O-Desmethyltramadol (O-DSMT)


O-Desmethyltramadol (O-DSMT) is an opioid that has primarily been used indirectly due to it being a primary metabolite of the analgesic tramadol. Since the 2010s it has occasionally been used on its own as it has been sold through the grey market/research chemical market.

It has a longer duration than tramadol and more typical opioid-like effects, although it still impacts monoaminergic systems to an extent that might somewhat influence the effect profile.

O-Desmethyltramadol = O-DSMT; Desmetramadol; O-Demethyltramadol

PubChem: 9838803

Molecular formula: C15H23NO2

Molecular weight: 249.354 g/mol

IUPAC: 3-[2-[(dimethylamino)methyl]-1-hydroxycyclohexyl]phenol



  • Light: 15 – 30 mg
  • Common: 30 – 50 mg
  • Strong: 50 – 70+ mg

Some people report using doses in the hundreds of milligrams, including via more intense routes of administration like intravenous. Because too little information exists about higher doses and non-oral routes, it is best to stick with common oral doses.

It does seem true that some users fail to receive any notable effects at common doses and have to use over 100 mg to receive a positive experience. Despite this, common doses are still recommended as they will usually be sufficient in someone without a tolerance.

A few reports of rectal administration exist. When using via this route nausea might be reduced and it may be somewhat more potent vs. oral use, so dosing needs to be adjusted accordingly.

Oral use is usually preferred to intranasal.



  • Total: 5 – 7 hours
  • Onset: 00:30 – 01:00

Experience Reports




O-DSMT has not been studied for medical uses, but it likely has efficacy in the same conditions as typical opioids like oxycodone and morphine. Also, based on research with tramadol, there’s some evidence indicating how much benefit people receive from tramadol is at least partly dependent on how much O-DSMT they produce (Kirchheiner, 2008 ; Stamer, 2007 ; Poulsen, 1996).

An animal study in rats did find O-DSMT had analgesic effects and those were blocked by a MOR antagonist (Valle, 2000). The effects were seen with the (R) enantiomer, as the (S) enantiomer produced no antinociception or respiratory depression at 2 to 10 mg/kg IV.

(Kirchheiner, 2008) – CYP2D6 status correlates with plasma levels of O-DSMT, pain reduction, and miosis.

  • Single dose of 100 mg tested in 11 UM status, 11 EM status, and 3 PM status metabolizers.
  • Results
    • Cmax of R,R-O-DSMT was significantly higher in the UM vs. EM group, with a mean difference of 14 ng/mL. Median R,R-Tramadol AUC was 786 μg*h/L in UM vs. 578 μg*h/L for EM.
    • Significantly higher pain threshold and pain tolerance along with greater miosis from tramadol in UM vs. EM. Pain was tested in healthy people via a cold pressure test. Differences in pupil diameter change between EM and UM were not significantly different (1.4 mm reduction vs. 2.2 mm).
    • Nearly 50% of UM had nausea compared with 9% of EM.
    • AUC of both tramadol enantiomers was 25% lower in UM vs. EM, but that was not statistically significant.
    • For the few PM individuals, there was no significant change in pain threshold or pain tolerance.
  • COI: Not reported

(Stamer, 2007) – CYP2D6 genotype does impact response to tramadol and clearly changes PK.

  • Germany. 174 patients received IV tramadol 3 mg/kg for postoperative analgesia following abdominal surgeries. The effect of PCA tramadol was studied for a 48-hour period after surgery.
  • Demographics: 18 were PM, 93 were IM, 68 were EM, and 8 were UM.
  • Results
    • Median AUC for (+)-O-DSMT was 0 ng*h/mL in PM, 38.6 in IM, 66.5 in EM, and 149.7 in UM.
    • In PM patients, the non-response rate was 4-fold higher compared to the other genotypes, while no significant difference in response was seen between the remaining genotypes.
  • COI: None. Some support from the R. Sackler Research Foundation.

(Poulsen, 1996) – CYP2D6 status is correlated with analgesia in humans

  • Denmark. Analgesic effect of 2 mg/kg oral evaluated in 15 EM and 12 PM individuals. DBRCT.
  • Results
    • In EM people, tramadol increased pressure pain detection and tolerance thresholds as well as the threshold for eliciting nociceptive reflexes after single and repeated stimulation of the sural nerve.
    • In PM participants, only thresholds to pressure pain tolerance and nociceptive reflexes after single stimulation were increased and the reflex threshold was increased less than in EM.
    • Serum concentration of (+)-O-DSMT 2 to 10 hours after tramadol ranged from 10 to 100 ng/mL in EM people compared to being below or around the 3 ng/mL detection limit in PM.
  • COI: Not reported

(Lehmann, 1990)

  • The median minimum effective concentration for postoperative analgesia with tramadol IV was 0.0362 μg/mL for O-DSMT, compared to 0.288 μg/mL for tramadol.


