Percutaneous vertebroplasty is a needle procedure in which bone cement is injected into a fractured vertebral body through a small skin puncture, guided by live X-ray, to stabilize a painful vertebral compression fracture. It is also one of the most genuinely contested procedures in spine care: several rigorous placebo-controlled trials found no meaningful advantage over a sham procedure, while a later trial restricted to very recent, severely painful fractures did find one.
What a vertebral compression fracture is
A vertebral body is the squat, load-bearing block at the front of each spinal segment. When bone density falls far enough, that block can fail under ordinary loads — bending to lift a bag, a hard sneeze, stepping off a curb. The front collapses while the back stays closer to its original height, producing the wedge shape seen on a side-view X-ray.
In the Rochester, Minnesota population study, the age- and sex-adjusted incidence of clinically diagnosed vertebral fracture was 117 per 100,000 person-years, roughly twice as high in women as in men, and 83% of those fractures followed moderate or no trauma.1 A fracture without a fall is a signal about bone quality. Many compression fractures are never diagnosed at all, because they hurt for a few weeks and then stop — which matters enormously when reading the treatment literature.
What the procedure actually involves
Vertebroplasty is not open surgery, despite the phrase “vertebroplasty surgery” being in common use. Nothing is cut open, no bone is removed, no hardware is implanted.
The patient lies face down. The skin and the periosteum — the sensitive lining of the bone — are numbed with local anesthetic. Practice on sedation varies between centers: some perform the procedure on local anesthetic alone, while others add light intravenous sedation, as the VAPOUR trial did.5 Under fluoroscopy (continuous X-ray imaging), a large-bore needle is advanced through the pedicle, the bony bridge connecting the back of the vertebra to the body at the front, until the tip sits inside the fractured vertebral body. Some operators use one pedicle, others both; some centers add CT guidance.
Polymethylmethacrylate (PMMA) bone cement is then mixed to a consistency thick enough to resist running, made radio-opaque so it shows on X-ray, and injected slowly while the operator watches it spread — watching above all for cement moving where it should not, into the spinal canal, the disc space, or a vein. If it starts to, the injection stops.
The cement hardens over a few minutes. It does not restore the vertebra’s original height; it fills and internally splints the fractured bone.
What recovery looks like
Most patients lie flat for a short observation period, commonly a couple of hours, and go home the same day. The mechanical logic of recovery is simple: there is no incision to heal and no implanted hardware to protect, and the cement is already at full strength before the patient leaves. What limits activity afterward is the fracture and the underlying bone, not the procedure.
Soreness at the needle entry site is common for a few days and is distinct from the fracture pain the procedure was meant to address. Whether pain relief arrives at all, and how quickly, is the contested question the trials below address — it is not something the recovery timeline settles.
How vertebroplasty differs from kyphoplasty
Kyphoplasty adds a step. Before cement goes in, a balloon is expanded inside the vertebral body to compact the trabecular bone and create a cavity, then removed; cement is injected into that cavity at lower pressure. The rationale is twofold: partially restore vertebral height, and reduce leakage by injecting into a preformed void rather than forcing cement through intact bone.
A meta-analysis of 16 comparative studies found kyphoplasty reduced the kyphotic wedge angle, increased postoperative vertebral body height, and lowered the risk of cement leakage (relative risk 0.62, 95% CI 0.47–0.80) compared with vertebroplasty — but produced no significant difference in pain or Oswestry Disability Index scores.9 The radiographic pictures differ. The patient-reported outcomes, in that analysis, did not.
What the evidence actually shows
In 2009 two placebo-controlled trials appeared in the same issue of the New England Journal of Medicine, and they reshaped the field.
Buchbinder and colleagues randomized 78 participants with one or two painful osteoporotic fractures under 12 months old and confirmed unhealed on MRI, to vertebroplasty or a sham procedure. Vertebroplasty produced no significant advantage on any outcome at any time point out to six months. Pain fell substantially in both groups; at three months the adjusted between-group difference was 0.6 points on a 0–10 scale (95% CI −0.7 to 1.8).2
The INVEST trial, led by Kallmes, randomized 131 patients with one to three painful fractures to vertebroplasty or a simulated procedure without cement. At one month there was no significant difference in disability (0.7 points on the Roland-Morris scale, P=0.49) or pain (0.7 points, P=0.19), and both groups improved immediately. By three months, 51% of the control group had crossed over to vertebroplasty versus 13% of the treatment group3 — a signal that fits neither camp’s argument neatly.
