Iboga

What it is

Iboga is the common name of Tabernanthe iboga, a forest shrub of Central and West Central Africa. Its root bark contains indole alkaloids, the best known of which is ibogaine, which can be extracted from the bark or made semi-synthetically from voacangine, a related alkaloid. The ibogaine in the main veteran study below was synthesized in South Africa from voacangine and not extracted from iboga root [1], and in a rat study discussed later it was made from voacangine extracted from root bark [2]. This matters for any claim about "iboga", because nearly all the clinical evidence concerns purified ibogaine hydrochloride and almost none whole root bark. One exception is a 30-person study in which five participants also received a crude root bark extract [13].

The liver enzyme CYP2D6 plays a large part in clearing ibogaine, and the body also produces noribogaine, a metabolite with effects of its own that lingers, with a half-life of roughly 28 to 49 hours according to the abstract of a healthy-volunteer study whose full text was not accessible [33] [6] [3]. Which compound drives which effect is not settled [2] [33], and that runs through the safety story below.

Ibogaine does not work like psilocybin or LSD. It binds with moderate to weak strength to many targets, including the NMDA receptor, opioid and sigma receptors and the serotonin and dopamine transporters [1] [6]. Sources disagree about the serotonin 2A receptor that most classic psychedelics act through, from low affinity or no significant role [32] [34] to agonism [14]. The experience is called oneirogenic, meaning dream-producing [1] [6].

What is known

The short version. Controlled evidence in healthy adults is almost absent. What exists is a handful of small drug-handling studies with no growth measure, one paper reporting minimal effects on mood ratings and psychological tests, and a registry record for a noribogaine study in healthy people that lists QT prolongation as an adverse event. The clinical literature is larger but mostly uncontrolled, and reports both signals of benefit and a serious cardiac risk. The claim that ibogaine raises BDNF and GDNF in people comes from rat experiments, where mature BDNF protein did not rise, and has not been measured in humans.

How little of the research involves healthy adults

Two published Phase 1 studies gave noribogaine or ibogaine to healthy volunteers, and neither full text was accessible, so what follows is what their abstracts report. In the first, 36 healthy, drug-free men received single oral noribogaine or placebo in four ascending steps, and the abstract reports no safety or tolerability problems and no mu-opioid effects [3]. In the second, 21 healthy volunteers received a single small oral amount of ibogaine after six days of placebo or paroxetine, which blocks CYP2D6, and the abstract reports it safe and well tolerated, with total exposure to ibogaine plus noribogaine about twice as high when the enzyme was blocked, and its authors suggested genotyping people awaiting treatment for CYP2D6 [4]. A separate paper on the same 21 volunteers reports in its abstract minimal influence on mood ratings and psychological tests two hours after the dose and no stimulant effects [36]. That is a null result on short-term mood and test performance, not a growth outcome.

A larger noribogaine study, sponsored by DemeRx NB and registered as NCT06480981, was a randomized, placebo-controlled Phase 1 study with repeated administration in 55 healthy people aged 18 to 45, with results posted on 22 January 2026. Four noribogaine groups of 10, 12, 12 and 9 people received rising amounts, and 12 received placebo. The record lists no serious adverse events but lists "ECG QT prolonged" in none of the 10 in the lowest group, 2 of 12 and 6 of 12 in the next two, 7 of 9 in the highest, and none on placebo. It gives no size of the QT change and no threshold, and no peer-reviewed paper was found [22]. A registered atai Therapeutics Phase 1/2 study, NCT05029401, includes healthy volunteers given ibogaine with QT modelling and has no posted results [22].

A systematic review of psychedelic treatments for substance use found five ibogaine studies with 166 participants, all in people being treated for a disorder, and no serious adverse events. That is weaker reassurance than it sounds: only 14 of the review's 37 studies (38 percent) reported adverse events at all, and the authors say it is unknown whether safety data were collected and left unreported [23].

Opioid dependence and other substance use (clinical populations)

A systematic review searching to December 2020 found 24 studies covering 705 people who received ibogaine or noribogaine and did not pool them because they were too different, which its authors say limits the evidence. By its tables, seven were case reports or series, eight observational or retrospective, six open-label and three double-blind controlled, two of the three testing noribogaine. The only controlled ibogaine study was a 20-person study in cocaine dependence that reported less craving than placebo at 72 hours and 24 weeks [9] [23]. The abstract describes the three double-blind studies differently from the tables and text. The reviewers concluded that the data suggest benefit for withdrawal and craving, and also reported severe medical complications and deaths [9]. A 2026 scoping review reports in its abstract three randomized trials: a 20-person cocaine pilot, the healthy-volunteer noribogaine study above and a 27-person noribogaine trial in opioid-dependent patients [15].

