Ketamine may help with chronic pain because it blocks the NMDA receptor, a glutamate receptor in the spinal cord and brain that keeps pain circuits amplified long after the original injury has healed. It is not an opioid and does not act on opioid receptors. It works on the amplifier rather than the volume knob, which is why it is considered mainly for pain that has become centrally driven rather than pain from an ongoing structural problem.
That is the theory. The evidence is narrower: the best trials are in complex regional pain syndrome (CRPS, also called RSD), the effects are modest and often temporary, and a 2025 Cochrane review of 67 randomized trials rated the certainty of evidence across chronic pain as a whole low to very low.6
What ketamine is
Ketamine is an anesthetic. At full doses it produces dissociative anesthesia; at the far lower doses used in pain medicine — often a tenth or less of that — it produces analgesia while the patient stays awake and conversational. Its primary action is non-competitive antagonism at the NMDA receptor. It has no affinity for GABA receptors, which separates it from sedatives and benzodiazepines. The liver converts it into norketamine,1 which is itself an NMDA receptor antagonist.10 In the United States it is a Schedule III controlled substance, FDA-approved as an anesthetic; use for chronic pain is off-label and governed by specialty consensus rather than by a label.2
The mechanism: NMDA receptors and central sensitization
Wind-up in the dorsal horn
When a peripheral nerve fires repeatedly, second-order neurons in the dorsal horn of the spinal cord respond more strongly to each identical input. This is wind-up, and the NMDA receptor is central to it. At rest the channel is plugged by a magnesium ion; sustained depolarization expels the magnesium, the channel opens, and calcium flows in, triggering changes that make the neuron more excitable and keep it that way.
Scale that up and you get central sensitization — the nervous system amplifying signals out of proportion to the input. Clinically it looks like allodynia (pain from a bedsheet or a breeze), hyperalgesia, and pain spreading beyond the original injury. Once that state is established, treating the tissue does not necessarily fix the pain, because the pain has stopped tracking it.
Why CRPS is the test case
CRPS is where this argument is strongest. Clinical and experimental evidence implicate NMDA receptor activity and glial activation in how CRPS is induced and maintained.4 If the receptor is doing the maintaining, blocking it should do more than mask pain — it should let the sensitized circuit reset.
There is some evidence for that. Pharmacokinetic-pharmacodynamic modeling in 60 CRPS-1 patients found analgesia persisting well after ketamine had cleared the plasma, with an estimated onset/offset half-life of about 11 days; the authors read this as ketamine initiating a cascade of desensitization in excitatory receptor systems that slowly abated.5 Relief that decays over weeks rather than hours would separate ketamine from an ordinary analgesic — but that reading rests on a model fitted to a single trial, not on a replicated finding.
What the trials actually found
In the foundational trial, 60 patients with CRPS-1 and a median disease duration of 7.4 years received a 4.2-day intravenous S(+)-ketamine infusion or placebo. Pain scores over 12 weeks were significantly lower with ketamine, and the lowest score came at the end of week 1 (2.68 versus 5.45 on a 0-10 scale). By week 12 the difference was no longer significant, and there was no functional improvement in either group. Psychomimetic side effects occurred in 76% of the ketamine group versus 18% of placebo.3 A second double-blind trial dosed four hours daily for ten days, capped at 0.35 mg/kg/h, and reported significant reductions across many pain parameters with none in the placebo group; both arms also received clonidine and midazolam, so the comparison is ketamine-plus-adjuncts against adjuncts alone.4
Now the counterweight. A 2022 meta-analysis of 20 randomized trials in 818 adults with CRPS-1 found ketamine reduced pain by about 8 points on a 0-100 scale, while bisphosphonates reduced it by about 24; the authors’ first-line recommendation was bisphosphonates.7 Note the base the ketamine number rests on: only 2 of those 20 trials tested ketamine, against 7 for bisphosphonates. A systematic review concluded evidence for ketamine is emerging and confirmatory trials are still needed.8
Broadest of all, the 2025 Cochrane review pooled 67 randomized trials of NMDA antagonists in chronic non-cancer, non-headache pain. It found no clear evidence that intravenous ketamine reduces pain intensity in the immediate, short, or medium term, rated that evidence low to very low certainty, and found it may increase adverse events.6 Most underlying studies were small, and blinding is hard when three quarters of the treatment group can tell they got something.
The reasonable reading: ketamine is a legitimate option for severe, centrally driven pain that has failed better-established treatments, but the effect is modest and time-limited, and anyone presenting it as a general answer for chronic pain is ahead of the data.
Route matters more than most patients are told
“Ketamine therapy” means at least three treatments with different pharmacology, evidence, and risk.
Intravenous
IV delivery bypasses first-pass metabolism, so plasma levels are predictable and can be titrated in real time against effect and side effects. This is the route used in nearly all the controlled trials, and where the evidence lives. It requires monitored administration, because the same dose control that makes it effective makes it possible to overshoot.
