Spinal traction has a plausible mechanical rationale for relieving stenosis symptoms, but the clinical evidence is split depending on where in the spine the narrowing occurs. For lumbar spinal stenosis, the most common form, rigorous trials have failed to show that traction adds meaningful benefit beyond standard exercise. For cervical stenosis and related nerve compression in the neck, the picture is more encouraging, with moderate-quality evidence that mechanical traction reduces pain and disability when combined with physical therapy. The gap between what traction does to the spine on an imaging table and what it does for the patient walking out the door is the central tension in this topic.
Why Traction Seems Like It Should Work
Spinal stenosis develops when the canal or openings where nerves exit the spine get narrower. In the lumbar spine, the usual culprits are thickening of the ligament that lines the back of the canal, arthritic changes in the facet joints, and bulging or collapsing discs.
1PubMed Central. Lumbar Spinal Stenosis: Pathophysiology, Biomechanics, and Innovations in Diagnosis and Management That narrowing squeezes nerve roots and disrupts blood flow around them, triggering pain, numbness, and the heavy-legged fatigue that makes walking difficult.2PubMed Central. Pathophysiology, diagnosis and treatment of intermittent claudication in patients with lumbar canal stenosis
Traction pulls the spine apart along its length, and imaging studies confirm that this actually opens space. In the cervical spine, researchers using MRI during graded traction found that foraminal area (the opening where the nerve exits) increased by about 6% at the lightest pull and nearly 19% at the heaviest, with foraminal height rising in step.3PubMed. Quantitative changes in the cervical neural foramen resulting from axial traction: in vivo imaging study A separate cadaver and imaging study found that cervical traction enlarged the cross-sectional area of the foramen to roughly 120% of the resting measurement at most spinal levels.4Spine. The Influence of Cervical Traction, Compression, and Spurling Test on Cervical Intervertebral Foramen Size Cadaveric work on manual cervical distraction also shows that pressure inside the disc drops during the pull, with the largest decreases occurring when the neck is gently flexed forward during traction.5PubMed Central. Intradiscal Pressure Changes during Manual Cervical Distraction: A Cadaveric Study
So traction demonstrably creates more room around compressed nerves, at least temporarily and at least in the cervical spine. The question is whether that temporary space translates into lasting symptom relief once the pull stops and the patient stands up.
Lumbar Stenosis and the Disappointing Trial Results
For the lower back, the answer has been largely no. A randomized trial comparing mechanical lumbar traction plus extension-based exercises against extension exercises alone found no evidence that adding traction made any difference. Patients in both groups improved, but the traction group did not improve more.6PubMed. The Effectiveness of Mechanical Traction Among Subgroups of Patients With Low Back Pain and Leg Pain: A Randomized Trial That finding held even when researchers looked at predefined subgroups thought to be the best candidates for traction.
A Cochrane systematic review of non-surgical treatments for lumbar stenosis with neurogenic claudication (the leg symptoms that flare with walking) evaluated multiple physical therapy trials. None of the trials that included some form of exercise demonstrated improved walking ability.7Cochrane Database of Systematic Reviews. Nonoperative treatments for lumbar spinal stenosis with neurogenic claudication While traction was not always the specific intervention tested, these findings form part of the broader evidence base showing that physical modalities for lumbar stenosis have struggled to outperform simple exercise programs in controlled settings.
An older systematic review of randomized controlled trials on traction for back pain more broadly found that the vast majority of studies were of poor methodological quality. Out of 13 trials in one strand of the analysis, only one scored high on quality metrics, and nine reported negative findings. Of the three that reported positive results, none were among the higher-quality studies.8Journal of Orthopaedic & Sports Physical Therapy. The efficacy of traction for back pain: a systematic review of randomized controlled trials The pattern is consistent: when trial design is rigorous and sham-controlled, lumbar traction tends not to separate from control treatments.
This does not mean traction never helps anyone’s low back. It means that across groups of patients, the average benefit is too small or inconsistent to detect reliably in trials. Some individuals clearly feel better after traction sessions. But the same can be said of many treatments that do not outperform placebo at the group level, and clinicians have to weigh that reality.
Cervical Stenosis Tells a Different Story
The neck is where traction has its strongest clinical case. A systematic review and meta-analysis of randomized trials in cervical radiculopathy (nerve root compression in the neck, which overlaps with cervical stenosis) found moderate-quality evidence that mechanical cervical traction combined with other physical therapy procedures is more effective than physical therapy alone. The effect size for pain reduction was large and statistically significant at both short-term and mid-term follow-ups. At long-term follow-up the effect size remained large but lost statistical significance, likely because the studies were small and varied in methods.9Physical Therapy. Cervical Radiculopathy: Effectiveness of Adding Traction to Physical Therapy—A Systematic Review and Meta-Analysis of Randomized Controlled Trials
Manual cervical traction (performed by hand rather than machine) also showed a large, significant effect on pain at short-term follow-up in the same review, though the evidence quality was rated lower because fewer studies examined it. The review’s overall conclusion was that adding mechanical traction to a physical therapy program for cervical nerve compression produces meaningful pain and disability improvements that a therapy program alone does not achieve.
