Open Versus Endoscopic Lumbar Pedicle Screw Fixation

Introduction:

With improved spinal instrumentation and surgical technique, the non-union rate for spinal fusion has dramatically declined in recent years. However, pseudoarthrosis remains common in certain patient populations such as the elderly, those who smoke and drink, those with metastatic and metabolic disorders, and those who do not comply with post-operative recommendations. In these populations and others, the non-union rate can be as high as 35% (1). In addition to non-union, complications associated with the harvest of autogenous bone such as donor site morbidity, blood loss, and increased operative time, has prompted the search for suitable alternatives (2,3). In addition to these problems, many spinal fusion surgeries are lengthy procedures that involve a relatively large surgical field, and are associated with a great deal of morbidity and a long hospital stay.

Recently, many new bone graft substitutes are being designed to incorporate or mimic one or more of the bone-forming components of autogenous bone to prevent the morbidity of the bone graft donor site. This is often achieved by combining a purified, recombinant form of an osteoinductive protein with an osteoconductive scaffold or matrix to attract, simulate, and support the formation of new bone by host osteogenic cells (1, 4-6). A number of spinal researchers and surgeons have applied this technology to spinal fusion with great success (1,4-19). As well as the development of the bone graft substitutes, improved spinal instrumentation and endoscopic surgical technique are also successfully used in spinal surgery (8, 20-27). By reducing the size of the incision, the surgeon reduces the iatrogenic trauma imparted on surrounding tissues, the risk of infection and the healing time.

In an attempt to eventually reduce the size of incision and dissection, and reduce the amount of bed rest each patient must endure, this study was designed to test novel graft substitute material, mineralized type I collagen (HealosÐ'®, Orquest, Mountain View, CA) plus recombinant human growth/differentiation factor-5 (rhGDF-5), in hope of increasing the rate and quality of fusion achieved. In addition, we have incorporated newly developed minimally invasive surgical technique.

Although both endoscopic spine surgery and the use of bone substitutes for spinal fusion are relatively new applications, they indicate the future direction of spinal surgery. The primary goal of this study is to test the efficacy of delivering graft materials and placement of pedicle screws in a minimally invasive manner using a posterolateral endoscopic approach and evaluate the capabilities of two graft materials to promote spinal fusion. The study also measures the fusion rate and histologic characteristics of two different graft materials following three different periods of implantation.

Materials and Methods:

Twelve skeletally matured sheep (weight range between 50 kg and 60 kg) underwent bilateral intertransverse process spinal fusions at the L4-5 level after approval from the Institutional Animal Care and Use Committee. All surgeries were performed on the L4-L5 level by using a posterolateral endoscopic approach and inserting transpedicular screws (Endius Inc, Plainville, MA). Two bone graft materials, autogenous iliac crest bone graft and bone graft substitute (HealosÐ'® plus rhGDF-5), were used on either the right or left side alternatively in each animal. No postoperative immobilization devices were utilized and all animals were allowed to move about their cages without restriction. Three groups were formed on the basis of the designated postoperative period: two months, four months, and six months. The animals were euthanized after designated time for gross inspection, manual palpation, histology, plain radiograph, and computer tomography (CT) evaluations.

Surgical Procedures.

Each animal was pre-anesthetized with appropriate dose for body weight of ketamine and diazepam and placed on the surgical table in the prone position. Isoflurane and intravenous fluids were administered throughout the remainder of the surgery. The sheep was shaved, prepared, and draped in a sterile manner. Lateral and anterior-posterior (AP) fluoroscopic view was used to determine the optimal incision location for the L4-L5 level. A #16 gauge 7-8 inch spinal needle was inserted 2cm lateral to midline at a 10-15 degree angle directed at the lateral aspect of the facet (AP view). In the lateral fluoroscopic view, the needle was directed at an angle that was parallel and adjacent to the superior end plate of the disc. The spinal needle was advanced with care under fluoroscopic guidance and docked on the lateral facet. A 2 cm longitudinal skin incision was made at the guide wire insertion position. Step dilators were passed over the guide wire sequentially dilating the muscle and fascia to 5, 10, 15 and 20 mm. A 20 mm FlexPosureв„Ñž endoscopic retractor is placed over the 20 mm step dilator. The distal skirt of the retractor was expanded to approximately 35 mm in the longitudinal direction. The endoscope was then positioned and secured laterally against the side of the retractor and was advanced to view the lateral facet. Soft tissue covering the lateral facet and adjacent structures was excised and aspirated using the MDSв„Ñž microdebrider with 4.5 mm diameter and 20 cm long shaver blade. Debridement continued to expose the two adjacent bony anatomy and landmarks for pedicle screw entry and transverse process were identified. A high-speed burr was then used to decorticate the transverse process and lateral wall of the facet joint to expose the bleeding cancellous bone. The retractor skirt and tube were moved and repositioned to be in line to act as a guide to optimally insert the pedicle screws. With lateral fluoroscopic guidance, the 5.5 mm diameter TriFixв„Ñž pedicle screws were passed through the opening created by this procedure and advanced through the pedicles, into the vertebral bodies. The graft materials (autogenous iliac crest bone graft and HealosÐ'® with rhGDF-5) were prepared as described later and placed either on the right or left side alternatively between the transverse processes in the paraspinal bed of each animal. The plate was then placed over the screws and secured with washers and nuts. The surgical site was examined for bleeding and hemostasis was achieved. The endoscope was removed and the retractor was withdrawn from the wound with a clockwise twisting motion. The fascial incision was closed with 2-0 absorbable sutures, and the skin was reapproximated with 2-0 absorbable sutures. The surgeon then performed the same operation on the contralateral side. The surgical site was accessed, and the pedicle screw