If you need a more accessible version of this website, click this button on the right. Switch to Accessible Site


You are using an outdated browser. Please upgrade your browser to improve your experience.

Close [x]

Follow Us

A Non-Surgical Approach to the Management of Lumbar Spinal Stenosis: A Prospective Observational Cohort Study

Donald R Murphy; Eric L Hurwitz; Amy A Gregory; Ronald Clary

Posted: 04/13/2006; BMC Musculoskeletal Disorders © 2006  Delank et al., licensee BioMed Central Ltd.


Background: While it is widely held that non-surgical management should be the first line of approach in patients with lumbar spinal stenosis (LSS), little is known about the efficacy of non-surgical treatments for this condition. Data are needed to determine the most efficacious and safe non-surgical treatment options for patients with LSS. The purpose of this paper is to describe the clinical outcomes of a novel approach to patients with LSS that focuses on distraction manipulation (DM) and neural mobilization (NM).
Methods: This is a prospective consecutive case series with long term follow up (FU) of fifty-seven consecutive patients who were diagnosed with LSS. Two were excluded because of absence of baseline data or failure to remain in treatment to FU. Disability was measured using the Roland Morris Disability Questionnaire (RM) and pain intensity was measured using the Three Level Numerical Rating Scale (NRS). Patients were also asked to rate their perceived percentage improvement.
Results: The mean patient-rated percentage improvement from baseline to the end to treatment was 65.1%. The mean improvement in disability from baseline to the end of treatment was 5.1 points. This was considered to be clinically meaningful. Clinically meaningful improvement in disability from baseline to the end of treatment was seen in 66.7% of patients. The mean improvement in "on average" pain intensity was 1.6 points. This did not reach the threshold for clinical meaningfulness. The mean improvement in "at worst" pain was 3.1 points. This was considered to be clinically meaningful.
The mean duration of FU was 16.5 months. The mean patient-rated percentage improvement from baseline to long term FU was 75.6%. The mean improvement in disability was 5.2 points. This was considered to be clinically meaningful. Clinically meaningful improvement in disability was seen in 73.2% of patients. The mean improvement in "on average" pain intensity from baseline to long term FU was 3.0 points. This was considered to be clinically meaningful. The mean improvement in "at worst" pain was 4.2 points. This was considered to be clinically meaningful. Only two patients went on to require surgery.
No major complications to treatment were noted.
Conclusion: A treatment approach focusing on DM and NM may be useful in bringing about clinically meaningful improvement in disability in patients with LSS.


Lumbar spinal stenosis (LSS) is a common and often disabling disorder that generally occurs in the sixth or seventh decade of life,[1] although it can uncommonly occur in younger individuals.[2] The incidence of this condition has been reported to be 8-11%,[3] with a slight preponderance in women.[1] LSS can lead to low back and leg pain, most typically via encroachment on the central canal, lateral recess, or lateral canal. The source of the encroachment is typically vertebral body osteophytes, hypertrophy of the ligamentum flavum or zygapophyseal joint, or a combination of these.[1] The posterior longitudinal ligament may be involved in some individuals.[4] The development of these degenerative changes is often accompanied by restriction of segmental mobility.[1]

One of the hallmarks of LSS is neurogenic claudication, in which the patient develops low back and/or leg pain after a period of walking that progressively worsens as walking is continued, with improvement or resolution when walking ceases and the patient sits or flexes the lumbar spine.[5]

LSS is one of the most common reasons for spine surgery in older people,[6] although little is known about the efficacy of surgical management of patients with LSS, particularly compared to non-surgical management.[7] It is generally felt that most patients with LSS should be managed non-surgically before considering surgical intervention,[8] but little is also known about what non-surgical approaches are most efficacious.

LSS can involve the central canal, the lateral recess, the lateral canal, or any combination of these.[6] This can lead to nerve root pain and dysfunction, i.e., radiculopathy. The pathophysiology of radiculopathy secondary to LSS is different from that of radiculopathy secondary to herniated disc (HD). In recent years it has increasingly become clear that much of the pain with acute radiculopathy secondary to HD is chemical, not compressive in nature.[9,10] The chemical inflammatory process with HD is initiated by the presence of nuclear material. But with LSS, it is likely that a different, or additional, mechanism that is involved in the production of nerve root pain.

Experimental evidence has suggested that chronic compression of the nerve root in LSS causes compromise of blood flow leading to congestion, ischemia, and intraneural edema.[11] This then leads to the development of periradicular fibrosis.[12] Increased pain with walking that is relieved with lumbar flexion (neurogenic claudication) is one of the hallmarks for LSS. Neurogenic claudication likely arises from increased metabolic demands of the nerve root in the presence of vascular compromise[13] and traction on the adhesed nerve root when lower extremity movement occurs during walking.[14] This may explain why the SLR is often negative in pts with LSS,[8] but is typically positive in patients with herniated disc. With LSS, compression, vascular compromise and perineural fibrosis dominate the pathophysiological picture, thus maneuvers that increase IVF pressure, i.e., extension,[15] or increase metabolic demands of the nerve root and movement of the fibrotic nerve root, as with walking, exacerbate the pain.

