Orthopedics

Materials and Methods

Between January 2005 and June 2010, the authors treated 18 patients with symptomatic osteoporotic vertebral compression fractures adjacent to lumbar instrumented circumferential fusion. The patient sample included 15 women and 3 men with an average age of 69.9 years (range, 53–85 years). Lumbar instrumented circumferential fusion involved traditional posterolateral fusion and posterior lumbar interbody fusion with pedicle screw fixation. The adjacent osteoporotic vertebral compression fracture was identified on radiographs and magnetic resonance imaging (MRI). All patients reported severe back pain that was not relieved by conservative therapies administered for more than 1 month, such as bed rest or the use of analgesics and braces. Fourteen patients reported minor trauma, such as a fall or slip. However, the remaining 4 patients reported no significant spinal trauma leading to the development of osteoporotic vertebral compression fractures. Exclusion criteria included patients with coagulopathy, infection, or major medical disease that were not suitable for percutaneous vertebroplasty. Patients with neurological deficit or evidence of adjacent stability or stenosis were also excluded. The 18 patients underwent percutaneous vertebroplasty with polymethylmethacrylate (PMMA) bone cement augmentation and were followed up for at least 18 months (range, 18–39 months). Dual-energy x-ray absorptiometry (DEXA) was used to measure bone mineral density before percutaneous vertebroplasty in each patient.

Percutaneous vertebroplasty was performed under local anesthesia. The vital status of the patient, including heart rhythm, blood pressure, and pulse oxygenation levels, was continually monitored by an anesthesiologist. The patient was placed in the prone position. Pillows supported the upper chest and pelvis to enable maximum spinal column extension. This postural reduction generally restored the body height of the fractured vertebrae. After careful skin sterilization and local anesthesia, an 11-gauge bone biopsy needle (Stryker Instruments, Kalamazoo, Michigan) was inserted using a mallet through the mid-portion of the pedicle into the collapsed vertebral body under fluoroscopic guidance. The needle tip was advanced until it reached the anterior one-third of the fractured vertebral body. The PMMA cement was prepared when the needle placement was adequate.

Forty grams of bone cement (Osteobond; Zimmer, Warsaw, Indiana) was mixed with 6 g of barium sulfate (Barytgen; Fushimi, Kagawa, Japan) for delivery into the fractured vertebrae. The cement mixture was polymerized at room temperature until it had the paste-like consistency necessary for controlled injection. The bipedicular approach was adopted for adequate distribution of bone cement and interdigitalization with the cancellous bone. The PMMA was slowly and continually injected through each pedicle using a 10-mL plastic syringe and a special screw syringe compressor (Biomedical Engineering Lab of E-D Hospital, Kaohsiung, Taiwan) (Figure 1). Injection of the PMMA bone cement was stopped when the cement filled the posterior one-third of the vertebral body or when the syringe compressor presented unusual resistance. Once injection was completed, the needle was removed, and hemostasis was achieved at the puncture site by applying gentle pressure.

Figure 1: The whole structure (A) and separated parts (B) of a special screw syringe compressor with a pressure sensor designed to detect unusual resistance during cement injection.

Most patients were discharged on postoperative day 1. All patients wore a rigid orthosis for 3 months. After discharge, the patients received regular follow-up at 1, 3, 6, and 12 months and annually thereafter. Clinical outcomes and radiographs were evaluated, and data were collected by the senior author (S.-C.Y.), who was the operating surgeon. During the follow-up period, all patients received alendronate or medication for osteoporosis, adequate calcium supplement, and nonsteroidal anti-inflammatory drugs; the dosages were gradually reduced according to the patient’s condition.

Plain radiographs were obtained for all patients immediately postoperatively and at each follow-up. The extent of vertebral body collapse was measured from the height of the maximum collapse on lateral radiographs, at the shortest distance measurable between the superior endplate and the inferior endplate. The percentage of vertebral body collapse and restoration after percutaneous vertebroplasty was calculated by comparing with the predicted normal vertebral body height. The predicted height was estimated by averaging the heights of the adjacent 2 vertebrae above and below the fractured vertebrae.

