Introduction
The term “adult spinal deformity” generally refers to any major coronal, sagittal, or biplane malalignment of the spine in a patient over the age of 18. The revision rate for primary adult deformity surgery patients is approximately 9% to 25%. Moreover, the revision rate following a revision adult spinal deformity procedure has been shown to be as high as 21%. Despite improvements in technology, including improved instrumentation and bone graft extenders, there has not been a trend toward lower revision rates. This is likely because improved techniques and medical management has allowed the deformity surgeon to take on more and more complex cases in older patient populations. This chapter aims to give an overview of revision surgery for the spinal deformity patient and arm the surgeon with the framework to both plan and execute these challenging cases.
Patient Assessment
Assessment of the revision deformity patient begins with a thorough history and physical examination with a keen understanding of the presenting complaint. Most commonly, this consists of pain located either regionally in the lower back, remotely in the lower extremities, or in the head. Local back pain may be a result of pseudarthrosis, fracture, infection, instrumentation failure, flatback syndrome, or adjacent segment degeneration. Lower extremity pain may result from new stenosis as a result of adjacent segment pathology or from a previous inadequate decompression or pseudarthrosis. Spinal deformity may be caused by fracture, most commonly a spondylolisthesis resulting from a fracture of the pars interarticularis secondary to an aggressively wide previous decompression. Other types of pain such as headache may occur from indolent cerebrospinal fluid leaks. The examiner should specifically ask about risk factors for pseudarthrosis including prior radiation exposure, use of disease modifying antirheumatic drugs, or heavy steroid use. Posture and decreased functional activity with mechanical back pain symptoms are other common presenting complaints particularly in those with iatrogenic flatback syndrome who cannot stand upright. Patients with iatrogenic coronal or sagittal deformity may complain of being stooped over and have limited walking capacity. In the history, particular attention should be given to the sequence of previous operations and their results. Did the patient improve following their previous surgery? If so, for how long and when did their current pain start?
Examination should begin with general assessment of the patient including gait and posture. Gait should be assessed for cadence and signs of any pathological changes including foot drop or Trendelenburg lurch. Posture should be assessed from both the front and side of the patient. The patient should be asked to stand as straight as possible with their hips and knees straightened. This should remove compensatory mechanisms for sagittal imbalance such as knee flexion and hip extension, accurately exposing the full extent of the spinal deformity. Furthermore, patients should be evaluated in the coronal plane, first by measuring for a leg length discrepancy. Apparent leg length (not accounting for soft tissue contractures and pelvic obliquity) may be measured using a tape measure from the umbilicus to the medial malleolus whereas true leg length may be measured from the anterior superior iliac spine to the medial malleolus. This would be done in addition to a scanogram for those with true leg length discrepancies. A complete neurological examination including strength and sensation should also be performed in conjunction with a reflex examination to rule out signs suggestive of thoracic myelopathy from adjacent segment degeneration with prior lumbar fusions.
