In a Tokyo spine clinic, a 68-year-old woman with mild lower back pain and no leg weakness receives an MRI that shows severe central canal stenosis. The surgeon recommends decompression surgery. She hesitates. She can still walk to the market and tend her garden. But the image—narrowed, compressed—looks alarming. This scenario plays out thousands of times across Japan each year, reflecting a persistent disconnect between what lumbar MRIs show and what patients actually feel.

Japan has one of the highest densities of MRI machines per capita among developed nations, and lumbar spine surgeons have grown accustomed to ordering scans early in the diagnostic workup. Yet a growing body of evidence suggests that the correlation between imaging findings and clinical symptoms is weak at best. As many as half of asymptomatic adults over 60 show signs of stenosis on MRI, and longitudinal studies have found that canal diameter alone predicts disability poorly. Despite this, the image often drives the decision to operate, leading to a clinical-practice gap that health systems are only beginning to address.

The MRI-Symptom Paradox in Japanese Spine Care

Routine lumbar MRI frequently reveals findings that appear severe—ligamentum flavum hypertrophy, facet joint degeneration, a spinal canal narrowed to less than 10 mm in diameter. In many cases, however, the patient reports only mild or intermittent symptoms. This paradox is well documented in the literature: a 2015 systematic review found that 30–50% of asymptomatic adults over 40 have MRI evidence of spinal stenosis. The prevalence increases with age, reaching nearly 80% in those over 70.

Japanese surgeons, trained to trust anatomical detail, often struggle to reconcile a dramatic image with a relatively functional patient. The pressure to act on what is visible can be intense. One orthopaedic surgeon at a university hospital in Osaka described the dilemma: “If I don’t operate on a patient with severe stenosis and they later develop cauda equina syndrome, I could be sued. The image is a liability.” This fear of litigation, combined with a culture that prizes definitive intervention, tilts the balance toward surgery.

Japan’s high MRI density—roughly 55 machines per million population, among the highest in the world—means that access to imaging is rarely a barrier. Patients often expect an MRI as part of their care, and some request one explicitly after seeing advertisements for spine clinics. Once the scan is done, the image becomes a focal point of the consultation. Surgeons may feel compelled to address what they see, even when symptoms are mild. The gap between radiological severity and functional impairment is not unique to Japan, but the country’s healthcare structure amplifies it. Reimbursement policies favor surgical over conservative care: a lumbar decompression procedure earns a hospital roughly ¥500,000 (about US$3,300), while a course of physical therapy brings in a fraction of that. This financial incentive, though rarely acknowledged openly, shapes clinical decisions.

Why Surgeons Trust the Image Over the Patient

Several factors explain why Japanese spine surgeons continue to rely heavily on MRI findings. Training programs emphasize anatomical diagnosis and surgical technique, with less focus on functional outcomes or shared decision-making. Residents learn to identify stenosis on axial and sagittal cuts, but they spend comparatively little time learning how to interpret a patient’s symptom narrative or administer validated disability scores like the Oswestry Disability Index.

Malpractice concerns also play a role. In Japan, the number of medical malpractice lawsuits has risen over the past two decades, and spine surgery is a relatively common target. A surgeon who opts for conservative management in a patient with “severe” stenosis on MRI risks being blamed if the patient later deteriorates. Defensive medicine—ordering scans and offering surgery to avoid litigation—is a real, if unmeasured, driver of over-treatment.

Patient expectations reinforce the pattern. Japanese media often portray spinal surgery as a modern, high-tech solution to back pain. Television programs feature dramatic recoveries after minimally invasive procedures, while conservative care is rarely highlighted. Patients who come to a clinic expecting a scan and a surgical solution may be disappointed by a recommendation for exercise and education. Some seek a second opinion until they find a surgeon willing to operate.

Reimbursement structures further entrench the imaging-first approach. The national fee schedule pays for MRI interpretation and for surgical procedures, but it provides limited reimbursement for the time-intensive counseling that conservative care requires. A 20-minute consultation to discuss watchful waiting or physical therapy generates far less revenue than a 90-minute operation. In a system where hospitals operate on tight margins, the financial logic pushes toward intervention.

Validated functional outcome measures are rarely used in routine practice. Many clinics do not systematically collect patient-reported outcomes before and after treatment. Without data on how patients actually fare, surgeons rely on their memory of successful cases and on the reassuring clarity of an MRI report. This cognitive bias—the availability heuristic—makes it easy to overestimate the benefits of surgery and underestimate the potential for improvement without it.

