The goal of this randomized controlled trial is to compare lumbar interbody fusion surgery with multidisciplinary rehabilitation in participants aged 20-65 years with persisting (≥ one year) low back pain. The main question it aims to answer is:

• Is lumbar fusion surgery superior to multidisciplinary rehabilitation in alleviating persisting low back pain?

Participants will be randomized to either lumbar interbody fusion surgery or a multidisciplinary rehabilitation program.

If randomized to lumbar fusion interbody surgery, the participants will:

• undergo radiologic examinations, including X-ray, MRI, and MRI spectroscopy
• provide blood samples at four intervals including postoperatively
• complete PROMs at five intervals
• have their activity monitored through the ActivePAL accelerometer
• undergo lumbar fusion surgery

If randomized to multidisciplinary rehabilitation, the participants will:

• undergo radiologic examinations, including X-ray, MRI, and MRI spectroscopy
• provide blood samples at three intervals
• complete PROMs at five intervals
• have their activity monitored through the ActivePAL accelerometer
• undergo multidisciplinary rehabilitation

Background: Low back pain (LBP) is the number one cause of disability worldwide. Chronic LBP (cLBP), also referred to as persisting low back pain, is defined as low back pain lasting at least twelve weeks, with a lifetime prevalence of about 23%. It represents extensive individual, societal, and financial burdens.

The etiology of cLBP is multifactorial with complex pathogenesis, and only a small proportion of patients with disabling cLBP have a well-understood pathophysiological cause. Over the last several decades, the biopsychosocial model of LBP has been the standard approach. Accordingly, current clinical guidelines recommend treatments such as advice to stay active, exercise, reassurance and coping strategies, analgesic medications, and avoiding bed rest. The existing treatments, however, have only small to moderate effects, and 85-90% of patients do not receive a pathoanatomical diagnosis (National Institute for Health and Care Excellence (NICE)).

Many pathophysiological hypotheses for cLBP have been proposed (e.g., annular tear, disc herniation, loss of disc height, facet joint degeneration, and Modic changes). For example, it has been hypothesized that cLBP with Modic changes is caused by bacterial infection. A Danish trial reported that antibiotic treatment was superior to placebo in selected cLBP patients with focal vertebral bone marrow changes (Modic changes graded by MRI of the lumbar spine). However, the Norwegian AIM (Antibiotic In Modic changes) study did not support this finding. Nonetheless, intervertebral disc (IVD) degeneration is generally accepted as a relevant cause of cLBP, especially among surgeons treating the condition. An experimental study of cultured human nucleus pulposus cells showed that acidic pH caused an increase in several pro-inflammatory, neurotrophic, and pain-related factors. This upregulation of inflammatory substances may, in turn, induce the ingrowth of nerve fibers into degenerative IVDs, possibly explaining how discs can become painful. Therefore, one possible approach to diagnose painful IVDs could be to measure IVD pH levels by MRI spectroscopy (MRS). The local inflammation in degenerative IVD is correlated with a systemic inflammatory response that can be measured in serum samples and is related to symptoms. Recent developments in diagnostics and biomarkers in the field may improve patient selection and support an individually tailored treatment.

A few randomized controlled trials have compared lumbar spine fusion with non-operative treatment, finding similar improvements in pain and disability for both treatment alternatives. This has led to guidelines recommending non-operative treatment.

The researchers in the current project aim to improve treatment, develop refined diagnostic assessments, and explore potential biomarkers in a multicenter randomized controlled trial (RCT), adding level-one evidence to the subject.

The overall concept is to assess whether there is a difference in treatment effect between modern lumbar interbody fusion (LIF) and multidisciplinary rehabilitation (MRE) for patients with cLBP. The researchers aim to:

• Evaluate the effect of LIF versus MRE at one-year follow-up (FU).
• Evaluate the treatment effectiveness at two-year FU in the cohort of patients crossing over from one treatment arm to the other at one-year FU. (Primary and secondary outcome measures will be the same as those listed in sections 1 through 12 of the 'Outcome Measures' section.
• Evaluate the effect of LIF versus MRE at one-year FU on physical activity, sleep disturbances, and circadian rhythm.
• Evaluate whether baseline MRS biomarkers are associated with treatment response after LIF at one-year FU.
• Evaluate whether baseline MRS biomarkers are associated with treatment response after MRE at one-year FU.
• Evaluate whether MRS biomarkers are associated with molecular biomarkers in blood samples collected at three time points.
• Evaluate if molecular biomarkers can identify patients with improvement after LIF at one-year FU.
• Evaluate if molecular biomarkers can identify patients with improvement after MRE at one-year FU.
• Evaluate, if feasible, the correlation between MRS biomarkers and perioperative pH measurement of the IVD.
• Evaluate the cost-effectiveness of LIF versus MRE from baseline to one-year FU The spinal surgeons will screen participants for trial eligibility by predetermined inclusion- and exclusion criteria and ensure that necessary imaging examinations (MRI and X-ray) have been done per protocol.

Patients will be invited to trial participation if all the inclusion criteria and none of the exclusion criteria are satisfied. Patients who may be relevant study candidates are set up for a new appointment with another surgeon associated with the project so that each potential participant receives an individual assessment by two different surgeons, ensuring a thorough evaluation and sufficient reflection period for the patient. To avoid undue pressure to participate in the study, one of the treating surgeons who have assessed the patient at the outpatient clinic will provide information about the project, and obtaining informed consent will be done by an independent third party, for example, a study nurse, a representative from rehabilitation or another surgeon associated with the project. To ensure comprehensive and unbiased information regarding both trial arms, representatives from the MRE arm will also have a face-to-face or electronic meeting with potential trial participants before their inclusion. General consent, as well as specific consent to the neuroscientific biobank, will be obtained. However, consent to the neuroscience biobank is not mandatory for study participation. The participant will be issued copies of the letter informing about the trial and the signed informed consent(s).

Eligible patients will be allocated in a 1:1 ratio between LIF and MRE, using a computer randomization procedure stratified by center. Block size and allocation sequence generation details will be provided in a separate document unavailable to those who enroll patients or assign treatment. Following screening, eligible participants will be randomized in a continuous manner. Treatment initiation will occur within three months of randomization for both treatment groups.

Baseline data collection will include sociodemographic variables (age, gender, BMI, ethnicity, educational level, work status), as well as information on physical workload, leisure activities, smoking habits, emotional distress (assessed with Hopkins Symptom Checklist-25 [HSCL-25]), fear avoidance beliefs (FABQ), Örebro Musculoskeletal Pain Screening Questionnaire (Short-form), and cLBP history (including duration and prior treatments such as spinal surgery, physiotherapy, and chiropractic therapy). At baseline, we will ask all participants to report their smallest worthwhile reduction of pain and disability (in percent, not specified to measurement instrument), what treatment group they hope to be randomized to, and how well they expect to be at one-year FU (Likert scale (1-7)). We will further ask all investigators at baseline to report what treatment allocation they think would be best for the individual participant if they were not participating in a study and to rate (Likert scale (1-7)) how well they expect the participant to be at one-year FU given the randomization.

Baseline sick listing and subsequent sick listing at each FU consultation will include documentation of complete work absence and part-time absence.

Prospective participants must undergo radiological examinations before study inclusion, including MRI and plain radiographs for measuring pelvic spine parameters. The MRI and radiographs must be obtained within six months before the initiation of treatment.

Hematological parameters (leucocytes, thrombocytes, hemoglobin (Hb)), sedimentation rate, CRP, electrolytes (Na and K), and measures of kidney (creatinine) and liver function (ALAT) will be assessed and registered at baseline if allocated to LIF intervention. Functional comorbidity index will be reported at screening and serve as a screening tool for other relevant illnesses together with physical examination at baseline.

The study will be monitored by the Clinical Trial Unit at Oslo University Hospital according to the standard by NorCRIN and Good Clinical Practice. Adverse events and serious adverse events will be registered at all study FUs. The study will focus on monitoring specific adverse events in individuals allocated to the LIF intervention. These events include the following:

• Deep and superficial wound infection and rupture of the surgical wound
• Hardware mechanical complications with loosening of pedicle screws and interbody implants before bony fusion of the addressed lumbar level are evidenced.
• Improper hardware placement
• Wrong-level surgery
• Dural tears leading to spinal headaches and possibly a pseudo-meningocele
• Nerve root injury possibly leading to chronic neuropathic pain, palsy, and dysfunction of the muscles affected.
• Postoperative spinal hematoma necessitating an acute surgical evacuation
• Cauda equina compression that may lead to transient or permanent dysfunction of the urinary bladder and bowel function.
• Potentially life-threatening hemorrhage by iatrogenic injury to large pelvic vessels
• Iatrogenic injury to the ureter necessitating additional urologic procedures
• Bowel perforation in case of ALIF
• Retrograde ejaculation due to iatrogenic injury to the lumbar autonomous nervous system (ALIF) Monitoring for these adverse events will involve clinical examination, blood sample analysis (including parameters such as Hb, CRP, and leukocytes), and radiologic imaging such as CT scans and MRI.

Blood samples for DNA genotyping (Genomics), methylation (Epigenetics), and RNA sequencing (Transcriptomics) will be collected at baseline, 6- and 12-months FU from participants who have consented to participate in the genetic analysis component of the study. Participation is optional. Participants who do not wish to participate in the genetic research may still participate in the study.

The researchers will also, at baseline, 6- and 12 months, measure a panel of 40 cytokines by duplicate serum analysis with a 40-plex Pro Human Chemokine multi-bead assay. Samples that can be used to analyze suspected molecular biomarkers in the future will also be collected.

For all participants throughout the study, the investigator and study site personnel will collect data about healthcare resource utilization associated with medical encounters. The cost-benefit analysis of LIF versus MRE will be assessed by QUALYs derived from EQ-5D-5L and hospital and community treatment costs. The following data will be collected:

• LIF: Time in theatre, duration of surgery, blood loss and transfusions, number of surgeons, anesthesiologists, and nurses, implant costs, length of sick leave, use of non-operative treatment (i.e., physiotherapy, rehabilitation), additional visits to the GP, additional follow-ups including imaging outside the study protocol.
• MRE: Medical consultations at the general practitioner or other primary care treatment, additional diagnostic tests (e.g., X-ray, MRI, CT, laboratory tests), medications (e.g., pain relievers, muscle relaxants, anti-inflammatories, antidepressants), time spent at treatment facilities, staffing costs, travel expenses (for patients traveling to rehabilitation institution for every treatment day or session), accommodation costs (for patients staying at the rehabilitation institution during the treatment period) and sick leave.

Inclusion Criteria:

1. Male and non-pregnant female patients between 20 and 65 years of age with persistent low back pain of at least one year's duration at inclusion

2. Received non-operative treatment in line with national [50] and international [49] guidelines, including at least self-management, exercise, and physical therapy, without satisfactory effect before study enrolment

3. Back-related disability: ODI 30 - 60 points at baseline

4. Back pain > leg pain

5. One- or two-level disc degeneration between L2 and sacrum with any of the following:

   • High-intensity zone (HiZ)
   • Modic changes
   • Severe disc height reduction exceeding 50% of the cranial disc

Exclusion Criteria:

1. Multilevel disc degeneration requiring intervention beyond two levels
2. Spondylolysis or lytic spondylolisthesis
3. History of previous spondylodiscitis
4. Previous lumbar fusion surgery
5. Scoliosis >20 degrees
6. Signs of a vertebral fracture at the planned level of fusion or its adjacent levels
7. Active smokers
8. Unlikely to adhere to treatment or complete follow-up (e.g., ongoing serious psychiatric disease, drug abuse, plans to move outside the catchment areas of the trial centers)
9. Significant nerve root compression assessed by MRI and clinical examination
10. BMI > 40
11. Not understanding the Norwegian language.
12. Generalized myalgia, including history or signs of fibromyalgia and myalgic encephalitis
13. Contraindications to MRI (e.g., cardiac pacemaker electrodes, metal implants in the eye or brain, claustrophobia).
14. Active cancer
15. Disabling chronic neurological disease (e.g., Parkinson's disease, ALS, MS)
16. Disabling osteoarthritis of the hip or knee (Kellgren & Lawrence grade III or higher)
17. Daily use of morphine equivalents ≥ 60mg or regular use of morphine-containing pain patches
18. Decline specific treatment arm

Outcomes assessors will not have access to information about participants' allocated treatment groups (lumbar interbody fusion or multidisciplinary rehabilitation)