Radiological assessment begins with 36-inch scoliosis films with both anteroposterior (AP) and lateral views with the patient standing with their arms in the supraclavicular fossa. Full body EOS thoracic kyphosis may also be useful for determining compensatory mechanisms of the pelvis and lower extremities. Additionally, supine or prone lateral films, AP films, and left and right bending films can help determine the flexibility of any deformity present. The respective curvatures of the spine should be measured including lumbar lordosis (LL), thoracic kyphosis, and cervical lordosis. LL should be measured from the cephalad end-plate of the L1 vertebra and the caudal end-plate of L5. Thoracic kyphosis should be measured from the cephalad end-plate of T4 and the caudal end-plate of T12. LL may be classified into four types based on the Rousslouly system. Sagittal alignment including the spinopelvic parameters should then be carefully measured. The most commonly used measurement is the sagittal vertical axis (SVA) which measures the horizontal distance (cm) between a C7 plumb line and the posterior superior corner of the S1 vertebral body and may either be positive (anterior to this point) or negative (posterior to this point). Sagittal alignment may also be measured using the T1 spinopelvic inclination (T1SPI), which is formed by an angle between a line connecting the center of the T1 vertebra and the bicoxofemoral joints and a vertical line. T1SPI is useful as it does not require calibration of the x-ray for magnification. The spinopelvic parameters most commonly measured are the pelvic incidence (PI), the pelvic tilt (PT), and the sacral slope (SS). The PI is a fixed parameter which measures the orientation of the sacrum relative to the pelvis and generally does not change after skeletal maturity. It is measured by an angle formed by a line connecting the center of the femoral heads to the center of the sacrum and a line perpendicular to the S1 end-plate. PT is measured by an angle formed between a line connecting the femoral heads to the center of the S1 end-plate and the vertical axis. SS angle is measured by line drawn along the superior S1 end-plate and the horizontal axis. Normal values for the SS, PT, PI, and LL have been described previously by Vialle et al. Surgeons should keep in mind the effect age has on sagittal parameters and that the target normal values change as the patients age increases. Specifically, lordosis tends to decrease with increasing SVA, PT, and TK.
Coronal alignment should be assessed by measuring the distance between a C7 plumb line and the center sacral vertical line (CSVL). In the coronal plane, pelvic obliquity is measured on an AP radiograph as an angle between a horizontal line and a line connecting two identical points on the pelvis (bilateral iliac crests, or most superior parts of the greater sciatic notch). Pelvic obliquity may be primary (i.e., caused by leg length discrepancy or hip osteoarthritis) or secondary (resulting from compensation to a severe lumbar coronal curvature). Thus, if pelvic obliquity is related to a leg length discrepancy, a shoe lift may fully or partially correct a flexible scoliotic lumbar curve and repeat x-rays should be performed with the lift.
Computed tomography (CT) scans should be used to assess the fusion status and screw placement of previous constructs. The CT scan of the pelvis should also be scrutinized for previous iliac crest harvests, in the event that iliac screws or further harvests are planned. If supine scoliosis x-rays are not available, the scout from the CT image may be used in its place to assess flexibility of the curve in the sagittal and coronal plane. CT myelography may be used for patients with metal artifact to gain an understanding of stenotic regions as well as the bony morphology. Magnetic resonance imaging (MRI) may be useful, particularly if done with contrast, to assess for previous epidural scarring and areas of stenosis.
Indolent infections may be a cause of chronic pain in revision cases and thus must be ruled out with careful laboratory evaluation including complete blood counts, erythrocyte sedimentation rate (ESR), and C reactive protein (CRP) when there is clinical suspicion. However, up to 9% of patients undergoing instrumentation revisions who are clinically asymptomatic may have an occult infection. Only roughly 40% to 46% will have an elevated CRP or ESR. Late infections tend to occur from indolent infections such as Propionibacterium acnes .
Approach Considerations
Epidural scarring may cause a significant roadblock to the revision deformity surgeon. Thus it is prudent to review previous approaches, and plot treatments that avoid going directly through scar tissue (i.e., an anterior approach if a previous posterior approach was performed). Moreover, if planning to remove a transforaminal lumbar interbody fusion (TLIF) cage, an anterior approach can be particularly useful. If going through the scar tissue is required, a systematic dissection should be employed. First the scar over midline should be left intact, as adjacent levels are exposed and previous instrumentation removed if necessary. Using care while removing hooks is critical because these may lie directly on dura and may have caused erosions and thinning. Following this, the adjacent levels should be instrumented. If landmarks are obliterated, fluoroscopy or navigation may be used. Healthy dura should be exposed above and below the epidural scar. If an osteotomy is planned, the scar should be excised by carefully developing a plane between scar and dura and peeling it off using a combination of Leksell rongeur and Cobb periosteal elevators. Epidural scar may fold during the closure of an osteotomy, compressing neural elements, and thus must be removed. If the dura is attached to the bone, an angled curette may be useful to develop a safe interval. However, if this is impossible, using a burr to approach from lateral to medial is another option.