(Valle, 2000) – O-DSMT has antinociceptive properties in rats and causes respiratory depression

  • (+)-O-DSMT given IV for 10 min led to dose-dependent respiratory depression shown by decreased arterial pH and higher arterial pCO2 level. Though the pH increase was only significant for the 10 mg/kg dosing condition, not 2 to 5 mg/kg.
  • MOR antagonist β-FNA pretreatment fully antagonized the antinociceptive effect of 2 mg/kg (+)-O-DSMT.
  • Significant antinociceptive effects were found with (+)-O-DSMT.
  • Antinociceptive and respiratory depressant effects were not significant with (-)-O-DSMT.
  • COI: Not reported



  • Euphoria
  • Mood enhancement
  • Anxiolysis
  • Physical euphoria/pleasant physical sensations
  • Sedation
  • Analgesia


  • Nausea
  • Vomiting
  • Impaired respiration
  • Reduced heart rate and blood pressure
  • Sweating
  • Cognitive impairment
  • Drowsiness
  • Itchiness
  • Constipation

Users who are seeking a classic opioid effect tend to prefer O-DSMT to tramadol, which makes sense when considering its pharmacology. Often it’s said to be pretty good in people who are relatively opioid naive and/or those who don’t have an opioid tolerance, whereas people who regularly use the strongest opioids may not find it very effective. It’s been compared favorably to codeine, tramadol, and kratom by a number of users.

It does not usually produce a substantial “rush” of euphoria and is therefore weaker in that respect than certain opioids, but it can reliably offer a relaxed, chill, and pleasantly sedated state. Relative to tramadol is can bring someone to a nodding-like state of relaxation and impaired consciousness, although too few reports of its effects exist to know if it as frequently produces a “nod” as other opioids.

Cognition will usually be somewhat impaired, leading to slowed thinking or brain fog. This is partly due to the sedating effect of O-DSMT not being counteracted by monoaminergic effects like it would be with tramadol and as a result, if someone is trying to receive mood enhancement and anxiolysis while still being functional, tramadol is sometimes a preferred drug.

Some mental stimulation is sometimes reported at light to common doses, especially early in the effect period, though this is less common than with tramadol.

Feelings of warmth and general coziness or fuzziness are often reported with common to strong doses and it can also be moderately numbing for sensations, though not nearly to the extent seen with dissociatives.

Itchiness is sometimes reported like with other opioids. It can be annoying, but in a fair portion of people the sensation is not very unpleasant and can even become pleasant upon itching. This may be combated by using an antihistamine with O-DSMT, though ones that affect the brain to a substantial degree, like DPH and hydroxyzine, will potentiate the sedative effects.

A strong sensation of labored breathing can be induced with strong doses. If you’re taking a common dose and notice a change in the sensation of respiration you most likely aren’t in any physical danger, but with an overdose the sensation of labored breathing is a sign of danger, which is why common doses are recommended.

Among the after effects are minor to moderate headache, edginess/irritability, decreased motivation, and insomnia for up to a few hours after it ends, so it’s better to dose earlier in the day if you don’t want sleep to be affected. By the following day the after effects are usually minor to non existent.


O-DSMT has two stereocenters, giving four potential stereoisomers, and it is sold as a mixture of enantiomers, R,R-O-DSMT and S,S-O-DSMT. The R enantiomer is thought to be the primary active substance since it has a very high affinity for the mu opioid receptor (MOR) and studies on tramadol confirmed the S enantiomer of that substance is largely less active than the R enantiomer.

When someone uses tramadol they will generate O-DSMT via O-demethylation, which is historically how most people have come into contact with the substance.

Drug testing

One study found that, like tramadol, it is cross-reactive with the EMIT II+ immunoassay for PCP (Hull, 2006). Tramadol itself is already pretty weak as a cross-reactive substance and O-DSMT is even weaker, so it doesn’t seem very likely that someone would test positive for PCP after using O-DSMT.

(Hull, 2006) – Cross-reactive with the PCP immunoassay, but less so than tramadol.

  • O-DSMT was confirmed to interact with EMIT II+ at 27 mAU/min, compared to 44 mAU/min for tramadol. That is lower than the 85 mAU/min cutoff for the PCP test.

Natural presence

In the 2010s tramadol was detected in soil in the far north of Cameroon. Although this was taken as a sign that it may be a naturally occurring drug, more comprehensive research found local farmers and users had been a source of contamination in the region, in large part due to animals being given the substance and then excreting it, leading to its presence in the soil and plants. The same situation was seen with some of tramadol’s metabolites, O-DSMT included, but like with tramadol it is no longer thought these drugs occur naturally (more information is available on the tramadol page).

(Kusari, 2014) – Tramadol and its metabolites were detected naturally in the far north of Cameroon.

  • Along with tramadol, soil analysis in the far north of Cameroon showed the presence of O-DSMT, like N-DSMT, and 4-hydroxycyclohexyltramadol, pointing to contamination by mammals in the region.


The strongest pharmacological evidence pertaining to O-DSMT shows it is a MOR agonist and that property fits with its reported effects in humans. It’s also the basis for tramadol having long been described as dual-action analgesic: tramadol inhibits pain as an SNRI and as an opioid because of its metabolism to O-DSMT.

Tramadol is easily classified as an SNRI because it has core actions of inhibiting serotonin and norepinephrine. This is mostly absent with O-DSMT. Bamigrade (1997) reported R,R-Tramadol significantly blocked serotonin uptake at 5 μM in vitro, while S,S-Tramadol and O-DSMT had no significant impact. Another in vitro study looking at noradrenergic activity showed tramadol significantly increased extracellular norepinephrine levels by 25% at 1 μM, while even at 10 μM O-DSMT only had a weak facilitatory effect of 17% (Driessen, 1993). Given how drastic the potency difference is between opioid activity and monoamine reuptake inhibition activity for O-DSMT, it effectively loses the SNRI characteristic of tramadol.

