The Long Argument Over Fusing a Painful Spine

Does spinal fusion work for back pain?

For chronic low back pain arising from wear in an otherwise sound spine, there is no confident yes — and that absence is the whole controversy compressed into a sentence. The operation is technically accomplished at what it does. What is disputed is whether what it does has anything to do with what hurts.

A fusion locks two or more vertebrae into a single block of bone, on the reasoning that a painful segment which no longer moves can no longer generate pain. The reasoning is elegant, and its weak point arrives immediately: identifying which segment is the culprit.

That step is far harder than it sounds, because the wear that marks a segment as suspicious is close to universal. Degenerative findings on CT and MRI — disc degeneration, bulges, protrusions — are highly prevalent among people with no symptoms whatsoever, and become more so with every decade: disc degeneration turns up in roughly 37% of pain-free 20-year-olds and 96% of pain-free 80-year-olds 1Ref 1Brinjikji W, Luetmer PH, Comstock B, et al. (2015).Systematic Literature Review of Imaging Features of Spinal Degeneration in Asymptomatic Populations.Supports that degenerative spine findings on CT/MRI are highly prevalent in pain-free people and rise with age — disc degeneration in 37% at age 20 to 96% at age 80 — so such findings often do not explain back pain..

A degenerated disc on a scan is closer to a description of age than a diagnosis of pain. Finding one does not establish that it is the source of anything.

What is actually in dispute

The disagreement has a precise boundary, and almost every public version of it ignores that boundary. It concerns adults with persistent low back pain and wear on imaging, whose nerves are not being compressed and whose spine is not unstable, deformed, broken, infected, or invaded by tumor. That is the largest group of people with bad backs, and the group for whom nobody can point to a broken part. That is where fusion is contested.

It is not contested elsewhere. A vertebra crushed by trauma, a segment destroyed by infection or tumor, a spine collapsing into deformity, a slip that has been shown to move — these are structural failures, and rebuilding a structure that has failed is a different proposition from silencing a joint that aches. The argument has never been about those, and it is not about them now.

Holding that boundary in view matters, because both camps get quoted outside it. "Fusion doesn't work" is a claim about the contested middle, not about a fractured spine. "Fusion gave me my life back" is usually a claim about one person, whose segment may genuinely have been the culprit. Neither settles anything for the person in the contested middle — where nearly everyone reading about this actually sits.

Both sides are looking at the same scan

A useful thing to understand is that this controversy is not a fight over the facts on the film. The surgeon proposing a fusion and the researcher questioning it see the same dark, flattened disc at the same level. They agree it is there. They agree it is worn. They disagree about what its presence licenses anyone to infer.

The evidence pushes hard against the intuitive reading. Because degenerative change is so common in people who feel nothing at all, and rises steeply with age, its mere presence carries very little diagnostic weight 1Ref 1Brinjikji W, Luetmer PH, Comstock B, et al. (2015).Systematic Literature Review of Imaging Features of Spinal Degeneration in Asymptomatic Populations.Supports that degenerative spine findings on CT/MRI are highly prevalent in pain-free people and rise with age — disc degeneration in 37% at age 20 to 96% at age 80 — so such findings often do not explain back pain.. A finding shared by most pain-free people of the same age cannot, on its own, explain why this particular person hurts. It is a background feature of a lived-in spine.

This is what makes the disagreement durable. The scan does not settle it, and the scan is the most vivid object in the room. A patient looking at their own MRI sees something unmistakably wrong and unmistakably permanent, and reaches the obvious inference — precisely the one the prevalence data will not support. No amount of looking harder at the image resolves it, because the question was never about the image.

The argument is really about a number on a questionnaire

Before asking whether fusion works, it is worth knowing what "works" is measured with, because the answer is considerably less dramatic than the word implies. Back pain trials, fusion trials among them, generally report the Oswestry Disability Index: a validated ten-section patient-reported measure, scored from 0 to 100%, that captures how much back pain interferes with lifting, walking, sitting, sleeping, travel, and social life 2Ref 2Fairbank JCT, Pynsent PB (2000).The Oswestry Disability Index.Supports the description of the Oswestry Disability Index as a validated 10-section patient-reported measure of low-back-pain-related disability scored from 0 to 100%, and how its scores are interpreted..

The Oswestry Disability Index (ODI) is the standard yardstick for back-related disability — 0 means the pain interferes with nothing, 100% means it dominates everything.

The index is a good instrument, and that is exactly why it clarifies things here. When a trial reports one group finishing a few ODI points ahead of another, that is a real measured difference on a validated scale. It is also, frequently, a difference a person would struggle to detect from inside their own week. A handful of points on a hundred-point disability scale is not the transformation the word "success" conjures when a surgeon and a patient use it across a desk.

So a great deal of the fusion argument is not a disagreement about data at all. Both sides accept the numbers. They disagree about whether a benefit of that size justifies an operation of this size — and that is a question about values, not about evidence, which is why more studies keep failing to end it.

