Herniated Disk: What the Evidence Says About Treatment

A herniated disk sounds like a mechanical failure, and the language on MRI reports does nothing to dispel that impression. “Disk extrusion at L5-S1 with moderate neural foraminal narrowing” reads like a structural problem requiring a structural fix. In most cases, it is not. The body is remarkably good at cleaning up herniated disk material on its own, and the evidence consistently shows that most people improve without surgery.

That last sentence is worth repeating, because the MRI report does not say it and the internet often fails to: the natural history of a herniated disk, even one producing significant sciatica, is improvement. Not for everyone. Not always on a timeline that feels reasonable when you are in pain. But in the large majority of cases, conservative management works.

What is happening

A spinal disk has two parts: a tough fibrous outer ring (annulus fibrosus) and a gel-like inner core (nucleus pulposus). A herniation occurs when the inner material pushes through a tear in the outer ring. The degree varies. A bulge is a broad-based outpouching that does not fully breach the annulus. A protrusion extends through the annulus but remains connected to the disk. An extrusion breaks through and may migrate away from the disk space. A sequestration is a free fragment, separated entirely from the parent disk.

If the herniated material contacts or compresses a nearby nerve root, the result is radiculopathy: pain, numbness or weakness along the path of that nerve, commonly experienced as sciatica. The L5-S1 and L4-L5 levels account for the vast majority of lumbar herniations, which is why sciatica typically radiates down the back of the thigh and into the calf or foot.

The pain from a disk herniation has two components. The first is mechanical compression: the herniated material physically pressing on the nerve root. The second is chemical inflammation: the nucleus pulposus contains inflammatory mediators that irritate the nerve even without heavy compression. This is one reason why the size of the herniation on imaging does not always correlate with the severity of symptoms. A small herniation with significant chemical irritation can be more painful than a large one that is not inflaming the nerve.

The key finding from clinical research is that the herniated material is often reabsorbed by the body over time. A systematic review of studies using serial MRI found that spontaneous regression occurs in a majority of cases, with larger herniations tending to reabsorb more completely [3]. This is counterintuitive: the bigger herniations, which cause more alarm, often have the better natural history. The immune system treats the extruded nucleus pulposus as foreign material and sends macrophages to break it down. Sequestered fragments, which have lost their blood supply from the parent disk, are particularly susceptible to resorption.

The imaging problem

Here is the part that alarmed patients rarely hear: disk herniations are extremely common in people with no back pain at all. A systematic review of imaging studies in asymptomatic adults found disk bulges in 30% of 20-year-olds, rising to over 80% of people over 80 [4]. Disk herniations (as opposed to bulges) were found in about 9% of asymptomatic 20-year-olds.

This means an MRI showing a disk herniation does not prove the disk is causing your pain. It may be an incidental finding, and treating it surgically when it is not the pain source would not help. This is one reason NICE guidelines advise against routine imaging for non-specific low back pain [5]. Imaging is appropriate when there are red-flag symptoms (cauda equina signs, progressive neurological deficit) or when symptoms have not improved after a reasonable period of conservative care and surgery is being considered.

The language on MRI reports often amplifies alarm. Terms like “severe,” “significant narrowing” and “disk degeneration” are radiological descriptions, not prognoses. They describe what the disk looks like, not what it will do. A patient reading their own MRI report without this context will almost always overestimate the severity of the finding, and that overestimation can itself slow recovery by increasing fear and avoidance of movement.

Conservative treatment

The evidence for conservative management is strong. A clinical review in the New England Journal of Medicine reported 80-90% improvement rates without surgery [1]. The timeline is typically 6 to 12 weeks for significant improvement, though residual symptoms may take longer to fully resolve.

Stay active. Bed rest worsens outcomes. Walk, move within tolerable limits, and gradually increase activity. The goal is not to push through severe pain but to avoid the deconditioning and fear-avoidance cycle that prolonged rest creates. NICE guidelines explicitly recommend staying active [5].

