Ketamine for Refractory Chronic Migraine: an Observational Pilot Study and Metabolite Analysis

This metabolite analysis and observation study (n=6) found that both lidocaine and ketamine infusions significantly reduced pain, with ketamine providing slightly more relief. But a week after each treatment, patients were back at the same level of pain. This is the first study to investigate if one of ketamine’s metabolites, (2R,6R)-hydroxynorketamine, could reduce pain.


Introduction: Patients with refractory chronic migraine have substantial disability and have failed many acute and preventive medications. When aggressive intravenous therapy is indicated, both lidocaine and (R,S)-ketamine infusions have been used successfully to provide relief. Retrospective studies have shown that both agents may be associated with short-term analgesia.

Methods: In this prospective, observational pilot study of 6 patients we compared the effects of lidocaine and (R,S)-ketamine infusions and performed metabolite analyses of (R,S)-ketamine to determine its metabolic profile in this population. One of (R,S)-ketamine’s metabolites, (2R,6R)-hydroxynorketamine, has been shown in animal studies to reduce pain but human studies in patients undergoing continuous (R,S)-ketamine infusions for migraine are lacking.

Results: All 6 patients tolerated both infusions well with mild adverse effects. The baseline mean pain rating (0-10 numeric rating scale) decreased from 7.5 ± 2.2 to 4.7 ± 2.8 by end of lidocaine treatment (p<0.05) but increased to 7.0 ± 1.4 by the post-discharge visit at 4 weeks (p>0.05 versus baseline). The baseline mean pain rating prior to ketamine treatment was 7.4 ± 1.4, which decreased to 3.7 ± 2.3 by the end of the hospitalization (p<0.05), but increased to 7.2 ± 1.7 by the post-discharge visit at 6 weeks (p>0.05 versus baseline). For the primary outcome the change in pain from baseline to end of treatment was greater for ketamine than lidocaine (-3.7 versus -2.8, p<0.05) but this has minimal clinical significance.

Discussion: Ketamine metabolite analysis revealed that (2R,6R)-hydroxynorketamine was the predominant metabolite during most of the infusion, consistent with previous studies.”

Authors: Eric S. Schwenk, Marc C. Torjman, Ruin Moaddel, Jacqueline Lovett, Daniel Katz, William Denk, Clinton Lauritsen, Stephen D. Silberstein & Irving W. Wainer


Psychedelics are known for their ability to relieve mental pain. From depression to PTSD, different molecules have proven to be effective in the reduction of suffering. Now researchers are asking themselves if the same can be done for physical pain.

Ketamine has been used for anesthetics and pain reduction purposes for many decades now. Studies find that ketamine reduces pain in the body. But what about some of the worst pain that people can experience, refractory chronic migraines.

On a scale of 1 to 10, people rate this pain at a (constant!) 7. Not all patients are being helped with the current methods and some are already turning towards psychedelics. Studies on LSD and psilocybin are showing favorable results.

The current study investigated the potential of ketamine to treat refractory chronic migraines. Ketamine was pitted against lidocaine, a local anaesthetic. The hypothesis is that a metabolite, a molecule that is formed when ketamine is being broken down, named (2R,6R)-hydroxynorketamine reduces pain as it has been shown to be effective in animal studies.

What made the pain go away?

  • The pain level decreased from 7.4 to 3.7 at the end of hospitalization
  • The numbers were slightly less pronounced for the lidocaine group (7.6 to 4.8)
  • But in both groups, six weeks later the pain was back at the same (high) level

The difference in pain between both groups was significant, with the ketamine group experiencing less pain after treatment. Alas, for patients this difference might not be clinically significant (it would then have to be 2.0 or greater contrast). What the study did find is that ketamine reduced pain within 3 days, whilst for lidocaine it took until the end of treatment (5 days).

The current study is very small (6 participants) but provides another data point that psychedelics can be a tool to help with some of the worst pain that people suffer from.


Corresponding Author:

Ruin Moaddel, Marc Torjman, and Irving Wainer are listed as co-inventors on patents for the use of ketamine metabolites in the treatment of depression, anxiety, anhedonia, suicidal ideation, and post-traumatic stress disorders.


In this prospective, observational pilot study, 6 patients with refractory chronic migraine received lidocaine and (R,S)-ketamine infusions. The pain rating decreased from 7.5 2.2 to 4.7 2.8 by end of lidocaine treatment, but increased to 7.0 1.4 by the post-discharge visit at 4 weeks.

Patients with refractory chronic migraine often have continuous pain and non-painful symptoms, substantial disability, and have failed multiple medications. (R,S)-ketamine has shown promise in retrospective studies and may be an option for these patients. (R,S)-ketamine is rapidly metabolized by cytochrome P450 enzymes into multiple metabolites, including (2R,6R;2S,6S)-hydroxynorketamine (HNK), which may have a therapeutic role in the treatment of refractory chronic migraine patients with a 5-day infusion of (R,S)-ketamine. A prospective, observational pilot study was performed to compare the analgesic benefit of a 5-day infusion of (R,S)-ketamine to retrospectively collected standard treatment data with lidocaine infusion in patients with refractory chronic migraine.

Patients with chronic migraine who met refractory criteria for class IV2 were eligible to participate in this prospective, observational study. All patients provided written informed consent and the study was approved by the Institutional Review Board of Thomas Jefferson University on March 20, 2019. Data were collected from the first hospitalization of patients receiving lidocaine and other IV agents, as well as from the first post-discharge office visit.