Sublingual
Swallowed ketamine is heavily metabolized before reaching the circulation. A lozenge study in six chronic pain patients — the authors called the findings preliminary — measured median bioavailability of 24% by both oral and sublingual routes, with extensive first-pass conversion to norketamine.10 A larger population pharmacokinetic study of a 50 mg S-ketamine oral thin film in 20 volunteers found bioavailability of about 26%, with roughly 80% converted to norketamine.11
That conversion is not purely a loss, since norketamine is an active NMDA antagonist. The sublingual route gives a different exposure profile rather than a simply weaker one: less parent drug, more active metabolite, a flatter curve. Note what those studies do and do not show — they measure how much drug reaches the blood, not whether the pain improves. On that second question the Cochrane review found no clear benefit for oral ketamine.6
Topical
Topical ketamine is the outlier, because it may not be working centrally at all. In a double-blind placebo-controlled crossover trial, 10% ketamine cream applied to the affected limb of 20 CRPS patients inhibited allodynia to light brushing and hyperalgesia to punctate stimulation, while plasma ketamine stayed below detectable limits — implying a peripheral action at cutaneous nociceptors. Touch thresholds were unchanged: it reduced the abnormal amplification without numbing the skin.9
That is a mechanistically interesting result from one small crossover study, and it should be read against the pooled evidence. The same Cochrane review found no clear evidence that topical ketamine reduces pain intensity in the immediate or short term, at low to very low certainty.6
Sublingual and topical formulations can be applied to a defined limb or taken at home without the infrastructure an infusion requires, which is why they get used outside research settings. That is a practical argument, not an evidence-based one. Route selection remains clinical judgment, not a settled question.
What a course involves, and what the risk is
A subanesthetic infusion is given with the patient awake and monitored. Blood pressure and heart rate typically rise, since ketamine stimulates the sympathetic nervous system. Patients commonly experience some mix of dissociation, visual distortion, dizziness, nausea, and detachment; these track the dose and clear with the drug. Consensus guidelines note that adverse events in most trials were few, but that higher doses and more frequent infusions carry more risk.2
Two harms are tied to repetition rather than a single exposure. Repeated prolonged infusions have caused drug-induced liver injury: in one series, three of six CRPS-1 patients receiving a second 100-hour infusion 16 days after the first developed elevated liver enzymes, which normalized within two months after stopping.12 Sustained heavy exposure — documented mainly in recreational users — can produce ulcerative cystitis and bladder fibrosis, with contracted bladder and hydronephrosis in long-term abusers.13 New urinary urgency, frequency, or pelvic pain during a course is a reason to stop and be evaluated, not to push through.
The consensus guidelines list a set of relative contraindications — situations calling for caution or avoidance rather than absolute prohibition: poorly controlled cardiovascular disease, severe liver disease, certain poorly controlled psychoses, elevated intracranial or intraocular pressure, pregnancy, and active substance abuse.2 A pharmacology review states plainly that long-term analgesic effects in chronic pain have not been demonstrated and that long-term safety questions remain unresolved.1
Where ketamine fits
The CRPS trial that produced significant pain relief produced no functional improvement3 — pain came down and the limb did not start working better on its own. That is the argument for treating a course as a window rather than a destination: it can lower the pain floor enough to make graded movement tolerable, and rehabilitation is what changes function. In CRPS, meta-analytic evidence favors bisphosphonates first,7 and sympathetic blockade, stellate ganglion blocks, and spinal cord stimulation each have their place. Ketamine belongs in the conversation after simpler options, not instead of them.
If you are in the St. Louis area
Padda Institute, Center for Interventional Pain Management, provides intravenous ketamine infusion as well as sublingual and topical formulations, with route chosen by diagnosis and tolerance. See ketamine therapy and complex regional pain syndrome treatment. Dr. Gurpreet Singh Padda, MD, MBA, MHP practices at 4477 Woodson Rd, Suite 100, St. Louis, MO 63134 and 12174 Natural Bridge Road, Suite 100, St. Louis, MO 63044. Phone: (314) 481-5000.
Frequently asked questions
Does ketamine cure chronic pain?
No. In the strongest CRPS trial the difference from placebo faded by week 12, and function did not improve.3 Where it helps, it lowers pain for a period measured in weeks rather than permanently, which can create room for rehabilitation — one option among several under pain treatments.
Is ketamine an opioid?
No. Its primary action is blockade of the NMDA glutamate receptor, not activation of opioid receptors.1 It is a controlled substance and requires the same care as any controlled medication, but it is a different pharmacologic class from opioid analgesics. See the pain management physicians page.
How long does relief last after an infusion?
