Conservative treatment for spine pain: what works, what doesn’t, and what carries risk.

Non-steroidal anti-inflammatory drugs (NSAIDs) are the most commonly prescribed medication class for spine pain worldwide. They inhibit cyclooxygenase enzymes (COX-1 and COX-2), reducing prostaglandin synthesis and thereby reducing pain and inflammation. They are effective for short-term symptom management. They are not disease-modifying — they do not repair discs, reduce herniations, or reverse stenosis. (Kuijpers, 2011; Tulder, 2000)

The practical answer: NSAIDs are a reasonable tool for managing acute painful flares — the 5 days when the pain from an acute disc herniation is at its worst, or the days after a long-standing stenosis suddenly worsens. They are not a chronic solution. A patient who has been taking NSAIDs daily for 6 months for spine pain has a poorly managed condition that requires reassessment, not dose escalation.

Muscle relaxants are a heterogeneous drug class. The centrally acting agents (cyclobenzaprine, baclofen, methocarbamol, carisoprodol) work primarily through CNS depression rather than direct action on muscle tissue. Their effect on acute muscle spasm is real but short-lived. Their risk profile for prolonged use is significant. (Kuijpers, 2011)

This question requires separating three distinct types of injections that are often conflated in patient conversations:

The critical point that all three have in common: none of them modify the structural cause of compression. A disc herniation that compresses a nerve does not retract because of an epidural steroid injection. The injection reduces the inflammatory component of nerve root irritation — which can be significant in acute radiculopathy — but the structural compression remains. In a patient who was going to improve spontaneously anyway (which the majority of disc herniations do), the injection accelerates the symptomatic course. In a patient with fixed structural compression, it buys time — with a cost.

Injection-based spine treatments are widely performed, widely underestimated in terms of risk, and in many centers inadequately monitored for complications. A 2022 prospective study by Voelker et al. systematically documented complications in a unicentric cohort — providing some of the most rigorous real-world data on the subject. The complications fall into three categories by severity. (Voelker, 2022)

The instinct to rest completely during severe pain is understandable. The evidence against prolonged bed rest is now one of the most consistent findings in the spine literature. The 2017 Cochrane overview of reviews on physical activity and exercise for chronic pain in adults (Geneen et al., covering dozens of systematic reviews) found that physical activity is consistently superior to rest for chronic pain, and that prolonged immobility leads to measurable physical and psychological deterioration. (Geneen, 2017)

The exception: cauda equina syndrome and acute cord compression require complete rest until surgical decompression. These are true emergencies where movement can worsen the neurological deficit. In every other scenario, the goal is graduated, progressive mobilization as soon as pain permits.

Back braces are one of the most recommended and least evidence-supported interventions in spine care. Their role is context-dependent. (Bell, 1988; Spratt, 1993; Prateepavanich, 2001)

This category encompasses treatments with widely varying evidence bases, cultural adoption, and risk profiles. The honest answer requires separating what the evidence shows for specific diagnoses, rather than giving a blanket endorsement or rejection.

The most important clinical point: spinal manipulation carries documented risks in structural spine pathology. High-velocity manipulation applied to a segment with active disc herniation and radiculopathy, severe stenosis, spondylolisthesis with instability, or osteoporosis can worsen the condition and has caused permanent neurological injury in documented cases. A positive response to manipulation in non-specific LBP does not generalize to structural pathology. (Mierau, 1987; Boswell, 2007)

Exercise is the single most evidence-supported non-pharmacological intervention for chronic spine pain. The distinction that matters clinically is between pain that signals damage and pain that is the expected response to exercise in a sensitized nervous system — commonly called “good pain” versus “bad pain.” Knowing which is which requires clinical judgment that cannot be reduced to a simple rule. (Geneen, 2017; O’Sullivan, 1997)

Exercise type matters for specific diagnoses. Core stabilization exercises have the strongest evidence for spondylolisthesis — they reduce shear forces at the unstable segment. Extension-based exercises (McKenzie protocol) may worsen stenosis and spondylolisthesis by reducing the already-narrowed canal diameter. Flexion-based exercises and cycling are generally well-tolerated in stenosis because they open the canal. Walking in the pool or cycling are often the first aerobic modalities for patients with significant claudication. (O’Sullivan, 1997; Kalichman, 2008)

The relationship between psychological factors and chronic spine pain is not a matter of the pain “being in the patient’s head” — a reductive and harmful framing. It is a matter of understanding how the nervous system modulates the experience and perception of pain signals that are structurally real. Psychosocial factors — what clinicians call “yellow flags” — are among the strongest predictors of who transitions from acute to chronic pain. (Manchikanti, 2008; Hall, 2019)

Cognitive behavioral therapy (CBT) has the highest level of evidence for chronic low back pain among psychological interventions — not as a replacement for structural treatment when it is indicated, but as an adjunct that improves outcomes significantly. A structural problem is not made worse by addressing the psychological component; it is made easier to manage. The biopsychosocial model of pain is not a concession to weakness; it is the most accurate description of how pain works.

Chronic pain is defined by duration: symptoms that persist beyond 12 weeks. Approximately 20% of patients who experience an acute episode of low back pain transition to chronic pain. Understanding why that 20% is different from the 80% who recover is one of the central questions in spine pain research. (Manchikanti, 2014; Simotas, 2000)

The transition from acute to chronic is not simply a matter of time passing. It involves neuroplastic changes in the spinal cord and brain that amplify pain processing — central sensitization. The persistent, unresolved structural stimulus (a herniation that does not reabsorb, a canal that does not decompress) drives this sensitization. But the psychological and behavioral factors described in the previous section independently accelerate the transition, regardless of the structural substrate. This is why two patients with identical MRI findings can have completely different pain experiences and clinical trajectories.