Introduction

It is estimated that 1.4 million vertebral compression fractures (VCFs) occur annually, causing pain and disability in patients worldwide.¹ The lifetime risk of a vertebral compression fracture in White women is 16%; in men, it is 5%. Historically, treatment of these fractures has been limited to analgesics, bed rest, and bracing. However, recently the development of vertebroplasty and kyphoplasty has provided physicians with additional treatment options for select vertebral compression fractures.

History of Vertebroplasty

Vertebroplasty was initially developed as an open procedure designed to augment the purchase of pedicle screws and to fill large voids from tumor resection. In 1984, however, at the University Hospital of Amiens, France, Galibert and Deramond performed the first documented percutaneous vertebroplasty.² The patient presented with severe cervical pain, and imaging demonstrated a large vertebral hemangioma encompassing the entire vertebral body of C2 with extension into the epidural space. After performing a C2 laminectomy to excise the epidural component of the lesion, a 15-gauge needle was inserted into the C2 vertebral body via an anterolateral approach, allowing injection of cement for structural reinforcement. The document of this case, as published in 1987, reports complete pain relief in this patient. Physicians at University Hospital Lyon continued to refine the percutaneous vertebroplasty technique as well as to expand its indications, using 18-gauge needles to inject polymethylmethacrylate(PMMA) into four patients with compression fractures. Since then, its popularity has spread dramatically.

Patient Selection/Indications

As with any procedure, the success of vertebroplasty relies heavily on the selection of appropriate patients and the skill of the operating physician. It is essential to identify patients with pain related to VCF, and exclude the other common sources of back pain in this population, including degenerative disk disease, spinal stenosis, facet arthropathy, or SI joint dysfunction. This process begins, of course, by taking a thorough history of the patient. It is particularly important to ascertain details regarding the timing of the onset of back pain, any known precipitating events, and those activities that worsen and alleviate the pain. Additionally, patients should be questioned regarding previous episodes of similar back pain, and the time until resolution of those symptoms. It is vital to understand the premorbid condition of the patient, as well as the impact of the back pain on activities of daily living. Finally, an assessment for allergies, anticoagulants, and medical problems, especially respiratory compromise, is essential to anticipate potential complications during the procedure. A thorough physical examination seeks to identify pain and tenderness to palpation at the level of radiographic abnormality. During the examination, the operator must pay attention to symptoms that may suggest pain from alternative sources.

Radiographic imaging plays an important role in the screening of patients for vertebroplasty. X-rays are often the first mode of imaging employed, due to their cost-effectiveness and ease of availability. In patients with a VCF, diffuse osteopenia and evidence of one or more compression fractures may be present. With neoplastic compression fractures, it may be possible to see focal lytic lesions or destruction of the bony trabeculae. A CT scan may be obtained for improved visualization of bony details. Used most often with pathologic fractures, CT may demonstrate expansion of the bony contours of the vertebrae and multilevel disease, both of which suggest an underlying malignancy. Preprocedure CT scanning also allows the operator to assess the integrity of the posterior wall of the vertebral body and pedicles, which if destroyed may be a source for significant complication.

MRI is particularly useful in the screening of patients with osteoporotic VCF due to its reported ability to discern the relative age of the fracture. Acute or subacute osteoporotic fractures up to 30 days old typically show evidence of bone marrow edema, with hypointense signal on T1-weighted and hyperintensity on T2-weighted and STIR sequences. At approximately 1 month after fracture, VCFs variably become isointense to normal bone marrow on T1- and T2-weighted sequences. Fully healed fractures are isointense to normal bone elements, or hypointense on T1 and T2 due to significant sclerosis. Recent studies have found a positive correlation between the MRI findings suggestive of a fracture less than 30 days old and clinical pain relief after vertebroplasty.

