Abstract
The incidence of vertebral fragility fractures and deformity increase steeply with age. Every sixth woman and every twelfth man will sustain a symptomatic vertebral fracture. Vertebral fractures result in pain, functional disability and decreased quality of life, which may last for several years, and may also affect mortality. The patient with an acute fracture should be examined with radiology for diagnosis. In case of a low-energy fracture, osteoporosis should be suspected and investigated. If the pain management fails, vertebroplasty or kyphoplasty could be considered. Braces may be used, but evidence for its effect is lacking. In the rare event of neurological compromise, or unstable fractures, surgical treatment should be considered. After vertebral fragility fractures, the risk for new fractures is high and secondary preventive measures advocated. The best evidence for secondary prevention is currently on medical treatment of osteoporosis.
Introduction
Vertebral fractures may occur in any part of the spine but are most common in the thoracic and lumbar regions.
Vertebral fractures tend to be more common in the transition between a more rigid part of the spine and a more flexible part, such as the transition between the thoracic and lumbar spine but are also common in the mid part of the kyphotic region of the thoracic spine .
The mechanism and force of injury determines whether a fracture occurs or not. If the force is larger than the skeleton can resist, a fracture occurs. The fracture displacement will be relatively larger when the bone quality is poor. In the osteoporotic vertebra, fracture may occur without any noticeable trauma. About one third of the patients seeking care for acute back pain could not recollect a trauma or other specific incident associated with the occurrence of the vertebral fracture .
Although most of these fractures are due to low-energy trauma, individuals with osteoporosis may sustain fracture types that resemble fractures in non-osteoporotic individuals, and individuals with osteoporosis may sustain high-energy fractures.
The radiological diagnosis of vertebral fractures may be difficult. Unlike other fractures, the age of a vertebral fracture is not always easy to determine, especially when osteoporosis is part of the etiology. Vertebral fragility fractures may be regarded as a continuum from a next to normal vertebra, to a crush compression fracture, with any degree of deformity in between.
It has been estimated that a large proportion of vertebral fractures are asymptomatic and that only 30–40% are discovered in the clinic . In a review of thoracic computed tomography scans made for other causes in individuals with a mean age of 61 years, one third had vertebral fractures .
There are numerous classifications available for vertebral fractures and vertebral deformity. Prevalence figures vary depending on the classification used . According to the classification by Genant et al. , a vertebral height loss of more than 20% may be considered a fracture. Wedge fractures are the most common type of fracture . Biconcave and crush fractures are more commonly a sign of osteoporosis than wedge fractures ( Fig. 1 ). Computed tomography has a higher sensitivity than plain radiography when diagnosing vertebral fractures . Magnetic resonance imaging is very sensitive when there is a need to determine whether the fracture is acute or not, and is often also the best method to exclude differential diagnoses such as infection and malignancy ( Fig. 2 ).
In this review, the term “vertebral fracture” is used for clinically symptomatic vertebral fractures. The term “vertebral deformity” is used for data from population based surveys, which may include non-symptomatic fractures and deformity associated with other disorders, for example Scheuermann’s kyphosis.
The burden of vertebral fractures
The incidence of vertebral fracture and deformity increases with age, and varies with country . Low-energy trauma was four times more common in persons 60 years or older sustaining thoracic and lumbar vertebral fractures than in younger persons . Odontoid fractures are among the most common cervical spine fracture in the elderly and the number is increasing , possibly due to an increasing elderly population.
The incidence of clinical symptomatic fractures increases steeply from the age of 50–80 ( Fig. 3 ). It has been estimated that the life time risk for a symptomatic vertebral fracture for a person of the age of 45 is 15% for a woman and 8% for a man . However, when looking at radiologically defined vertebral deformity the prevalence is similar in men and women or slightly more prevalent in men than in women .
The incidence of patients with fractures in the thoracic and lumbar spine treated at hospitals in Sweden has been estimated to 30/100,000 persons per year, with about two thirds of the patients being 60 years or older . The corresponding incidence of cervical spine fractures between 1998 and 2011 was 8/100,000 persons per year, with an incidence of 5/100,000 for those being up to the age of 59 and 20/100,000 for those aged 60 years or older .
