Lumbar Spinal Stenosis

Typical symptoms

The characteristic clinical presentation is neurogenic claudication. Symptoms may be bilateral, unilateral, or markedly asymmetric:

• Pain, heaviness, tingling, or numbness in the buttocks and legs when walking or standing for extended periods
• Relief on sitting or leaning forward ("shopping trolley sign")
• Limited walking distance, often varying from day to day
• Frequently better tolerance of cycling or uphill walking than upright walking or going downhill
• Possible weakness or unsteadiness in the legs
• Back pain may be present but is not always the dominant feature

Walking distance does not necessarily become progressively shorter. With mild to moderate, tolerable symptoms, the course often remains stable over years; spontaneous improvement is also possible.

Diagnosis: the patient comes before the MRI

Assessment begins with a thorough history and clinical examination. I am interested not only in pain intensity and walking distance, but also in the precise pattern of aggravating and relieving factors, neurological deficits, balance, hip mobility, and leg circulation. Vascular disease, hip osteoarthritis, and peripheral neuropathy can produce similar symptoms — and may co-exist with stenosis.

Depending on the situation, the following investigations follow:

• MRI: standard investigation for visualising the spinal canal, nerve roots, discs, and ligaments
• CT: particularly useful for bony detail or when MRI is not possible
• Standing radiographs: when spondylolisthesis, scoliosis, or spinal malalignment is suspected
• Walking or stress test: for a more objective assessment of functional limitation

Imaging should explain the clinical working diagnosis and localise the affected segment. Severe narrowing on MRI can produce few symptoms; conversely, less dramatic imaging with a matching clinical pattern may be clinically significant.

Flexion-extension radiographs are not part of routine assessment and are rarely used in my practice. They may provide additional information for a specific surgical planning question, but no single measurement drives the treatment decision.

Surgical techniques

Decompression

The aim of surgery is to restore adequate space for the compressed nerves. Thickened ligament and small amounts of bone or joint tissue are selectively removed. Load-bearing structures are preserved wherever possible.

Decompression can be performed microsurgically, through tubular retractors, or endoscopically. For appropriate indications, all these techniques can achieve good outcomes. What matters most is not the label of the technique, but complete decompression at the correct level without unnecessary tissue damage or destabilisation.

Endoscopic techniques can reduce blood loss and hospital stay in selected patients. However, the randomised evidence base is smaller and covers shorter follow-up than for established microsurgical techniques; general clinical superiority has not been demonstrated. Surgery under local or regional anaesthesia is also only suitable for selected indications and patients.

Is fusion also needed?

In most cases, no. For straightforward degenerative stenosis without independent instability, decompression alone is usually sufficient. Degenerative spondylolisthesis does not, according to current evidence, automatically mean that screws, rods, and fusion are necessary.

The larger randomised trials with five-year data show no clinically relevant superiority of routine additional fusion in typical patients with degenerative spondylolisthesis. Fusion also entails a longer and more demanding procedure. A smaller older US trial (SLIP) found an advantage in a general quality-of-life measure and fewer re-operations in a narrowly selected patient group; however, no statistically significant difference was found in the spine-specific ODI. This opposing position is relevant but does not outweigh the larger and more recent data.

Fusion remains appropriate when there is an independent rationale — for example, significant deformity, a complex revision situation, or when adequate decompression would require removal of so much of the load-bearing joint structures that destabilisation is to be expected. Genuine mechanical instability that is convincing in the overall clinical and radiological picture may also play a role. The term "instability" lacks a uniform definition; the decision must therefore be derived from the overall clinical picture and the anticipated additional benefit of fusion.

What can realistically be expected?

Decompression primarily improves leg symptoms and walking capacity; non-specific back pain responds less reliably. In the large NORDSTEN-SST trial, approximately 70% of patients achieved a clinically relevant functional improvement at five years; approximately 10% required a further spinal procedure during this period. These group-level figures are not an individual guarantee. The duration and extent of nerve damage, co-morbidities, and functional baseline all influence outcomes.