Interpretation of MRI findings in a patient with a malignant VCF is more challenging. While STIR sequences with fat suppression may be helpful to show edema, there may be heterogeneous or diffuse areas of hyperintensity on STIR or T2-weighted imaging. Some authors have suggested a pattern of hypointensity or isointensity on diffusion-weighted sequences. In any case, evidence of abnormal signal in the posterior elements, an expansion of the contour of the vertebral body or posterior elements, or any associated epidural/extravertebral soft tissue mass suggests an underlying malignancy.

Bone scintigraphy may be employed to detect a relatively recent fracture in patients who cannot tolerate an MRI. Increased radiotracer uptake has been correlated with positive clinical response to vertebroplasty. However, this technique is limited by the fact that the bone scan may show increased tracer uptake for up to 12 months after fracture. Thus this method must be correlated with corresponding anatomic imaging.

After a careful history, physical examination, and assessment of radiographic imaging, the physician must then determine not only that the source of the patient’s pain is indeed a VCF, but also that this fracture is amenable to vertebroplasty. The primary indication for vertebroplasty is the alleviation of pain associated with a VCF due to osteoporosis or tumor. Repeated studies have demonstrated superior pain relief with treatment of acute or subacute fractures. Perhaps most notable is the non–industry-sponsored, double-cohort by Alvarez et al³, which compared vertebroplasty to nonoperative treatment for VCFs. He found statistically significant differences at 3 months follow-up. Wardlaw et al⁴ published a randomized controlled trial comparing balloon kyphoplasty with nonsurgical care for VCFs. He too demonstrated a significant improvement in the intervention group at 1 month. Some now advocate the treatment of VCF within days of injury if the pain is so severe as to require parenteral narcotics and hospitalization. Late treatment, 6 months to years after the initial injury, is less likely to completely relieve pain, but symptomatic improvement has been noted in some studies.

Technique

As with any procedure, the process begins with proper patient selection (as discussed earlier) and the procurement of informed consent. A thorough discussion of the risks and benefits of the procedure is not only an ethical necessity, but it also helps to assess and modulate the patient’s expectations of clinical results.

Prior to the procedure, a thorough review of the patient’s medical history (to allow the identification of potentially complicating disease processes), medications (to assess for anticoagulants), and physical examination allows for necessary procedural modifications to ensure the safety of the patient. Once optimized, the patient may be brought to the procedure suite. Preoperative antibiotics, typically one gram of cefazolin, should be given 30 minutes prior to the commencement of the procedure.

Anesthesia for vertebroplasty is commonly a combination of conscious sedation and a local anesthetic. It is often advantageous to administer partial doses of sedation prior to positioning to decrease patient discomfort and anxiety. General anesthesia may be used if the patient is unable to tolerate prone positioning with only sedation due to pain or psychological disability; however, this adds risk and substantial cost to the procedure. Once secured in the prone position, radiographic imaging must be properly aligned. Fluoroscopy is the most commonly used modality. While biplanar fluoroscopy machines are now available, allowing fast real-time visualization of the procedure, these are expensive and are not readily available to many physicians. If single-plane imaging is used, it is imperative to obtain orthogonal projections to allow reliable assessment of needle positioning. CT scanning may be used as an adjunct to fluoroscopy: however, alone it does not allow for real-time monitoring of needle placement or cement injection. Also, it may require general anesthesia to limit patient movement. Indications for CT scan use include cervical or high-thoracic fractures necessitating the visualization of the carotid/jugular complex and vertebral vessels, sacral insufficiency fractures, and pathological fractures with risk of tumor displacement. In these cases, if fluoroscopy is not used, then cement must be injected in very small aliquots and scans should be performed frequently to assess for leakage.

The approach should be determined preoperatively based on the location of the lesion and its etiology. Preoperative CT scanning may be employed to help make this assessment.

Injection Materials

The ideal filler material for use in vertebroplasty and kyphoplasty must demonstrate good biocompatibility, adequate biomechanical strength and stiffness, and radiopacity for use in fluoroscopically guided procedures. Additionally, the material must be amenable to easy preparation, and possess appropriate flow and polymerization or crystallization characteristics.