Surgeons should remember that the strongest risk factor for a dural tear is revision spine surgery. Moreover, attempting to perform an osteotomy through a large bed of epidural scarring can increase the risk for neurological complication, particularly in the thoracic spine.
Sagittal Considerations
Flatback Syndrome
Flatback syndrome results from the failure to maintain normal lordosis. It may either be primary, occurring as a result of disc degeneration, or secondary following an arthrodesis or decompression procedure which results in a positive sagittal balance. This syndrome was first described by Doherty in a patient who had undergone Harrington distraction instrumentation for fusion of scoliosis with resulting loss of lordosis. Patients generally complain of an inability to stand erect, back pain which worsens with upright activity, and even pain in the neck from compensatory hyperlordosis. Flatback syndrome may be the result of primary or secondary etiologies. Primary etiologies include diseases such as ankylosing spondylitis, degenerative scoliosis, and congenital abnormalities. The etiology of secondary flatback syndrome include hyperlordotic fusion, distraction instrumentation, pseudarthrosis with loss of correction, and collapse of the lumbar vertebral body from fracture and postlaminectomy kyphosis. More recently, standalone interbody devices inserted into the lumbar spine have been described to occasionally result in graft subsidence and sagittal decompensation. The goals of spinal realignment should be to correct the SVA to less than 50 mm, PI-LL to less than 10 degrees, and the PT to less than 20 degrees. Because the PI is a fixed pelvic parameter which does not change with position and is proportional to the LL, it may be used as a target for lordosis restoration. In planning reconstructive surgery for sagittal malalignment, a clear understanding of compensatory mechanisms is critical. Pelvic retroversion, which is a compensatory mechanism that results in an increased PT in an attempt to maintain sagittal alignment, must be accounted for, as the sagittal deformity may be much greater than it appears. Thoracic hypokyphosis is another compensatory mechanism which can recoil following a corrective surgery, in one study increasing on average 13 degrees following a lumbar pedicle subtraction osteotomy (PSO). Roussouly et al. suggest that their type 4 lordosis patients, those with an exaggerated ratio of lordosis length to kyphosis length (20:40), are at increased risk for reciprocal changes following an osteotomy realignment.
In patients with long sweeping deformities or flexible deformities, Smith-Peterson osteotomies (SPOs; resection of the posterior column between facet joints) may be used at multiple levels to correct sagittal alignment. In patients where the deformity is rigid (so-called fixed sagittal imbalance) or focally angular, the use of a PSO may be necessary and can achieve up to 30 to 40 degrees of sagittal correction. In general, a surgeon can achieve approximately 10 cm of sagittal correction. Similarly, a single PSO generally obtains up to 30 degrees of sagittal correction and is equivalent to three SPOs. However correction of the SVA is significantly less with three SPOs (5 cm) compared with a single PSO (11 cm). Osteotomy placement is critical, as the lower osteotomies in the lumbar spine have a greater capacity to reduce PT compensation. Care must be taken if a patient demonstrates increased SVA without increased compensatory PT, as this could indicate a hip flexion contracture, severe lumbar stenosis, or a focal or systemic problem with the extensor muscles. Moreover, overaggressive osteotomies can be deleterious in the elderly, as they are unable to recalibrate to dramatic changes in their sagittal alignment. In such cases, bringing the SVA over the femoral heads is usually sufficient.
Pseudarthrosis
Pseudarthrosis may contribute or coexist with flatback syndrome. Indeed, spinal malalignment may predispose a patient to pseudarthrosis. In such cases, the pseudarthrosis may be exploited to generate lordosis without the use of an osteotomy, through an anterior approach combined with a posterior approach. In a series of 38 patients with lumbar pseudarthrosis, anterior-posterior revision showed a very low persistent rate of pseudarthrosis for both structural autografts (0%) and allografts (6%) and was able to achieve a good correction of flatback syndrome deformity at 13 degrees average (range 4 to 53 degrees).