Clinical outcomes were evaluated by asking patients to quantify their pain on a visual analog pain scale (VAS) at each follow-up appointment: 0 means no pain and 100 means the most pain possible. At 1-year follow-up, the patients’ conditions were evaluated on the basis of pain, activity, and analgesic use, which were categorized as excellent (4), good (3), fair (2), or poor (1) according to modified Brodsky’s criteria.¹¹ Patients were also asked whether the procedure had relieved the pain for which they were treated, if they were satisfied with the procedure, and whether they would undergo percutaneous vertebroplasty again if needed. Paired t test and Wilcoxon signed rank test were used to compare the clinical outcomes and radiolographic findings pre- and postoperatively. A value of P<.05 was considered statistically significant.

Discussion

Instrumented posterior lumbar interbody fusion, an alternative circumferential fusion, can restore disk height, decompress neural elements, immobilize the unstable intervertebral disk, and achieve anterior support for load bearing through a single posterior approach.^1–7 Recently, this method has been widely applied as an effective procedure for treating patients with degenerative lumbar spine disease. However, the development of pathology at the mobile segment next to the lumbar spinal fusion has been a complication of spinal fusion and has been termed adjacent segment disease.^8–10 Spondylolisthesis, instability, hypertrophic facet joint arthritis, herniated nucleus pulposus, and stenosis are reported frequently. Less commonly reported findings include scoliosis and vertebral compression fracture. Surgical treatment of adjacent segment disease usually focuses on the adequate decompression of neural elements and subsequent extension of the fusion. Patients with adjacent osteoporotic vertebral compression fracture usually present with intractable back pain with no symptoms of spinal stenosis. Therefore, no need exists for decompression. In addition, the number of instrumented fusions has increased to strengthen the securing effect in patients with osteoporosis.^12–14 This often requires a wider wound and more screws, thereby increasing operative and anesthesia time, risk, and cost.

Percutaneous vertebroplasty with PMMA is a minimally invasive technique that has been used to augment benign spinal compression fractures, hemangiomas, multiple myelomas, lymphomas, and vertebral metastatic lesions since the 1980s and is becoming the standard procedure for osteoporotic vertebral compression fractures because of its simplicity and efficacy.^15–18 Significant pain relief can be achieved in more than 90% of the patients with painful osteoporotic vertebral compression fractures, according to most studies.^19–22 Few reports have addressed adequate management in patients with symptomatic osteoporotic vertebral compression fractures following lumbar instrumented circumferential fusion.

The bone mineral density scores of all patients in the current study indicated osteoporosis. All fractures developed at the cranial vertebrae adjacent to the instrumented circumferential fusion segment. A high stress gradient seems to exist between a fused segment and the adjacent mobile segment. Therefore, a compression fracture can develop in an osteoporotic vertebra or in a relatively strong vertebra. Several studies suggest that bone mineral density decreases in the vertebra cranial to or at the level of the fusion segment.^23–25 An animal model of anterior and posterior column instability was developed to allow in vivo observation of bone remodeling and arthrodesis following spinal instrumentation. McAfee et al^23,24 demonstrated that the rigidity of the spinal instrumentation could lead to device-related osteoporosis of the vertebrae. Bogdanffy et al²⁵ studied 15 patients who underwent a combined anteroposterior L4-S1 spinal fusion and examined their bone mineral density using DEXA scan. They reported that the bone mineral density at L3, one level above the fusion, and at L2, two levels above the fusion, significantly decreased 3 months postoperatively and continued to decrease at 6 months postoperatively. These changes can be attributed to postoperative immobilization, altered biomechanics secondary to the arthrodesis.

