Cementing a Fractured Vertebra, and Why the Sham Trials Mattered

What the cement procedures actually do

Vertebroplasty threads a needle into a collapsed vertebral body and injects bone cement, which hardens in place. Kyphoplasty adds a step first: a balloon is inflated inside the bone, making a cavity and attempting to push the collapsed vertebra back toward its old height. Both are done through the skin with imaging guidance, and both are aimed at the pain.

The injury they address is a vertebral compression fracture — the front of a vertebral body giving way and wedging, most often in bone thinned by osteoporosis. It does not take much: a cough, a lift, or stepping off a kerb harder than expected can be the whole story, which is part of why the pain arrives with so little explanation attached.

The logic offered for the cement is easy to believe: the bone is broken, the cement makes it solid, solid things do not hurt. Hold that thought. Every procedure below had a story exactly that tidy.

Why "better the next day" cannot answer the question

Here is the problem with the evidence that is easiest to gather. A person has a procedure at their worst moment, and afterward they are better. That sequence is compatible with the cement working, and it is equally compatible with the cement doing nothing at all, because several things improve pain on their own timetable and all of them are running at once.

Three of them, specifically. People seek a procedure when pain is at its peak, and a peak is the point from which things are most likely to move down — so the measurement after almost any intervention looks better than the one before. Expectation does real work: a needle, a theatre, and a serious person in a lead apron are a powerful thing to receive. And time passes, which for many painful conditions is itself a treatment.

"She had it and she was better the next day" is the most persuasive form of evidence there is, and one of the least reliable.

What a sham surgery trial is, and why anyone would run one

A sham-controlled surgical trial gives the control group everything except the part being tested. They are consented, sedated, brought to theatre, positioned, and given the incision or the needle stick; they hear the same sounds and the same commentary; they recover on the same ward with the same instructions. The only thing they do not get is the step supposed to work.

The design exists because everything around a procedure is itself an intervention. Comparing an operation against a waiting list measures the operation plus the theatre plus the expectation plus time. Sham surgery subtracts all of that.

The ethics are less exotic than they sound. Participants consent knowing they may receive the sham, and a committee approves the trial only where the question is genuinely unsettled — where nobody yet knows whether the people getting the real procedure are the lucky ones. That bar is why placebo surgery trials are rare, and why most procedures have never faced one.

That last fact is worth sitting with. A procedure never sham-tested has not been shown to fail. It has not been shown to work either. Only one of those is a verdict.

What happened when two orthopaedic operations met a placebo

Two widely performed orthopaedic operations have been put through exactly this test, and both came out worse than their reputations. Neither is a spine procedure, and neither says anything directly about cement. What they establish is that a plausible mechanism, a confident surgeon, and a corridor of grateful patients are all compatible with an operation that does nothing its placebo does not.

The knee. In people aged 35 to 65 with a degenerative medial meniscus tear and no osteoarthritis, arthroscopic partial meniscectomy was no better than sham surgery for relieving symptoms 1Ref 1Sihvonen R, Paavola M, Malmivaara A, et al. (FIDELITY) (2013).Arthroscopic Partial Meniscectomy versus Sham Surgery for a Degenerative Meniscal Tear.In patients aged 35-65 with a degenerative medial meniscus tear and no osteoarthritis, arthroscopic partial meniscectomy was no better than sham surgery for symptom relief — used as one of the two worked examples of what a sham-controlled surgical trial can reveal.. At the time, arthroscopy for meniscus tear was among the most commonly performed orthopaedic operations in the world, and the mechanical story was impeccable: there is a torn flap in the joint, so remove the torn flap.

The shoulder. Arthroscopic subacromial decompression provided no clinically important benefit over placebo — an arthroscopy without the decompression — or over no treatment, for subacromial shoulder pain 2Ref 2Beard DJ, Rees JL, Cook JA, et al. (CSAW) (2018).Arthroscopic subacromial decompression for subacromial shoulder pain (CSAW): a multicentre, pragmatic, parallel group, placebo-controlled, three-group, randomised surgical trial.Arthroscopic subacromial decompression gave no clinically important benefit over placebo (arthroscopy only) or no treatment for subacromial shoulder pain — used as the second worked example of a plausible mechanical operation that did not beat a convincing placebo.. It rested on a story every bit as tidy as the cement's: a bone spur is rubbing on a tendon, so take the spur away.

Both relieved pain. Both were followed by grateful patients. Neither beat a convincing fake. That is what the CSAW trial and the FIDELITY trial were for.

The method does not always find against the operation

This is not a machine for discrediting surgery, and reading it that way would be its own error. Randomised trials of orthopaedic procedures also find in favour of the operation, and when they do the finding carries the same weight as the negative ones. The point of the method is that it produces an answer nobody controls in advance — including the people hoping the answer is no.