Evidence That Symptoms Often Don't Match the Scan

The mismatch between imaging and symptoms is not a rare anomaly; it is a well-replicated finding. A landmark study published in the New England Journal of Medicine in 1994 found that 28% of asymptomatic volunteers had a herniated disc on MRI, and the proportion rose with age. Subsequent research on spinal stenosis confirmed the pattern: a 2010 meta-analysis reported that the pooled prevalence of asymptomatic stenosis was 21% in adults under 60 and 57% in those over 80.

Longitudinal studies have tracked patients over time and found that MRI findings change little even as symptoms fluctuate. In the Framingham Heart Study offspring cohort, lumbar spine MRI findings were poorly correlated with back pain or functional limitation after adjusting for age and body mass index. Canal diameter, the most commonly reported measure of stenosis, explained less than 5% of the variance in walking distance or pain scores.

Some clinicians have turned to dynamic MRI or upright imaging, hoping to capture stenosis that appears only when the patient is weight-bearing. Yet these advanced techniques have not improved predictive value. A 2018 study by Kanno et al. comparing supine and upright MRI in patients with suspected stenosis found that the two methods agreed on the presence of severe stenosis in only 60% of cases, and neither predicted symptom severity better than the other.

Age-related changes in the spine are nearly universal after 60. Disc degeneration, facet joint arthritis, and ligament hypertrophy are part of normal aging, not necessarily signs of disease. When these findings are labeled as “pathology” on a radiology report, both surgeon and patient may interpret them as requiring action. The result is a cascade of interventions that may not improve outcomes.

Counter-arguments exist, of course. Some surgeons argue that MRI findings are more predictive in patients with classic neurogenic claudication—pain that worsens with walking and eases with sitting—than in those with nonspecific back pain. They point to studies showing that patients with severe stenosis on MRI are more likely to benefit from decompression than those with mild or moderate stenosis. For instance, a 2016 analysis of the SPORT trial data found that among patients with severe stenosis (canal area < 100 mm²), those randomized to surgery had a significantly greater improvement in walking ability at two years compared with those treated nonoperatively. Similarly, a 2020 cohort study from the Japanese Society for Spine Surgery reported that patients with central stenosis and a cross-sectional area below 75 mm² had a 70% probability of achieving a meaningful clinical improvement with surgery, versus 40% with conservative care. These findings suggest that in carefully selected patients—those with classic claudication and severe anatomical narrowing—MRI can help identify who is most likely to benefit from an operation.

Yet even in this subgroup, the correlation is modest, and many patients improve without surgery. The challenge lies in distinguishing those who will deteriorate without intervention from those who will do well regardless. The Fukushima study, discussed below, provides a useful perspective.

The Fukushima Study on Conservative Care Outcomes

One of the most instructive pieces of evidence comes from a cohort study conducted at Fukushima Medical University. Researchers enrolled more than 200 patients with lumbar spinal stenosis confirmed on MRI and tracked them for two years. All patients received a standardized conservative care program that included exercise, education, and, in some cases, epidural steroid injections. Surgery was offered only if symptoms failed to improve.

The results were striking: roughly two-thirds of patients reported significant improvement in pain and walking ability at the one-year mark, and the improvement was sustained at two years. The MRI grade at baseline—whether mild, moderate, or severe—did not predict who would improve. Patients with the tightest canals were just as likely to benefit from conservative care as those with milder narrowing.

Only 12% of patients in the Fukushima cohort underwent surgery during the follow-up period. Among those who did, the decision was driven more by symptom severity and patient preference than by imaging findings. The study’s lead author, a spine surgeon, noted that the findings challenged the automatic assumption that severe stenosis requires surgical correction. “We should not operate on the MRI; we should operate on the patient,” he said in an interview.

The Fukushima study is not an outlier. Similar results have emerged from randomized trials in the United States and Europe. The Spine Patient Outcomes Research Trial (SPORT), a landmark US study, found that patients with spinal stenosis who received surgery had modestly better outcomes than those treated nonoperatively, but the difference narrowed over time, and many patients in the nonoperative group crossed over to surgery. Importantly, the SPORT trial enrolled only patients with symptoms severe enough to consider surgery, not those with incidental findings on MRI.