Further confirming the importance of MOR, Sevcik (1993) tested the impact of the tramadol and O-DSMT enantiomers on locus coeruleus activity using rat brain slices in vitro. With differing potencies, all four drugs could significant reduce locus coeruleus activity. The effect of R,R-O-DSMT was nearly abolished just with 0.1 μM of the opioid antagonist naloxone, while additional use of rauwolscine (an adrenergic antagonist) was required to block the impact of R,R-tramadol and S,S-O-DSMT. The (R) enantiomer of O-DSMT was unaffected by rauwolscine and even the (S) enantiomer only showed partial attenuation, with full attenuation being seen once naloxone was present.

Over time some additional actions of O-DSMT have been indicated. Tramadol and O-DSMT both inhibit TRPA1(transient receptor potential ankyrin 1) at relevant concentrations, though tramadol is more potent (Miyano, 2015). TRPA1 is involved in sensation and blocking it has some antinociceptive effects, although not in a way that clearly explains most of what tramadol or O-DSMT do.

Racemic O-DSMT has a 19 nM affinity for MOR, compared to the substantially higher 12,000 nM affinity seen with tramadol (Volpe, 2011). The (R) enantiomer of O-DSMT has a much higher affinity (3 nM vs. 674 nM). Other studies have given similar affinity values for the drug.

Direct inhibition of Substance P, which is involved in pain transmission, was seen in vitro, with O-DSMT inhibiting the activity of Substance P to 71% of its normal level at a low concentration of just 0.1 μM (Minami, 2011). Substance P is a neurotransmitter released within afferent nerve fibers into the spinal cord and it plays a role in the excitatory synaptic input required to ultimately produce pain sensation.

There may be a serotonergic component to the analgesia offered by O-DSMT, although that could very well be indirect given the pain transmission system is complex and monoamines are involved throughout key parts of it, therefore disturbing monoamine activity could reduce the impact of O-DSMT but in an indirect manner. Yanarates (2010) demonstrated in mice that a spinal serotonergic lesion induced by 5,7-DHT significantly blocks the antinociceptive effect of tramadol and O-DSMT. Specifically administering a 5-HT7 antagonist also blocked the antinociception, whereas ketanserin and ondansetron did not attenuate the effects, suggesting a role of 5-HT7, but not 5-HT2 and 5-HT3 receptors.

5-HT2C antagonism may not play a role in tramadol or O-DSMT’s analgesia, but O-DSMT has been shown to at least bind to that site and could be causing some effect as a result. In vitro research using rat 5-HT2C expressed in Xenopus oocytes showed O-DSMT inhibited serotonin-evoled calcium-activated chloride currents, with even 0.1 μM of O-DSMT reducing the current to 74.9% of control (Horishita, 2006).

An in vitro study has shown inhibition of muscarinic M1 receptor activity (Nakamura, 2005), although in humans the effects of tramadol and O-DSMT largely don’t align with those of an antimuscarinic and this effect, if present in humans, may be minimally relevant at any typical dose.


(Grond, 2004)

  • Racemate Tramadol
    • MOR: 2.1 μM
    • DOR: 57.6 μM
    • KOR: 42.7 μM
    • NE uptake inhibition: 0.78 μM
    • Serotonin uptake inhibition: 0.9 μM
  • R,R-Tramadol
    • MOR: 1.3 μM
    • DOR: 62.4 μM
    • KOR: 54.0 μM
    • NE uptake inhibition: 2.51 μM
    • Serotonin uptake inhibition: 0.53 μM
  • S,S-Tramadol
    • MOR: 24.8 μM
    • DOR: 213 μM
    • KOR: 53.5 μM
    • NE uptake inhibition: 0.43 μM **** (most potent NE blocker of the listed drugs)
    • Serotonin uptake inhibition: 2.35 μM
  • R,R-O-DSMT
    • MOR: 0.0034 μM
    • DOR: NR
    • KOR: NR
    • NE uptake inhibition: NR
    • Serotonin uptake inhibition: NR

(Wentland, 2009)

Note: This is an outlier study in multiple respects, giving tramadol and O-DSMT high affinity for δ and κ opioid receptors, not just MOR. For now this claim should be viewed with skepticism. More research is needed.

  • Tramadol
    • MOR (hot ligand: DAMGO): 1.6 μM
    • DOR (hot ligand: Naltrindole): 0.0094 μM
    • KOR (hot ligand: U69,593): 0.014 μM
  • O-DSMT
    • MOR (hot ligand: DAMGO): 0.0086 μM
    • DOR (hot ligand: Naltrindole): 2.9 μM
    • KOR (hot ligand: U69,593): 0.450 μM


(Yanarates, 2010) – 5-HT7 receptors in the spinal cord appear connected to the antinociceptive and antihyperalgesic effects of tramadol and O-DSMT