In the first vertebroplasty procedures, PMMA bone cement mixed with a contrast agent, typically barium sulfate, was injected into vertebral bodies under image guidance.⁵ Since Charnley first reported its use in 1960, PMMA bone cements have been used by orthopedic surgeons for the fixation of both plastic and metal components in joint replacement and, less often, in the stabilization of pathological fracture. Early studies showed maintenance of the bond between the prosthesis and the PMMA with no evidence of harmful systemic effects. Thus, PMMA is now widely used throughout orthopedics.

Advantages to PMMA include its familiarity for operating physicians, its ease of handling, and its cost-effectiveness. Also, PMMA shows good biomechanical strength and stiffness and evidence that it is relatively bioinert. For this reason, as of April 2004, the FDA has approved the labeling of certain brands of PMMA for the treatment of pathological fractures of the vertebral body resulting to osteoporosis and tumor. However, several disadvantages to PMMA have become apparent. Perhaps most notable is PMMA’s lack of osteoconductivity. As such, there is no potential for remodeling or integration into the surrounding bone. Histologic studies have reported a thin fibrous membrane surrounding the PMMA injected into vertebral bodies, providing further evidence of the lack of osseous integration. Therefore, PMMA relies solely on the bulk effect of injected cement for strength and stability. Additionally, there have been theoretical concerns regarding the high polymerization temperature of PMMA, though to date there has been no clear evidence to support this. Finally, as is well documented in the arthroplasty literature, PMMA is associated with potential monomer toxicity. The molecule is known to be arrhythmogenic and cardiotoxic at the volumes used in knee and hip replacement. For this reason, many authors recommend limiting vertebroplasty or kyphoplasty to two or three levels at any surgical setting.

The limitations of PMMA cement have led researchers to seek alternative filler materials. The primary characteristic of these novel products is their osteoconductivity. The best studied of these synthetic bone substitutes is the class of calcium phosphate cements. As osteoconductive agents, these possess the potential for resorption of cement and replacement with new bone, effectively restoring vertebral body bone mass. Studies have shown evidence of osteoclastic resorption of the cement and fragmentation with vascular invasions and bony ingrowth.⁶ Histologic results show direct bone apposition suggestive of remodeling. Like PMMA, calcium phosphate fillers initially function as bulk-filling agents. However, due to their osteoconductive capabilities, their strength is gradually reinforced by new bone formation. Biomechanical testing of calcium phosphate cements has verified their ability to restore the mechanical integrity of the vertebral body.

Calcium sulfate, also known as plaster of Paris, has been investigated as a potential filler material. Long used as a bone graft substitute, calcium sulfate is injectable, osteoconductive, and cures with a limited exothermic reaction. Histologic and radiographic analysis has shown progressive resorption of the cement and osteoblastic rimming of the newly woven bone. However, there is concern that the material is too rapidly resorbed, leading to lack of stability during the remodeling process.

Calcium phosphate and calcium sulfate cements share several common problems. Both materials have a low viscosity as well as handling characteristics that are different from PMMA and thus are unfamiliar to most orthopedic surgeons. The cost of these products is also well above that of PMMA. Finally, these products, as ion suspensions, have thixotropic properties. Thus, the material is susceptible to separation within the delivery tube, making injection difficult.

Finally, novel composite materials, such as the cross-linked resin and glass ceramic particles of Cortoss, by Orthovita, have been approved by the FDA as potential alternative fillers. Proposed advantages of these materials include constant flow characteristics, inherent radiopacity, lower polymerization temperature, and mechanical strength properties that exceed those of PMMA. Animal studies have demonstrated its osteoconductive capacity. The potential for these composite fillers is still being defined.

