Article Text

Role of cementoplasty in the management of compression vertebral body fractures
  1. M Hamady,
  2. S Sheard
  1. St Mary’s Hospital, London, UK
  1. Dr M Hamady, St Mary’s Hospital, Praed Street, London W2 1NY, UK; mohamad.hamady{at}


Osteoporotic vertebral compression fractures cause pain, reduced mobility and consequently poor quality of life, and as such have a significant impact on health resources. Their prevalence can be expected to increase with the ageing population. Until recently, only conservative management has been available to alleviate pain and improve mobility. Originally developed in the 1980s to treat vascular malformation in the spine, vertebroplasty offers an interventional method for improving symptoms of vertebral compression fractures. Percutaneous vertebroplasty involves stabilisation of the fractured vertebral body using cement which is introduced via a needle under image guidance. Kyphoplasty is a more recent development in which a balloon is inflated within the fractured vertebral body in order to correct any loss of height before cement stabilisation. There is a lack of major randomised controlled trials on either procedure at present, but evidence of their safety and efficacy is increasing, suggesting that both are more effective than conservative management, with low risk of complications.

  • vertebroplasty
  • balloon kyphoplasty
  • compression fractures

Statistics from

Compression vertebral body fracture secondary to osteoporosis remains a significant clinical problem. In the UK an estimated 150 000 osteoporosis related fractures occur each year, resulting in an annual cost of approximately £750 million (€830 million, US$1100 million). There are 40 000 symptomatic vertebral body fractures each year. These cause substantial morbidity and mortality as well as social and health services expenditure.1 2 Moreover, the financial impact on the health service is expected to amplify, given the evidence of increasing age specific fracture incidence and the ageing population.3

Traditional treatment of painful compression vertebral body fracture includes oral and parental analgesia, back bracing and muscle relaxants.4 Calcitonin may also have some analgesic effect in these patients.5 Although most of the osteoporotic vertebral body fractures remain subclinical or respond to conservative treatment, some of them may continue to have severe back pain, restricted mobility and functional deterioration.

The comorbidities associated with vertebral body fracture include deep venous thrombosis, acceleration of osteoporosis, loss of height, respiratory problems and emotional problems secondary to chronic pain.

Vertebroplasty, and more recently kyphoplasty, have emerged as interventional methods to alleviate pain, prevent further vertebral body collapse and improve mobility, thereby shortening recovery time, eliminating the need for prolonged nursing and rehabilitation, and also cutting down the amount of analgesia required.

This article discusses the recent advances in interventional spine augmentation and pain relief and presents the current evidence and future directions.


Percutaneous vertebroplasty is a minimally invasive technique in which cement is injected under image guidance in the vertebral body to strengthen and to stabilise the fractured spine. Dr Deramond, a French interventional radiologist, performed the first vertebroplasty case in 1984 in France, for the treatment of a painful aggressive haemangioma of the spine.6 Since then the use of vertebroplasty has expanded to involve not only spinal fractures of various aetiology, but also fractures or impending fractures secondary to various causes in different parts of the body such as hip and sacroiliac joints. The cement used in both procedures, polymethyl methacrylate (PMMA), has a long history of safety in orthopaedic practice since the early 1960s.7


Kyphoplasty represents a more recent development and involves a modification of the vertebroplasty technique, in which an inflation balloon (bone tamp) is used within a collapsed vertebral body in order to partially restore the height, followed by stabilisation by injecting cement into the resulting cavity.


The main indications for vertebroplasty are:

  • Painful osteoporotic vertebral compression fracture(s) refractory to medical treatment. Failure of medical therapy is defined as minimal or no pain relief with the administration of physician prescribed analgesics or achievement of adequate pain relief only with narcotic dosages that induce excessive intolerable sedation, confusion or constipation. Associated major disability such as inability to walk, transfer, or perform activities of daily living is almost always present.

  • Painful vertebral fracture or impending fracture related to benign or malignant tumour, such as haemangioma, myeloma or metastasis.

  • Painful vertebral fracture associated with osteonecrosis (Kummell’s disease).