  • Mice. Nociception was assessed via radiant heat tail-flick and plantar incision tests. The serotonergic pathways in some mice were lesioned with intrathecal 5,7-DHT.
    • Selective 5-HT7 , 5-HT2 , and 5-HT3 antagonists were given in some tests.
  • Results
    • Antinociception from both tramadol and O-DSMT was significantly diminished in serotonergic lesioned mice. Intrathecal 5-HT7 antagonist administration blocked tramadol and O-DSMT effects.
    • Ktanserin and ondansetron failed to reverse the antinociceptive and antihyperalgesic effects, indicating a lack of involvement of 5-HT2 and 5-HT3 receptors.
    • Plantar insicion produced significant induction of thermal hyperalgesia. Tramadol and O-DSMT both significantly improved hyperalgesia. 5-HT7 receptors were involved in that action since intrathecal 5-HT7 antagonism failed to influence hyperalgesia on its own, while blocking the effect of tramadol and O-DSMT.
      • Ketanserin, on the other hand, did not change the efficacy.
  • COI: None

In vitro

(Miyano, 2015) – Tramadol and O-DSMT suppress TRPA1 activity, but not TRPV1. Tramadol is more potent than O-DSMT.

  • Tramadol and O-DSMT at 0.01-10 μM failed to increase intracellular Ca2+ in HEK293 cells on their own, regardless of expression of hTRPV1 or hTRPA1 compared with capsaicin or allyl isothiocyanate (AITC), respectively. Using either as pretreatments before capsaicin did not affect capsaicin’s TRPV1-mediated activity.
  • Pretreatment with tramadol 0.1-10 μM and O-DSMT 1-10 μM significantly suppressed AITC-induced Ca2+ increases in HEK293 cells expressing hTRPA1 and a patch-clamp study confirmed that tramadol and O-DSMT (10 μM) decreased inward cation currents induced by AITC.
    • Neither induced a dose-dependent or large change in Ca2+ on their own, but some doses of tramadol (0.01 and 0.1 μM) and O-DSMT (0.1 and 10 μM) induced a slight increase in Ca2+.
    • Pretreatment was effective with 5 min but not 30 sec of pretreatment.
  • COI: None. Supported by Japanese institutional sources.

(Volpe, 2011)

  • In vitro MOR binding with recombinant human MOR conducted against 2 nM labelled DAMGO.
    • Tramadol: 12,486 nM
    • Racemic O-DSMT: 18.59 nM
    • R,R-O-DSMT: 3.359 nM
    • S,S-O-DSMT: 674.3 nM

(Minami, 2011) – Inhibition of Substance P receptors in Xenopus oocytes

  • In vitro. Xenopus oocytes expressing Substance P receptors. O-DSMT inhibited the action of 100 nM Substance P to 71.0% at 0.1 μM, to 73.6% at 1 μM, and to 56.7% at 10 μM.

(Horishita, 2006) – Inhibition of 5-HT2C receptors at relevant concentrations

  • In vitro. Rat 5-HTC receptor was expressed in Xenopus oocytes.
  • Results
    • O-DSMT inhibited serotonin-evoked Ca2+-activated chloride currents in oocytes expressing 5-HT2C.
      • 0.1 μM reduced the current to 74.9% of control, 1 μM reduced the current to 57.7% of control, and 10 μM reduced the current to 35.3% of control.
    • O-DSMT altered the dissociation constant for serotonin without changing maximum binding, indicating competitive inhibition.
  • COI: Funding from institutional sources.

(Hara, 2005) – Minimal impact on glycine receptors or NMDAR.

  • Human recombinant neurotransmitter-gated ion channels expressed in Xenopus oocytes. Glycine, and NMDA receptors expressed.
  • Results
    • Neither tramadol nor O-DSMT had a significant impact on glycine receptors.
    • Tramadol and O-DSMT inhibited the glutamate-concentration response curve without changing the ED50 or Hill coefficient for NMDAR, indicating non-competitive inhibition.
      • Tramadol IC50 of 16.4 μM and O-DSMT IC50 16.5 μM.
    • Neither tramadol nor O-DSMT influenced basal currents of any receptors tested in the study.
  • COI: Partly supported by Grants-In-Aid for Research from the Ministry of Education, Science, and Culture of Japan.

(Nakamura, 2005) – Inhibition of muscarinic M1 receptor activity in vitro

  • In vitro. M1 and M3 receptors expressed in Xenopus oocytes.
  • Results
    • O-DSMT (0.1 – 100 μM) inhibited ACh-induced currents in oocytes with an IC50 of 2 μM for cells expressing M1 receptors, while it did not alter currents in cells expressing M3 receptors.
      • O-DSMT inhibited ACh-induced currents to 77.6% at 0.1 μM, to 72.9% at 1 μM, and to 58.7% at 10 μM.
    • O-DSMT inhibited the binding of QNB to M1 receptors while not changing binding to M3 receptors.
  • COI: Supported by institutional sources.

(Gillen, 2000)

  • Cloned human MOR expressed by membranes from CHO cells.
    • R,R-O-DSMT: 3.4 nM
    • S,S-O-DSMT: 240 nM
    • Racemic tramadol: 2,400 nM
    • A GTPγS binding assay (agonist-induced stimulation of GTPγS binding) showed the most active drug was R,R-O-DSMT.

