Verteboplasty
the Rise and Fall of a Convenient Truth
Vertebroplasty- transcript
Perhaps I should title this presentation, the rise and probable imminent fall of the medical procedure, Vertebroplasty. Like all medical treatments and procedures, Vertebroplasty, involves an original story, a hypothesis that begins as a tentative exploration and evolves over time with practice and medical research to become standard medical practice.
Never without research to validate the story, most medical research generally only proves its worth. That is the intention of the medical research. And of course, there is business. Much business.
Eventually, hopefully, good solid Scientific Research comes to the rescue to truly decide either way...But is the cavalry too late.....
This story began for me when I was journal surfing in an online university library. I came across an intriguing editorial in the Spine Journal written by the editor in chief himself, Dr Eugene Carragee, The title read: 'The Vertebroplasty Affair' subtitled 'the mysterious case of the disappearing effect size.'
Even better was the very first sentence, invoking the ever omnipotent detective Sherlock Holmes "Looking back Watson, it may have seemed to good to be true. There were the usual clues".
Now I was truly intrigued and fully engaged in the Vertebroplasty Affair. I could sense a controversial, interesting, and revealing story. I had to see it through thoroughly.
Eloquently, he continues:
'Societies and governing bodies praised the clever innovation, sounded the trumpet and sotto voce, both perfunctorily and judiciously called “the usual suspects” to take the stand
i.e. - high quality, randomized, clinical trials- to give testament to anecdotal assumptions.' and 'It took a decade to round them up'.
I favour and admire the creativity and art in medicine but in the end it all must stand up to the highest scientific research standards. This involves observation, question and experiment as we were introduced to by the father of modern science, Galileo.....
Observation that sees something new and different,
Question that challenges the story
and Experiment that changes the story.
And Finally, a willingness to trust the research even though
it goes against current popular belief.
2. Vertebral Compression Fractures
Vertebral Compression Fractures or VCF's are a frequent occurrence in the United States, Europe and Japan, at least there is data collated for Vertebral Compression Fracture in these countries.
750 thousand persons suffer a Vertebral Compression Fracture every year in the United States and 1.4 million in the 3 regions combined.
In this photograph we see a Vertebral Compression Fracture.
Notice how the normal shape of the vertebra is deformed, that is crushed.
There are at least three types of Vertebral Compression Fracture the Wedge fracture being the most common at 51% of all vertebral fractures, the Bi-concave fracture representing 17% and Crush fractures at 7% of all vertebral fractures.
The most prevalent vertebral level for a fracture is T12 and L1. Together they comprise 58% of all Vertebral Compression Fracture. Overall thoracic and lumbar regions are equally prone to Vertebral Compression Fracture.
Vertebral Compression Fracture is the most common fracture of all osteoporosis fractures. The culprit trauma is from lifting especially with outstretched arms and most frequently by far, from falls.
Vertebral Compression Fracture affects one quarter of women over 50. Only one third become clinical and painful.
Unrecognized fractures show up frequently at other times often without any recall of major trauma. Once a woman has a Vertebral Compression Fracture and they are over 65 then they have a one in four chance of re-fracture within the next 5 years.
VCF is thought to be a self limiting condition from which significant recovery occurs. However, a Vertebral Compression Fracture represents the beginning of a year of decline in health for that person.
Consequences for vertebral fracture include: Spinal deformity in the form of increased kyphosis, loss of height, spinal rigidity and an increased number of days being bed ridden.
Just one Vertebral Compression Fracture reduces respiratory function by 9%. These fractures are often severely painful for 6 weeks With a Visual Analog Scale for pain at 7 out of 10 and many develop chronic spinal pain for one year or more.
One study followed a population over 22 years to find that those who suffered a Vertebral Compression Fracture died sooner than their neighbours at a significant rate – of 95.1 per 1,000 compared to 62 per 1,000.
Treatment for a Vertebral Compression Fracture is traditionally quite simple- it involves a combination of bed rest, bracing, and medications. And more recently we have seen the appearance and rapid expansion of a new treatment and the topic of this presentation, the procedure called Vertebroplasty.
The cost of Vertebral Compression Fractures to the US in 2002 was estimated at $12-18 billion dollars.
3. What exactly is a vertebroplasty?
On this slide we see on the left a vertebral fracture and on the right that same fracture now has a large black mass in place of the fracture. The black material on Xray is poly methyl metha crylate. It also referred to as PMMA. It is a synthetic polymer or Acrylic cement or simply Bone cement.
Vertebroplasty employs the use of a large needle which is inserted and then The PMMA is injected to fill the fractured vertebra. Which as you can see, then clearly restores the shape and normal height of the vertbera. Vertebroplasty was first reported in the literature in 1987 when a French team Galibert et al provided a preliminary report on the use of vertebroplasty for spinal fracture.
Jensen et al in 1997 reported that 47 patients with spinal fracture treated with vertebroplasty had 90% pain relief within 24 hours. In 1999 Martin et al similarly had great success with 80% improvement for 40 patients with 68 fractures between them. Again of those showing improvement reported complete relief of pain. Further, Deramond et al in 1998 report and I quote: “Immediate and complete relief of pain in 90% of patients”. Diamond and co compared vertebroplasty to conservative treatment. 55 patients denied the procedure and were therefore included as the non randomized control group. The study took pain scores at 24 hours, 6 weeks, between 6 and 12 months. At 24 hours the vertebroplasty group reported 53% pain relief and 29% functional improvement, Whilst the conservative group Showed Zero change for both pain and function.
However, both groups vertebroplasty and conservatively managed showed no differences at 6 weeks and beyond.
This is a critical article in my review of the literature for these authors choices of time intervals of reported pain. I will come to this later…
So vertebroplasty is on a roll, it works magic in just 24 hours. In fact, it becomes very popular in the United States as recorded by Medicare.
Every person over 65 and enrolled in Medicare constitutes a large widespread population for obtaining statistics on medical conditions and procedures. For every 1,000 persons enrolled in Medicare 0.73 will receive a vertebroplasty. This is the national average. For the 277 regions defined by mediacre the number of vertebroplasty procedures performed has doubled in the 6 year period between 2001 and 2006.
The lowest rates for this procedure are 0.12 to 0.13 per one thousand enrollees which occurred in Albany, Manhattan and Newark New Jersey. There were 105 of the 277 regions that were 25% lower than the national average. And there were 66 regions 30% above the national average. The highest rates were at 3.0 per one thousand enrollees. Topeka, Kansas, Fort Wayne Indiana, and Asheville North Carolina all fell into this group.
To put this clearly into perspective for the regions compared above, if you live in Topeka, Kansas 35 persons per ten thousand Medicare Enrollees will receive a Vertebroplasty But if y ou live in Manhattan Only one person per ten thousand of enrollees will receive a Vertebroplasty.
That is 30 times difference between these regions. There are likely many reasons for the differences between the rates of vertebroplasty performed in these regions.
4. This Vertebroplasty Affair
In examining this Vertebroplasty affair, I have had the benefit of hind sight and the excellent gold standard research of Buchbinder and Kallmes along with some expert interpretation, but also along the way I have been struck but several aspects of this condition and the quality of research designs.
After reading the Carragee editorial and the gold standard research papers I brought the topic of Vertebroplasty to the physical therapy department at work.
Immediately, one colleague of a great many years of experience said, "I just tell them that their pain will be much better in three days". That was the on the shop floor experience of 40 years. That resonated loudly with the Kallmes paper and the statement on page 574:
‘The two study groups had substantial improvement in back-related disability and pain immediately (3 days) after the procedure, with similar improvement in the two groups. The improvement in each group at 3 days was maintained at 1 month.'
Thirdly, Klazen et al 2009 reported on 431 patients with vertebral fractures. 229 improved within the first week and therefore did not receive a vertebroplasty, that's greater than 50% of presenting patients improved immediately and did not require intervention. The remaining 202 patients were randomly split into two groups:
101 patients were assigned to receive a Vertebroplasty And the other 101 patients were assigned to conservative management. Ten percent of this second group later switched to Vertebroplasty.
Hence 90 percent of the 330 eligible patients improved significantly with in one month without intervention by Vertebroplasty. And 75% of patients with vertebral fracture presenting to the six hospitals participating in this research improved without intervention.
I ask the Question?
"What might have happened to the other 101 patients assigned to Vertebroplasty if they did not receive the procedure?"
Did they need it?? So, just wait 3 days and many will be better. Wait one month and they will all likely be better.
Hence a research study that pits Vertebroplasty against Conservative management and sets up the design of measurement for pain at 24 hours and then at 6 weeks is glossing over true observed data.
This is not Science. This research design is poor by ignorance or favour and will give misleading results about the natural history and need for intervention for Vertebral Fracture and the true effects of Vertebroplasty.
Which is why I have called Vertebroplasty a Convenient Truth. It has been set up to work better than conservative management when in fact it is no better. As such it is neither efficacious or ethical but merely: Convenient.
At this point I want to remind you all about what is good evidence and good research and what is not.
The standards for research quality and power are set by
The Centre for Evidence-based Medicine in Oxford UK.
Five levels of research have been defined with Level one being at the top of the list.
Level one, randomized controlled trails is utilized to determine the effectiveness of a treatment procedure.
Level two involves Cohort Studies where a group of people are followed over time; one with a condition and the other
The Framingham Heart study is an example of a cohort study, where beginning back in 1948 with 5,209 adult subjects have been followed for their heart health, and making available data for the effect of exercise, diet and medication on the cardiovascular disease.
Level 3:
In level 3 case-controlled studies two matched groups are followed. One with a condition and one without- eg smoking versus non smoking.
Level 4:
Case series follows patients after similar treatments but is not as powerful as Level3 case controlled studies.
Level 5:
Level five is the lowest of the list of research power and simply involves expert opinion.
Hulme et al in 2006 reviewed 1000 pubmed articles for vertebroplasty and kyphoplasty (actually this is not true, he does not state that they reviewed 1,000 articles.- If in fact you perform a pubmed search you will come up with 1000 papers for vertebroplasty). They found only 69 to be of scientific value (true). The majority of these were case series. That is level 2 research where patients were followed after a procedure.
There were no studies at level 3,2 or 1. The level one studies came in 2009 and found no difference between Vertebroplasty and a Sham procedure.
The Buchbinder and Kallmes articles invoked Massive Medical Outrage within the medical world of vertebroplasty advocates. D espite level one research, expert opinions abounded, "I have performed 1000 Vertebroplastys", "Volume of PMMA does matters depsite the research", "Vertebroplasty works: We treat a totally different population", "This makes no sense", "Bad news or sham news".
Several letters to the editor were scientifically replied to by the Buchbinder and Kallmes groups. They criticized the standard of research on vertebroplasty up until the time of their papers titling their replies to the medical outrage as when randomized placebo-controlled trials clash with common belief.
Firstly:
the favourable natural history of vertebral fractures they all get better anyway despite what you do.
secondly;
regression to the mean severe pain will improve upon repeated measurement
third:
the placebo effect is accentuated by invasive treatment
fourth:
in unblinded assessments patient expectation is likely to
be influenced by passionate advocates for the treatment
fifth:
those lost to follow up typical have a worse outcomes than
those that don't get lost
In 2009 Buchbinder and her group published a paper in the New England Journal of Medicine reporting on the effect of Vertebroplasty compared to a Sham procedure.
In the exact same edition of the same journal and in fact the very immediate subsequent article was a similar study by the Kallmes group.
Both studies were randomized, blinded, sham controlled clinical trials
The sham versus vertebroplasty proceeded thus:
the patent was laid prone a needle inserted into the periosteum of the pedicles and lidocaine and bupivacaine were infiltrated into the region
Then in the case of the Vertebroplasty procedure an 11 to 13 gauge needle
was used to penetrate the target vertebra and PMMA was injected into the fractured suite.
In the case of the Sham, no penetration occurred only the odour of PMMA was released into the operating room.
As stated previously, all patients in the Kallmes study were substantially better after three days and that improvement was maintained at one month.
So Dr Eugene Carragee writes:
"....if the original case-series reports were too good to be true...."
"....these high-quality studies are certainly too true to be good...."
"....for current practice patterns in the US..."
Vertebroplasty is billed as a "minimally invasive" procedure, however we must remember that it is invasive and it comes with some risks:
-PMMA has turned up in the lungs, heart and nervous system
-some studies report increased facture to adjacent vertebrae after vertebroplasty
-and not all practitioners were finding 90% of their patients were NOT all better.
PMM A also "forever precludes healing and remodelling at the fracture site"
Despite the current potent research, Dr JAmes Weinstein of the journal spine laments:
"...it remains to be seen whether there will be a paradigm shift in the treatment of vertebral compression fractures with vertebroplasty.."
As a final word concerning Vertebroplasty: perhaps a good original story
but followed by poor science and in the end possibly a medical mistake to invest so heavily professionally, emotionally and financially without valid research.
So I say, BEWARE the story.
Beware the original idea that goes on un-researched by the most potent level beware the story becoming accepted medical practice even dogma by habit or intention rather than research and mostly I encourage every single practicing medical professional to be their own scientist and to trust their observations and have the courage to question the story and even propose the experiment that tests the story.
|