One open-label example is a 30-person observational study of opioid detoxification with ibogaine at two private, fee-charging clinics in Baja California, Mexico, supported by the Multidisciplinary Association for Psychedelic Studies (MAPS). At one month 15 of the 30 reported no opioid use in the previous 30 days, and 10 reported none at three months, but only 20 of the 30 were available for the one-month analyses. Without a comparison group, the numbers cannot separate the compound from the setting and the motivation of people who seek it. The paper reports no clinically significant cardiovascular events under three-lead ECG monitoring with trained staff for at least 24 hours, and gives no QT figures [13]. A 2023 systematic review of 31 articles, mostly case reports, series and observational studies, found some efficacy but worrying cardiotoxicity and mortality, drew no conclusions about efficacy or an optimal amount, and described most high-mortality cases as occurring in home or covert settings [14].

A meta-analysis of about 28 animal studies (27 in its abstract) shows reduced drug self-administration, short-lived motor impairment and, in some experiments, cerebellar cell loss, which appeared with injection into the abdominal cavity at a high amount and not with oral dosing. The authors found no animal study of heart rate, rated reporting quality as poor, and advise against drawing conclusions about clinical efficacy [17].

Veterans with brain injury: the MISTIC study (clinical population)

MISTIC, short for Magnesium-Ibogaine: the Stanford Traumatic Injury to the CNS protocol, is the best-known recent human study, observational and open-label, with no control group. It enrolled 30 male Special Operations veterans, most with mild traumatic brain injury and many with psychiatric diagnoses at baseline: 23 met criteria for PTSD, 15 for major depression and 14 for an anxiety disorder. Of 34 people assessed, 33 enrolled and 30 completed. Treatment took place at Ambio Life Sciences, a clinic in Tijuana, Mexico, which the veterans arranged independently with a nonprofit grant, and Stanford did not supply the drug. Three authors are affiliated with Ambio, which comes from the author metadata because the competing-interests statement could not be read [1].

Disability on the WHODAS 2.0 (the primary outcome) fell from 30.2 to 19.9 immediately afterward (d = 0.74) and to 5.1 at one month (d = 2.20). Clinician-rated PTSD, depression and anxiety all dropped at one month, with effect sizes above 2.0, and memory and attention tests showed improvement and no decline. The authors list the limits themselves: no control group, a protocol that included preparation, coaching and on-site complementary activities, veterans who chose to travel so expectation could not be separated from the drug, at least two participants who returned to heavy psychosocial stress and lost some of the gain, follow-up only to one month, and exploratory analysis of suicidal thinking. They call for controlled trials [1].

Three later papers used the same cohort. According to abstracts, the only text read for two, stronger mystical-type experiences went with larger PTSD reductions, a correlation inside one uncontrolled group [18], and brain imaging showed changes in blood flow and connectivity [19]. The third, an imaging study of 22 to 25 of the men, was read in full. Average cortical thickness rose from 2.08 to 2.13 millimetres, but only at a trend level after correction, and the authors found no significant correlation between the structural changes and the change in disability, say it is unclear whether the changes are adaptive, and collected no biological samples, so growth factors were not measured. Four authors are listed as inventors on Stanford intellectual property relating to magnesium-ibogaine [20].

Where the BDNF and GDNF claim comes from

BDNF and GDNF are proteins that support the survival and growth of nerve cells, and a claim often attached to ibogaine says it raises both. The source is a rat study in which male rats received a single abdominal injection of ibogaine at one of two amounts or a vehicle, six animals per group (four for protein), and four brain regions were examined 3 or 24 hours later. At 24 hours, GDNF gene activity rose only at the higher amount, in two regions, and BDNF gene activity rose in all four regions at the higher amount and in two at the lower. Protein told a narrower story: mature GDNF protein rose about two-fold in one region at the higher amount only, mature BDNF protein did not rise in any region, and the precursor proBDNF rose in one. The authors note that gene activity and protein are usually only weakly linked [2].