  • Unstable compression fracture with demonstration of movement at the wedge deformity.

  • Conditions in which reinforcement of the vertebral body or pedicle is desired before a posterior surgical stabilisation procedure in patients with osteoporosis or with vertebral tumour resulting in bone loss/destruction.

  • Patients with multiple compression deformities resulting from osteoporotic collapse in whom further collapse would likely result in pulmonary compromise, gastrointestinal tract dysfunction, or altered centre of gravity with associated increased risk of falling as a result of deformity of the spine.

  • Chronic traumatic fractures in normal bone with non-union of fracture fragments or internal cystic changes.

Indications for kyphoplasty are:

  • Painful osteolytic or osteoporotic compression fractures (Phillips) including progressive vertebral collapse.


Absolute contraindications

  • Osteomyelitis or systemic infection.

  • Uncorrectable coagulopathy.

  • Radicular pain or radiculopathy caused by a compressive syndrome not related to vertebral body collapse. However, preoperative vertebroplasty may be indicated if a spinal destabilisation procedure is planned.

  • Tumour extension into the vertebral canal or cord. Nevertheless, in patients with neural or nerve root compression from ingrowth of a tumour, vertebroplasty may be performed in conjunction with laminectomy or with radiation therapy.

  • Lack of surgical backup or lack of efficient patient monitoring facilities.

  • Prophylaxis in osteopenic patients who have no evidence of vertebral fracture(s) and no planned spinal destabilisation procedure.

Relative contraindications

  • Spinal canal compromise of ⩾20% by retropulsed fragment.

  • Fractures extending into the posterior cortex because of increased risk of cement leakage. However, in our experience, accurate positioning of the needle in the anterior one third of the vertebral body minimises this risk.

  • Complete or >90% vertebral body collapse.

  • If the patient has longstanding back pain for >1 year.

  • Young age.


The diagnosis of vertebral body fracture is usually confirmed by x ray. A detailed history and physical examination are needed to correlate the site of pain and tenderness to the level of compression fracture as well as to exclude radiculopathy. Magnetic resonance imaging (MRI) examination or a bone scan is essential to confirm the involved vertebral fractures and the nature of the cause, especially if there is more than one vertebral body compressed (figs 1 and 2). On the MRI scan high signal changes on the T2 or STIR (short tau inversion recovery) sequences will help to localise the vertebral fractures that may respond to treatment with percutaneous vertebroplasty or kyphoplasty (fig 3). Most experienced operators prefer to carry out percutaneous vertebroplasty in vertebrae which have not been extensively destroyed and maintain at least one third of their original height.

Figure 1

Sagittal magnetic resonance image showing T8 vertebral wedge fracture in an 80-year-old man.

Figure 2

Radioisotope scanning of the same patient as in fig 1 showing increased uptake at the same vertebral level.

Figure 3

A 75-year-old woman with severe pain at the level of L3. Magnetic resonance imaging scan showing compression fracture of L3. Note previous vertebroplasty of L4 and L5.



Percutaneous vertebroplasty can be performed under local anaesthesia combined with conscious sedation using midazolam and fentanyl as day case procedure. The procedure is performed in the interventional radiology suite under strict sterile measures. Patients are usually in hospital for overnight stay. In cases of severe vertebral collapse or difficult anatomy, the procedure is performed under computed tomography (CT) and fluoroscopic guidance with the patient lying in the prone position.

The needle (10–15 gauge) for injecting the cement is placed into the vertebral body under radiological guidance, most commonly through the pedicle. Coaxial biopsy could be done if tissue diagnosis is needed. A small amount (3–5 ml) of acrylic bone cement (PMMA) is then injected into the affected vertebra until resistance is met or until cement reaches close to the posterior wall of the vertebral body (figs 4 and 5).

Figure 4

(A) Lateral fluoroscopic image shows the 10 G needle in the partially collapsed vertebra through the transpedicular approach. (B) Lateral fluoroscopic image showing the distribution of bone cement (polymethyl methacrylate) within the vertebral body. The patient had almost complete pain relief.

