic analysis of the patient with AIS should include long-cassette standing anteroposterior (AP) and lateral radiographs of the spine. The position of the patient’s arms should also be noted on the radiograph, as different positions may lead to alterations in sagittal balance.12 A set of flexibility films should also be collected, and may include right and left side-bending, push-prone, or traction radiographs.13 –16

As discussed in Chapter 5, multiple radiographic measurements are needed to fully understand the patient’s radiographic deformity. The magnitude and flexibility of each spinal curve will determine whether a curve is structural and should be included in the fusion. Identifying and measuring specific vertebrae will assist in selecting specific fusion levels. On the AP film, the C7 plumbline (C7PL) should be drawn as a vertical line from the center of C7 distally, and the CSVL should be drawn from the midpoint of the sacrum up to the proximal stable vertebra.17 The apical vertebral translation (AVT) is the distance from the apical vertebrae to the CSVL in the lumbar spine or to the coronal C7PL in the thoracic spine. The apical vertebral rotation can be documented with the Nash-Moe index.18

The stable vertebra is the vertebra most bisected by the CSVL. The neutral vertebra is identified as the vertebra with a symmetrical location of the pedicle shadows within the outline of the vertebral body. The end-vertebra is the vertebra most tilted from the horizontal. The identification of these vertebrae has had poor interobserver and intraobserver reliability among reviewers. In one study, 50 consecutive surgically treated cases of AIS were reviewed by 16 scoliosis surgeons. Interobserver reliability for the end-, neutral, and stable vertebrae had kappa values of 0.45, 0.32, and 0.52, respectively.19 Potter et al20 reviewed the reliability of identifying of the end stable, and neutral vertebrae in plain radiographic studies. One hundred consecutive surgically treated cases of AIS were evaluated at several intervals by multiple spine surgeons. Intraobserver reliability was good-to-excellent for the stable, neutral, and end-vertebrae (kappa = 0.69 to 0.88, 0.65 to 0.73, and 0.74 to 0.91, respectively). However, interobserver reliability was poor (kappa = 0.26 to 0.39). The identification of these levels is crucial for selecting fusion levels. This variability in identifying the stable, end, and neutral vertebrae increases the variability seen in selecting fusion levels.

Certain clinical parameters, such as shoulder balance, can also be assessed on AP films. The radiographic interpretation of shoulder position is done in several ways including the measurement of the T1 tilt, coracoid height, and first rib clavicle height. Kuklo et al21 reviewed 112 AIS patients with a proximal thoracic (PT) curve >20 degrees to identify radiographic parameters that would predict postoperative shoulder balance at 2 years of follow-up. The clavicle angle, as formed by the intersection of a horizontal line and the tangential line connecting the highest two points of each clavicle, provided the best preoperative radiographic prediction of postoperative shoulder balance ( Fig. 13.2 ). How this translates into selecting the optimal proximal fusion level is still not clear.

**Fig. 13.2** Illustration of the clavicle angle. (From the Spinal Deformity Study Group Radiographic Measurement Manual, page 56, published by Medtronic, Inc., 2004.) Abbreviations: CHRL, clavicle horizontal reference line; CRL, clavicle reference line.

Operative Algorithm

The principles of surgical treatment for scoliosis are based on outcome measures including pulmonary function testing, radiography, cosmetic appearance, functional outcome, range of motion, and aerobic studies.22 –27 Classification and understanding of the patient’s spinal-curve type is crucial in avoiding many postoperative complications, including decompensation. The system set forth by Lenke et al10 provides a comprehensive classification of curve patterns and a template for the surgical management of AIS. A thorough understanding of the Lenke classification system is essential before determining the vertebral levels for spinal fusion.

Selection of Fusion Levels Anterior Spinal Fusion

Anterior spinal fusion (ASF) with instrumentation is typically reserved for cases olf scoliosis in which only one curve is being treated; this specifically applies to cases of Lenke type 1 and type 5 curves. Awareness of the flexibility of the compensatory curve and its response to treatment of the main curve is critical in anterior surgery for scoliosis. The anterior approach, both open and thoracoscopic, has received increased attention in the past decade.21,28–30 The anterior approach provides excellent curve correction and may facilitate the inclusion of fewer levels in the fusion mass than would a posterior approach.4,31,32 Instrumentation of fewer levels is especially possible for flexible scoliotic curves.33 Selection of fusion levels typically includes all segments within a curve, from the end vertebra cranially to the end-vertebra caudally.34 –36 Depending on the approach, either a single- or dual-rod technique can be utilized. We currently employ a dual-rod technique with screw fixation at every level in our open approach.

Hall and co-authors33,37 advocated short instrumentation for flexible thoracolumbar/lumbar (TL/L) curves. If the apex of the curve is a vertebral body, they advocated fusing one vertebra above and one vertebra below the curve. If the apex is a disc space, the fusion and instrumentation would be performed two vertebral levels above and two levels below the apex. Using this technique, Hall and co-authors demonstrated an initial correction of 87%, declining to 67% at 2 years. The satisfaction rate in their study of 17 patients was 88%. To achieve these results, Hall et al recommended overcorrecting the instrumented vertebrae.

Application of any surgical technique requires assessment of the patient’s overall coronal and sagittal balance as well as the clinical deformity. For example, an anterior procedure may lead to worsening of the kyphosis in kyphotic thoracic curves. If a patient has a high right shoulder preoperatively, selective anterior fusion of a left Lenke type 5 curve may increase the patient’s shoulder asymmetry.