The number of vertebral fractures in the United States has been estimated be about 700,000 yearly, and the annual cost to be 13.8 billion USD, corresponding to 18,000 USD per vertebral fracture , but it is likely that the total annual cost is an underestimation due to the large number of undetected vertebral fractures and the changing demography with a larger proportion of elderly.
In Sweden, it has been estimated that the fracture related cost for patients hospitalized for vertebral fractures is 12,000 Euros in the first year, which was only slightly lower than the figure for patients with a hip fracture .
The burden of vertebral fractures
The incidence of vertebral fracture and deformity increases with age, and varies with country . Low-energy trauma was four times more common in persons 60 years or older sustaining thoracic and lumbar vertebral fractures than in younger persons . Odontoid fractures are among the most common cervical spine fracture in the elderly and the number is increasing , possibly due to an increasing elderly population.
The incidence of clinical symptomatic fractures increases steeply from the age of 50–80 ( Fig. 3 ). It has been estimated that the life time risk for a symptomatic vertebral fracture for a person of the age of 45 is 15% for a woman and 8% for a man . However, when looking at radiologically defined vertebral deformity the prevalence is similar in men and women or slightly more prevalent in men than in women .
The incidence of patients with fractures in the thoracic and lumbar spine treated at hospitals in Sweden has been estimated to 30/100,000 persons per year, with about two thirds of the patients being 60 years or older . The corresponding incidence of cervical spine fractures between 1998 and 2011 was 8/100,000 persons per year, with an incidence of 5/100,000 for those being up to the age of 59 and 20/100,000 for those aged 60 years or older .
The number of vertebral fractures in the United States has been estimated be about 700,000 yearly, and the annual cost to be 13.8 billion USD, corresponding to 18,000 USD per vertebral fracture , but it is likely that the total annual cost is an underestimation due to the large number of undetected vertebral fractures and the changing demography with a larger proportion of elderly.
In Sweden, it has been estimated that the fracture related cost for patients hospitalized for vertebral fractures is 12,000 Euros in the first year, which was only slightly lower than the figure for patients with a hip fracture .
Clinical presentation
A patient with a symptomatic vertebral fracture presents with a more or less sudden onset of pain at the fracture site or close to it. The pain decreases with rest and increases with activity. The pain may involve a severe functional restriction, and especially the elderly may be unable to perform activities of daily living. The pain may occasionally radiate anteriorly. In the case of previous fractures, kyphosis in the thoracic spine may be present and the rib cage very close or even in contact with the iliac crest, which may be associated with pain.
Differential diagnoses such as pneumonia, ruptured aortic aneurysm, spondylodiscitis, urinary tract infection, myocardial infarction, and pathological fractures, or other painful malignant processes should be looked for. Many of these can often be confirmed or excluded during history taking and clinical examination.
It is uncommon that an osteoporotic thoracic or lumbar vertebral fracture affects the spinal cord or nerve roots by retropulsion of parts of the fractured vertebrae. If the patient complains of radiating pain in the legs, sensory loss in the legs, or loss of muscular strength or balance, or urinary or fecal incontinence, an MRI of the spinal column may be indicated and should be performed without delay if there are signs of spinal cord affection. If there is a significant encroachment of the spinal canal, emergency decompressive and stabilising surgery may be indicated ( Fig. 4 ). Surgery in the osteoporotic spine is challenging, and the risk of later failure due to poor bone quality is large. Multiple anchoring points are used and special implants for better purchase in osteoporotic bone are available. Augmentation of the vertebraes with bone cement is sometimes performed to allow better screw purchase .
Risk factors
Among risk factors for vertebral fracture are high age, prior falls or fractures, smoking, oral glucocorticoids, inactivity, underweight, chronic obstructive pulmonary disease, rheumatoid arthritis and ankylosing spondylitis . A patient with low back pain after trauma, older age or use of corticosteroids can be considered having “red flags”, which indicate a need to perform diagnostic imaging for the suspicion of a vertebral fracture. A combination of these red flags increases the diagnostic accuracy .