It varies widely and is not predictable in an individual. Modeling in CRPS-1 patients estimated an onset/offset half-life of roughly 11 days, with relief outlasting the treatment period.5 Some patients get weeks; some get very little. The decision to continue is made against measured response — part of the evaluation under CRPS treatment.
Is sublingual or topical ketamine as effective as an infusion?
They are not equivalent, and not simply weaker versions of one thing. Sublingual bioavailability is roughly a quarter of the dose taken, with most of the rest converted to norketamine.1011 Topical ketamine appears to act peripherally on the skin, with plasma levels too low to measure.9 Both have a thinner evidence base than infusion: pooled analysis found no clear benefit for either oral or topical ketamine.6 They are used because they are practical for a defined limb or for home dosing; see ketamine therapy.
Will I be sedated or put under?
No. Subanesthetic infusions are given with the patient awake and monitored, not under general anesthesia. Dissociative effects are common and dose-related, and resolve as the drug clears.3 Arrange for someone else to drive. The same awake, monitored approach applies to other procedures here, including nerve blocks.
What should I try before considering ketamine?
For CRPS, meta-analytic evidence favors bisphosphonates first,7 and sympathetic blockade has its own supporting trials.8 Depending on the pain generator, an epidural steroid injection may be more appropriate. Start with a diagnosis-first evaluation: request an appointment or use the contact page.
Sources
- Peltoniemi MA, Hagelberg NM, Olkkola KT, Saari TI. Ketamine: A Review of Clinical Pharmacokinetics and Pharmacodynamics in Anesthesia and Pain Therapy. Clinical Pharmacokinetics. 2016;55(9):1059-77. PMID 27028535
- Cohen SP, Bhatia A, Buvanendran A, et al. Consensus Guidelines on the Use of Intravenous Ketamine Infusions for Chronic Pain From the American Society of Regional Anesthesia and Pain Medicine, the American Academy of Pain Medicine, and the American Society of Anesthesiologists. Regional Anesthesia and Pain Medicine. 2018;43(5):521-546. PMID 29870458
- Sigtermans MJ, van Hilten JJ, Bauer MCR, et al. Ketamine produces effective and long-term pain relief in patients with Complex Regional Pain Syndrome Type 1. Pain. 2009;145(3):304-311. PMID 19604642
- Schwartzman RJ, Alexander GM, Grothusen JR, et al. Outpatient intravenous ketamine for the treatment of complex regional pain syndrome: a double-blind placebo controlled study. Pain. 2009;147(1-3):107-15. PMID 19783371
- Dahan A, Olofsen E, Sigtermans M, et al. Population pharmacokinetic-pharmacodynamic modeling of ketamine-induced pain relief of chronic pain. European Journal of Pain. 2011;15(3):258-67. PMID 20638877
- Ferraro MC, Cashin AG, Visser EJ, et al. Ketamine and other NMDA receptor antagonists for chronic pain. Cochrane Database of Systematic Reviews. 2025;8:CD015373. PMID 40819842
- Fassio A, Mantovani A, Gatti D, et al. Pharmacological treatment in adult patients with CRPS-I: a systematic review and meta-analysis of randomized controlled trials. Rheumatology (Oxford). 2022;61(9):3534-3546. PMID 35104332
- Duong S, Bravo D, Todd KJ, Finlayson RJ, Tran DQ. Treatment of complex regional pain syndrome: an updated systematic review and narrative synthesis. Canadian Journal of Anaesthesia. 2018;65(6):658-684. PMID 29492826
- Finch PM, Knudsen L, Drummond PD. Reduction of allodynia in patients with complex regional pain syndrome: A double-blind placebo-controlled trial of topical ketamine. Pain. 2009;146(1-2):18-25. PMID 19703730
- Chong C, Schug SA, Page-Sharp M, Jenkins B, Ilett KF. Development of a sublingual/oral formulation of ketamine for use in neuropathic pain: Preliminary findings from a three-way randomized, crossover study. Clinical Drug Investigation. 2009;29(5):317-24. PMID 19366273
- Simons P, Olofsen E, van Velzen M, et al. S-Ketamine Oral Thin Film-Part 1: Population Pharmacokinetics of S-Ketamine, S-Norketamine and S-Hydroxynorketamine. Frontiers in Pain Research. 2022;3:946486. PMID 35899184
- Noppers IM, Niesters M, Aarts LPHJ, et al. Drug-induced liver injury following a repeated course of ketamine treatment for chronic pain in CRPS type 1 patients: a report of 3 cases. Pain. 2011;152(9):2173-2178. PMID 21546160
- Castellani D, Pirola GM, Gubbiotti M, et al. What urologists need to know about ketamine-induced uropathy: A systematic review. Neurourology and Urodynamics. 2020;39(4):1049-1062. PMID 32212278
Dr. Gurpreet Singh Padda, MD, MBA, MHP
