Complications Related to Lumbar Pedicle Screw Instrumentation
Scott G. Tromanhauser
Joel W. Jenné
OVERVIEW
Instrumented lumbar procedures are performed to augment fusion and provide stability, although the specific indications and added benefits of instrumentation are routinely debated. Historically, instrumentation of the spine began with wiring techniques by Hadra in 1891 (1) to stabilize a fracture-dislocation of the cervical spine. As knowledge of anatomy and biomechanics has improved, the concepts and designs of instrumentation have developed throughout the last century, for both anterior and posterior procedures, including hooks and rods, sublaminar wiring, transpedicular fixation, facet screws, interbody cages, anterior screws with plates or rods, as well as various combinations thereof. Techniques to implant such devices have also been modified or created, varying from traditional open to endoscopic approaches for anterior procedures, minimally invasive tube surgery and the Wiltse approach for posterior procedures and, recently, the introduction of surgical navigation systems to assist with placement of spinal instrumentation.
In order for patients to achieve the maximum benefit from any given surgical procedure, the surgeon must have a clear understanding of the indications and risks associated with the selected procedure. Minimizing risk is the foundation of the Hippocratic Oath, “First do no harm.” To do so, one must have a thorough understanding of the relevant anatomy, the procedure itself, and the risks and benefits associated with the use of instrumentation and its alternatives, and then one must apply this knowledge to the correctly selected patient.
The role of instrumentation in lumbar procedures remains somewhat controversial with certain lumbar pathologies. Of the recent prospective randomized trials, conflicting conclusions have been made regarding instrumented versus noninstrumented fusions in the lumbar and lumbosacral spine. Fischgrund et al. (2) reported successful fusion rates of 82% and 45% in instrumented versus noninstrumented degenerative lumbar spondylolisthesis with spinal stenosis, respectively. In a prospective, randomized trial, Zdeblick (3) reported arthrodesis in 65% of noninstrumented, 77% of semirigid, and 95% of rigidly instrumented cases. Conversely, Thomsen et al. (4) found no significant difference in both outcomes and fusion rates in instrumented and noninstrumented fusions with arthrodesis achieved in 68% and 85%, respectively.
Thorough, critical review of the literature is necessary to draw any significant, applicable conclusions regarding the effects of surgical treatment and outcomes. Bono and Lee (5) performed a critical analysis of trends in fusion for degenerative disc disease, concluding that greater use of technology has been noted in the literature and trends toward increased fusion rates; however, the clinical and functional benefits are not yet clear. The authors also point to multiple weaknesses in the spinal literature, with articles lacking information such as the type of instrumentation used, graft location, fusion rate, graft source and brace type, or simply failing to state clearly the study methodology. As such, reviewing and comparing outcomes (which are also quite variable) between studies is limited.
Regardless of these recognized weaknesses in determining
the optimum surgical indications for instrumentation, its role still exists, and the surgeon must recognize potential complications when using instrumentation to augment fusion. The goal of this chapter is to cover the incidence of complications related to lumbar pedicle instrumentation, the relevant anatomy, and how to prevent and manage these complications when they occur.
the optimum surgical indications for instrumentation, its role still exists, and the surgeon must recognize potential complications when using instrumentation to augment fusion. The goal of this chapter is to cover the incidence of complications related to lumbar pedicle instrumentation, the relevant anatomy, and how to prevent and manage these complications when they occur.
Incidence of Complication
Pedicle screw fixation offers many advantages from a biomechanical standpoint. Segmental fixation adds significant strength and rigidity to the spine. These biomechanical advantages are not realized without potential risks. Placement of pedicle screws requires knowledge of the anatomy, landmarks, and inherent limitations of the selected instrumentation to limit complications. Such complications can occur intraoperatively, postoperatively, or during long-term follow-up.
Intraoperative complications include dural tears, nerve root and vascular injuries, as well as facet and pedicle violations. The incidence of dural tears with instrumentation is relatively low, with rates cited at 0.1% of screws placed (6) or 0% to 6% of patients (7,8,9,10), and can often be directly repaired with minimal long-term morbidity. Dural injuries may occur during attempted placement of pedicle screws through a breach of the pedicle or unintentional injury to an exposed dural sac from a slipped instrument.
In complications related to pedicle screw placement, the greatest focus has been on injury to adjacent nerve roots from a pedicle breach. Anatomically, there is very little room for error medially and inferiorly to the pedicle. Misplaced screws or pedicle fractures in these regions have a propensity to cause dural injuries, root injury, or radicular symptoms, which may or may not be reversed with screw removal (Fig. 12.1). Misplaced screws recognized intraoperatively can be redirected and usually salvaged. The literature reports misplaced screw rates varying between 1% and 18% of screws (11,12,13,14,15,16,17,18) and between 1% and 9% of patients (19,20,21,22,23,24) Percentages are unknown for misplaced screws that were subsequently redirected or for those patients who remain asymptomatic despite a malpositioned screw. Rates of nerve root irritation range from 0% to 1% of screws (25,26) and from 0.6% to 10% of patients (27,28,29,30).
Facet violations through direct penetration or abutment of the adjacent facet with the screw head or rod may negatively affect results. Violation of the caudal facet is essentially without clinical significance, as this facet is involved in the fusion. However, the most cephalad level is outside the desired fusion levels and injury to this facet or impingement from hardware may be a source for symptomatology (Fig. 12.2). Superior facet joint violations have been reported in 24% to 35% of patients and in 15% to 23% of screws (31,32). Violations were noted twice as frequently on the left side, and almost half involved a single-level L5 screw, with the authors concluding that limited dissection of single-level fusions and the caudal, medial trajectory of the L5 pedicle may make accurate placement slightly more difficult for a right-handed surgeon on the patient’s left side (32).
Short-term, postoperative complications are either related to previously unrecognized pedicle violation and secondary root irritation or early hardware failure. Early hardware failure may be the result of technical insertion of the instrumentation (e.g., cross-threaded set screws, failure to tighten set screws) or mechanical failures related to screw or rod design. Complications related to hardware failure were initially emphasized in the literature, including screw fracture and screw-rod interface loosening (33,34,35,36). Advances in implant design have minimized such complications, which are now infrequently reported.
Pseudarthrosis is seen in long-term follow-up. Clinical features including persistent pain or radiographic findings such as poor fusion mass, segment motion, or lucency around screws and screw fracture may lead to the diagnosis of a failed fusion (Fig. 12.3). Pseudarthrosis rates vary depending on the indication for instrumented fusion from 68% to the “almost 100%” success of Roy-Camille et al. (37,38,39,40,41,42). There exists significant variability in the literature for fusion rates, as many studies rely on various radiographic parameters and not re-exploration of the fusion mass, thus creating a high margin of error (43).
Finally, no surgical case is without risk of infection which may present immediately or delayed following the procedure. Documented infection rates vary between 1% and 8% (44,45,46,47,48,49,50,51) for instrumented lumbar fusions. These rates are generally considered to be higher than the quoted 1% rate for discectomy and 2% for noninstrumented fusions (52,53,54,55,56,57,58,59).