Age and previous fracture predict new vertebral fractures independently of bone mineral density. In women with prevalent vertebral deformities, the risk of contracting a new vertebral deformity in a year is about 20% . Women with a prevalent vertebral deformity carry a fivefold risk of contracting a new vertebral deformity in a year, with an up to 12-fold risk if two or more previous vertebral deformities exist .
Men and women with vertebral fracture or vertebral deformity have an increased risk of sustaining other types of osteoporotic fractures, for example hip fracture . A risk increase of 2.3 for hip fracture and 2.8 for any fracture has been reported, when compared to individuals without vertebral fractures .
Increased bone turnover and increased high sensitivity CRP have been reported to be risk factors for vertebral fractures , but their use for fracture prediction in the clinic has not been established.
The prevalence of osteoporosis increases with age. The importance of osteoporosis as a contributing cause of fracture increase with higher age . A one standard deviation decrease in bone density, or ultrasound, increases the relative risk for any type of fracture by 1.5. A site specific measurement has an even better predictive ability, and the corresponding relative risk for a measurement in the lumbar spine for vertebral fracture is 2.3 .
Even if low bone density is a risk factor for vertebral fractures and deformity, up to half of the fractures occur in non-osteoporotic individuals .
Outcome of vertebral fractures
Pain from a vertebral fracture is initially intense. Recovery is most prominent during the first 3 months. After that period recovery is slow. In a longitudinal cohort study, patients that had visited an emergency unit due to vertebral fracture were followed for a year. The average pain at three weeks was 70 out of a maximum of 100 on a von Korff pain score. The same score was 60 after 12 months . Crush fractures, severely deformed fractures and thoracic fractures had the worst prognosis, with a lower magnitude of improvement than other fractures. Patients with a vertebral fracture had more bodily pain, as determined by the SF-36, both at two months and two years after the fracture, compared to controls from the population . Controls in a trial comparing kyphoplasty and non-surgical care for osteopenic acute vertebral fractures had an initial visual analogue scale back pain (0–10) of 7, which decreased to 4 after one year , similar to other trials .
The reduced quality of life from vertebral fractures has been reported to be persistent for several years, and longer than for other fractures associated with osteoporosis . The prevalence of previous fractures (vertebral or other osteoporotic fracture) has a negative impact on the effect of quality of life in case of new vertebral fractures . Vertebral deformity, i.e. not clinically recognised fractures, are also associated with increased back pain and limitation of function .
Vertebral fracture has been associated with a shortened life-span, but is to some extent, but not all, explained by co-morbidity . An estimated survival of 61% compared to an expected of 76%, 5 years after a thoracic and lumbar vertebral fracture has been reported , with a higher mortality in men than in women . Prevalent vertebral deformity has been reported to be associated with a modest shortened life-span in women, but this excess can at least to some extent be explained by other factors associated with poor health . Of patients sustaining an odontoid fracture in the cervical spine and being 65 years or older, 18% were deceased within the first 12 months after the fracture .
Pain management
Pain management in patients with vertebral fractures should start with acetaminophen (paracetamol). A second choice or addition is non-steroidal anti-inflammatory drugs (NSAID). Due to the risk of gastro-intestinal and cardiovascular side effects, which concerns both the traditional non selective NSAIDS and the newer cyclooxygenase 2-inhibitors, the use of NSAID should be cautious, especially in the elderly. The addition of codeine or tramadol may be sufficient in cases with moderate pain and may be exchanged to morphine, oxycodone and similar opiods if needed .
The possible pain-reducing effect of drugs used to treat osteoporosis has been studied. One small study reported a beneficial short-term effect on pain with intravenous clodronate . A randomised double blind trial in 32 patients with acute vertebral fractures reported that pamidronate given during three consecutive days had a better effect on pain than placebo. The effect lasted 30 days .
