Introduction
Degenerative disease of the lumbar spine is a common cause of disability and pain, with lumbar spondylolisthesis affecting 11.5% of the US population and lumbar stenosis affecting more than 200,000 adults in the United States. Spinal fusion surgery is indicated in select patients who have failed conservative medical management of symptomatic degenerative disease of the lumbar spine. The ultimate goal of surgery is to relieve pain, improve disability, and provide stability via bony fusion to prevent motion at the degenerated segment(s). Rates of lumbar fusion surgeries in the United States are rising substantially, with a 220% increase in the number of lumbar fusion surgeries performed from 1991 to 2001. Although later analyses show a slight plateau in rates of lumbar fusion since the 1990s, complex fusion procedures are becoming more frequent, with a 15-fold increase in complex lumbar fusions from 2002 to 2007.
Pseudarthrosis, also known as nonunion, is a relatively common complication encountered after lumbar fusion surgery that can be catastrophic to the integrity of the fusion construct. Complications including mechanical instability, instrumentation failure, and proximal and distal junctional failure/kyphosis may occur as a result of nonunion. Furthermore, patients with pseudarthrosis may experience further pain and disability in addition to their presenting symptoms that can lead to the progression of the disease and the need for reoperation.
In this chapter, the authors review the definition of lumbar pseudarthrosis, risk factors for developing pseudarthrosis, methods of prevention, diagnostic methods for evaluation, and the mechanical biology of bone healing.
Lumbar Pseudarthrosis Definition, Epidemiology, and Diagnosis
Pseudarthrosis, or nonunion, is defined as the failure of postoperative bony fusion after spinal surgery, resulting in the potential for mechanical instability. Pseudarthrosis may or may not be symptomatic. Symptoms, when they do occur, primarily consist of low back pain and spinal deformity. Fig. 23.1 demonstrates an example of a pseudarthrosis. Heggeness and Esses proposed a posterior lumbar pseudarthrosis classification system in 1991, with four different morphologies depending on bony fusion construct geometry: atrophic, transverse, shingle, and complex, with atrophic being the most common subtype.
The true incidence of pseudarthrosis is difficult to properly assess because of the number of patients who may be asymptomatic or have undiagnosed failure of bony fusion. Martin et al. reported that 471 of 2345 patients who underwent lumbar fusion in one series from 1990 to 1993 required reoperation after fusion, and 111 patients (23.6% of 471 patients who required reoperation or 4.7% of 2345 total) underwent reoperation for pseudarthrosis. In a systematic review of lumbar pseudarthrosis, Chun et al. reported an incidence of 5% to 35% after lumbar fusion surgeries and noted that the incidence increased as the number of levels fused increased.
Several randomized trials have been completed to discern whether pseudarthrosis rates/rates of fusion are lower in patients who undergo instrumented fusion versus decompression and fusion alone and to discern whether the choice of surgical approach affects lumbar fusion rates. Fischgrund et al. randomized patients to decompression and posterolateral fusion alone or to decompression and fusion with instrumentation; they found that arthrodesis occurred by 2 years in 82% of the instrumented cases versus 45% of the noninstrumented cases ( P =.0015). At follow-up of at least 5 years, the patients with solid fusions had better pain/clinical outcomes than the patients who did not.
Regarding fusion rates by surgical approach, Christensen et al. reported that patients randomized to circumferential fusion (anterior lumbar interbody fusion and posterolateral fusion) had higher rates of solid fusion (92% vs. 80%, P < .04) than patients who underwent posterolateral fusion. However, Lee et al. performed a systematic review of randomized trials comparing fusion rates of different surgical approaches and found that no firm conclusions could be drawn from the available data on surgical approaches and lumbar fusion rates.
Imaging Modalities to Assess Pseudarthrosis
Static radiographs may be helpful in diagnosis of pseudarthrosis but are limited in their diagnostic accuracy. Although static radiographs may rule out pseudarthrosis across a segment if there is demonstration of clear bridging bone either in the disc space or posterolaterally between the transverse processes, they do not evaluate motion across the segment, and therefore cannot definitively demonstrate nonunion. In contrast, dynamic radiographs (e.g., flexion-extension films) are more helpful in evaluating pseudarthrosis, which is commonly defined as persistent motion across the involved segment. Zero to five degrees of motion difference on flexion-extension films is considered normal in a given fused motion segment; however, any degree of motion above that range may be indicative of nonunion, although not necessarily diagnostic for it. Interestingly, many argue there is limited utility in obtaining routine postoperative radiographs in this patient population. Yamashita et al. investigated this in 63 patients who underwent lumbar instrumented fusion; they concluded that plain radiographs should be performed as indicated clinically rather than routinely after instrumented lumbar fusion and that the majority of asymptomatic patients do not require routine postoperative plain radiographs.
Computed tomography (CT), particularly thin-cut CT with multiplanar reconstruction, is the most widely used imaging modality to assess fusion status. CT is preferred for its widespread availability, high resolution of bony detail, multiplanar evaluation, and relatively low cost (e.g., vs. magnetic resonance imaging or nuclear medicine studies). CT has a high specificity with regard to lumbar fusion status (78%–86% specificity when compared with the gold standard of intraoperative findings).