(Sevcik, 1993) – O-DSMT reduces locus coeruleus activity via MOR agonism

  • In vitro using rat midpontine brain slices. Antagonism studies using α2 adrenoreceptor blocker rauwolscine or opioid antagonist naloxone.
  • With differing potencies, the enantiomers of Tramadol and O-DSMT reduced locus coeruleus neuronal activity. The most potent drug was (+)-O-DSMT.
    • IC50 values:
      • (+)-O-DSMT: 0.15 μM
      • (-)-O-DSMT: 2.1 μM
      • (+)-Tramadol: 11.6 μM
      • (-)-Tramadol: 6.0 μM
  • The effect of (+)-O-DSMT was nearly abolished by 0.1 μM naloxone alone. While (+)-Tramadol (30 μM) and (-)-O-DSMT (10 μM) only became inactive if both naloxone 0.1 μM and rauwolscine 1 μM were present.
    • Whereas the effect of (-)-Tramadol was not changed by naloxone yet was blocked by rauwolscine.
    • (+)-Tramadol activity was greatly reduced by naloxone and abolished once there was a combo of naloxone and rauwolscine. Yet rauwolscine alone failed to significantly attenuate the activity if no naloxone was presne .
    • Effect of (+)-O-DSMT was unaffected by rauwolscine but nearly abolished just from naloxone.
    • Effect of (-)-O-DSMT was moderately attenuated by rauwolscine alone but disappeared in the presence of both rauwolscine and naloxone. Naloxone alone markedly diminished (-)-O-DSMT activity.
  • Norepinephrine 30 μM had a hyperpolarizing effect. This was potentiated by (-)-Tramadol 100 μM but it was not potentiated by (+)-O-DSMT 10 μM.
  • (+)-O-DSMT 10 μM hyperpolarized cells, which was abolished by naloxone.
  • COI: Not reported


Half-life: 6.7 hours (Grond, 2004) or 9 hours (Dayer, 1994)

Tmax: 3 hours

Cmax (after 100 mg of tramadol): ~25% of the tramadol concentration, which would be around 0.075 μg/mL. This will vary significantly based on CYP2D6 phenotype and may be affected by OCT phenotype.

AUC: ~25% that of tramadol (Grond, 2004). This will vary significantly based on CYP2D6 phenotype.

Renal impairment doubles the half-life of O-DSMT (Miotto, 2016).

Whereas tramadol’s effects are impacted by genetic differences in at least a couple areas, namely CYP2D6 and OCT1, the impact of genetics on O-DSMT may be reduced. When people receive opioid effects from tramadol that is mostly because they are converting it, with the help of CYP2D6, to O-DSMT. Taking O-DSMT bypasses that and should give a more consistent effect profile if pure opioid-like properties are what the user desires.

Organic cation transporter 1 (OCT1)

Because O-DSMT itself is still going to be metabolized and uptake into the liver for that metabolism can vary based on OCT1 status, there can still be some inter-individual variability there. OCT1 mediates the uptake of the drug into the liver, as it’s a transporter expressed on the membranes of hepatocytes. Polymorphisms are common, like with CYP2D6. For example, around 9% of Europeans have a “poor transporter” profile correlating with higher levels of drugs taken up by OCT1, which probably extends to O-DSMT, allowing those users to get stronger effects per dose.

There’s some evidence OCT1 really is relevant in the effects of tramadol and likely, by extension, O-DSMT. Stamer (2016) demonstrated a poor transporter phenotype correlated with nausea/vomiting and Tzvetkov (2011) found miosis was increased in poor transporters. Both findings point to reduced OCT1 activity causing more opioidergic effects because O-DSMT is not being metabolized as quickly.

Tzvetkov (2011) also reported in vitro membrane penetration was strong for tramadol even when OCT1 was absent, suggesting differences in OCT1 will specifically affect the pharmacokinetics of O-DSMT, whose penetration is low without OCT1.

Multiple studies have show a person’s OCT1 profile affects the pharmacokinetics of O-DSMT (Stamer, 2016 ; Tzvetkov, 2011 ; Matic, 2016).

(Stamer, 2016) – OCT1 function affects O-DSMT PK and effects.

  • 205 patients scheduled for elective open abdominal or urological surgery: 19 had 0 active OCT1 alleles, 82 had 1 functional allele, and 104 had 2 functional alleles
    • Analgesic regimen consisted of 3 mg/kg tramadol (max 250 mg) IV before termination of anesthesia along with dipyrone 1 gram. Then the regimen in the post-anesthesia care unit was additional IV tramadol doses as needed and subsequent PCA for a max of 48 hours.
  • Results
    • Cumulative tramadol use via PCA was lowest in patients with 0 active alleles vs. the other groups. Plasma AUC for O-DSMT were 111.8 ng*h/mL with 0 active alleles, 80.2 ng*h/mL with 1 active allele, and 64.5 ng*h/mL with 2 active alleles.
    • Postoperative nausea and vomiting were reported in 47.3% of OCT1 poor transporters in first 24 h compared to 29.3% of intermediate and 28.9% of extensive transporters.
  • COIO: None. Supported by R Sackler Research Foundation and by a grant of the German Research Foundation.