For femoroacetabular impingement syndrome, hip arthroscopy led to modestly better patient-reported hip function at twelve months than personalised physiotherapist-led conservative care, at substantially higher cost 3Ref 3Griffin DR, Dickenson EJ, Wall PDH, et al. (UK FASHIoN) (2018).Hip arthroscopy versus best conservative care for the treatment of femoroacetabular impingement syndrome (UK FASHIoN): a multicentre randomised controlled trial.For femoroacetabular impingement syndrome, hip arthroscopy produced modestly better patient-reported hip function at 12 months than personalised physiotherapist-led conservative care, at substantially higher cost — used as the counterweight showing rigorous trials also find in favour of surgery, and to distinguish a conservative-care comparison from a placebo comparison.. It found for the operation — modestly, and at a price, which is what a real answer usually looks like rather than a rout.

That trial also carries a distinction that does most of the work on this page. It compared surgery against conservative care, not against a placebo, so everyone knew which group they were in — a weaker design for the question of whether the active step does anything. The hip impingement surgery vs physical therapy comparison answers "is surgery better than the alternative?" A sham trial answers "is the surgery doing what we think?" A procedure can pass one and fail the other.

How "did it work" becomes a number

Before a trial can say whether something worked, it has to turn "worked" into a number, and the number is a choice. For back-related disability the usual instrument is the Oswestry Disability Index, a validated ten-section questionnaire about how much back pain interferes with ordinary activity, scored from 0 to 100 percent 4Ref 4Fairbank JCT, Pynsent PB (2000).The Oswestry Disability Index.The Oswestry Disability Index is a validated 10-section patient-reported measure of low-back-pain-related disability scored from 0 to 100% — used to explain how spine-procedure trials operationalise 'did it work' into a number.. Every claim about a spine procedure's benefit is a claim about something like that scale.

This is where confusion is manufactured, usually without anyone intending it. A trial can report a difference that is statistically significant — unlikely to be chance — and clinically invisible, meaning nobody in it would have noticed. Both can be true of one result, and a brochure prints the first.

The useful question is never "was there a difference?" It is "how big, on what scale, and would I have felt it?"

What the scan settles, and what it does not

A scan showing a collapsed vertebra does establish that a vertebra has collapsed. What it does not establish is that the collapse is the source of the pain, and in an older spine the gap between those two statements is wide — because an older spine on a scan is crowded with findings, most of them also present in people who feel nothing at all.

Disc degeneration appears in about 37% of pain-free 20-year-olds and about 96% of pain-free 80-year-olds, and degenerative findings like bulges and protrusions are common in people with no pain at all — such findings frequently do not explain back pain 5Ref 5Brinjikji W, Luetmer PH, Comstock B, et al. (2015).Systematic Literature Review of Imaging Features of Spinal Degeneration in Asymptomatic Populations.Degenerative spine findings on CT/MRI are highly prevalent in pain-free people and rise with age (disc degeneration about 37% at age 20 to about 96% at age 80), and often do not explain back pain — used, explicitly as evidence about degenerative findings rather than fractures, to show that an older spine's imaging is crowded with findings that would be there regardless..

A fresh compression fracture is a different finding from disc degeneration, and that research is not about fractures. The transferable point is about reading: the scan of a painful older back shows a great deal, and most of it would be there regardless. Which finding generates the pain is a clinical judgment made with the person in the room, not one the picture makes.

That judgment is upstream of everything else. If the cemented level is not the one generating the pain, no property of the cement can help. Which vertebra is responsible therefore comes before whether the procedure works, and it is fair to ask a proceduralist to justify that out loud.

When a spine fracture is not a question about evidence

There is a version of this that is not an evidence question at all. Some spine fractures are not osteoporotic collapses — they are the first sign of a cancer or an infection, or they arrive with compression of the nerves. Those are assessed urgently by a spine service rather than weighed against physiotherapy, and sorting that out comes before any conversation about cement.

The signs used to screen for these are the red flags, and they perform less well individually than their reputation suggests: most have high false-positive rates alone. A few do raise the probability of fracture meaningfully — older age, prolonged corticosteroid use, significant trauma 6Ref 6Downie A, Williams CM, Henschke N, et al. (2013).Red flags to screen for malignancy and fracture in patients with low back pain: systematic review.Most individual red flags for spinal fracture or malignancy have high false-positive rates, though some — older age, prolonged corticosteroid use, significant trauma — raise the post-test probability of fracture; used for how red flags are interpreted in combination rather than as a checklist.. That is why they are read in combination and in context, not as a checklist where one hit means a scan.

Where a procedure is clearly the right call, the argument is not about pain scores at all. New weakness in the legs, numbness across the saddle area, or loss of bladder or bowel control alongside a spine fracture is a structural emergency, and the operation that follows is not a treatment for pain — it is the removal of pressure from a nerve, against a clock. A fracture caused by a tumour or an infection is treated for the tumour or the infection.