The study asked about drug-seeking and not personal growth, is small, used males only and did not test the metabolite separately, so it is unknown whether ibogaine or noribogaine is responsible [2]. MISTIC measured none of these, and BDNF and GDNF appear in it only in a background sentence [1]. The imaging paper and the Ibogaine Experience Scale paper repeat the idea in their introductions without saying the measurements came from rats [20] [32]. Even in humans, blood may not show what happens in the brain: a randomized crossover trial of LSD in 45 healthy adults recruited and 43 analysed found no effect on serum or plasma BDNF at any time point, and its authors note that blood BDNF mostly does not originate in the brain [21]. No human study found for this review measured BDNF or GDNF after ibogaine or iboga.

Safety in healthy adults

The evidence on safety has to be read in four layers, because they come from different people.

The two published Phase 1 studies above reported no safety problems in their abstracts, in small, short, single-administration designs, the first in men only, and they cannot detect rare events [3] [4]. The registry results for the 55-person noribogaine study are less reassuring: QT prolongation was listed in a growing share of people as the amount rose, from none in the lowest group to 7 of 9 in the highest and none on placebo, with no serious adverse events listed and no size of effect given [22]. Exposure also varies between people: paroxetine doubled total exposure in the ibogaine study abstract [4], and in 14 patients clearance differed more than tenfold between the lowest and highest CYP2D6 activity scores [33].

In clinical populations under monitoring, the cardiac signal is consistent. Ibogaine blocks the hERG potassium channel of the heart. In cells carrying human hERG channels it reduced the current reversibly and in proportion to concentration, with a half-maximal effect near 4 micromolar, and the authors note that blood levels in people fall in the low micromolar range [10]. That paper tested ibogaine only, in cells, with no metabolite and no animals. Two further laboratory papers, read only as abstracts, are cited for the mechanism of the block [11] and for noribogaine and voacangine [12], so claims about those compounds are not independently confirmed here.

In a Dutch study of 14 patients with opioid use disorder, 12 men and 2 women, screened to exclude heart disease and monitored with repeated electrocardiograms for at least 24 hours, half reached a corrected QT interval (QTc) above 500 milliseconds, a level generally treated as high risk, and both women were among them. Six of 14 were still above 450 milliseconds after 24 hours according to the abstract and table, but the results text gives 29 percent, which would be four, so the paper contradicts itself. Eight patients received intravenous magnesium and seven had a heart rate below 60. Every patient developed ataxia that resolved within 24 hours, and some needed support to walk. No torsades de pointes, a dangerous heart rhythm that a prolonged QT interval can lead to, occurred, but the study was far too small to detect it, and the recording was not continuous. The amount given was at the low end of those used earlier, so toxicity may be underestimated, and the authors concluded that use outside a well-controlled medical setting should be avoided [6].

A later analysis of the same 14 patients found that the QTc rise followed a saturating curve against ibogaine levels, with an estimated maximum of about 68 milliseconds, and the authors postulate that the prolongation comes mainly from ibogaine, though the two compounds cannot be fully separated because both were present [33]. Registry results for noribogaine in healthy volunteers [22] show QT events with the metabolite alone, so a lower cardiac risk for noribogaine is not established. According to its abstract, a noribogaine trial in 27 opioid-dependent patients found a QTc rise that grew with exposure, with the largest average effects from about 16 to 42 milliseconds [5]. According to its abstract, a 2022 systematic review of adverse events in 18 studies ranked QTc prolongation the most common acute effect and reported longer-lasting cardiac, psychiatric and neurological effects in some people [8].

MISTIC looks different, although three of its authors are affiliated with the clinic that gave the treatment. Its authors report no clinically meaningful QT prolongation, defined in the paper as prolongation qualitatively detectable on monitoring, no serious adverse events, and transient cerebellar signs in all participants that resolved within 24 hours, with headache in 40 percent and nausea in 23 percent [1]. The protocol included magnesium infusions, screening that excluded cardiovascular, liver and kidney problems, and a doctor, nurse or EMT with advanced cardiac life support certification at a ratio of at least one to two patients. The two studies cannot be compared directly, since they differ in population, protocol, concomitant drugs and QT assessment, and MISTIC's QT monitoring was visual observation of the ECG, not a formal QT study [1] [6]. According to its abstract, a review of 12 clinical sources concluded that controlled settings with trained staff and cardiac, psychiatric and medical monitoring are essential for safety [16].