Figure 5

(A) Fluoroscopy image shows polymethyl methacrylate (PMMA) bone cement flowing through the 10 G needle into the vertebral body. (B) Anteroposterior fluoroscopic image of the completed procedure with PMMA cement successfully distributed across the midline.


Kyphoplasty follows much the same principles. However, general anaesthesia is needed with overnight hospital admission. A larger sized cannula (7 G) is inserted into the affected vertebral body under radiological guidance, through which the inflatable bone tamp is introduced and inflated. Inflation is stopped when either the fracture is adequately reduced, maximum pressure is reached, or cortical bone is reached. The balloon is removed and the ensuing cavity is filled with an equal volume of cement (toothpaste-like viscosity) to reduce leakage of the bone cement into adjacent structures or vasculature. In some cases antibiotics are also included in the injected material if the patient is immunocompromised. The procedure generally takes 1–2 h. CT may be used several hours after injection to assess vertebral body filling and to detect any leakage of the cement. Non-steroidal anti-inflammatory drugs or other oral pain killers can be prescribed for 2–4 days for local pain control. However, most patients experience considerable pain relief within 24 h after the procedure.

National Institute for Health and Clinical Excellence (NICE) guidelines for percutaneous vertebroplasty emphasise that the procedure should only be used in cases of pain refractory to conservative treatment. They recommend the following:

  • That the procedure is only undertaken where there is access to a spinal surgery service.

  • That each case should have been discussed by a multidisciplinary team including a radiologist and a spinal surgeon.8


The patient is maintained recumbent for 2 h after the procedure (90% of the cement’s ultimate strength is obtained in 1 h). Vital signs and neurologic evaluations (focused on the lower extremities) are monitored every 15 min for the first hour. After 2 h and when recovered from the anaesthetic, the patient can be mobilised.

If an increase in pain, significant change in vital signs, or a neurologic change occurs, an immediate evaluation should be started to determine the cause, including a neurological and surgical consultation.


The mechanism of pain relief experienced by patients who undergo vertebroplasty has not been established. While many investigators have concluded that the pain relief is due primarily to the mechanical support provided by the bone cement, some have hypothesised that other factors, such as nerve damage resulting from heat generated during the exothermic polymerisation of PMMA, or from toxicity caused by the unpolymerised monomer, may also play a role in reducing pain perception.9 10


The overall risk associated with percutaneous vertebroplasty is relatively low (1–3% for benign disease and up to 10% with metastatic disease).9 11 The most common complication is pain and tenderness at the site of injection. This usually resolves within 24–48 h with paracetamol or ibuprofen.

The most serious complication is cement leakage into either neurological or vascular structures. Evidence of leakage into surrounding spaces on subsequent imaging has been reported to be as high as 41%, but this is largely asymptomatic.12 Leakage into the epidural space or neural foramen can result in neuralgia or cord compression. Such cases, though rare, if symptomatic should be treated with emergency surgical decompression. Gangi et al reported 15 cases of epidural leak in 868 cases of percutaneous vertebroplasty, and of these only three cases had neuralgia without spinal cord compression.13 Evans et al treated 488 patients with only two patients complaining of uneventful radicular pain with no cord compression.14 Leakage of cement into the venous system may lead to embolisation of the pulmonary vasculature. This is not usually symptomatic or associated with any clinical sequelae.13

Although, as a newer procedure, there are fewer data on kyphoplasty, the published literature reports a lower rate of adverse events. The rates of cement leakage have tended to be lower: 11.3% in a study by Ledlie and Renfro, with no symptoms as a result; and 8% in Liebermann et al, equally clinically insignificant.15 16 This is postulated to be due to the compression of surrounding cancellous bone, and also the ability to inject a more viscous cement into the cavity.