Posterior Spinal Fusion

Type 1: Main Thoracic Curves

Type 1 main thoracic (MT) curves are the most commonly treated form of AIS.38 Although these curves can be treated with either ASF or posterior spinal fusion (PSF), posterior instrumentation and fusion remain the gold standard.39 Although Chapter 17 provides an in-depth review of the Lenke type 1 curve, some basic principles and rules will be reviewed here.

To determine the upper instrumented vertebra (UIV), clinical shoulder balance needs to be assessed. As previously discussed, the clavicle angle appears to be the most reliable preoperative indicator for shoulder assessment. For a patient with a right MT curve and right shoulder elevation, proximal extension of spinal fixation to T4 or T5 is appropriate. Exclusion of the upper thoracic segments will allow left shoulder elevation to occur with correction, and will produce level shoulders postoperatively. For a patient with level shoulders, extension of fixation to T3 or T4 is often indicated. Cranial extension of the posterior spinal segmental instrumentation (PSSI) to include the upper thoracic segments will facilitate control over the left shoulder height and maintain shoulder balance postoperatively. For a patient with left shoulder elevation, extension of fixation to T2 is usually necessary to eliminate postoperative shoulder elevation. This proximal extension of the spinal fixation levels will allow intraoperative compression of the upper left thoracic segments to lower the left shoulder and correct the patient’s preoperative imbalance.

The senior author of this chapter (L.L.) usually selects the lowest instrumented vertebra (LIV) as the lowest vertebra touched by the CSVL for lumbar curves with an A modifier in the Lenke classification system. Most commonly, this is the vertebra proximal to the stable vertebra (stable-1), or occasionally the vertebra two levels proximal to the stable vertebra (stable-2). PSSI in type 1 curves is best suited for patients with a normal or hyperkyphotic sagittal modifier. For type 1B curves, the LIV (the most cephalad vertebra intersected or bisected by the CSVL) will usually be located in the thoracolumbar junction.

The most controversial curve pattern treated with selective fusion of the thoracic spine is the Lenke type 1C pattern. The lumbar curves in this curve pattern are large and deviated, and are flexible (side-bending Cobb angle <25 degrees). Care must be taken and further evaluation done when the decision is made to perform a selective thoracic fusion on a type 1C curve. The Cobb-angle measurements, AVR, and AVT of both the thoracic and lumbar curves must be evaluated before deciding on whether or not to do a selective fusion. When the Cobb angles ratios of the MT-to-TL/L curves, and the respective ratios of their AVR and AVT values, are >1.2, a selective fusion is a possible surgical option, with a low incidence of lumbar decompensation or “adding-on” of caudal vertebral segments to the patient’s spinal curvature.40 Also, the thoracolumbar junction must lack kyphosis (i.e., the T10-L2 sagittal Cobb angle must be <10 degrees).11 Clinical evaluation is also important in deciding whether or not to perform a selective fusion. The patient should have a thoracic-to-lumbar scoliometer ratio >1.2, and the right shoulder should be higher than or level with the left shoulder.11 Additionally, the patient should demonstrate a significant clinical thoracic prominence and minimal flank deformity ( Fig. 13.3 ).40 With the patient erect, thoracic truncal shift should be much more visible than the lumbar shift. These findings on physical examination reinforce the concept of the thoracic curve being more structural than the lumbar curve. Skeletal maturity does not factor into this decision.41

**Fig. 13.3 (A)** A 13 + 11-year-old girl with 59-degree thoracic and 56-degree lumbar AIS curves. The MT curve side-bends to 23 degrees and the lumbar curve bends to 14 degrees and is nonstructural. The T5 to T12 kyphosis was +21 degrees, therefore the Lenke curve classification is type 1CN.

Although these characteristics have been outlined in the literature, variability and differences in opinion remain with regard to surgical decisions. Newton et al42 examined this variability by reviewing 203 patients with Lenke type 1B or 1C curves treated at five different sites of the Harms Study Group (HSG). The lumbar curve was included in from 6 to 33% of spinal fusions performed, depending on the study site. Factors that increased the performance of lumbar fusion included a larger preoperative lumbar curve, displacement of the apical vertebra from the CSVL, and a smaller thoracic-to-lumbar curve ratio ( Fig. 13.4 ).

The advantages of ASF or PSF for Lenke type 1 curves have also been investigated. Potter et al43 compared their results with ASF and PSF done with thoracic pedicle screws (TPS) for Lenke type 1 curves. The patients in this study were followed for an average of 3 years postoperatively. Patients who had PSF with TPS had ~1.2 more levels fused than did their matched cohorts who had ASF. However, a greater correction of scoliosis was achieved in the thoracic and lumbar spine in the group that had PSF. The rotational correction was greater in the PSF group with TPS. This yielded a significant improvement in the rib hump and other radiographic parameters in this group. However, PSF with TPS was associated with a decrease in thoracic kyphosis, whereas ASF tended to increase thoracic kyphosis in these patients.44 Currently, we approach all Lenke type 1 curves posteriorly. By avoiding disruption of the chest wall, we obtain excellent corrections without the deleterious effects on pulmonary function seen with anterior approaches. Posterior techniques, such as Ponté osteotomies, are currently being investigated to determine their role in minimizing the tendency to increase the kyphosis effect of all pedicle screw thoracic constructs.