VERTOS IV, published in 2018, randomized 180 patients with acute osteoporotic fractures to vertebroplasty or a sham procedure and followed them for 12 months. Mean pain fell significantly in both groups at every follow-up point, and the difference between groups was never statistically significant. Analgesic use dropped equally in both.6
The 2018 Cochrane review pooled 21 trials. Against placebo, it found high- to moderate-quality evidence of no clinically important benefit: mean pain was 5 points in the placebo group and 0.7 points better with vertebroplasty (95% CI 0.3 to 1.2) — an absolute reduction of 7%, against a minimal clinically important difference of 15%. That interval excludes zero, so a small average difference was detectable; it sits well below the threshold at which a patient would notice. Effects did not differ between acute and subacute pain. The review also concluded that open trials comparing vertebroplasty with usual care likely overestimated benefit.7
The counter-evidence, and why timing may matter
VERTOS II, an open-label trial, compared vertebroplasty with conservative treatment in fractures under six weeks old and found a large advantage at one month and one year.4 Being unblinded, it carries exactly the bias Cochrane flagged. But one of its numbers stands regardless of blinding: of 431 patients screened as eligible, 229 — 53% — had spontaneous pain relief during the assessment period and never entered the trial. More than half got better on their own, without any procedure.
VAPOUR tested vertebroplasty in the patients most likely to benefit. It enrolled 120 patients with osteoporotic fractures less than six weeks old and severe pain (numeric rating scale 7 or higher out of 10), randomized them to vertebroplasty or a masked placebo procedure, and used a deliberate high-volume vertebral fill technique. At 14 days, 44% of the vertebroplasty group had pain below 4 out of 10 versus 21% of the placebo group — a difference of 23 percentage points (95% CI 6–39; p=0.011).5
A later subgroup analysis suggested the effect concentrated in fractures under three weeks old: 51% versus 20% reaching mild pain at 14 days.8 That is exploratory and hypothesis-generating, not established — but it is the honest explanation of why VAPOUR diverged from earlier trials, which enrolled fractures mostly older than the window VAPOUR targeted.
Put plainly: the evidence does not support vertebroplasty as a routine treatment for compression fracture pain. It leaves open a narrower question — whether patients with a very recent fracture and severe, poorly controlled pain benefit — where one good trial says yes.
Who is considered a candidate
The features that recur across the trials finding benefit: an osteoporotic vertebral compression fracture, recent onset (weeks, not months), severe pain that has not responded to non-procedural management, MRI evidence that the fracture is unhealed and is the level generating the pain, and pain that reproduces on palpation over that segment.
The trial evidence argues against the procedure when the fracture is already healing, or when imaging does not match where the patient hurts. Separately, active infection and uncorrected bleeding disorders are reasons not to put a needle into bone at all. Fractures from tumor, myeloma, or trauma are a different clinical problem.
Vertebroplasty treats a mechanical problem in one bone. It does nothing about the disease that broke the bone. Any patient having this procedure for an osteoporotic fracture needs a parallel treatment plan for the osteoporosis itself, or the next fracture is simply waiting.
Risks and honest expectations
Cement leakage is the characteristic complication. Most leaks are visible on imaging and cause no symptoms; the ones that matter go into the spinal canal or the venous system. Serious adverse events reported in the trials Cochrane included were osteomyelitis, cord compression, thecal sac injury, and respiratory failure. Because these events were few, the review could not establish whether vertebroplasty meaningfully raises the risk of new symptomatic vertebral fractures (relative risk 1.29, 95% CI 0.46 to 3.62) or of other serious adverse events.7 VERTOS IV reported two adverse events in its vertebroplasty arm: one respiratory insufficiency, one vasovagal reaction.6
No honest account of this procedure promises a cure or a specific success rate. In the trial finding the largest benefit, fewer than half the treated patients reached mild pain at two weeks.5 Pain also improves substantially without any procedure — that happened in every placebo arm above. The real question is whether waiting is tolerable, not whether the procedure is the only path.