(Matic, 2016) – OCT1 genotype significantly affects the PK of O-DSMT in infants

  • The Netherlands. 50 infants requiring analgesia received an IV tramadol dose of 2 mg/kg followed by continuous infusion of 5-8 mg/kg/d.
  • Results
    • The prevalence of 2 inactive OCT1 alleles was 4%.
    • SLC22A1/OCT1 genotype significantly correlated with the O-DSMT/Tramadol ratio. Infants with under 2 functional alleles had a higher O-DSMT/Tramadol ratio than infants with 2 functional alleles.
    • Pairing CYP2D6 and OCT1 genotype showed there was a 57.8% higher O-DSMT/Tramadol ratio in those with 2 or more functional CYP2D6 alleles paired with under 2 OCT1 alleles.
  • COI: None. Funded by institutional sources.

(Tzvetkov, 2011) – OCT1 polymorphisms lead to different O-DSMT PK and different tramadol response

  • In vitro test using HEK293 cells and a human study using healthy people.
  • Human study involved 24 healthy volunteers. 14 had two active OCT1 alleles, 8 had one active allele, and 2 had two inactive alleles.
  • In vitro
    • O-DSMT showed low membrane permeability without transporters, while tramadol did show permeability.
    • Tramadol uptake into HEK293 cells was not altered by OCT1 overexpression. While OCT1 overexpression increased O-DSMT uptake 2.4-fold. The increase was reversed by OCT1 inhibitors and absent when inactive non-functional OCT1 variants were overexpressed.
  • Human
    • Concentration of tramadol in plasma was independent of OCT1 genotype. Volunteers with non-functional OCT1 polymorphisms had significantly higher plasma levels of O-DSMT and significantly prolonged miosis.
    • AUC0-24 for (+)-O-DSMT decreased with the number of active alleles:
      • 805.5 μg*h/L with zero
      • 679.4 μg*h/L with one
      • 420.2 μg*h/L with two
  • COI: None


There is very little history specifically related to O-DSMT because it has not been marketed on its own by pharmaceutical companies and comparatively little research has been conducted on it relative to tramadol.

It appears to have first shown up on its own in the 2000s and then it became a somewhat popular substance around the 2010s, with grey market/research chemical vendors selling it online. Analysis of commercial incense products purchased in 2008 and 2009 revealed O-DSMT was present alone (n=1) or with caffeine (n=2) in three of 140 samples (Dresen, 2010).

Bodily samples analyzed by Sweden’s STRIDA project, which began in 2010, showed 9 of 103 drug samples were positive for tramadol or O-DSMT (Helander, 2013). 3 exclusively contained O-DSMT, 4 had both, and 2 were positive just for tramadol.

In Germany, a user of an alleged kratom product called “Krypton” tested positive for O-DSMT and kratom alkaloids, but not for tramadol (Arndt, 2011).

(Arndt, 2011) – A Krypton user tested positive for kratom alkaloids and O-DSMT.

  • Germany. Testing the urine of an opiate-addicted female. 33-year-old female with an extensive substance use history was successfully detoxed and admitted to a rehab center. After ~3 months, she showed an altered clinical picture with miosis, agitation, and moderate euphoria.
  • Kratom alkaloids and O-DSMT, but not tramadol or N-DSMT, were detected.
  • COI: Not reported

Analysis of 251 samples submitted to Sweden’s STRIDA from 2010 to 2015 revealed a single case of powder containing O-DSMT, yet another “Krypton” branded sample suspected to contain either kratom alkaloids or O-DSMT had no detectable psychoactive substance in it (Backberg, 2018).


O-DSMT continues to be used, but it is not a very popular opioid, in part because of limited availability and a somewhat high price, according to user reports. It has occasionally shown up in products, such as those alleged to contain kratom, perhaps to increase the strength of the mixture, but overall it is not a prevalent substance.

Legality (As of January 2019)

Note: This list is not exhaustive. Always check your local laws to verify how O-DSMT is treated in your region.

Australia: Not specifically controlled

Canada: Uncontrolled

United Kingdom: Class A

United States: Uncontrolled



Few detailed reports of its overdose effects are available but there is good reason to suspect it more readily leads to a stereotypical opioid-like overdose than tramadol. This would include effects like impaired consciousness, coma, respiratory depression, nausea, constipation, slowed heart rate and/or blood pressure, and miosis.

Because there may still be a small amount of monoaminergic effects, atypical overdoses may produce tachycardia, hypertension, and non-hypoxic seizures, but there is far more reason to anticipate typical opioid effects instead.

It is reasonable to suspect O-DSMT can be fatal when large doses are used or when it is combined with other depressants, such as alcohol, benzodiazepines, and barbiturates. Case reports do not exist, but that is more likely due to a relatively low level of use than to a strong safety profile. Given this, avoiding depressant combinations and strong+ doses is highly recommended.

