Options of Thoracic Idiopathic Scoliosis Fixation and Fusion

INTRODUCTION
For many years, the gold standard for fixation for a thoracic curve in idiopathic scoliosis was a single Harrington rod applied through a posterior approach, with a distraction rod on the left side. See Figures 1A & B. This yielded a "good result" in most cases. Posterior approach refers to exposing the laminae and posterior elements (Figures 2 and 3). Anterior approach refers to exposure of the vertebral bodies (Figures 4A & B and 5).

Figure 1A, Single Harrington rod applied through a posterior approach	Figure 1B, Single Harrington rod applied through a posterior approach

Figure 2, Posterior Approach	Figure 3, Posterior Approach

Figure 4A, Anterior Approach	Figure 4B, Anterior Approach
Figure 5, Anterior Approach

EVOLUTION
However, advances have been made centered on trying to obtain more correction and to get patients back to their previous activities more quickly with less reliance on bed rest, cast or brace.

A next step in the evolution was that of using two rods for a thoracic curve and achieving segmental fixation (fixation at each level) with either Wisconsin wires or sublaminar wires. A distraction rod was used on the left side and a neutralizing Luque rod on the right side. See Figures 6A–D.

Figure 6A, Preoperative	Figure 6B, Side View
Figure 6C, 2 Harrington Rods	Figure 6D, Sublaminar Wires

The next evolutionary stage then was that of CD (Cotrel–Dubousset) instrumentation using multiple hooks with both rods in an effort to get even stronger fixation and to "derotate" the spine somewhat and also to reduce and correct the lordosis that often goes along with right thoracic curves. See Figures 7A–D.

Figure 7A, Preoperative	Figure 7B, Side View
Figure 7C, CD Instrumentation	Figure 7D, Multiple Hooks

Step 3 has now evolved into a concept of anchors at the top and the bottom of the curve with wires in between, either Wisconsin wires or sublaminar wires. In most cases, the distal anchors described are hooks or pedicle screws, and the proximal fixation points are hooks which are left loose until the final fixation. See Figures 8A–D for x–rays, Figures 8E & F for clinical result in the patient.

Figure 8A,Preoperative	Figure 8B, Side View
Figure 8C, Wisconsin Wires	Figure 8D, Side View

	Figure 8E, Preoperative
		Figure 8F, Postoperative

Step 4 of the evolutionary phase for posterior treatment has been that of using pedicle screws more liberally throughout the thoracic and upper lumbar spine in order to achieve more fixation, to reduce the number of levels being fused and to reduce the number of junctional problems. Junctional problems refer to breakdown or kyphosis that develops in segments above or below the instrumented vertebrae. See Figures 9A–D.

Figure 9A, Preoperative	Figure 9B, Side View
Figure 9C, Pedicle Screws	Figure 9D, Pedicle Screws Side View

Anterior instrumentation has been popular for thoracolumbar and lumbar curves for several years. See Figure 10.

Figure 10 Anterior Instrumentation

Recently, anterior instrumentation for a thoracic curve has become popular. The advantages of the anterior procedure are the following:

(1) Saving one or two levels distally;
(2) Another alternative to correct the lordotic component of the thoracic curve;
(3) Some suggestion of slightly better correction and slightly better derotation. See Table 1.

Table 1. ANTERIOR APPROACH
Cons	Pros
disrupts shoulder muscles	different incision
pseudarthrosis rate not clear	saves 1–2 fusion levels
effect on pulmonary function	slightly better correction
harder to revise

The anterior technique has evolved from use of the threaded rod, to use of the solid rod, to potentially the use of two solid rods anteriorly in certain patients. See Figures 11A–D. The potential advantages of the anterior approach are:

(1) A different scar;
(2) Preservation of the posterior spinal extensor muscles. See Table 2.

Figure 11A, Preoperative	Figure 11B, Side View
Figure 11C, Vertebral Body Screws	Figure 11D, Side View

Table 2. POSTERIOR APPROACH
Cons	Pros
disrupts posterior muscles	pseudarthrosis rate known
implant prominence in thin patient	easier to revise
small incidence of junctional deformities	very stable
1–2 more fusion levels than with anterior segmental spinal instrumentation

The anterior approach is still invasive, though, and does involve cutting through and then sewing up shoulder muscles such as the serratus anterior and the latissimus dorsi. The anterior approach has evolved further now into doing it with an endoscopic/video–assisted thoracoscopic technique. See Table 3.

Table 3. THORACOSCOPIC APPROACH
Cons	Pros
longer operation	less postop soreness
smaller implants	less effect on pulmonary function!??
postop brace needed	different scars
pseudarthrosis rate not known

Rather than a single incision, three to five "stab" incisions and otherwise the same surgery with a screw at each level and a single rod.

Further, many surgeons and patients decide to have either an internal or external thoracoplasty done at the time of the correction to reduce the size of the rib hump. We do know that anterior surgeries done with an open thoracotomy and also thoracoplasty techniques do diminish pulmonary functions which then take about two years to recover.

In the modern day, virtually no spine surgeons in North America use Harrington distraction. Surgeons use either open or endoscopic anterior approaches or posterior approaches with bilateral rods, hooks, wires and/or pedicle screws. Patients usually stand and walk the day after surgery and do not wear casts or braces.

At present there is no strong data to suggest that one technique is superior to the others. Each technique has the potential to work quite well for the patient as it does in most cases, but each of the techniques has limitations.

An incidence of pseudarthrosis (fusion not healing) still does exist with all techniques. With any posterior technique, there is a potential for implant prominence. With any anterior technique, there is a potential for reduced pulmonary function. Most surgeons do not brace the patient postoperatively except after the anterior endoscopic techniques. In five to ten years, it may become clearer that one technique is superior to the others. It is also quite possible in five to ten years that we will not decide that the various posterior and anterior techniques are all very equivalent.

Still the goal is to achieve a solid and stable fusion with the top and the bottom of the fusion in acceptable sagittal alignment and parallel to the shoulders and the pelvis.