The hypothesis is that in the treatment of low back pain (LBP) radiating to the leg, the long-term results of prolotherapy are more effective than those of the current conventional treatment: epidural steroid injections (ESI). This research will examine the efficacy of prolotherapy injections versus epidural steroid injections for the treatment of low back pain radiating to the leg. This is a randomized, unblinded study, in which patients seen in the principle investigator's pain clinic will be randomly divided to receive treatments from either the experimental, prolotherapy group, or the active control, ESI group.

The primary goal of the study is to investigate the long-term efficacy of prolotherapy for patients with LBP referring to the leg. The secondary goal is to compare this efficacy with that of epidural steroids. The reason for exploring options other than ESI is that the latter have proved to be disappointing. Comprehensive reviews have been written on interlaminar and transforaminal epidurals which basically show poor long-term results of longer than 3 months. In addition, ESI carries risks of neurological damage, epidural hematoma and infection. However, ESI continues to be the most widely used treatment worldwide. Research also shows that with increasing age, there is an increased incidence of ligament laxity, spondylolisthesis and angulation which may in turn lead to nerve impingement and deterioration in function with time. Sprained and strained ligaments are themselves capable of referring pain down the leg even as far as the ankle. Research shows that much of the pain referred down the leg is not from impinged nerve roots but from other soft tissues, such as the above-mentioned ligaments; these must be addressed and treated not only to treat pain but in order to improve function.

The term prolotherapy is otherwise known as proliferative regeneration therapy and is aimed at doing just the opposite of cortisone, namely, to strengthen the structures injected, usually ligaments. Prolotherapy solutions are also used to treat partially torn tendons, as in the case of partial rotator cuff tears. Research on prolotherapy has shown that this treatment mode produces varying results in the treatment of low back pain and carries fewer risks than epidurals. One can infer from this that it may provide a safer and better long-term treatment method than ESI. Yelland's review shows that prolotherapy works for the treatment of LBP if this treatment method is combined with other measures such as exercises or manipulations. In this study, patients with low back pain radiating to the leg will be randomized to receive either epidural steroid injections or prolotherapy injections using a solution made up of 20% dextrose. In light of the results of the research quoted, it was decided to give exercise instructions tailored to every patient's condition. Both patients from the experimental and the control groups will receive this instruction in order to avoid the presence of another confounding variable.

A precondition to being included in the trial is having either a CT or MRI of the lumbar spine within the previous 18 months and not having any of the exclusion criteria cited below. Once included in the trial, patients will be randomized into the study and control groups. All epidural injections will be performed under fluoroscopy, and radiocontrast dye will be injected to verify that the injectate will be given in the correct place. Patients in this group will receive 3 interlaminar epidural steroid injections approximately 4 weeks apart. The solution injected will be made up of 80mg methylprednisolone acetate with bupivacaine. The level injected will depend on the clinical picture.

All of the prolotherapy dextrose injections will be performed under ultrasound guidance. Prolotherapy patients will receive 5 sessions approximately 4 weeks apart. In each session, 6 injections in different areas of the lumbosacral spine and sacroiliac ligaments will be injected with 20% dextrose solution using a 25 gauge spinal needle. The targeted structures include the following: the facet joint capsular ligaments, interspinous ligaments, and some of the sacroiliac ligaments, depending on the clinical assessment. The clinical picture will determine what levels will be injected in each session.

As described below, patients will be assessed prior to the study and after the study regarding their pain and function.

Because several patients fail to improve and continue to suffer from severe pain, it was considered unethical to keep patients in their original grouping and prevent crossover for a period of 12 months.

Crossover will be permitted if all of the following conditions apply:

1. At least 1 month has passed since the last treatment in the original allocation group
2. NRS pain levels are at least 6 out of 10
3. Pain and function have not improved with the originally allocated treatment
4. The patient requests the crossover. Participants will receive the same treatments in the crossover groups as patients who were originally allocated to those groups Therefore, patients who cross over to the epidural side will receive up to 3 epidural injections approximately 1 month apart. Participants who cross over to the prolotherapy grouping will receive up to 5 treatment sessions also approximately 1 month apart.

Follow-up of crossover participants will also be performed in the same manner and by the same independent investigator as for non-cross-over participants at 1, 3, 6 and 12 months after the last crossover treatment.

Inclusion Criteria:

• Pain radiating down one or both legs or to the groin of at least 12 weeks' duration
• Patients with disc lesions with radiating pain to the leg(s)
• Mild spinal stenosis

Exclusion Criteria:

• History of back surgery
• Recent history (less than 2 years) of active malignancy
• Recent fracture in the lumbar spine or pelvis of less than 12 months
• Active locus of infection in the body
• Coagulation disorders, and current anticoagulation therapy, excluding aspirin
• Chronic medication with corticosteroids and NSAIDS (which are said to possibly neutralise the effect of prolotherapy) - the latter must be stopped 24 hours prior to the first treatment session
• Recent injection of cortisone for back pain or any other pathology elsewhere in the body- patients must wait 2 weeks before commencement of the study
• Concurrent significant depressive illness or evidence of catastrophisation, fibromyalgia
• Concurrent history of active autoimmune disease or inflammatory joint disease evidence of a peripheral neuropathy

NOTE:

If any of the above illnesses appear during the time of the treatment in any patient, the patient will be withdrawn from the trial as treatment may be detrimental to his or her health. In addition, follow-up is not relevant to compare a diseased patient with any of the above with patients who are free of the above illnesses.

When all of the follow-up is completed - hopefully within 12 months

No masking is possible at this point in time.