(Elkalioubie, 2011) – Near-fatal response in an ultrarapid metabolizer who overdosed

  • France. 22-year-old female who was previously healthy and then found unconscious. Repeated episodes of cardiac arrest requiring CPR and immediate admittance to ICU where refractory circulatory shock was diagnosed, requiring extracorporeal circulatory support by venoarterial membrane oxygenation. Hypothermia of 35.5°C was noted along with mydriasis.
  • Echocardiography showed severe biventricular failure. Hypoglycemia at 0.57 g/L, lactic acidosis at pH of 6.85 and lactate of 18 mM/L. Renal failure was seen with creatinine of 22 mg/L and urea 0.54 g/L, but normal hepatic and coagulation parameters.
  • Routine tox screen of serum and urine at ICU admission: negative for alcohol, drugs, and other poisons. Only positive for tramadol in both matrices.
  • Despite tramadol identification, naloxone was not used or justified. Within the first 8 h of admission: Patient developed severe liver failure with profound coagulopathy. Signs of liver failure abated after 36 hours and continuous EEG showed no signs of hepatic or post-anoxic encephalopathy.
  • Discharged after 35 days in hospital.
  • Toxicology
    • Initial blood level for tramadol: 3.22 mg/L, which was fairly low considering the reputed dose of 4.5 grams.
    • Apparent elimination half-life was 16 hours, with tramadol persisting over the therapeutic level for 72 hours.
    • Calculation of tramadol/O-DSMT ratio was 2.54 and tramadol/N-desmethyltramadol yielded a ratio of 11.4.
  • Genotyping predicted UM phenotype for CYP2D6. Also, she was on ketoconazole, a CYP3A4 inhibitor, which was present in her system at an enzyme-inhibiting concentration. The data indicate she could have excessively produced O-DSMT, while having a low production of N-desmethyltramadol.
  • COI: Not reported


Typically in the tramadol overdoses that exist, which are rare, tramadol itself is the dominant drug in the body and the concentration of O-DSMT is lower. But in some instances where the tramadol-related death is linked with respiratory depression, cardiac arrest, and bradycardia, for example, metabolism to O-DSMT has been blamed. So it’s possible that an opioid-like fatality from tramadol is much more likely in extensive or ultrarapid CYP2D6 metabolizers.

A series of fatalities involving kratom alkaloids and O-DSMT occurred in the late 2000s and early 2010s, though all the fatalities also involved additional drugs (Kronstrand, 2011).

  • (Kronstrand, 2011) – Series of nine fatalities involving a combination of kratom alkaloids and O-DSMT, always with other drugs present.
    • Background
      • A commercial preparation called Krypton has been shown to contain kratom alkaloids along with O-DSMT.
    • Sweden. Since November 2009, the authors identified 9 cases of fatal overdose in which postmortem blood samples were positive for both mitragynine and O-DSMT in the absence of tramadol, indicating direct O-DSMT use and therefore the cases were interpreted as involving Krypton.
    • All fatalities involved other drugs, among those being antidepressants, ethanol, and benzodiazepines, which could have contributed. Lung congestion w/ higher lung weight was consistently found (except for one case), supporting respiratory depression and opioid-type overdose.
    • Some cases had a history of ordering Krypton or at least a prior history of drug abuse.
    • Toxicology
      • O-DSMT in blood: 0.4 – 4.3 μg/g
      • Mitragynine: 0.02 – 0.18 μg/g
    • COI: Not reported


Seizures may less of an issue in overdose compared to tramadol, though this has not been confirmed. The (S) enantiomer of O-DSMT did have the ability to cause seizures in rats, but minimal research was done because similar doses also caused a lot of respiratory depression (Potshka, 2000).


(Ahmadimanesh, 2018) – There is a significant correlation between tramadol/O-DSMT metabolite plasma concentration and seizure occurrence.

  • Iran. 120 tramadol-intoxicated patients were split into seizure and no-seizure groups. All tramadol poisoned patients referred to the ED of Loghman-Hakim hospital in Tehran during a 3-month period were studied.
  • Seizures were observed in 41.6% of the population. The other common complications were nausea and vomiting in over 40%.
  • Seizures occurred in the first 3 hours in 72% of patients. Seizure was significantly correlated with concentrations of tramadol, O-DSMT, and N-desmethyltramadol, and history of previous seizures.
  • Average concentration of N-desmethyltramadol was significantly higher in males.
  • Plasma level of O-DSMT was correlated with the onset of seizure.
  • Median values of the estimated ingested doses were 1000 mg in both groups, with very large SDs of over 1000 mg. The minimum reported dose associated with seizure, which occurred in 3 patients, was 200 mg.
  • Co-ingestion
    • 15 patients co-ingested benzodiazepines. None had seizures despite 3 having a history of seizure due to tramadol poisoning and epilepsy.
    • Co-ingestion of other opioids significantly correlated with a lower risk of seizure.
  • COI: None. The study was fully supported by the Tehran University of Medical Sciences.


Some human and animal evidence points to a risk of liver toxicity (Arafa, 2018), though tramadol has been used for decades in humans, including long-term, and no clear evidence of a severe impact on liver health exists. Despite what has been shown, hepatotoxicity may not be a real concern in humans when used moderately, though more research is needed for a definitive answer.