If you are in the St. Louis area
Padda Institute evaluates vertebral compression fractures at its offices in St. Louis and Bridgeton, Missouri. For local detail on evaluation, imaging, and scheduling, see the vertebroplasty for compression fractures page or the kyphoplasty page.
Frequently asked questions
Is vertebroplasty considered surgery?
No. It is done through a needle under X-ray guidance with local anesthetic — no incision, no bone removal, no implanted hardware. It belongs to the category of image-guided interventional procedures rather than open spine surgery, like most options listed under pain treatments.
How long does the pain relief last?
The trials measured outcomes out to six and twelve months, so anything beyond that is not established. When relief does occur, it comes from stabilizing that specific fractured bone, and nothing in the procedure prevents the next fracture at another level — which is why osteoporosis treatment has to run alongside it. See the vertebroplasty page for how that evaluation is structured.
Should I choose vertebroplasty or kyphoplasty?
Comparative data show kyphoplasty restores more vertebral height and leaks less cement, but has not been shown to produce better pain or function scores. The decision usually turns on fracture morphology and height loss. The kyphoplasty page describes where the balloon step is worth adding.
What happens if I do nothing?
A large share of acute vertebral compression fractures settle on their own — in VERTOS II, 53% of screened patients improved spontaneously before randomization. Doing nothing procedural is a legitimate option, particularly when pain is controllable. See Dr. Gurpreet Singh Padda, MD, MBA, MHP on how treatment decisions are framed.
Why do some doctors say vertebroplasty does not work?
Because several well-designed placebo-controlled trials, and the Cochrane review pooling them, found no clinically important benefit over a sham procedure. That position is evidence-based, not dismissive. The counter-position rests mainly on VAPOUR, which studied a narrower group. Both readings can be held honestly — bring the question to your evaluation, starting at the contact page.
What if my pain is not from the fracture?
This is the most common reason vertebroplasty fails to help. Back pain in older adults frequently comes from facet joints, discs, the sacroiliac joint, or compressed nerve roots, and an old fracture visible on imaging may be incidental rather than the pain generator. Each of those points to a different treatment — targeted joint or nerve procedures for the first three, and for a smaller group with persistent nerve-related pain, options such as spinal cord stimulation.
Sources
- Cooper C, Atkinson EJ, O’Fallon WM, Melton LJ. Incidence of clinically diagnosed vertebral fractures: a population-based study in Rochester, Minnesota, 1985-1989. J Bone Miner Res. 1992. PMID 1570766
- Buchbinder R, Osborne RH, Ebeling PR, et al. A randomized trial of vertebroplasty for painful osteoporotic vertebral fractures. N Engl J Med. 2009. PMID 19657121
- Kallmes DF, Comstock BA, Heagerty PJ, et al. A randomized trial of vertebroplasty for osteoporotic spinal fractures. N Engl J Med. 2009. PMID 19657122
- Klazen CAH, Lohle PNM, de Vries J, et al. Vertebroplasty versus conservative treatment in acute osteoporotic vertebral compression fractures (Vertos II): an open-label randomised trial. Lancet. 2010. PMID 20701962
- Clark W, Bird P, Gonski P, et al. Safety and efficacy of vertebroplasty for acute painful osteoporotic fractures (VAPOUR): a multicentre, randomised, double-blind, placebo-controlled trial. Lancet. 2016. PMID 27544377
- Firanescu CE, de Vries J, Lodder P, et al. Vertebroplasty versus sham procedure for painful acute osteoporotic vertebral compression fractures (VERTOS IV): randomised sham controlled clinical trial. BMJ. 2018. PMID 29743284
- Buchbinder R, Johnston RV, Rischin KJ, et al. Percutaneous vertebroplasty for osteoporotic vertebral compression fracture. Cochrane Database Syst Rev. 2018. PMID 30399208
- Diamond T, Clark W, Bird P, et al. Early vertebroplasty within 3 weeks of fracture for acute painful vertebral osteoporotic fractures: subgroup analysis of the VAPOUR trial and review of the literature. Eur Spine J. 2020. PMID 32170438
- Wang B, Zhao CP, Song LX, Zhu L. Balloon kyphoplasty versus percutaneous vertebroplasty for osteoporotic vertebral compression fracture: a meta-analysis and systematic review. J Orthop Surg Res. 2018. PMID 30348192
Dr. Gurpreet Singh Padda, MD, MBA, MHP
