This prospective observational cohort study was conducted after approval by the local ethics committee and was performed based on the Consolidated Standards of Reporting Trials (CONSORT) guidelines. Inclusion and exclusion criteria were as follows:

1. Inclusion criteria

• Mild to moderate lumbar disc herniation (occupying <30% of spinal canal confirmed on MRI scan), contained or uncontained.
• Sciatica +/- lower back pain correlating to the lumbar disc herniation visualised on MRI.
• Symptoms of sciatica ongoing for at least 6 weeks.
• Non-response to conservative management (physiotherapy/analgesia)
• >18 years of age.

1. Exclusion criteria

• Large disc prolapse (>5mm or occupying >30% of spinal canal)
• Evidence of cauda equina or cord compression.
• Significant structural deformity of spine (e.g. spondylolisthesis/scoliosis/spinal fracture)
• Acute illness
• Pregnancy
• Injury litigation

No randomisation took place. Patients agreeing to take part in the study were offered one of two treatment options; caudal epidural injection alone (which would act as the comparative control group) or the novel combination treatment involving APLD, intradiscal ozone and caudal epidural in combination (which would act as the comparative experimental group). Possible risks and benefits of each treatment option were discussed and the patient was given time to make their decision. Informed consent was obtained.

Patient’s demographic data and data characterising the patient’s discogenic sciatica was either provided by patients or was obtained by a review of MRI reports and scans. The techniques of the two treatment options are described below:

1. Caudal epidural injection

Procedure performed under local anaesthetic with fluoroscopic guidance and with patient in prone position. Using sterile technique, access to the epidural space was obtained and confirmed using contrast. Mixture of 9ml of lidocaine and 6mg of betamethasone injected.

1. APLD, intradiscal ozone and caudal epidural injection in combination (APLD/Ozone/Epidural)

Procedure performed under local anaesthetic with fluoroscopic guidance and with patient in prone position. First, APLD was performed of the affected disc using the Stryker Dekompressor probe.  Disc access was gained using a cannula; placement confirmed using fluoroscopy and probe introduced. The probe was connected to the mechanical aspirator which aspirates part of the NP. Each herniation was decompressed for 3 minutes. After completion of APLD, a spinal needle is inserted into the disc space and position confirmed. The oxygen-ozone mixture (ozone) is produced in real time by a medical ozone generator with a concentration of 30µg. 10-15ml are injected into the disc. Finally patients received a caudal epidural injection as previously.

The primary outcome measure was the McNab clinical outcome score (McNab score) and secondary outcome measures were the improvement in VAS scores for back pain and sciatica and occurrence of complications. Each patient completed a VAS score for their back pain and sciatica prior to their procedure as a baseline measurement. Patient follow up occurred, via telephone interview, 1 month after their procedure and also when the study ended. VAS scores and the McNab clinical outcome score were recorded along with any complications which had occurred.

The software used for statistical analysis was Stata 12.0 (Stata-corp, College Station, Texas). Patient characteristics were summarized and significant differences in characteristics between treatment groups were calculated using the two sample T-test or the Pearson Chi squared test.

An analysis of the efficacy of both treatment groups was performed using the following outcome measures:

1. Improvement in VAS of LBP 1 month after treatment
2. Improvement in VAS of LBP at final follow up
3. Improvement in VAS of sciatica at 1 month after treatment
4. Improvement of VAS of sciatica at final follow up
5. McNab clinical outcome score at 1 month after treatment
6. McNab clinical outcome score at final follow up

Comparisons of outcome results were made between APLD/Ozone/Epidural group and the caudal epidural group using a two sample T-test. A P-value < 0.01 indicated a statistically significant difference between the groups. The McNab score was also analysed as a categorical variable to give the percentage of patients within each treatment group with successful outcomes and the percentage with good-excellent outcomes.

The following subgroups were then analysed:

1. Patients aged less than 50 years (<50yrs)
2. Patients aged over 50 years (>50yrs)
3. Male patients
4. Female patients
5. Patients with sciatica that was worse than LBP (Sciatica>LBP)
6. Patients with LBP which was greater than sciatica (LBP>Sciatica)
7. Single level disc disease
8. Multi level disc disease
9. Patients with symptoms ongoing for less than 1 year (<1yr)
10. Patients with symptoms ongoing for over 1 (>1yr)
11. Patients with contained disc herniation
12. Patients with a non-contained disc herniation
13. Patients with signs of nerve root compression on MRI
14. Patients with no signs of nerve root compression on MRI

Within each subgroup comparisons of outcome results were made between APLD/Ozone/Epidural group and the caudal epidural group using a two sample T-test. The McNab score was also analysed as a categorical variable to give the percentage of patients within each treatment group with successful outcomes.

A power calculation demonstrated that 37 patients receiving the APLD/Ozone/Epidural treatment and 74 patients receiving caudal epidural treatment will give 84% power to detect a 30% difference in the outcome measures.