A meta-analysis of calcitonin given to patients with an acute vertebral fracture pain found a pain-reducing effect during the first weeks of treatment . However, the long-term use of calcitonin has recently been associated with cancer risk. Therefore, the use of calcitonin has been restricted by the European Medicines Agency and is no longer available for the treatment of osteoporosis in Europe.
In a randomised controlled trial teriparatide was compared to risedronate. Subjects were women with back pain likely to be due to vertebral fractures. During a 6 month follow up, back pain decreased in both groups, but there was no difference in back pain between women treated with teriparatide or risedronate .
In a meta-analysis concerning women with postmenopausal osteoporosis and men with idiopathic or hypogonadal osteoporosis, teriparatide reduced the risk of any back pain (relative risk 0.66) compared to a pooled comparator group consisting of patients with placebo, alendronate, or hormone replacement therapy alone . The effect seemed to last up to 30 months after drug discontinuation .
Specific interventions
Brace treatment
There are several types of spine braces or orthoses. Body jackets are custom made moulded braces for unstable fractures. Hyperextension, or three point, braces are used for more stable fractures in the lower thoracic and upper lumbar area. There are also various types of soft braces providing varying degrees of stability.
The aim of the brace treatment is to stabilise the fracture and thereby prevent an increased deformity (most often kyphosis) and alleviate pain until the fracture heals. However, possible side effects of braces include muscular atrophy, skin irritation, restricted respiration and cost.
There are few studies comparing brace treatment with no brace treatment. One randomised study of 47 patients (mean age 39 years) with burst fractures (AO type A3) treated with a thoracolumbar sacral orthosis or no orthosis did not show any difference during a one year follow up, as assessed by the Roland Morris Disability questionnaire and radiology . A similar result was found in another study on 23 patients (mean age 40 years) with AO type 3 burst fractures when comparing a thoracolumbar sacral orthosis with no orthosis .
The experience at our clinic, and shared by others, is that many elderly individuals with poor soft tissue experience a great amount of discomfort when treated with a brace. Therefore, braces have been specifically designed for the treatment of painful osteoporotic vertebral fractures. Two studies have been reported by the designers of the Spinomed brace. This is a light weight brace with pads and metal splints carried as a backpack. Women ( n = 62, mean age 72 years) with vertebral fractures, pain and kyphosis were randomized to the Spinomed brace or no treatment. The brace was worn 2 h per day. Back extensor strength was the primary end-point, and was significantly better after 6 months usage in the intervention group than in the controls. A similar result was found for well-being and pain. A second study on 108 women reported similar results .
Vertebroplasty/kyphoplasty
The percutaneous injection of bone cement (polymethyl metacrylate; PMMA) into vertebrae was initially described in the treatment of hemangiomas and malignancy . The same technique was later applied in patients with painful vertebral compression fractures. The procedure is performed under local or general anesthesia. The patient is placed in the prone position. One or two needles are inserted into the vertebral body. During vertebroplasty bone cement is injected without any direct reduction of the fracture. In kyphoplasty, two balloons are inserted into the vertebral body and expanded. A cavity is created and a reduction of the deformity is achieved. The cavity is filled with bone cement ( Fig. 2 ).
Several prospective uncontrolled trials reported that 70–90% of patients with osteoporotic vertebral fractures achieved pain relief after vertebroplasty . During 2009 two randomized trials reported lack of benefit with vertebroplasty when compared to a sham treated control group . Several professional associations thereafter discouraged the use of vertebroplasty and kyphoplasty. Even if well designed, these studies have raised some questions. The sham group received local anesthetic injection, which may reduce the pain associated with the vertebral compression . The studies also included patients with a duration of pain due to a vertebral fracture of up to one year, and protocols were changed during the time of the study; shorter follow-ups and fewer patients were included than originally planned.
Other randomised studies without a sham procedure as control have been published. A recent meta-analysis found 7 unique randomised controlled studies comparing a treatment group (cement injection) and a control group . Of these, 6 were vertebroplasty studies and 1 a kyphoplasty study, and they included the two previously mentioned studies with a sham control group. In all, the number of patients that had been treated with cement injection was 440, and the number of controls was 437.
