Douglas Wardlaw, honorary professor, Robert Gordon University, Aberdeen, Scotland, talked to Spinal News International about data for the use of balloon kyphoplasty for the management of vertebral compression fractures.
Compared with percutaneous vertebroplasty, what are the potential benefits and disadvantages of balloon kyphoplasty?
Compared with percutaneous vertebroplasty, the benefits are that patients have a better fracture reduction, better quality of life, less pain and live longer. The perceived disadvantage is the increased cost of the procedure. However taking into account the quality of life benefits, there may in the long run be little or no difference.
Data from Kallmes et al1 and Buchbinder et al,2controversially, raised questions about the effectiveness of percutaneous vertebroplasty . Have they affected the perception of balloon kyphoplasty (ie, have they raised questions about its effectiveness?)
They did initially, but then the flaws in these studies became apparent. Both studies were very slow to recruit due to patient selection and because of a lack of patients wishing to enter. They were therefore a highly selected group and not at all the same as routine patients. Neither Kallmes nor Buchbinder had a control group with standard care (common practice) to assess the placebo effect of the sham intervention. Fractures were older (less than one year compared with less than three months) than in the FREE and Vertos II studies.3,4,5
Additionally, some of the pain observed may have been due to the deformity and the change in the spinal mechanics, which may have be relieved by local anaesthetic injection (marcaine was used in the Kallmes study and lidocaine in the Buchbinder study). There was also a significant difference in the crossovers in the vertebroplasty patients in the Kallmes study compared with placebo at three months, suggesting the placebo effect of the sham procedure was wearing off as one would expect. The FREE and Vertos II showed a significant effect at two and one years, way beyond the expected time frame of the placebo or sham effect. One has to remember that patients in a study may have “special” treatment or may feel themselves “special”, which may in itself have an effect like a “sham treatment or placebo effect”. It is therefore important to have a control group for this reason.
Fritzell et al6recently published a study that “could not document that balloon kyphoplasty was cost-effective”. What is your view about this?
Using 12-month EQ-5D scores from the FREE study, Ström et al7 compared cost-effectiveness between balloon kyphoplasty and non-surgical management in patients in the UK hospitalised for painful vertebral compression fractures. The study showed that the cost per quality adjusted life year (QALY) gained was £8,840 (€11,155.62). This cost is below the amount per QALY that society is willing to pay (WTP) in the UK, which is between £20,000 and £30,000 (approx. €25,000—38,000).
Both the Ström and Fritzell analyses did not take into account differences between the two procedures in the rate of subsequent fractures and mortality risk, which has the potential to show kyphoplasty as more cost effective. The FREE two-year data showed there is no difference in subsequent fracture rate, and studies show that following balloon kyphoplasty treatment, there is a significant reduction in mortality risk compared to vertebroplasty and conservative care. The Fritzell paper was a subset of 63 Swedish patients from the FREE trial (300 patients). Quality of life was assessed with EQ-5D, and costs were determined according to hospital billing systems, which were generally based on Swedish national costing guidelines. After two years, societal costs were significantly greater for balloon kyphoplasty than for non-surgical management. The cost per QALY gained by using balloon kyphoplasty instead of medical management was Swedish Karona (SEK) 884,682 (€92,154 and USD $134,043). This cost is above the maximum WTP amount per QALY gained for Swedish society (SEK 600,000 [€62,500, USD$ 90,901]). They were unable to conclude that balloon kyphoplasty is more cost-effective than non-surgical management in the Swedish population studied.
However, the randomisation of the FREE trial was not stratified according to centre, but was according to trial entry at time of entry, and so the resulting population was not a true randomisation. Moreover, a sensitivity analysis using the entire patient population from the FREE trial found a cost/QALY at a level that was cost effective. Clearly the smaller sample size loses power.
The lead author of the study, Peter Fritzell, told Spinal News International that balloon kyphoplasty “could be used in carefully selected patients with osteoporotic fractures, but not as a routine procedure”. Do you agree? If so, which patients do you think are the best candidates for balloon kyphoplasty?
It depends on what you mean by “routine”. I believe that patients with significant osteoporotic fractures—greater than say 15% anterior height loss—that have significant pain (that correlates with MRI/x-ray) continuing after a week or so following onset of pain can be considered candidates for treatment. Fractures should be monitored to see if they have progressive height loss and if so, patients should be offered balloon kyphoplasty.
Is there an optimal timeframe in which balloon kyphoplasty should be performed— ie, is there a cut-off point after the onset of fracture symptoms that the procedure would not be effective?
I think the fractures should be treated early to achieve the optimum height restoration, and I think this needs further study. Certainly, fractures that are more than three months old and where the pain has reduced and reached a steady state will not be helped by balloon kyphoplasty. If there is progression of the deformity, then there is delayed healing of the fracture and balloon kyphoplasty should be offered.
What are the remaining questions about balloon kyphoplasty?
The next few years will provide data on the effects of balloon kyphoplasty on the occurrence of subsequent fractures, the cost effectiveness of the procedure, and the relationship between restoration of vertebral height and effects on pain and mobility. The KAVIAR study, is a prospective, randomised, non-blinded trial conducted in the United States and Canada to compare the effects of balloon kyphopalsty and vertebroplasty in patients with vertebral compression fractures. The primary outcome measure of the study is the percentage of patients who experience subsequent incident fractures 12 to 24 months after surgery. This study will also assess healthcare use and quality of life data to calculate the relative cost effectiveness of balloon kyphoplasty and vertebroplasty.
The CEEP study, conducted in the USA compares the cost effectiveness and efficacy balloon kyphoplasty and vertebroplasty in patients with vertebral compression fractures sustained within the previous 12 months. There are two other prospective randomized studies, the OSTEO-6 (a prospective randomised comparative study of balloon kyphoplasty, vertebroplasty and conservative management in acute osteoporotic vertebral compression fractures less than six weeks’ old) and the STIC2 (prospective randomised study of balloon kyphoplasty and vertebroplasty in subacute osteoporotic vertebral fractures older than six weeks). Both are evaluating the relationship between restoration of vertebral height and improvement in pain and they aim to determine whether outcomes differ in patients with older and younger fractures.
Do you think in the future, balloon kyphoplasty will replace vertebroplasty or is there a place for both vertebral augmentation procedures?
There is a place for both. Late fractures (>3 months), which have healed with the deformity unchanged from the initial one, with persistent pain, should be offered a local anaesthetic and/or steroid injection and if pain persists, vertebroplasty should be offered. Of course there is a wide variation and each case will have to be assessed independently. However, the above studies hopefully will provide more definitive evidence.
References
1. Kallmes et al. N Eng J Med 2009; 361: 569–79
2. Buchbinder et al. N Eng J Med 2009; 361: 557–68
3. Klazen et al. Lancet 2010; 376: 1085–92
4. Wardlaw et al. Lancet 2009; 373: 1016-24
5. Boonen et al. J Bone Miner Res 2011;26:1627–37
6. Fritzell et al. Spine 2011; 36: 2243–51
7. Strom et al. Acta Orthop 2008; 79: 269–80