Primary outcome data were obtained from the electronic medical record (Epic) for lidocaine hospitalizations and from prospective patient assessment for ketamine hospitalizations. Patients with refractory chronic migraine were admitted for 5 days to Methodist Hospital for lidocaine infusion.

Ketamine infusions were started at 10 mg/h and increased every 4-6 hours in increments of 5-10 mg/h up to the point of intolerable adverse effects or 1 mg/kg/h, whichever came first.

A patient with complex regional pain syndrome received a 5-day continuous ketamine infusion. Blood samples were collected at baseline, 24 hours, 72 hours, and at the end of the infusion, which was approximately 120 hours. Venous blood samples were collected from patients and stored at -70°C until they were shipped to an external laboratory for analysis. Ketamine and its metabolites were determined using a previously described achiral method.

A sample size of 6 pairs was calculated to detect a significant difference in pain between ketamine and standard treatment with lidocaine using a paired t-test assuming a mean difference of 4 points with standard deviation (SD) of 2.5 in the 0-11 numerical rating scale for pain from pre- to post- treatment.

Comparisons of pain ratings between hospitalizations were performed using the Kruskal Wallis test. Correlations between mean daily pain ratings and ketamine metabolite concentrations were also performed using the Pearson correlation coefficient.

Demographics and Pain Outcomes

A total of 7 patients were screened for eligibility over a 6-month period, and 6 were enrolled in the study. Two patients had a diagnosis of depression. Five out of six patients received dihydroergotamine in addition to lidocaine during the initial hospitalization, and all had complete data for all pain outcomes. The mean lidocaine infusion rate was 1.6 0.8 mg/min on day 1 and 2.5 0.7 mg/min on day 4.

All 6 patients had complete pain data, including baseline, end of treatment, and post-discharge pain ratings. Ketamine treatment resulted in a statistically significant decrease in pain rating (0-10 numeric rating scale) compared to lidocaine treatment.

All 6 patients experienced some AEs during treatment with ketamine, but none required complete discontinuation of ketamine.

All 6 patients had complete ketamine metabolite and concurrent medication data for analysis. The circulating plasma concentrations of (R,S)-ketamine, (R,S)-norketamine, (R,S)-DHNK, and hydroxynorketamine were correlated with mean daily pain ratings. The mean concentrations of (R,S)-ketamine, (R)-norketamine, (S)-norketamine, (2R,6R)-HNK, (2S,6S)-HNK, and (R,S)-DHNK all increased from baseline to 24 h, with (2R,6R)-HNK reaching the highest concentrations at 72 h and the end of infusion.

In this pilot study, patients with refractory chronic migraine received either (R,S)-ketamine or lidocaine infusions. Ketamine provided greater short-term pain relief, but the time course of pain relief differed between the infusions, and patients returned to baseline pain 6 weeks after both treatments. Patients with refractory chronic migraine often have disabling pain and have failed a minimum of 3 classes of medication. IV infusions can potentially improve short-term pain in this challenging headache population, and further prospective study is warranted. (R,S)-ketamine is a chiral phencyclidine derivative that was initially developed as an anesthetic agent. Pharmacokinetic and pharmacodynamic studies following acute or sub-chronic (R,S)-ketamine administration demonstrated that the antinociceptive activity produced by the continuous administration of (R,S)-ketamine could not be explained by the plasma concentrations of the target compounds.

A method was developed to quantify all of the major (R,S)-ketamine metabolites. The results indicated that (2R,6R;2S,6S)-hydroxynorketamine was a major circulating metabolite and that its plasma concentration was far greater than (R,S)-ketamine. A continuous 5-day infusion of (R,S)-ketamine increased the concentration of (2R,6R;2S,6S)-HNK, which was the predominant circulating compound, and decreased the concentration of (R,S)-ketamine and (R,S)-norketamine, which peaked at 24 or 48 hours followed by a rapid decline. Previous studies had established that (2R,6R)-HNK has analgesic properties, so the plasma samples were reanalyzed using an enantioselective chromatographic method.

(2R,6R)-HNK plasma concentrations exceeded (R,S)-ketamine concentrations at 72 h in 5 of the 6 patients and in 4 of the 6 patients on Day 5. This is consistent with previous findings that (2R,6R)-HNK has a larger systemic exposure than (2S,6S)-HNK. Previous studies have identified CYP2A6 as the primary enzyme involved in the transformation of (R,S)-norketamine to (2R,6R;2S,6S)-HNK, and have demonstrated that CYP2A6 and CYP2B6 are both induced by (R,S)-ketamine. However, the induction of CYP2B6 was not observed in the present study.

The results of this study demonstrate that the major circulating compound at 72 h and at the end of the infusion was (2R,6R)-HNK, and that these time points coincide with the optimum pain relief. Further studies should be conducted to determine whether (2R,6R)-HNK could be a viable therapeutic agent for chronic pain conditions. Our results are generalizable to other clinical settings where 5-day ketamine infusions are used, but further research is necessary to determine the role of these metabolites in pain, particularly during prolonged infusions.

Patients with refractory chronic migraine have continuous pain and substantial disability. Lidocaine and (R,S)-ketamine infusions have been shown to reduce short-term pain and “break the cycle” of constant symptoms, but (2R,6R)-HNK may hold promise as an analgesic without (R,S)-ketamine’s psychomimetic adverse effects.

Ketamine is a dissociative anesthetic used in the treatment of chronic pain. It has been shown to be effective in treating pain associated with complex regional pain syndrome.

Study details

Compounds studied

Topics studied

Study characteristics
Open-Label Within-Subject