A retrospective study of home cervical traction using a simple over-the-door device found symptomatic relief in about 81% of patients with mild to moderately severe cervical spondylosis syndromes. Patients with the mildest symptoms had a 100% improvement rate, those with moderate symptoms improved about 77% of the time, and even patients with radiculopathy improved in 90% of cases.10PubMed. Efficacy of home cervical traction therapy The study was retrospective and uncontrolled, so those numbers reflect a combination of traction benefit and natural symptom fluctuation. Still, the consistency across severity grades is notable.
Why would traction work better in the neck than the low back? One reason is anatomy. The cervical spine is lighter and more mobile, so relatively modest traction forces can open the foramina by a meaningful percentage, as the imaging studies confirm. The lumbar spine bears the full weight of the upper body, and the forces required to distract it appreciably are much larger. Some clinicians also point out that lumbar stenosis tends to be more structurally advanced by the time it causes symptoms, with bony overgrowth and ligament thickening that traction forces cannot reverse.
Does the Type of Traction Device Matter?
The variety of traction devices on the market is staggering. A review of spinal traction history cataloged more than a dozen distinct lumbar traction methods alone, ranging from sliding-table devices and motorized intermittent traction units to inversion tables, seated traction systems, and manual flexion-distraction techniques.11Journal of Vertebral Subluxation Research. A History of Spine Traction All of them use some combination of axial distraction or flexion-based forces, but they differ in angle, force magnitude, and whether the pull is continuous or intermittent.
For the cervical spine, a randomized trial compared three approaches: exercise alone, exercise with clinic-based mechanical traction, and exercise with a home over-the-door traction unit. At six months, patients in the mechanical traction group had meaningfully lower disability scores than both other groups. The over-the-door group fell in between, and by twelve months the mechanical traction group still maintained a statistically significant advantage over exercise alone.12Journal of Orthopaedic & Sports Physical Therapy. Exercise only, exercise with mechanical traction, or exercise with over-door traction for patients with cervical radiculopathy, with or without consideration of status on a previously described subgrouping rule: a randomized clinical trial This suggests that clinic-based mechanical devices, which allow more precise force control, produce better outcomes than simpler home setups. The over-the-door device is not useless, but patients relying on it alone may be getting a diluted version of the benefit.
A separate randomized trial found that both Saunders-device cervical traction and high-intensity laser therapy improved pain and function in patients with cervical spondylosis, both immediately and in the medium term after treatment.13PubMed Central. The Effectiveness of Cervical Spondylosis Therapy with Saunders Traction Device and High-Intensity Laser Therapy: A Randomized Controlled Trial Since both active treatments worked, this trial does not tell us much about traction versus doing nothing, but it does indicate that standardized traction devices produce consistent, measurable improvements.
Who Is Most Likely to Benefit
One reason traction trials often come back negative at the group level is that the patients selected for those trials are heterogeneous. Not everyone with spinal stenosis has the same anatomy, symptom pattern, or stage of disease, and lumping them all together can wash out genuine benefits in subgroups.
Researchers have tried to address this by developing clinical prediction rules. One study identified four variables that together dramatically changed the probability of responding to mechanical lumbar traction: not working a manual labor job, low fear-avoidance beliefs about physical activity, no neurological deficit, and being over 30. When all four were present, the probability of short-term improvement with traction jumped from about 19% to roughly 69%.14PubMed Central. A clinical prediction rule for classifying patients with low back pain who demonstrate short-term improvement with mechanical lumbar traction That is a huge shift, and it suggests that traction’s poor group-level results may hide real benefit in specific patients. The catch is that this prediction rule has not been extensively validated, so it is more of a hypothesis-generating finding than a reliable screening tool.
The absence of neurological deficit in that rule is interesting. You might expect that patients with more nerve compression would benefit most from a treatment designed to open space around nerves, but the opposite seems true for lumbar traction. Patients with frank weakness or reflex changes may have structural compression too severe for traction forces to meaningfully relieve. Cervical traction research partly mirrors this: the retrospective study of home traction found high improvement rates across severity levels, but the few patients who worsened were in the moderate-severity group, suggesting that some degree of structural disease responds unpredictably to pulling forces.10PubMed. Efficacy of home cervical traction therapy
Safety Considerations
Traction is generally considered a low-risk treatment. The forces involved in cervical traction are modest (typically in the range of 5 to 15 kg for mechanical devices), and lumbar traction usually uses somewhere around 35 to 50% of body weight. Serious complications are rare but not zero. Recognized contraindications include unstable spinal fractures, active infections or tumors affecting the spine, severe osteoporosis, and conditions where the spinal cord is already severely compressed and any manipulation might worsen it.
In the cervical traction studies, a small number of patients reported symptom aggravation. In the home traction study, five out of 44 patients with moderate cervical spondylosis felt their symptoms were worsened by traction.10PubMed. Efficacy of home cervical traction therapy That roughly 11% aggravation rate in the moderate group is worth knowing about, because it means cervical traction is not a universally benign intervention even when it helps the majority.