Some previous reports described junctional fractures that developed after long instrumented fusion.^14,26,27 Hart et al²⁶ reported the use of prophylactic vertebral augmentation for the prevention of proximal junctional collapse cranial to multilevel lumbar fusion. They suggested that women older than 60 years undergoing lumbar instrumented fusion may benefit from prophylactic vertebroplasty or kyphoplasty. Lattig²⁷ used bone cement augmentation at the uppermost screws and the first mobile vertebra in multilevel adult deformity surgery to prevent adjacent segment failure. He concluded that additional vertebroplasty of the neighboring 1 to 2 noninstrumented levels in long spinal deformity fusion can reduce the number of more invasive revision procedures that are required later in elderly patients with osteoporosis. However, only 3 patients in the current study had 4-level fusion, and 15 patients underwent short instrumented fusion (Table 1). Some patients developed 2- or 3-level adjacent vertebral compression fractures. This may be attributed to the elderly patient group with rigid instrumented circumferential fusion.

The authors do not perform prophylactic vertebroplasty to avoid proximal junctional fractures because the true prevalence or incidence of developing vertebral compression fractures adjacent to the instrumented fusion remained unclear. Approximately one-third of the osteoporotic vertebral compression fractures are symptomatic and require intensive treatment. Percutaneous vertebroplasty is simple and effective, but it is not a risk- or complication-free procedure. Several acute complications and delayed sequelae are reported in the literature. The authors suggest that salvage procedures should be reserved for patients who require revision surgery.

A detailed history and physical examination must be conducted to determine the need for revision surgery. In the current study, intractable back pain was the dominant complaint, rather than sciatica or intermittent claudication. Radiographs initially indicated a collapsed vertebra adjacent to or neighboring the fusion segment. Preoperative MRI was crucial for obtaining more information and making an accurate diagnosis. Magnetic resonance imaging and radiographs should be performed to determine the need for percutaneous vertebroplasty and to achieve the best result. The clinical outcomes of the patients in the current study are encouraging and similar to those treated for simple or usual osteoporotic vertebral compression fractures. An average VAS reduction of 53 points occurred after percutaneous vertebroplasty. Two patients with 4-level fusion and 1 patient with 3-level fusion continued to have back pain that influenced their activities of daily living. Residual back soreness was commonly reported and is possibly caused by osteoporosis, flat back syndrome, and increased stress of neighboring motion after instrumented circumferential fusion. Some patients with more complicated adjacent segment disease, such as adjacent instability and stenosis or burst fracture with significant neural compression, were excluded from this study. A wider revision surgery with extension of the decompression and instrumentation was required for this excluded patient group.

Adjacent osteoporotic vertebral compression fractures are common after vertebroplasty or kyphoplasty with bone cement augmentation. However, few studies have addressed adjacent osteoporotic vertebral compression fractures after spinal fusion. Increased stiffness and rigidity of the lumbar instrumented circumferential fusion altered the load transfer biomechanics to the adjacent vertebrae. The early failure and further collapse of the adjacent nonfusion level may result from a stress-riser effect and a substantial disparity in biomechanical properties between the flexible and fixed lumbar spine. In a case-control study with a mean follow-up of 10 years, Toyone et al²⁸ assessed the long-term prevalence of vertebral fractures after lumbar spinal fusion with instrumentation. They reported that adjacent-level fractures developed within 8 months postoperatively and remote-level fractures developed between 8 and 22 months postoperatively. Ahn and Lee²⁹ analyzed the characteristics of adjacent vertebral compression fractures following lumbar spinal fusion. Great variation existed in the time from fusion to fracture, with an average of 38.1 months (range, 5–120 months) in their series. The intervals of index surgery and fractures in most of the current patients seem to concentrate within 1 year. The other 8 adjacent vertebral compression fractures occurred more than 2 years later.

The sample size in the current study was not large enough for the authors to perform meaningful multivariate analyses to accurately identify predictor variables. Although no statistical conclusion could be made, minor trauma during activities of daily living might be a potential and unpredictable risk factor that could explain the significant difference among these studies.

Conclusion

The elderly patients in the current study who underwent lumbar instrumented circumferential fusion were at risk of developing adjacent vertebral compression fracture because of the increased stress of the rigid fusion construct. Percutaneous vertebroplasty is a minimally invasive and effective method to treat this kind of adjacent segment disease. Performing preoperative MRI is important for clearly identifying the pathology. In patients with poor bone stock, postoperative education and medication to treat osteoporosis are mandatory for achieving good long-term results.