The US Food and Drug Administration (FDA) has published the results of its online reporting system, which showed higher reported rates of adverse events following kyphoplasty than have been published in the literature, and higher rates compared with vertebroplasty. Spinal cord compression requiring surgical intervention, and pedicle fractures, were reported in association with kyphoplasty more frequently than with vertebroplasty, perhaps due to the larger gauge needles required for the procedure. In total they have received 52 reports over a 5 year period for both procedures, with four deaths in total. Two reported deaths were associated with the treatment of eight or more vertebral levels at once.17

Bone infection or fracture is an extremely rare complication since this procedure has to be performed under strict sterile conditions. There have been two case reports of symptomatic pulmonary embolism and to our knowledge only one published case of fatal pulmonary embolus after vertebroplasty.1820

Fat embolism (bone implantation syndrome) has been reported in patients undergoing hip arthroplasty using cement injection. However, the amount of PMMA injected during orthopaedic surgery is remarkably larger than the injected dose in these procedures.21 Local trauma due to needle puncture of adjacent structures (dura, lungs, pleura) and pedicle fractures can rarely occur. Allergic or idiosyncratic reactions (hypotension, arrhythmias) to radiographic contrast agent or PMMA are exceedingly rare.

There is a concern that percutaneous vertebroplasty and kyphoplasty may be associated with increased fracture rate in vertebra in the vicinity or remote to the treated vertebra due to excessive intradiscal leak or to changes in biomechanical properties of the spine.22 23 This has been shown in ex vivo trials and small scale clinical studies, but this risk has not been validated by a controlled trial.24 As patients with an osteoporotic vertebral body fracture are already at high risk of developing a second one, other conservative and medical lines of management of osteoporosis should accompany invasive treatment.


Since 1986 there has been a growing number of percutaneous vertebroplasty procedures performed around the world.13 14 2527 Gangi et al treated 868 vertebra and achieved pain relief in 78% of the osteoporosis cases, 83% in patients with vertebral tumours, and 73% in patients with haemangiomas.13 The incidence of disc leak was 1.7% and neuralgia 0.3%. The average follow-up time in this study was 2.7 years (range 2–7 years). In another controlled study 488 patients were treated with PMMA vertebroplasty, achieving 90% pain relief and significant improvement in ambulation and ability to perform activities of daily living (ADL).15 In 2002, McGraw et al studied 100 patients undergoing vertebroplasty. Ninety-seven per cent reported immediate improvement in pain, and 93% had continued pain relief at follow-up (6 months or longer). There were two minor adverse events, only one of which was directly related to the procedure.28

A recent preliminary randomised controlled trial (RCT) entitled VERTOS compared short term (2 week) differences between vertebroplasty and optimal pain medication. They found that vertebroplasty patients had less pain, better quality of life and lower analgesia use than those on conservative treatment. Patients in the “conservative treatment” arm of the trial who went on to have vertebroplasty also had significant improvement after the procedure.29

A second RCT, VERTOS II, is in progress, aiming to recruit 200 patients to assess the cost effectiveness of percutaneous vertebroplasty versus conservative management at 12 months with respect to pain, quality of life, complications, secondary fractures, and mortality.30

Main messages

  • Osteoporotic vertebral compression fractures cause significant morbidity and mortality, particularly in an ageing population.

  • While previously management was restricted to conservative measures only, interventional procedures have now been developed to allow stabilisation of the fractured vertebral body with the aim of reducing pain and improving mobility.

  • In vertebroplasty, cement is injected into the fractured vertebral body, while in kyphoplasty this is preceded by inflating a balloon within the vertebral body to correct loss of height before stabilisation.

  • Research to date suggests that both procedures are safe and effective in reducing pain and improving quality of life; however, there are few randomised controlled trials, and very few data comparing kyphoplasty with vertebroplasty.

There is still a distinct lack of RCTs on kyphoplasty. An early study by Lieberman et al in 2001 showed promising results in 70 kyphoplasty procedures conducted in 30 patients.16 Forty-seven per cent restoration of the vertebral body height was achieved in 70% of the cases. In addition, there was remarkable improvement in both physical function and pain with no significant complications. Ledlie and Renfro followed up 117 patients for 2 years after kyphoplasty.15 Sixty-eight per cent of patients reported complete pain relief after 1 week, increasing to 86% after 6 weeks, which was maintained until the 2 year follow-up. They also reported a significant improvement in mobility and reduction in analgesia requirements. Kasperk et al treated 60 patients for painful osteoporotic vertebral fractures. Forty patients underwent kyphoplasty, and this group had a 12% increase in vertebral height, compared with further loss of height in the conservatively treated group. Pain and mobility were significantly improved in the kyphoplasty group.31

Current research questions

Since both procedures are relatively new, most of the ongoing research is in the form of randomised controlled trials to assess their long term safety and efficacy compared with conservative management.