Complications

In a patient with an osteoporotic VCF treated by percutaneous vertebroplasty, the incidence of complication necessitating surgical intervention is estimated to be less than 1%.⁷ In patients with neoplastic VCF, 2.7% to 5.4% will require surgery to manage a complication of vertebroplasty. In this population, less significant complications are estimated to occur in up to 10% of patients. This increased risk is likely due to an increased risk of cement extravasation due to cortical breaks in the vertebral body.

The most common complication of vertebroplasty is approximately 72 hours of mild local tenderness. More severe pain localized to the needle site may be due to hematoma or bruising. This can be minimized with 5 minutes of manual compression after removal of the cannula. More common in patients with an underlying malignancy, is dermatomal or radicular pain. Typically, no specific treatment is needed and NSAIDs are used to treat pain. Occasionally, a brief course of either oral or local steroid injections may be necessary to relieve the pain. Such complications may be monitored and treated conservatively so long as there are no associated motor deficits or bladder or bowel incontinence. The etiology of radicular pain may be from spinal cord or nerve root compression due to retropulsion of tumor fragments or extravasation of cement. At its worst, paraplegia may occur by this mechanism.

Extravasation of cement into the epidural veins can cause a cement embolism to the lung. As in hip and knee arthroplasty, the pressurized injection of cement into the vertebral body can also cause a fat embolism. While the majority of these emboli are asymptomatic, they can be particularly problematic in patients with pre existing pulmonary conditions, such as COPD. Further respiratory complications can be induced by inaccurate placement of the needle, which can cause a pneumothorax. As in all invasive procedures, there is a risk of bleeding, which is more common in the parapedicular approach due to the large paraspinous vessels. Infection is exceedingly rare. Finally, there have been reports of deaths attributed to vertebroplasty and kyphoplasty. These seem to be due to severe cement allergy or pulmonary failure in patients with preoperative pulmonary compromise.

Nejm Randomized Controlled Trials

The August 6, 2009 issue of the New England Journal of Medicine presented two randomized studies seeking to assess the efficacy of vertebroplasty for pain relief in osteoporotic vertebral fractures. In Buchbinder et al,⁸ enrolled patients with one or two painful osteoporotic VCFs less than 12 months old and unhealed, as confirmed by MRI, were randomized to either vertbroplasty or a sham procedure. Outcomes were assessed up to 6 months. They concluded that there was “no beneficial effect of vertebroplasty over a sham procedure at 1 week or at 1, 3, or 6 months among patients with painful osteoporotic vertebral fractures.” Kallmes et al⁹ randomly assigned 131 patients with one, two, or three painful osteoporotic VCFs thought to be less than 1 year old to either vertebroplasty or a similar sham procedure. Outcomes were assessed up to 3 months. Of note, MRI was only employed if the age of the fracture was “unknown.” This study concludes that “improvements in pain and pain-related disability associated with osteoporotic compression fractures in patients treated with vertebroplasty were similar to improvements in a control group.”

Upon further examination of these studies, several important criticisms have been raised.¹⁰ These are discussed below.

Conclusion

The introduction of vertebroplasty, and now kyphoplasty, has provided physicians with additional options for the treatment of VCFs. Specifically, vertebroplasty has been shown to be indicated for the acute and subacute treatment of VCFs due to osteoporosis and malignancy. Contraindications to the procedure include asymptomatic VCFs; painful VCFs that are improving with conservative medical treatment; traumatic VCFs in the young, nonosteoporotic patient; and patients with osteomyelitis, an uncorrectable coaguloapthy, or an allergy to any component of the procedure.

While the technique and its indications are continuing to evolve, numerous studies, including Alvarez et al, have suggested that vertebroplasty is a successful and safe procedure to alleviate the pain associated with acute and subacute osteoporotic or neoplastic VCFs. The recent Kallmes and Buchbinder articles do not support the previous robust benefits demonstrated in other studies; however, their efficacy must be questioned in light of the numerous criticisms stated above, most notably the use of a local anesthetic injection as an unproven sham procedure.