Counts of deaths differ by source and cannot be added together. A forensic review of fatalities outside West Central Africa from 1990 through 2008 found, according to its abstract, which is all we could read, 19 people who died between 1.5 and 76 hours after taking ibogaine. The abstract says the evidence did not suggest a characteristic neurotoxic syndrome, that in 12 of 14 cases with adequate postmortem data advanced medical illness, mainly cardiovascular, or other commonly abused substances explained or contributed to the death, and that other apparent risk factors included seizures during withdrawal and uninformed use of ethnopharmacological forms [7]. The systematic review of clinical studies describes two fatalities among its 24 studies, one of them in New Zealand. It also collected deaths and emergencies from other sources and, after removing duplicates, counted 58 cases, 20 emergencies and 38 deaths, with concurrent drug use in 34.5 percent, ibogaine given for opioid use disorder in 70.7 percent, and most events involving heart rhythm disturbances. This is a cumulative count from mixed sources and not a rate [9]. A clinical primer cites 33 publicly reported deaths [34]. Case reports cited in the Dutch paper include seizures, mania and one suicide during underground treatment [6].

A fourth layer comes from people who used ibogaine outside research. In a 2026 US national online survey of 1,964 adults, 74 reported ibogaine use in the previous 12 months. Of these, 45.1 percent attributed a worsening of physical health to it, 34.5 percent a worsening of mental health, and 74.2 percent used health services, against 6.2, 4.7 and 16.2 percent among psilocybin users. Among the 59 who had used it in the previous three months, 94.8 percent attributed at least one of 13 listed adverse effects to it, including thoughts of death or suicide (59.3 percent). The sample is small, the attribution self-reported, and the group not healthy: 52.0 percent used it for a medical symptom and 10.4 percent had ever had a schizophrenia diagnosis. A survey cannot show cause, and it is not evidence that healthy adults are harmed at these rates [35].

Taken together, the data describe a compound with a documented cardiac liability, a long-acting metabolite that also produces QT events in healthy volunteers, variable metabolism between people, and deaths in people with other risk factors, in a literature where death counts are cumulative and safety reporting is thin. No study has established a screening threshold that would make exposure acceptable for a person with no diagnosis. This page describes what was observed in supervised research and does not describe, and Metanoia does not suggest, any way to reduce risk outside that setting.

Legal status

These entries come from official legal pages opened on 2 October 2026 and are grade 1 unless marked otherwise. They describe the law and grant no permission, and whether research with a controlled substance is allowed, and under what authorization, is not assessed here.

United States (federal). Ibogaine is a Schedule I hallucinogenic substance, listed at 21 CFR 1308.11(d), item 21, with DEA code 7260, and the entry names Tabernanthe iboga among its other names. The entry was read on 2 October 2026 in the eCFR, up to date as of 1 October 2026 (accessed 4 October 2026), and the item number and DEA code were checked in that text. Only paragraph (d) was read, and the 2026 amendments to the section, the latest on 21 September in paragraph (h), did not touch the ibogaine entry [24].

Texas. Senate Bill 2308 was signed on 11 June 2025 and took effect immediately. It directs the state health commission to fund a consortium to run FDA drug-development trials of ibogaine. The enrolled text states no dollar amount, and it and the bill history were read in full as source text [25]. In December 2025 the lead institutions announced a $50 million award for a two-year program called IMPACT. The announcement says SB 2308 provides the $50 million, which the enrolled text does not state, and where the amount was appropriated was not checked. It is the institutions' own announcement and is graded 3 [26].

Colorado. House Bill 26-1325 was approved on 4 June 2026, the date printed in the signed text (Session Laws, chapter 372), and is listed as taking effect on 12 August 2026 subject to a referendum petition, which was not checked. It creates an ibogaine research pilot program in the state's Behavioral Health Administration with up to five sites, each needing a benefit-sharing plan with Indigenous communities that documents the source of the ibogaine, and the pilot provisions take effect only if the Administration receives $150,000 in gifts, grants, donations or equivalent in-kind items before 1 January 2028. Its definition of natural medicine includes semi-synthetic ibogaine made from Tabernanthe iboga or Voacanga africana, subject to board approval. State law does not change the federal Schedule I status. The ibogaine sections of the signed text were read, and the sections on licensing, healing centers and psilocybin were not [27].