  • Is one procedure preferable to the other in terms of efficacy, complication rates, or cost effectiveness?

  • What is the mechanism of pain relief?

  • What is the optimum point in the natural history of a vertebral fracture at which to intervene?

  • There is also continued interest in developing bone cement substances.

Voggenreiter used balloon kyphoplasty to treat 30 patients with vertebral compression fractures, largely focusing the study on the correction of vertebral height which was found to be influenced by patient positioning and bone tamp inflation in equal measures. All patients reported immediate pain relief.32 Ledlie and Renfro reported significant lasting improvement in vertebral body height at 2 years.15 However, subsequent studies have reported that the degree of correction of kyphosis in balloon kyphoplasty does not correlate with the level of symptomatic relief reported, so the significance of this is uncertain.31 33


Thus far, only small non-randomised studies comparing vertebroplasty and kyphoplasty have been published, but these have been subjects of a number of meta-analyses. The consistent conclusion is that the two procedures are comparable in terms of pain reduction, quality of life improvement, and safety. DeNegri et al followed 26 patients and showed that both vertebroplasty and kyphoplasty were effective in pain reduction, with no significant difference between the groups.34

A recent meta-analysis by Gill et al compared the efficacy (in terms of pain reduction) of kyphoplasty and vertebroplasty for a total of 1309 patients. Both procedures achieved comparable pain reduction, measured using visual analogue scales (VAS), immediately post-procedure. Similarly at follow-up there was no significant difference between reported pain levels. They suggest this indicates that the cement stabilisation is the key element.35

Taylor et al performed a meta-analysis of data in 2006 to assess not only efficacy but also the relative safety of the two procedures. They concluded that both were safe and effective compared to medical treatment alone, and that kyphoplasty had a lower adverse event profile.36 Hulme et al performed an analysis of 69 individual studies, and reached similar conclusions. They reported difficulty in measuring relative improvements in patient function, and an insufficient number of RCTs on either procedure.37

Percutaneous vertebroplasty stabilises the vertebral body and relieves pain; however, it does not restore vertebral height. Kyphoplasty, first reported in 2000, has become popular as a way of both reducing and stabilising a vertebral compression fracture. An increasing number of early studies have shown kyphoplasty to be at least as effective as vertebroplasty in relieving symptoms, while having a lower adverse events profile. However, a few studies have raised questions about the durability of the vertebral height correction, with one ex vivo trial by Kim et al showing progressive loss of height with repeated axial loading, compared with no additional loss of height in vertebroplasty.38 Ledlie and Renfro reported significant lasting improvement in height at 2 years.15 It is also still uncertain at what stage in the natural history of a compression fracture it is best to intervene with kyphoplasty.

Further controlled studies are needed to evaluate the long term efficacy of this new technique.

Further research is also being carried out on improving new cement/bone filling materials such as hydroxyapatites and bis-gamyl methacrylate, especially biomaterials, that convert to bone or stimulate bone growth.39 Biodegradable materials are also under investigation.

Percutaneous vertebroplasty and kyphoplasty are minimally invasive techniques for pain management and improvement of mobility in patients with vertebral body fractures. The published data show good short and midterm results with high success and low complication rate. However, long term efficacy needs to be proven. The most critical elements for a successful procedure are proper patient selection, correct needle positioning, and high profile image quality and operator experience. The decision to perform either procedure should be made by a multidisciplinary team because the choice between vertebroplasty, kyphoplasty, surgery, radiation therapy, medical treatment, or a combination of these therapies depends on a number of factors. These factors include the local and general extent of the disease, the spinal level involved, and the pain experienced by the patient as well as his or her neurological condition, the state of health, and life expectancy.


View Abstract


  • Competing interests: None declared.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.