Pain management. NSAIDs for the inflammatory component, short-term only. They address the chemical inflammation around the nerve root and can make movement more tolerable. Neuropathic pain medications (gabapentin, pregabalin) are sometimes prescribed for radicular symptoms, though the evidence for their use in sciatica specifically is mixed. Oral corticosteroid courses are sometimes used for acute radiculopathy, with some evidence for short-term symptom reduction, but no evidence for long-term benefit.

Direction-specific exercises. If extension centralizes your pain (moves it from the leg toward the back), McKenzie-type exercises have evidence for this specific subgroup. The movement is a prone press-up: lying face down, placing the hands at shoulder level and pressing the upper body up while keeping the hips on the floor. If extension worsens leg symptoms (peripheralization), this is the wrong direction. Flexion-based approaches or neutral spine exercises may be more appropriate. The key is assessing your directional preference, not following a generic protocol.

Time. This is the treatment most people undervalue. The body reabsorbs herniated disk material over weeks to months [3]. The acute pain typically peaks in the first 2 to 4 weeks and improves significantly by 6 to 12 weeks. Knowing this does not make the pain less real, but it does change the decision-making calculus around surgery.

Surgery

The Peul trial randomized 283 patients with sciatica from disk herniation to early surgery or continued conservative care. Surgery provided significantly faster pain relief in the short term. At one year, outcomes were statistically similar [2]. The SPORT observational study found similar patterns: both groups improved substantially, with surgery offering a faster trajectory [6].

There is an important nuance in the SPORT data. The randomized arm of the trial had significant crossover: many patients assigned to conservative care eventually opted for surgery, and some assigned to surgery chose to wait. This makes the as-treated comparison more informative than the intention-to-treat analysis in some respects, and it suggests that the decision between surgery and conservative care is not binary. It is a judgment call, made over time, about how much pain a person is in and how quickly they need relief.

This does not mean surgery is unnecessary. It means it is a reasonable choice for people with severe, persistent symptoms who prefer faster relief and accept the surgical risks, and it is not a failure to avoid it. The evidence supports both paths for the majority of disk herniations.

Surgery is urgent for:

• Cauda equina syndrome: bladder or bowel dysfunction, saddle-area numbness, bilateral leg weakness. This is a surgical emergency. The window for intervention is hours, not days.
• Progressive motor deficit: worsening foot drop or leg weakness despite conservative care. This indicates ongoing nerve damage that may not recover if compression continues.

What the evidence does not support

Chiropractic high-velocity manipulation for acute disk herniation with radiculopathy is not recommended by NICE and has limited evidence of benefit in this specific presentation. The concern is not theoretical: forceful manipulation of a spine with an acute herniation carries a risk of worsening the protrusion.

Spinal decompression tables and traction devices lack consistent evidence. These devices apply intermittent distraction to the lumbar spine with the theory of creating negative intradiscal pressure to retract the herniation. The theory is plausible, but the trials have not shown reliable benefit over sham treatment or natural history.

Prolonged opioid use for disk pain is not recommended due to limited efficacy for neuropathic pain and significant risks including dependence, tolerance and opioid-induced hyperalgesia. Short courses may be appropriate for severe acute pain, but they should not become the long-term management strategy.

Epidural steroid injections have mixed evidence for disk-related sciatica. They may provide short-term relief in some patients, but systematic reviews have not found consistent long-term benefit, and NICE guidelines recommend them only in the context of a specialist assessment when other treatments have not provided adequate relief [5].

Red flags: when to see a doctor

Cauda equina syndrome is a surgical emergency. Its symptoms are bladder or bowel dysfunction (inability to urinate or loss of bowel control), numbness in the saddle area (inner thighs and groin), and bilateral leg weakness. If you develop any of these symptoms, get to an emergency department. This is not a “wait and see” presentation.

Progressive foot drop (difficulty lifting the front of the foot) or rapidly worsening leg weakness also warrant prompt assessment, even without bladder or bowel symptoms. These indicate significant nerve root compression that may require surgical intervention to prevent permanent damage.

When to see a doctor. If your back pain is severe, getting worse, or accompanied by bladder or bowel changes, leg weakness, or numbness in the groin or inner thighs, seek medical care promptly. A GP or physical therapist can assess whether imaging or specialist referral is appropriate.