Possible complications include dural tear with cerebrospinal fluid leak, haematoma, infection, nerve injury, thrombosis, and general anaesthetic and cardiovascular risks. Individual risk depends heavily on the extent of surgery and overall health. Additional fusion increases operative duration and physical demands and therefore requires a clear anticipated incremental benefit.

What does the research say?

Moderate stenosis does not inevitably worsen

Wessberg and Frennered followed patients with moderate, non-operated symptoms over a mean of 3.3 years. Leg and back pain remained stable or improved in the large majority; only 7% crossed over to surgery. This applies to tolerable symptoms — not to rapidly progressive neurological deficits.

Wessberg P, Frennered K. Eur Spine J. 2017. doi:10.1007/s00586-017-5075-x

Structured physiotherapy can help

A randomised trial of 259 patients found that combined manual therapy and individualised exercise produced better short-term outcomes than group exercise or standard medical care; at six months, differences between groups were no longer detectable. A second trial in patients already meeting surgical criteria found no difference between decompression and structured physiotherapy at two years — however, 57% of the physiotherapy group crossed over to surgery during follow-up.

Schneider MJ et al. JAMA Netw Open. 2019. PMID 30646197 · Delitto A et al. Ann Intern Med. 2015. PMID 25844995

Corticosteroid injections have limitations

In 400 patients with central lumbar stenosis, adding corticosteroid to local anaesthetic alone provided no clinically relevant benefit for leg pain or function at six weeks. This does not exclude targeted infiltration in individual cases, but argues against uncritical repeated series.

Friedly JL et al. N Engl J Med. 2014. PMID 24988555

Decompression can provide substantial benefit when the indication is right

In SPORT, operated patients showed greater improvement over four years in the as-treated comparison than those managed non-operatively. Due to extensive treatment crossover, the randomised intention-to-treat analysis was less definitive; SPORT provides important evidence of efficacy but does not mean that every symptomatic stenosis requires surgery.

Weinstein JN et al. Spine. 2010. PMID 20453723

Routine fusion offers no additional benefit in most cases

The Swedish Spinal Stenosis Study found no advantage of additional fusion at two years — regardless of whether degenerative spondylolisthesis was present. At five years, several secondary outcomes were actually more favourable after decompression alone; re-operation rates did not differ.

Försth P et al. N Engl J Med. 2016. PMID 27074066 · Karlsson T et al. Bone Joint J. 2024. PMID 38945544

The NORDSTEN-DS trial specifically studied patients with degenerative spondylolisthesis. At two and five years, decompression alone was non-inferior to decompression with instrumented fusion; the rate of further procedures over five years was also similar.

Austevoll IM et al. N Engl J Med. 2021. doi:10.1056/NEJMoa2100990 · Kgomotso EL et al. BMJ. 2024. PMID 39111800

The smaller SLIP trial randomised 66 selected patients with stable Grade I spondylolisthesis. It found an advantage for fusion in the general physical SF-36 score and fewer re-operations, but no statistically significant difference in the spine-specific ODI; at four years, follow-up data were available for only 68% of participants. It represents an important counterpoint, but provides less robust evidence than the larger multicentre trials.

Ghogawala Z et al. N Engl J Med. 2016. PMID 27074067

For decompression, outcomes matter — not the label

NORDSTEN-SST randomised 437 patients to three tissue-preserving decompression techniques. At five years, there were no relevant differences in function, pain, or further procedures. A smaller randomised trial showed comparable two-year outcomes for fully endoscopic and tubular microsurgical decompression. Technique may therefore be chosen individually — provided the nerves are fully decompressed and load-bearing structures are preserved.

Hermansen E et al. Eur Spine J. 2025. PMID 39448401 · Kotheeranurak V et al. Eur Spine J. 2023. PMID 37010607