A non-surgical approach that attempts to target the unique pathophysiology of LSS may be best able to rapidly improve pain and function in these patients. Such a treatment strategy would attempt to mobilize the segment(s) involved, decompress the involved nerve root(s) and mobilize the involved nerve root(s) to break up periradicular adhesion, thus releasing nerve root entrapment, and restoring vascular function. It would appear that maintaining intersegmental and nerve root mobility would then be important in order to maximize the long term benefit of treatment.

The purpose of this study was to assess, using rigorous outcome measures, the results of a non-surgical management strategy for patients with LSS that focuses on distraction manipulation (DM) and neural mobilization (NM). Theoretically, these methods were employed in order to improve motion segment mobility (DM) and nerve root mobility (NM). It is not known whether these modalities actually create these effects, and this study does not evaluate these theoretical mechanisms. But the outcome of a strategy that focused on these methods was assessed. This strategy has not previously been evaluated.


Patients were treated according to the usual protocol utilized at the Rhode Island Spine Center for patients with radiculopathy secondary to LSS. The primary interventions, which were utilized in all patients, were:

• Distraction manipulation (DM) - This is a manipulative technique developed by Cox.[18] Although other forms of manipulation are believed to be effective in patients with LSS,[19] no form other than DM was used with the patients in this study.

In applying DM, the patient lay prone on a table that allows for distraction of the spine through inferiorward and flexion movement of the lower body (figure 1). This maneuver has been demonstrated to decrease intradiscal pressure[20] and is believed to create vertebral motions and increase the intervertebral foramen.[21]


Figure 1. The application of distraction manipulation.

• Neural mobilization (NM) - This a manual and exercise oriented method that is theorized to mobilize nerve roots that are suspected to be the source of nerve root pain.[22,23] Distal mobilization was applied by having the patient lie supine while the doctor or therapist dorsiflexed the ankle and flexed the hip with the knee extended. The leg was raised until the practitioner felt the "barrier",[24] i.e., the point at which tension is initially felt. The foot is then moved alternately into plantar flexion and dorsiflexion repeatedly for several cycles.

• Exercises that are taught to the patient and which are designed to compliment the DM and NM by mobilizing the lumbar spine and the involved nerve root(s). These included the "cat and camel" exercise[25] in which the patient is quadruped and alternately flexes and extends (within the comfort level) the cervical and lumbar spine, and "nerve flossing" exercises,[25] which attempt to mobilize the involved nerve roots and there associated peripheral nerves.

DM and NM and the related exercises are the constants of treatment in this study - all patients are treated with these methods. In addition, certain patients may also have been had other modalities included in their individual programs, such as mobilization exercises and spinal stabilization exercise.[26,27] While the frequency and duration of care were determined on an individual basis, patients were generally seen 2-3 times per week for 3 weeks initially, after which the first follow up (FU) reexamination was performed, which included the primary outcome measures (see below). This was typically followed by either continued frequency of 2 times per week or a reduction in frequency to 1 time per week, though some patients who are fully recovered were released after the first FU reexamination to 3 week FU.

This was a practice-based project in which the data gathered were those data that are collected as part of the routine of practice at the Rhode Island Spine Center. Also, the treatments provided each patient were those that are provided in the routine care of patients with LSS at the Rhode Island Spine Center. No experimental procedures were used and no personally identifiable information on any patient is presented. The study protocol was reviewed and approved by the HIPAA compliance officer of the Rhode Island Spine Center. Because of this, it was not deemed necessary to obtain formal approval from an Institutional Review Board.


The combination of DM and NM may be a safe and effective approach for patients with LSS. Because the sample size is relatively small and there is no control group, firm conclusions regarding this cannot be drawn. The outcome of this approach compares favorably with other non-surgical treatments, and treatment with DM and NM may be a viable non-surgical option before considering surgery for LSS. This approach deserves closer scrutiny in the form of randomized controlled trials.

Authors' Contributions

DRM conceived of the idea of compiling the data for publication, was one of the treating practitioners and was the principle author of the manuscript. ELH was responsible for statistical analysis, helped with design and presentation, and contributed to the writing of the manuscript. AAG was one of the treating practitioners, helped with data gathering and organization, and contributed to the writing of the manuscript. RC helped with data gathering and organization and contributed to the writing of the manuscript.

Pre-Publication History

The pre-publication history for this paper can be accessed here: