Background
Revision lumbar spine surgery is predominantly associated with several distinct pathological processes resulting from an initial index surgery. For example, in the treatment of disc herniations, residual stenosis and recurrent disc herniations are well described in the literature. There is an estimated 5% to 25% chance of a recurrent disc herniation at the same level following discectomy. In addition, approximately 30% of patients undergoing primary lumbar microdiscectomy surgeries undergo revision lumbar decompression for progressive degenerative pathologies. Furthermore, Heindel et al. found that 38.5% of patients undergoing revision microdiscectomy went on to have lumbar fusion surgery within 4 years of their revision surgery. Because it is estimated that approximately 90,000 Medicare patients undergo primary discectomy surgery annually in the United States and because thousands more undergo primary lumbar decompression for degenerative stenosis, there are a significant number of patients who will require revision surgery based on the natural history alone.
Adjacent segment degeneration, also termed adjacent segment disease (ASD) when clinically symptomatic, most commonly occurs at levels adjacent to a previous lumbar fusion and is also an important factor to consider in revision surgery. Adjacent segment degeneration is seen radiographically in up to 92.2% of patients, and ASD is estimated to occur in 5% to 31% of patients following lumbar fusion surgery, with an increased risk of pathology developing at the thoracolumbar junction or lumbosacral junction. The rate of ASD occurrence doubles approximately every 2 years following spinal fusion, with an approximate incidence of ASD of 1.7% at 2 years postoperatively and increasing to 9% of patients at 8-year follow-up. Furthermore, iatrogenic conditions that develop after fusions include pseudarthrosis and implant failure. These pathologies may occur either early or late after an index intervention. In addition, flatback syndrome represents an iatrogenic condition related to fusing the patient with insufficient lordosis, resulting in either a mismatch of pelvic incidence and lumbar lordosis or in converting a compensated flat back into an uncompensated flatback.
Finally, wound complications are multifactorial and may be seen in patients with poor nutrition or diabetes, which are most common in patients undergoing long-segment fusions or revision surgery. The surgeon must be familiar with these pathological processes resulting from primary interventions. The surgeon should understand that upfront preventative strategies (e.g., medical optimization preoperatively, proper positioning, preoperative planning, appropriate bone “carpentry,” meticulous wound closure) may mitigate some of these conditions.
Adjacent Segment Pathology
The majority of patients who undergo lumbar spine surgery for degenerative spinal pathology can be expected to have some progression of disease at adjacent levels over time. Moreover, lumbar fusions have resulted in an increasing pathological entity, adjacent segment pathology (or radiographic evidence of degenerative change at segments rostral or caudal to the index fusion levels), that is related to the index surgery ( Fig. 4.1A,B ). Adjacent segment degeneration may be characterized as the development of new or progressive stenosis, listhesis, or disc space collapse at a level(s) immediately above or below the index levels of fusion, and it may be asymptomatic. Adjacent segment degeneration that becomes clinically symptomatic is defined as ASD. There are multiple proposed risk factors in the development of ASD. One study by Zhong et al. evaluated patients undergoing instrumented fusion for spondylolisthesis and found that concomitant decompression at segments adjacent to the fusion or preexisting spinal stenosis immediately rostral to the fused levels were significant predictors of ASD. It is theorized that the disruption of the most rostral or caudal facet complex during dissection of the posterior elements may lead to iatrogenically induced hypermobility of the unfused segments, resulting in ligamentous hypertrophy and stenosis. It has been proposed that if care is not taken during dissection to preserve the inter- and supraspinous ligaments between levels, there may be weakening of the posterior tension band, resulting in increased motion of the facets at the levels adjacent to the fusion causing ligamentum flavum hypertrophy and further stenosis. Additionally, the disc space adjacent to a fusion may become overloaded by increased mechanical stress, leading to degeneration. Other proposed mechanisms that alter the biomechanics at the adjacent levels include the length of the fusion and stiffness of the construct. Finally, at the most severe end of the spectrum of adjacent segment pathology, proximal junctional failure and kyphosis may occur and may be attributed to alterations in segmental biomechanics at the interface between the rigid instrumented spine and mobile adjacent segments. The indication for revision surgery in the setting of adjacent segment degeneration is symptomatic degenerative changes; such revision should be performed with the understanding that further intervention may continue to accelerate degeneration at the adjacent unfused levels.
Pseudarthrosis
Pseudarthrosis is defined as a “false joint” or nonunion, which may be seen following arthrodesis surgery. This radiographic diagnosis is usually based on x-rays or computed tomography (CT) imaging and is usually associated identified by incomplete bony fusion across an intervertebral segment or continued motion of the segment. In addition, a haloing effect may be present around the screws, strongly suggesting micromotion. There are several potential etiologies for the development of a pseudarthrosis: inadequate decortication, poor bone remodeling at the molecular level from inadequate substrate (i.e., neurofibromina), a limited bony surface area to facilitate fusion, significant motion despite instrumentation, smoking, poor nutrition, or excessive micromotion. Pseudarthrosis is commonly observed at the level of three column osteotomies, infected spines, fractured rods, or at L5–S1 owing to the poor healing environment and the unique biomechanical stresses at L5–S1. Radiographic pseudarthrosis may or may not be clinically significant. Some patients develop pain at the site of a pseudarthrosis and require revision fusion surgery. If the pseudarthrosis is present at a junctional level, such as T11–L1 or L5–S1, there may be yet another fusion failure even after revision surgery attributed to kyphosis and lack of global sagittal balance. Although clinical symptoms are commonly seen in patients with pseudarthrosis, some patients may present without symptoms.
Implant Failure
A strict definition of implant failure was defined by Okamoto and colleagues as “fracture of a metal component, such as the screws or rods, or the disassembly of fixed constructs.” Okamoto further delineates screw failures as: screw ploughing (translation perpendicular to the long axis of the screw), loosening (radiolucent halo of ≥1 mm around the screw), pull-out (translation parallel to the long axis of the screw), cut-out (ploughing leading to disruption of the pedicle or vertebral body wall or end plate), or screw breakage. With respect to cage or interbody device placement, subsidence, migration, collapse, or breakage are additional types of implant failures. With respect to lumbopelvic fixation, implant failure is typically evident at the caudal end of a construct or across the lumbosacral junction ( Fig. 4.2A,B ). These findings are thought to be because of the long moment arm across a fixed multisegment construct with the highest stress acting at the distal levels as well as the high translational forces present at these levels with axial loading. Implant failure may be seen in both short- and long-term follow-up. Early implant failure (<3 months) may result from poor bone quality or poor preoperative planning with regard to alignment of the spine and sheer forces placed upon the implants. Additionally, in patients with poor bone quality, the sheer forces resulting from the very stiff construct against soft bone may precipitate implant failure. Delayed implant failure may be found with pseudarthrosis. Smith et al. evaluated adult spinal deformity patients and found that 6.8% of patients had symptomatic rod fracture on follow-up, with the majority of rod fractures occurring within 12 months before arthrodesis was expected. Pain is the most common presenting symptom after a rod fracture. The most common site of rod fracture is at the level of a three-column osteotomy because of the significant instability that ensues after such an osteotomy. Rod fractures can be clinically significant in the setting of concomitant pseudarthrosis, pain, or wound issues. Typical indications for reoperation after implant failure related to cage placement include severe migration, improper placement, or endplate destruction that would increase the risk of pseudarthrosis. Reoperation for screw issues as described above may be indicated with either radiographic findings or clinically significant symptoms.
Flatback Syndrome
Flatback syndrome was first described by Takemitsu in 1988. Flatback syndrome is defined as loss of normal lumbar lordosis. This may result in a pelvic incidence and lumbar lordosis mismatch, with symptomatic low back, hip, and thigh pain from sagittal imbalance. Patients with flatback syndrome attempt to compensate for an increased sagittal vertical axis (SVA) with pelvic retroversion and increased pelvic tilt. Typical presenting symptoms include difficulty with horizontal gaze, debilitating pain, early fatigue, and an exaggerated forward stooped posture. The condition has been described in people who have prolonged standing or stooped posture during work activities. The loss of disc space height and multilevel wedging of vertebral bodies can progress to flatback syndrome or lumbar kyphosis ( Fig. 4.3 ). However, a significant portion of presenting patients come to evaluation with iatrogenic flatback syndrome from prior fusion. There are several potential etiologies of flatback syndrome, including inadequate lumbar lordosis induced into a surgical construct or adjacent segment degeneration with disc space collapse and loss of lordosis at unfused lumbar segments. Additionally, it has been shown that transforaminal lumbar interbody fusion (TLIF) may lead to loss of segmental lordosis, resulting in flatback syndrome. In the era of distraction Harrington rod placement for adolescent idiopathic scoliosis, many patients also went on to develop flatback syndrome secondary to the distraction that was performed for coronal plane correction. Such patients had compensated flat backs because they were young, but as they aged, their compensation ability declined, resulting in symptom manifestation years after their index surgeries. The most common indication for intervention is significant impairment of quality of life in patients who have exhausted nonsurgical care and no longer have compensatory mechanisms (by increasing their pelvic tilt) to stand upright. When compensatory measures—hyperextension of unfused segments and pelvic retroversion—have been exhausted, it has been shown that sagittal imbalance then becomes an important driver of pain and disability. Bess et al. have reported that patients with flatback syndrome have significant impairment in health-related quality of life comparable to patients suffering from such chronic diseases as autoimmune conditions, heart failure, and diabetes. Glassman et al. have also reported that symptom severity concomitantly increases with worsening positive sagittal imbalance. These patients generally warrant operative intervention for correction of their iatrogenic deformity because their quality of life is so impaired. Additional indications for surgery include symptomatic lumbar stenosis with radiculopathy or neurogenic claudication resulting from malalignment and degenerative disc space collapse that is refractory to conservative measures. However, the decision to operate must include an appropriate evaluation of age, comorbidity status, bone density, and the tolerance of the patient and family for the high-risk flatback deformity correction. In a study evaluating age and complications following adult spinal deformity correction, Smith et al. reported a 71% rate of major and minor complications in patients aged 65 to 85 years compared with a 17% complication rate for 24- to 44-year-old patients. Although the complication rate of these operations is significant, elderly patients were found to have the most significant improvement in Oswestry disability index (ODI) and leg pain scores. Ultimately, the decision to operate on symptomatic flatback syndrome must account for patient disability, risk tolerance, realistic expectations, and patient and family understanding of the high complication rates.