Proximal Junctional Kyphosis
Proximal junctional kyphosis (PJK) is defined as kyphosis at the proximal end of a construct greater than 10 degrees and is a cause of revision surgery owing to sagittal decompensation. There is no clear consensus on its etiology, but factors such as overcorrection, osteopenia, fracture, posterior tension band destruction, construct rigidity, pelvic fixation, and age have all been implicated. Consideration should be given to the anchor type at the upper instrumented vertebra. The use of hooks has been associated with a decreased risk of PJK compared with pedicle screws at the upper instrumented vertebrae (UIV). The rate of acute proximal junctional failure (an increase in proximal kyphosis of 15 degrees or more, UIV or UIV+1 fracture, or failure of upper instrumentation) after adult spinal deformity is 5.6%. Failures occur most commonly as a result of fracture or soft tissue failure. Fractures of the UIV have been linked to increased UIV angles (the angle of the UIV relative to the horizontal in the sagittal plane).
Indications to perform revision surgery for PJK include progressive deformity, intractable local back pain, or new neurological deficit. In general, the patient’s deformity will reduce with prone positioning; however, 10 to 15 lbs of Gardner-Wells traction may be applied to facilitate the reduction. Fusing to beyond the apex of kyphosis is typically necessary. Kim et al. reviewed a series of 32 patients who required revision for PJK and had a 2-year follow-up period. Both Oswestry Disability Index (ODI) and Scoliosis Research Society (SRS) questionnaire scores improved significantly from prerevision to final follow-up; however, 16% required a second revision surgery for further PJK.
Coronal Considerations
Patients with a pelvic obliquity resulting from a leg length discrepancy may have a compensatory coronal curve that can be rigid or flexible. Those which are flexible may correct with the addition of a shoe lift, whereas rigid curves will not. In rigid curves, the correction should either balance the spine perpendicular to the oblique pelvis, essentially ignoring the pelvic obliquity, or correct the spine to a level pelvis if future correction is planned or a shoe lift is tolerable.
Complications
The overall short-term and long-term complication rates between revision and primary procedures for deformity correction appear to be similar. Lapp et al. showed that the overall complication rates for revision versus primary anterior-posterior lumbar fusion for deformity were similar, with primary procedure patients having an increased functional and diminished pain level compared with revision patients. Despite this, revision patients had superior levels of satisfaction. The long-term complication rates in this study were 44% for primary procedures and 35% in the revision group. The average decrease in pain was 42% in the primary group versus 49% in the revision group, whereas the overall functional score showed 80% improvement in primary patients versus 71% in the revision patients. However, in a nationwide database study, Diebo et al. showed that revision spinal deformity surgery patients had higher procedure-related complications, neurological complications, infection and wound-related complications, and longer hospital courses compared with primary fusion patients. Instrumentation failure at the lumbosacral junction is important in long fusion constructs and is a common site of failure. Failure of S1 screws or L5–S1 cage collapse is particularly common in patients with long fusions to the sacrum without distal fixation (bilateral iliac screws). Thus using anterior interbody devices in combination with bicortical S1 screws and bilateral iliac screws should be considered in long fusions to the sacrum. Additionally, Annis et al. showed in a retrospective series that the use of low dose BMP2 (2.5 mg) and iliac fixation at the L5–S1 junction made the use of an interbody device at this level unnecessary.
Pseudarthrosis may occur at anywhere from 10% to 21% and may occur commonly at osteotomy sites. Another potential complication with realignment surgery is serious neurological injury. Buchowski et al. noted that 11% of patients undergoing a PSO sustained an injury such as loss of motor strength of two or more grades or loss of bowel and bladder control, which were permanent in 2.8% of cases.
Conclusion
In conclusion, revision of previous adult spinal deformity surgery can be especially challenging; however, with careful planning and meticulous technique these patients can be some of the most satisfied in spine surgery. As more primary fusions are performed, the number of revision cases will continue to grow, necessitating improved approaches to dealing with these cases.
References
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