Critics of the Fukushima study point out that it was observational, not randomized, and that patients who chose surgery may have been systematically different from those who did not. Yet the study’s strength lies in its real-world design: it reflects what happens when patients are offered conservative care first, with surgery reserved for those who truly need it. The low surgery rate suggests that many operations could be avoided.

Operative Rates Vary Widely Across Japanese Prefectures

If MRI findings alone determined the need for surgery, one would expect operative rates to be relatively uniform across Japan. They are not. A 2019 analysis of administrative data found a threefold variation in lumbar decompression rates across the country’s 47 prefectures. The highest-rate prefecture performed roughly 120 decompressions per 100,000 residents per year; the lowest performed fewer than 40.

What explains the variation? Not population need. The prevalence of disabling low back pain and leg symptoms is similar across prefectures. Instead, the variation correlates with supply-side factors: regions with more MRI machines per capita and more spine surgeons per capita have higher surgery rates. The pattern strongly suggests supply-sensitive care—where the availability of resources drives utilization rather than clinical necessity.

Clinical practice guidelines, issued by the Japanese Orthopaedic Association and other professional bodies, recommend a trial of conservative care before considering surgery for spinal stenosis. Yet adherence to these guidelines is uneven. In high-rate prefectures, surgeons may be more willing to operate early, perhaps because they have shorter waiting lists or because local practice norms favor intervention. In low-rate areas, conservative care may be more readily available, and patients may be more reluctant to undergo surgery.

The variation raises uncomfortable questions about equity and appropriateness. A patient with identical symptoms and MRI findings could receive surgery in Tokyo but be managed conservatively in rural Hokkaido, depending on where they live. Without better data on outcomes, it is impossible to say which approach is more appropriate. But the sheer magnitude of the variation suggests that many operations are performed where they are not needed, and some are withheld where they are.

Efforts to reduce unwarranted variation have gained traction in recent years. The Japanese government has begun publishing regional data on surgical rates, and some insurers have introduced prior authorization requirements for lumbar MRI. But changing deeply ingrained practice patterns takes time.

Bridging the Evidence-Practice Gap: What Works

Several strategies have shown promise in aligning clinical practice with the evidence. Shared decision-making tools, such as decision aids that present the risks and benefits of surgery versus conservative care in plain language, have been shown to reduce surgical rates by 20–30% in some settings. Patients who use these tools often choose less invasive options than they initially expected.

Functional outcome tracking before and after MRI ordering could help contextualize imaging findings. If a patient with severe stenosis on MRI reports minimal disability on a validated questionnaire, the surgeon can use that information to justify a conservative approach. Some Japanese hospitals have begun implementing routine collection of the Oswestry Disability Index and the Zurich Claudication Questionnaire, but adoption remains patchy.

Delayed imaging for uncomplicated low back pain is another evidence-based strategy. International guidelines recommend against routine MRI for nonspecific low back pain in the first four to six weeks, as most patients improve spontaneously. Yet many Japanese patients receive scans earlier, sometimes on the first visit. Encouraging a trial of conservative care before imaging could reduce the number of incidental findings that lead to unnecessary surgery.

Japan’s Choosing Wisely campaign, launched in 2015, has promoted conservative first-line care for low back pain and discouraged routine imaging. The campaign, modeled on the US initiative, has produced patient-friendly lists of procedures to question. However, awareness among both clinicians and patients remains modest, and the campaign has not yet had a measurable impact on imaging rates.

Orthopedic societies are updating appropriateness criteria for lumbar MRI. The Japanese Orthopaedic Association, in collaboration with the Japanese Society for Spine Surgery and Related Research, has published new guidelines that emphasize the importance of correlating imaging findings with clinical symptoms. The guidelines recommend against surgery based on imaging alone and call for a trial of conservative care in most cases. Whether these guidelines will change practice remains to be seen.

For related reading on how clinical practice gaps affect patient outcomes in other settings, see Bipolar Patients in Rural Ethiopia and Kenyan Hypertensive Patients.

Ultimately, reducing over-reliance on MRI will require changes at multiple levels. Surgeons must learn to trust the patient’s story as much as the image. Patients must understand that a narrowed canal on MRI does not inevitably require surgery. And the healthcare system must realign incentives to reward outcomes rather than procedures. Until then, the paradox will persist: a scan that shows too much, and a patient who may need less.

Disclaimer: This article is for informational purposes only and does not constitute medical advice. Individuals with spine-related symptoms should consult a qualified healthcare professional for diagnosis and treatment options.