Where fusion has been tested, and what it was measured against

Direct randomized evidence about fusion is thinner than the operation's popularity would suggest, and the clearest trial addresses an adjacent question rather than the contested middle itself. It asked whether adding an instrumented fusion to a decompression helped people with lumbar spinal stenosis — a condition where nerves genuinely are crowded. The answer was no. The fused group did no better clinically at two years or at five than those who had decompression alone, while the larger operation cost more and demanded more of the patient 3Ref 3Försth P, Ólafsson G, Carlsson T, et al. (2016).A Randomized, Controlled Trial of Fusion Surgery for Lumbar Spinal Stenosis.Supports that adding instrumented fusion to decompression for lumbar spinal stenosis did not improve clinical outcomes at two or five years versus decompression alone, while increasing cost and operative burden..

That result deserves reporting precisely, because it gets stretched both ways. It does not show that fusion never helps anyone. It shows that in a population where fusion was routinely added as a precaution, the addition bought nothing measurable and charged for the privilege.

The other half of any verdict is the comparison. Exercise therapy for chronic low back pain probably reduces pain and improves function against no treatment, usual care, or placebo, with small-to-moderate effects 4Ref 4Hayden JA, Ellis J, Ogilvie R, Malmivaara A, van Tulder MW (2021).Exercise therapy for chronic low back pain.Supports that exercise therapy probably reduces pain and improves function in chronic non-specific low back pain compared with no treatment, usual care, or placebo, with small-to-moderate effects., and guidelines place non-drug treatment first for back pain — exercise, multidisciplinary rehabilitation, and related approaches — with medication in a secondary role 5Ref 5Qaseem A, Wilt TJ, McLean RM, Forciea MA (American College of Physicians) (2017).Noninvasive Treatments for Acute, Subacute, and Chronic Low Back Pain: A Clinical Practice Guideline From the American College of Physicians.Supports that guidelines place non-pharmacologic treatment — including exercise and multidisciplinary rehabilitation — first for low back pain, with medication in a secondary role.. Small-to-moderate deserves an honest reading: rehabilitation is not a cure, and anyone who has finished the program still hurting knows it. But once the yardstick is a few ODI points, an unglamorous rehab program becomes a genuinely difficult opponent to beat.

Why a contested operation stays common

If the evidence is this equivocal, the obvious question is why fusion rates look the way they do — and the answer is not that surgeons are cynical. Low-value care for low back pain, spanning unnecessary imaging, opioids, injections, and surgery, is widespread internationally, and clinicians have publicly called for it to be reduced 6Ref 6Buchbinder R, van Tulder M, Öberg B, et al. (2018).Low back pain: a call for action.Supports that low-value care for low back pain — unnecessary imaging, opioids, injections, and surgery — is widespread globally and has been the subject of published calls to reduce it.. Fusion is one entry on a long list, and the list exists because ordinary pressures, not bad actors, produce it.

Several of those pressures point the same direction. Someone in year three of severe pain has exhausted patience along with treatments and wants something decisive done. A scan has already produced a finding that looks exactly like an explanation 1Ref 1Brinjikji W, Luetmer PH, Comstock B, et al. (2015).Systematic Literature Review of Imaging Features of Spinal Degeneration in Asymptomatic Populations.Supports that degenerative spine findings on CT/MRI are highly prevalent in pain-free people and rise with age — disc degeneration in 37% at age 20 to 96% at age 80 — so such findings often do not explain back pain.. An operation is concrete, scheduled, and feels like action, while another course of exercise feels like being told to wait again. And the person who improves after a fusion remembers it vividly, while the person who would have improved anyway has no way of knowing that they did.

Being offered a fusion is not a sign that someone is trying to sell an operation. More often it means a clinician has run out of gentler things to offer and wants to help.

Where the argument ends

There is a version of this article that finishes with "avoid the knife," and it would be wrong. Fusion is the right operation, and sometimes the only one, where a spine has genuinely failed as a structure — a vertebra broken by trauma, destroyed by infection or tumor, a progressive deformity, or an instability that has been demonstrated rather than assumed. In those situations the surgery restores something measurably broken, and none of the debate above applies.

The low back red flags that point toward that territory — fever, unexplained weight loss, a cancer history, a significant injury, new bowel or bladder trouble — move a back problem out of the contested middle and into medicine nobody is arguing about.

For the contested middle, the useful posture is neither refusal nor consent but specificity. Which segment is being fused, and what implicates it beyond its appearance on a film? What does success look like in points on a scale, and how many? What has been tried first, and for how long? And what does the operation ask in return — lumbar fusion recovery, the hardware, the months of spinal fusion healing time — set against a benefit measured in single digits 3Ref 3Försth P, Ólafsson G, Carlsson T, et al. (2016).A Randomized, Controlled Trial of Fusion Surgery for Lumbar Spinal Stenosis.Supports that adding instrumented fusion to decompression for lumbar spinal stenosis did not improve clinical outcomes at two or five years versus decompression alone, while increasing cost and operative burden.?

Those questions are not obstruction, and a good surgeon welcomes them. They are how a person finds out whether they are the patient this operation was designed for — the only version of the question that has an answer.