For lumbar traction, the risk profile is similarly mild but includes temporary increases in pain, muscle spasm, and in very rare cases worsening of disc herniation. People with spinal stenosis who also have spondylolisthesis (a vertebra that has slipped forward) need particular caution, since traction could theoretically increase the instability. Any traction program should start with low forces and progress gradually while monitoring symptoms.
Where Traction Fits in a Treatment Plan
When physical therapists are surveyed about how they actually treat lumbar spinal stenosis, traction is not at the top of the list. In one study, the treatments most frequently advocated by therapists were flexibility exercises, stabilization exercises, and strengthening, each endorsed by more than 80% of respondents. Heat, ice, acupuncture, and joint mobilization followed. Patients themselves reported receiving massage, strengthening, flexibility work, and heat or ice most commonly.15IOS Press / PubMed Central. Physical therapy treatment options for lumbar spinal stenosis Traction is used, but it tends to be one tool among many rather than a centerpiece.
This matches the evidence. For lumbar stenosis, the foundation of conservative management is exercise, particularly flexion-based movements that open the spinal canal (think: leaning forward, cycling, seated activities). Traction might offer short-term comfort for some patients and could be tried as a supplement, but building a treatment plan around it is hard to justify given the trial data. For cervical stenosis and radiculopathy, the case for including traction is much stronger, especially mechanical traction delivered in a clinical setting as part of a broader rehabilitation program.
Inversion Tables and Consumer Devices
Inversion tables, gravity boots, and over-the-door cervical traction units are widely sold for home use, and many people with stenosis try them hoping to avoid surgery. Inversion therapy applies traction by flipping the body partially or fully upside down, letting gravity pull the spine apart. This is one of the many axial distraction methods cataloged in the traction literature, and it does produce some spinal decompression. But there are specific concerns for people with stenosis.
Being inverted raises blood pressure and intraocular pressure, which makes it risky for anyone with uncontrolled hypertension, glaucoma, or cardiovascular disease. For lumbar stenosis specifically, full inversion puts the lumbar spine into extension (arching backward), which is the position that typically worsens stenosis symptoms by further narrowing the canal. The people most likely to benefit from lumbar decompression in a flexed position are getting the opposite posture on an inversion table. Partial inversion at a mild angle may avoid some of these problems, but the forces are correspondingly lower and the decompression effect smaller.
Over-the-door cervical traction units fare better in the evidence, as noted in the trial data, but they produce smaller and less consistent benefits than clinic-based mechanical devices. If you are considering a home device, it is worth at least starting with supervised sessions to learn the correct positioning and force range before translating it to your living room.
The Placebo Problem in Traction Research
One persistent challenge in evaluating traction is that it is nearly impossible to blind. Patients know whether they are being pulled or not, which makes placebo-controlled trials difficult. Some researchers have used sham traction (applying the device at very low forces that should not produce a mechanical effect), but patients can often tell the difference. This matters because spinal stenosis symptoms fluctuate naturally, and any treatment that involves lying down, relaxing, and receiving hands-on clinical attention is likely to produce some improvement regardless of its specific mechanical action.
The systematic review of lumbar traction trials highlighted this exact issue: studies ranged from poor to moderate quality, and the ones with better design tended to show negative results.8Journal of Orthopaedic & Sports Physical Therapy. The efficacy of traction for back pain: a systematic review of randomized controlled trials This is a pattern seen across many physical medicine interventions, where early, enthusiastic reports give way to more sobering findings as methodology improves. It does not necessarily mean traction is pure placebo, but it does mean that some of the benefit patients report is likely attributable to the therapeutic context rather than the traction force itself.
For cervical traction, the evidence is more robust partly because the trials that have been done are of somewhat better quality and the effect sizes are large enough to emerge even with imperfect blinding. When a meta-analysis finds a large, significant effect using conservative statistical methods, the finding is harder to dismiss as pure placebo, even if the trials were not perfectly blinded.9Physical Therapy. Cervical Radiculopathy: Effectiveness of Adding Traction to Physical Therapy—A Systematic Review and Meta-Analysis of Randomized Controlled Trials
When Traction Cannot Replace Surgery
Severe spinal stenosis, particularly in the lumbar spine, eventually causes progressive neurological loss: worsening weakness, bladder or bowel dysfunction, or unrelenting pain that no longer responds to conservative measures. At that point, traction and other non-operative treatments are not a substitute for surgical decompression. The decision to operate is typically based on the severity of neurological compromise and the degree to which symptoms interfere with daily life, not on whether traction has been tried first.
That said, many people with mild to moderate stenosis never need surgery. Symptoms wax and wane, and some people stabilize or improve with time, exercise, and activity modification alone. Traction occupies an uncertain middle ground in this picture. For cervical nerve compression, it can be a genuinely useful non-surgical tool that reduces pain enough to keep someone functional and out of the operating room. For lumbar stenosis, the evidence says it is unlikely to be the deciding factor, though it may provide short-term comfort that makes an exercise program more tolerable. The honest summary is that traction’s value depends heavily on where your stenosis is, how advanced it is, and whether you are the kind of patient who tends to respond, a determination that medicine still cannot make reliably for most individuals.