In the short time follow up (12 weeks or less) significant beneficial effects on pain, functional outcome and health related quality of life were seen in the treated patients when compared to the controls. The longer follow up (26 weeks or more) showed a similar pattern. The differences were larger at the short time follow up than at the long time follow up. The mean difference in the visual analogue scale for back pain was more than 3 which exceeded the minimum clinically important difference .
Overall, adverse events did not differ between treated patients and controls. One of the risks that have been feared to be associated with injecting stiff bone cement in a vertebral body is adjacent level vertebral fractures. In the meta-analysis by Andersen and colleagues there was no difference in new fracture risk between the treated group and the control group, with 19% of the treated patients and 19% of the controls contracting new vertebral fractures . However, follow-up times were short and exceeded one year in only one study.
At present there seems to be no clinically important difference between vertebroplasty and kyphoplasty, with a reduction of a mean preoperative visual analog scale of 8 to about 3 points postoperatively in both vertebroplasty and kyphoplasty patients . Cement leakage is a common complication, and more common in vertebroplasty than in kyphoplasty. The reason is the low viscosity cement and the higher injection pressure used in vertebroplasty than in kyphoplasty. However, symptomatic cement complications are few. A slightly higher risk for complications has been reported with vertebroplasty than with kyphoplasty , but does not seem to affect outcome . Postoperative infections are rare after these percutaneous procedures, but may be life-threatening .
Vertebroplasty and kyphoplasty should not be performed in patients with a ruptured vertebral posterior wall due to the risk of cement leakage in the spinal canal, and may not be technically possible to perform in patients with severely crushed vertebras.
The equipment needed for balloon kyphoplasty is more expensive than for vertebroplasty, and kyphoplasty is also more often performed under general anesthesia . In a Swedish study, balloon kyphoplasty was not more cost-effective than standard medical treatment . In a simulation model the cost-effectiveness of kyphoplasty, vertebroplasty and non-surgical management was compared. The study found kyphoplasty to be more cost-effective than vertebroplasty and non-surgical management . The model had a large degree of uncertainty and the results should be interpreted with caution. This result contrasts the findings in randomised controlled trials, in which kyphoplasty was not more effective than vertebroplasty, and kyphoplasty not more cost-effective than standard medical treatment .
Surgery
Only a minority of elderly patients with thoracic and lumbar fractures are treated surgically . Minor cervical spine fractures may be left untreated. Other fractures are treated with a cervical neck orthosis, or if unstable, with surgery. Odontoid fractures are among the most common cervical spine fracture in the elderly . Treatment is debated. A cervical neck orthosis provides relative stability but leads in up to 85% of cases to a pseudoarthrosis, but not all are symptomatic. Surgical treatment leads to bony union in almost all cases, but since these patients often belong to a frail part of the population, surgery is not always a possible option .
An indication for surgery in the cervical, thoracic and lumbar spine is an unstable fracture, sometimes with neurological compromise, most often associated with high-energy trauma such as falls and traffic accidents . However, the number of fall-induced fractures with associated spinal cord injuries in the elderly has been reported to be increasing , indicating a growing need of surgical treatment in the elderly.
Exercise
Exercise has been advocated to increase bone density , and to prevent falls and fractures . A recent Cochrane report studied whether exercise could improve outcome after osteoporotic vertebral fractures. The authors found 7 randomised or quasi-randomised studies that compared exercise or physical therapy with non-exercise or placebo. The studies included in total 484 female and 4 male participants. The studies were diverse. Some studies reported improved pain, physical performance and quality of life during treatment. However, study data could not be pooled and the quality was in general low. No clear conclusions could be drawn .
Related posts:
Hip fracture as the tracer condition
Multi-disciplinary care of the patient with acute hip fracture: How to optimise the care for the elderly, traumatised patient at and around the time of the fracture to ensure the best short-term outcome as a foundation for the best long-term outcome
Principles of osteoporotic fracture treatment
Prevention of incident fractures in patients with prevalent fragility fractures: Current and future approaches
How can you prevent falls and subsequent fractures?
Secondary prevention and estimation of fracture risk
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
