Osteoarthritis (OA) is a degenerative joint disease and is the most common form of arthritis. Pain reduction and functional recovery are the key elements of the clinical management of OA. Current treatment guidelines recommend a combination of pharmacological and non-pharmacological treatments. However, these are not always effective, with nearly 20% of patients not responding to any standard therapy, including joint replacement.

The mechanisms of pain relief are not well understood and are complicated by the remarkably large placebo effect, and inter-individual variation. There is no objective criteria for predicting whether a patient will respond to a given treatment

Duloxetine, an antidepressant drug, has proven effectiveness in various chronic pain syndromes including knee OA. The effect is however limited and only clinically relevant in around half of the trial patients. Importantly, it is currently unclear how and in whom duloxetine alleviates chronic pain.

Advanced MRI techniques use strong magnetic fields and radio frequency signals to generate metabolic, anatomical and functional brain images (fMRI).

Remifentanil is a potent analgesic agent whose analgesic effect has been well characterised in healthy volunteers, including fMRI studies showing modulation of activation of regions in the brain related to pain processing. Nevertheless, the neural correlates of remifentanil effects have not yet been investigated in chronic pain patients.

The aim of this research is to use a combination of multimodal MRI, genetic and psychometric assessments to identify the mechanisms of pain relief in knee OA patients, following treatments with duloxetine and remifentanil, in a placebo controlled condition. With this we also aim to identify genetic, anatomical and brain activity predictors of treatment outcomes.

The main hypotheses are:

• Analgesic response to duloxetine treatment can be predicted using a range of baseline brain imaging markers and QST.
• Analgesic response to duloxetine is mediated by modulation of neural networks underpinning emotional control.
• Duloxetine-induced changes in brain activation differ between responders and non-responders.

This study is expected to last for two years. It is funded by Arthritis Research United Kingdom and forms part of a wider scientific investigation, using translational methodologies, to enhance the understanding of arthritis pain and to improve its treatment.

Inclusion Criteria:

• Radiographically defined OA knee changes with knee pain
• Must have self-reported knee pain
• Able to give informed consent
• Over 35 years old
• Male or female
• Females not pregnant or lactating and using effective contraception

Exclusion Criteria:

• People with any known contraindication to MRI like

  • Intraocular metallic foreign bodies;
  • Intracranial aneurysm clips;
  • Cardiac pacemakers and defibrillators;
  • Cochlear implants;

• People with a significant head tremor;

• People with potential metal foreign bodies due to previous accidents;

• Breastfeeding or pregnancy, confirmed by pregnancy test;

• People that are felt to be unfit for the MRI scan according to the judgement of medically qualified personnel, either on the research team, or the patient's clinical team. (eg. due to back pain, claustrophobia, acute sickness etc.) This includes patients with signs of impaired temperature regulation such as an extremely high fever;

• Patients with large tattoos, specifically in the head, neck or shoulder region;

• Persons that do not have the capacity to consent;

• Aged less than 35;

• Major medical, neurological and psychiatric co-morbidities;

• Other significant medical condition;

• Metallic agents embedded within the body (ie. Shrapnel, surgical pins);

• Refusal by participant to general practitioner (GP) being informed;

• Have uncontrolled narrow-angle glaucoma;

• Have recently taken monoamine oxidase inhibitor (MAOI) or Mellaril® (thioridazine);

• Taking fluvoxamine, ciprofloxacin or enoxacin;

• Taking St. John's Wort, a herbal treatment (Hypericum perforatum);

• Taking other medicines containing duloxetine;

• Have liver disease or severe kidney disease;

• Currently on antidepressant treatment, including treatment for pain with tricyclic agents such as amitryptiline;

• Have recently taken monoamine oxidase inhibitor (MAOI) or Mellaril® (thioridazine);

• Taking tramadol;

• Known hypersensitivity, allergy or intolerance to one of duloxetine's components;

• Unwillingness to take caution in relation to use of other centrally active substances such as alcohol and sedative drugs;

• Current treatment with potent inhibitors of CYP1A2 like fluvoxamine;

Participants undergoing acute treatment (remifentanil or placebo) have, in addition to the stated above, the following exclusion criteria:

• Taking morphine
• Known hypersensitivity, allergy or intolerance to one of remifentanil's components or other fentanyl - analogues
• Current treatment with cardiac depressant drugs such as beta-blockers and calcium channel blocking agents