(Arafa, 2018) – O-DSMT generation from tramadol is associated with hepatotoxicity in long-term use

  • Background
    • α-GST is a sensitive and good hepatic enzyme in the assessment of earlier hepatocellular damage vs. aminotransferases.
    • Hepatotoxicity has been reported as a rare side effect affecting under 0.1% of patients. Some evidence indicates a high rate of tramadol abusers have atypical AST and LDH results indicative of impaired liver function.
  • Egypt. 60 patients given chronic tramadol treatment for a mean of 3.9 years. Dose of 200-400 mg/d. They were compared to healthy, drug-naïve controls. CYP2D6 duplication allele carriers (UM status) followed by EMs had significantly higher levels of O-DSMT in urine along with higher serum lipid peroxidation and lower levels of total antioxidants.
    • Comedications including narcotic analgesics, NSAIDs, local anesthetics, hypnotics, paracetamol, and corticosteroids were excluded.
  • Significant increases in serum hepatic damage markers including α-GST levels and liver function enzyme activities compared to no significant increase vs. controls for PMs.
    • α-GST, ALT, AST, ALP, and GGT were all affected.
  • Tramadol caused more moderate to severe hepatotoxicity in more significant CYP2D6 metabolizers.
  • COI: None


Tolerance can develop very fast with daily use, especially when compulsively using it multiple times per day. The best effects may be hard to obtain after just a few days of heavy use, which is a major cause of people becoming highly dependent on the substance since they continue chasing the progressively harder to obtain positive effects.

Within a week of beginning heavy daily use, and especially within a few weeks, withdrawal will be an issue upon stopping administration. The withdrawal effects match those seen with other opioids, although they may tend to be weaker in intensity when compared to what occurs with diamorphine, for example.

Among the withdrawal effects are sweating, vomiting, a strong ill flu-like feeling, runny nose, chills, lethargy, and headache. At least 3 to 5 days of a moderate to strong ill/flu feeling should be expected.

Though more reports are needed to confirm how O-DSMT compares to other opioids, the mental effects are reportedly less severe than what’s seen with the classically strongest opioids, meaning withdrawal could be less depression and despair-inducing.


Arndt, T., Claussen, U., Güssregen, B., Schröfel, S., Stürzer, B., Werle, A., & Wolf, G. (2011). Kratom alkaloids and O-desmethyltramadol in urine of a “Krypton” herbal mixture consumer. Forensic Science International, 208(1–3), 47–52. https://doi.org/10.1016/j.forsciint.2010.10.025

Bäckberg, M., Jönsson, K. H., Beck, O., & Helander, A. (2018). Investigation of drug products received for analysis in the Swedish STRIDA project on new psychoactive substances. Drug Testing and Analysis, 10(2), 340–349. https://doi.org/10.1002/dta.2226

Dresen, S., Ferreirós, N., Pütz, M., Westphal, F., Zimmermann, R., & Auwärter, V. (2010). Monitoring of herbal mixtures potentially containing synthetic cannabinoids as psychoactive compounds. Journal of Mass Spectrometry, 45(10), 1186–1194. https://doi.org/10.1002/jms.1811

Helander, A., Beck, O., Hägerkvist, R., & Hultén, P. (2013). Identification of novel psychoactive drug use in Sweden based on laboratory analysis-initial experiences from the STRIDA project. Scandinavian Journal of Clinical and Laboratory Investigation, 73(5), 400–406. https://doi.org/10.3109/00365513.2013.793817

Horishita, T., Minami, K., Uezono, Y., Shiraishi, M., Ogata, J., Okamoto, T., & Shigematsu, A. (2006). The tramadol metabolite, O-Desmethyl tramadol, inhibits 5-hydroxytryptamine type 2C receptors expressed in Xenopus oocytes. Pharmacology, 77(2), 93–99. https://doi.org/10.1159/000093179

Kronstrand, R., Roman, M., Thelander, G., & Eriksson, A. (2011). Unintentional fatal intoxications with mitragynine and o-desmethyltramadol from the herbal blend krypton. Journal of Analytical Toxicology, 35(4), 242–247. https://doi.org/10.1093/anatox/35.4.242

Matic, M., De Wildt, S. N., Elens, L., De Hoon, J. N., Annaert, P., Tibboel, D., … Allegaert, K. (2016). SLC22A1 /OCT1 Genotype Affects O-desmethyltramadol Exposure in Newborn Infants. Therapeutic Drug Monitoring, 38(4), 487–492. https://doi.org/10.1097/FTD.0000000000000307

Minami, K., Yokoyama, T., Ogata, J., & Uezono, Y. (2011). The Tramadol Metabolite O-Desmethyl Tramadol Inhibits Substance P– Receptor Functions Expressed in Xenopus Oocytes. Journal of Pharmacological Sciences J Pharmacol Sci, 115, 421–424. https://doi.org/10.1254/jphs.10313SC

Nakamura, M., Minami, K., Uezono, Y., Horishita, T., Ogata, J., Shiraishi, M., … Sata, T. (2005). The effects of the tramadol metabolite O-desmethyl tramadol on muscarinic receptor-induced responses in Xenopus oocytes expressing cloned M1or M3receptors. Anesthesia and Analgesia, 101(1), 180–186. https://doi.org/10.1213/01.ANE.0000154303.93909.A3

Sevcik, J., Nieber, K., Driessen, B., & Illes, P. (1993). Effects of the central analgesic tramadol and its main metabolite, O‐desmethyltramadol, on rat locus coeruleus neurones. British Journal of Pharmacology, 110(1), 169–176. https://doi.org/10.1111/j.1476-5381.1993.tb13788.x

Valle, M., Garrido, M. J., Pavon, J. M., Calvo, R., & Troconiz, I. F. (2000). Pharmacokinetic-pharmacodynamic modeling of the antinociceptive effects of main active metabolites of tramadol, (+)-O-desmethyltramadol and (-)-O-desmethyltramadol, in rats. J Pharmacol Exp Ther.

Category Tag