When pseudarthrosis is suspected clinically, but plain radiographs and CT are inconclusive, other imaging modalities such as 99m Tc-labeled diphosphonate nuclear medicine studies and positron emission tomography (PET) may be helpful to assess bony metabolic activity, which may correlate to bony fusion occurrence. However, these imaging studies are expensive, require extensive resources for acquisition and proper interpretation, and have not been shown to be superior to other imaging modalities for the assessment of pseudarthrosis specifically.
Choudhri et al. published an American Association of Neurological Surgeons/Congress of Neurological Surgeons (AANS/CNS) Joint Section on Disorders of the Spine & Peripheral Nerves guidelines paper on the radiographic assessment of lumbar fusion status with several recommendations, including: (1) static radiographs are not recommended to assess fusion status; (2) CT with fine-cut images is recommended and is appropriately sensitive for assessing fusion status after both posterior and anterior lumbar fusions; and (3) lack of facet fusion is more suggestive of pseudarthrosis than is absence of bridging posterolateral bone alone. In a similar guideline publication, Dhall et al. analyzed the correlation between radiographic fusion status and functional postoperative outcome and concluded that there is moderate evidence demonstrating a positive association between radiographic fusions and improved clinical outcomes.
Mechanical Biology of Bone Healing and Fusion
Bone healing is a complex biological and mechanical process. Briefly, bone is derived of organic components (collagen, osteocytes, osteoblasts, osteoclasts, neurovascular networks) as well as calcium-based inorganic components (primarily hydroxyapatite [HA]). After fractures or intentional disruption of the bony cortex, as is the case in spinal fusion surgery, osteogenesis occurs in the void that remains between bony fragments. This process is highly dependent upon a number of factors including hormones, cytokines, proteins (e.g., bone morphogenetic protein [BMP]), mineral availability, and mechanical stimuli. The bony fusion process from the initial injury to the beginning of solid fusion usually takes approximately 6 weeks but may take as long as 6 months for mature, solid bone to fully form.
The term “osteoconductive agents” refers to agents that provide a scaffold onto which bony development may occur (e.g., ceramics), whereas “osteoinductive agents” refers to agents that aid in the formation of new bone in a heterotopic locality (e.g., demineralized bone matrix [DBM], BMP). In general, osteoconductive substitutes may be more suitable in the anterior spine after rigid immobilization, whereas osteoinductive substitutes are indicated as substitutes for posterolateral spine fusion given their ability to promote fusion between decorticated bone.
Patient Risk Factors for Pseudarthrosis
Numerous patient factors have been identified as risk factors for the development of pseudarthrosis, including smoking, radiation exposure, osteoporosis, rheumatoid arthritis, nonsteroidal anti-inflammatory drug (NSAID) use, inflammatory arthritides (such as ankylosing spondylitis), and poor nutrition.
Smoking has been identified as a risk factor for nonunion because of its known impairment of the bone healing and fusion process. Bydon et al. reported that pseudarthrosis rates were significantly higher in a smoking cohort than in a nonsmoking cohort (29.17% vs. 10.92%; P =.019) in patients undergoing two-level posterior lumbar fusion, but no difference between smokers and nonsmokers was observed for single-level fusions. Phan et al. found that the rate of failed fusion was significantly greater for smokers than nonsmokers (odds ratio [OR] 37.10; 95% confidence interval [CI] 3.79–365.20; P =.002) in a cohort of patients undergoing anterior lumbar interbody fusion. Animal studies in rabbits have demonstrated that nicotine alone (e.g., in patch or gum form) may be harmful to bone graft revascularization even in nonsmokers. Therefore, careful patient selection and counseling on nicotine use is warranted in patients undergoing lumbar fusion.
Osteoporosis has also been associated with fusion failure. Osteoporotic bone is inherently less mechanically stable than normal bone because of its lower density and poor pull-out strength when subjected to instrumentation. Therefore, selection of longer, large-diameter screw instrumentation is indicated in these patients. In addition, many advocate for treatment of the osteoporosis before elective lumbar fusion surgery if that is possible. Also, placing screws to maintain a position, rather than to achieve a deformity correction, is preferable in osteoporotic patients.
NSAID use has been associated with an increased risk of pseudarthrosis. Dodwell et al. conducted a meta-analysis of NSAID use and nonunion risk, finding that the pooled OR for nonunion with NSAID exposure was 3.0 (95% CI 1.6–5.6); however, this meta-analysis was not specifically investigating pseudarthrosis after lumbar fusion but rather nonunion after bony fracture. Li et al. conducted a meta-analysis on the use of ketorolac and its effect on pseudarthrosis after thoracolumbar posterolateral fusion, concluding that adults who had ketorolac administered for longer than 2 days and at doses 120 mg/day or more were at a greater risk of pseudarthrosis (OR 4.75; 95% CI 2.34–9.62; P < .001). However, Urrutia et al. found that the use of ketoprofen after lumbar fusion at a single level in an animal model did not decrease the fusion rate.
Poor nutritional status, including that secondary to alcohol abuse, has been implicated in increasing the risk of nonunion. There is risk of hypovitaminosis D, hypocalcemia, and hypophosphatemia in patients with poor nutritional status, all of which impair proper bony fusion from occurring optimally.
Bone Graft Materials
Bone graft materials have been used for decades to provide additional mechanical support to the lumbar fusion construct, as well as to biologically induce bony fusion via direct stimulation of bony healing or via the presence of a structural support scaffolding for bony ingrowth to occur. Table 23.1 displays various bone graft materials and their respective properties.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
