Chronic pain affects 1 in 4 US adults, and many cases are resistant to almost any treatment. Deep brain stimulation (DBS) holds promise as a new option for patients suffering from treatment-resistant chronic pain, but traditional approaches target only brain regions involved in one aspect of the pain experience and provide continuous 24/7 brain stimulation which may lose effect over time. By developing new technology that targets multiple, complimentary brain regions in an adaptive fashion, the investigators will test a new therapy for chronic pain that has potential for better, more enduring analgesia.

A diverse array of chronic pain syndromes are refractory to almost all treatment but involve pathological activity in similar brain regions. This suggests therapeutic potential for deep brain stimulation (DBS) for refractory pain disorders, but despite early promise, long-term efficacy is lacking. Current DBS devices are limited in anatomical reach, targeting only a subset of the distinct brain regions known to be important. Further, DBS therapy is bluntly applied in an "open-loop," continuous fashion without regard to underlying physiology. As a result of these shortcomings, DBS for pain is often ineffective or shows diminished effect over time. Loss of therapeutic effect may be due to nervous system adaptation or a failure of stimulation to accommodate patient- specific dynamics of pain processing. DBS could be significantly improved by seeking individually optimized brain targets or by using neural biomarkers of pain to selectively control stimulation when it is needed ("closed-loop" DBS). Better brain targets would also address the different dimensions of pain such as somatosensory (location, intensity and duration), affective (mood and motivation) and cognitive (attention and memory). The main goal of this study is to test the feasibility of personalized targeting of brain regions that support multiple pain dimensions and to develop new technology for "closed-loop" DBS for pain. The study team will develop data-driven stimulation control algorithms to treat chronic pain using a novel device (Medtronic Summit RC+S or Percept RC) that allows longitudinal intracranial signal recording in an ambulatory setting. By building this technology in an implanted device, chronic pain DBS is tailored to each patient and will advance precision methods for DBS more generally.

Beginning with an inpatient trial period, subjects with various refractory chronic pain syndromes will undergo bilateral surgical implant of temporary electrodes in the thalamus, anterior cingulate, prefrontal cortex, insula and amygdala. These regions have been implicated in the multiple dimensions of pain. The goal of the trial period is to identify candidate biomarkers of pain and optimal stimulation parameters for each individual, and to select subjects who show likelihood to benefit from the trial. A subgroup of 6 such patients will then proceed to chronic implantation of up to 3 "optimal" brain regions for long-term recording and stimulation. The invstigators will first validate biomarkers of low- and high-pain states to define neural signals for pain prediction in individuals (Aim 1). The investigators will then use these pain biomarkers to develop personalized closed-loop algorithms for DBS and test the feasibility of performing closed-loop DBS for chronic pain in weekly blocks (Aim 2). Finally, the investigators will assess efficacy of closed-loop DBS algorithms against traditional open-loop DBS or sham in a double-blinded crossover trial (Aim 3) and measure mechanisms of DBS tolerance. Our main outcome measures will be a combination of pain, mood and functional scores together with quantitative sensory testing. Successful completion of this study would result in the first algorithms to predict real-time fluctuations in chronic pain states and development of a new therapy for currently untreatable diseases.

Inclusion Criteria:

1. Age 22-80 years old

2. Clinical diagnosis of a refractory chronic pain syndrome including

   1. post-traumatic pain syndromes (e.g. root avulsions, nerve crush injuries, spinal cord injury)
   2. postsurgical pain syndromes (e.g., postmastectomy syndrome, post-thoracotomy syndrome, phantom limb pain, post-surgical spinal pain)
   3. postherpetic neuralgia
   4. complex regional pain syndrome
   5. atypical facial pain
   6. central pain syndromes (e.g. post-stroke pain, multiple sclerosis pain, post-radiation pain)
   7. post-radiation plexopathy

3. Two or more years or more of medically refractory severe pain

4. Average daily pain for the past 30 days reported as >6 on a 0-10 numeric rating scale (NRS)

5. Pain that fluctuates over a range of at least 3 points on the NRS

6. Patient has failed at least two pain medications from different classes as determined by a neurologist or pain management specialist with stable doses of medications for 30 days prior to baseline visit.

7. Lack of a surgically correctible etiology for the pain as determined by 2 independent surgeons

8. Ability to speak / read English

9. Capable of understanding and providing informed consent

10. Absence of significant cognitive impairment - score of 25 or greater on the Montreal Cognitive Assessment (MoCA)

11. Successful detection of pain biomarkers or positive symptomatic response to inpatient stimulation trial period if performed.

Exclusion Criteria:

1. Major medical co-morbidities increasing the risk of surgery including uncontrolled hypertension, coagulopathy, severe diabetes, major organ system failure, active infection or history of implant related infections, immunocompromised state or malignancy with < 5 years life expectancy
2. Presence of cardiac pacemakers/defibrillators, implanted medication pumps, intra-cardiac lines, any intracranial implants (e.g., aneurysm clip, shunt, cochlear implant, electrodes) or other implanted stimulators not compatible with RC+S system
3. Pregnancy or breast feeding: all women of child bearing potential will have a negative urine pregnancy test prior to undergoing their surgical procedure.
4. Active depression (BDI > 20), Suicide attempt \

There will be assigned active intervention and non-active intervention periods for each patient throughout Phase 2 and 3 of the study. The participant will not know if he/she is in the active or non-active period of the phase.