New Zealand. The Medicines Regulations 1984, Schedule 1, Part 1 (prescription medicines), list both ibogaine and noribogaine in the legislation website version current as at 10 July 2026. Both were located by text search of the full regulations page, and the regulations were not read end to end. A prescription listing does not by itself show that any product is approved [28].

Other jurisdictions were not verified against primary instruments and are not described. Statements that ibogaine is permitted or unregulated in particular countries circulate widely on clinic and advocacy sites, and they are not used here.

What is not known

Open scientific questions

We found no published controlled trial of ibogaine or iboga in healthy adults with growth-type outcomes. Human data from healthy volunteers are limited to two pharmacology studies in 57 people, a mood and test-performance paper on the same 21 of them, a 55-person noribogaine study whose only results are in a registry record, and a registered ibogaine Phase 1 study with no posted results [3] [4] [22] [36]. The cardiac data come mostly from patients, although the registry shows QT events in healthy volunteers given noribogaine, and MISTIC's screening excluded people with the problems that raise risk. No work found sets a heart-safety threshold for a healthy volunteer, and large differences in clearance with CYP2D6 activity mean a threshold would likely have to be individual [33].

On the clinical side, most studies are open-label, small and short, and a 2022 systematic review found them too different to pool [9]. MISTIC followed participants for a month, so durability, the role of the surrounding program and expectation are unresolved, and blinding is hard because the experience is unmistakable and long [1]. Whether the acute experience drives the later change, as the reported mystical-experience correlation suggests, is untested [18]. With so many binding targets it is unknown which matter for any outcome, and the growth-factor route has only rodent evidence behind it [1] [2]. Personal growth appears only as self-report: a few of 13 people interviewed about underground ibogaine use described it [31].

A 70-item Ibogaine Experience Scale was developed in 2025 from 499 clients of one private clinic in Tijuana, the clinic that also provided the MISTIC treatment [1]. It rates the subjective experience within 48 hours, and its factor of introspection and personal change is a rating of that experience and not a measured growth outcome. No equivalent instrument exists for growth outcomes in healthy people after ibogaine, and the scale has not been tested as a predictor of outcomes [32].

What Metanoia asks

Before anything else, is research in healthy adults possible at all? If it is, what cardiac screening, medication review and metabolic information would make it responsible? The answer may be that it is not, and an honest statement of that would itself be a result. Next, in a program like MISTIC, how much of the change belongs to the compound and how much to preparation, setting, group experience and expectation? The material also matters, because whole root bark, extracted ibogaine and semi-synthetic ibogaine are different things to record, which is what the first of Metanoia's four record-keeping standards is for. Last comes sourcing and communities, since Colorado's pilot law already treats benefit-sharing with Indigenous communities as a condition [27].

Any future human work by Metanoia would start from four record-keeping standards (what the material was, what exposure occurred, the surrounding context, and the outcome) and an outcome measure, still in development, addressed to the twelve dimensions of human potential the research asks about. None of that infrastructure has been built for this substance yet.

History and tradition

The tradition itself is grade 4, traditional and ethnographic, and reaches this page only through secondary descriptions. No primary ethnography was read for this page.

Iboga is used in Bwiti, an African religious tradition practiced in Gabon. Clinical papers describe it variously as religious, spiritual and healing ceremonies [1] and coming-of-age rituals [6], and the IUCN page says the roots are used in initiation ceremonies of groups such as Bwiti, citing a source from 1969 [29]. A study of the cultural importance of plants in Benin and Gabon (99 interviewed, 96 analysed) found iboga in the Gabon lists as a plant product, a sacred wood, and as a species. The paper does not describe ritual or use or report how often iboga was named, and its authors note that secrecy around Bwiti and low literacy, with 22 lists given orally or through a third person, may have led to an underestimate [30].

The modern history of ibogaine is told mostly in the addiction literature. One paper credits Howard Lotsof with discovering its anti-addictive properties in 1962 through self-experimentation, stated as fact in its introduction with a book-chapter citation and not independently checked [31], and a systematic review dates it to the 1960s [14].

The global IUCN Red List assessment of Tabernanthe iboga, read directly on 2 October 2026, lists it as Least Concern, assessed in 2018, with a stable population trend. The area of occupancy is only 408 square kilometres, and the threats section of the assessment page was empty when read [29]. CITES status was not checked, and no source read for this page measured harvest pressure in Gabon or the effect of demand on communities there, so those concerns remain plausible but unverified.

The book on all twelve dimensions   Back to the twelve dimensions