Dose-response study of N,N-dimethyltryptamine in humans. I. Neuroendocrine, autonomic, and cardiovascular effects

Earlier human and animal research on classical hallucinogens (for example LSD, psilocybin, mescaline) implicated serotonergic mechanisms but varied widely in subject selection, administration methods, and outcome measurement. N,N-dimethyltryptamine (DMT) is an indole hallucinogen found in plants and animals and is notable for its very rapid onset and short duration. Because existing human data were limited and heterogeneous, the investigators sought a more systematic characterisation of DMT's subjective effects, linked to dose, in a cohort of experienced hallucinogen users. Strassman and colleagues aimed to map the psychological effects of graded intravenous DMT doses and to pilot a new instrument, the Hallucinogen Rating Scale (HRS), developed from interviews with experienced DMT users. The study therefore set out to (1) describe dose-related subjective phenomena across clinically derived symptom clusters, and (2) evaluate whether the HRS and statistical clustering or factor analysis could quantify and distinguish dose effects for future psychopharmacological comparison studies.

Participants were recruited by word of mouth and screened to exclude current Axis I disorders (except one subject with an adjustment disorder), ongoing medical illness, and long-term medication use. One participant developed a recurrence of major depression during the protocol and was withdrawn; the analysed sample comprised 11 experienced hallucinogen users (10 men, 1 woman) with prior hallucinogen exposures ranging from six to hundreds. The investigators also classified subjects by prior MDMA (“Ecstasy”) exposure into high (≥5 uses, n=6) and low/no exposure (≤2 uses, n=5) groups to explore possible modulatory effects. The design combined an initial nonblind phase (each subject received two nonblind intravenous infusions of DMT at 0.04 and 0.4 mg/kg) with a double-blind, randomised, saline placebo-controlled crossover phase. Double-blind treatments were saline and IV DMT fumarate at 0.05, 0.1, 0.2, and 0.4 mg/kg. Doses were administered as a 30-second IV infusion followed by a 15-second saline flush. Sessions were separated by at least 1 week for men and 1 month for the single female subject, studied in the early follicular phase. Measures relevant to this paper focused on subjective effects. The Hallucinogen Rating Scale (HRS) was drafted from interviews of experienced DMT users describing smoked free-base DMT and refined during the study; the final version used in the double-blind phase contained 126 items scored 0–4 (0 = not at all; 4 = extremely). The HRS was administered after physiological sampling and removal of IV lines (about 30 minutes post-infusion). Physiological monitoring included rectal temperature and blood sampling until 60 minutes post-injection; autonomic and cardiovascular measures were reported in a companion article. Statistical analysis used PC-SAS Version 6. Screening for dose effects employed one-way ANOVA (P<.05) on individual HRS items; items meeting this threshold were retained. Items were grouped in two ways: a priori clinical ‘‘clusters’’ (Somaesthesia, Affect, Perception, Cognition, Volition, Intensity) and principal components factor analysis (VARIMAX rotation) specifying six factors to parallel the clusters. Main inferential tests were one-way repeated measures ANOVA across dose on cluster/factor scores. A two-way repeated measures ANOVA, with repeated factor dose and between-subjects MDMA-history class, tested whether prior MDMA exposure influenced responses.

Data analysed were from 11 subjects. The principal behavioural finding was a clear dose-response: 0.2 and 0.4 mg/kg IV DMT produced near-instantaneous visual hallucinations, bodily dissociation, and marked shifts in mood that supplanted ordinary mental content; auditory phenomena occurred in about half the subjects. Effects emerged rapidly (often during or immediately after the 30-second infusion) and resolved quickly, with time course paralleling DMT blood levels described in the companion report. Lower doses (0.05 and 0.1 mg/kg) were not hallucinogenic and produced primarily somaesthetic and affective changes; the 0.1 mg/kg dose in particular produced uncomfortable physical tension and was reported as the least pleasurable. Qualitatively, the 0.4 mg/kg dose — particularly when administered nonblind as the first high dose — produced an overwhelming ‘‘rush’’ described variously as intense, rapidly developing, and briefly anxiety-provoking (some subjects used metaphors such as a ‘‘freight train’’). This rush often progressed to loss of bodily awareness, dissociation, and an alternation or coexistence of fear and euphoria. Visual imagery predominated and was vivid across eyes-open and eyes-closed conditions; reports included formed images (figures, objects), complex geometric and kaleidoscopic patterns, and greatly intensified colours. Auditory effects were generally unformed sounds (high-pitched whining, chattering), while somaesthetic reports included chest constriction, sensations of flying or falling, thermal shifts, and occasional pleasurable genital sensations without orgasm. Cognitively, subjects retained an observing ego in many cases but experienced altered reality testing and reports of ‘‘other intelligences’’ or beings. Volitional control was markedly reduced at the highest doses. On the HRS, 75 of 126 items showed a significant dose effect by one-way ANOVA and were retained for further analysis. Cluster-level analyses showed that five of six clinical clusters discriminated placebo from the lowest DMT dose (0.05 mg/kg). Two clusters (Intensity and Cognition) discriminated across all doses. The statistically separable ‘‘hallucinogenic breakpoint’’ lay between 0.1 and 0.2 mg/kg for four clusters (Intensity, Affect, Perception, Cognition). Comparatively, principal components factors were more sensitive at distinguishing the hallucinogenic 0.2 mg/kg from the non-hallucinogenic 0.1 mg/kg dose (six of six factors versus four of six clinical clusters). A two-way repeated measures ANOVA found no significant effect of prior MDMA exposure on HRS cluster or factor responses across doses.

The investigators interpret their findings as demonstrating a steep dose-response for IV DMT in experienced hallucinogen users, with 0.2 and 0.4 mg/kg producing rapid, intense visual phenomena, bodily dissociation, and large affective shifts, while lower doses produced mostly somaesthetic and mood effects. They note that subjective descriptions aligned with prior intramuscular and field reports of DMT, and that the rapid onset and brief duration are consistent with pharmacokinetic and animal data cited by the authors. The HRS showed preliminary utility in quantifying dose-related subjective effects, and both clinical clustering and principal components analysis provided better discrimination of dose effects than the biological (neuroendocrine, cardiovascular, autonomic) measures reported in the companion article. Strassman and colleagues discuss possible serotonergic mechanisms underlying DMT effects, mentioning putative roles for 5-HT2 and 5-HT1A receptor subtypes and noting that differential activation of receptor subtypes might help explain qualitative differences at lower versus higher doses. They also link rapid onset and short duration to DMT's transport into cortex and rapid metabolism by monoamine oxidase. Comparisons with LSD emphasise differences in onset and time course rather than in quality of experience, although cross-tolerance data are limited. Key limitations acknowledged by the authors include the preliminary nature of the HRS (it was derived from and tested in experienced DMT users and requires independent validation), the sample characteristics (small size and selection of experienced, motivated volunteers), and the influence of ‘‘set’’ and ‘‘setting’’ on hallucinogen effects. The authors also note that the HRS in this report assessed clusters quantitatively but did not parse affect into qualitative subtypes (for example, separating euphoria from anxiety) and that generalisability to other hallucinogens or to subjects without prior hallucinogen experience remains to be established. They recommend further work using receptor-specific manipulations, broader subject samples (including clinical populations), and validation of the HRS against independent data.