A randomised placebo-controlled study of the effects of lysergic acid diethylamide microdosing (15 μg) on pain perception in healthy volunteers

Classic psychedelics act primarily at serotonergic receptors and have attracted renewed interest for potential therapeutic uses beyond psychiatry, including analgesia. Earlier observational and small experimental studies report pain relief after psychedelic use in conditions such as cluster headache, phantom limb pain and migraine, and one controlled healthy-volunteer study found increased tolerance to experimentally induced pain after a low (20 μg) dose of LSD. Microdosing—regular ingestion of sub-perceptual or mildly perceptual doses, often about one tenth of a full dose—has become popular and is reported anecdotally to reduce pain, but no controlled study had evaluated a repeated microdosing schedule for analgesia prior to this work. Cavarra and colleagues set out to test whether a microdosing regimen of 15 μg LSD, given twice weekly for 2 weeks (four administrations total), would increase pain tolerance and reduce subjective painfulness, unpleasantness and stress in healthy volunteers exposed to an experimental pain task. The study also examined physiological (blood pressure, heart rate) and subjective drug-effect measures, and whether these related to any analgesic effects; 15 μg was selected to minimise alterations in consciousness while aiming to retain potential analgesic action, given prior findings at 20 μg but not 10 μg.

The trial used a randomised, double-blind, placebo-controlled, parallel-group design. Two groups received either 15 μg LSD base in 0.6 mL 96% ethanol or an ethanol placebo, administered orally at 10:00 on Mondays and Thursdays for 2 consecutive weeks (four doses). Pain testing occurred on the first and fourth dosing days at 1 h and 5 h post-administration, with a follow-up test day in week 4 that replicated the procedure but without drug administration. Healthy volunteers aged 18–65 were recruited and screened; inclusion required English proficiency and body mass index 18–28. Key exclusions included current psychotropic medication, recent psychedelic exposure (past 3 months), substance use during the study, pregnancy/lactation, psychiatric history or family history of psychosis. Participants were urine- and alcohol-tested on visit days. The study was approved by the relevant ethics committee and registered in Dutch trial registries. The primary outcome was pain tolerance measured by the Cold Pressor Task (CPT): immersion of the dominant hand in 2.5–3°C water with a maximum duration capped at 180 s. Participants were unaware of the cap; time of immersion was used as the primary endpoint. Secondary outcomes were VAS ratings (0–100 mm) of painfulness, unpleasantness and stress assessed immediately after each CPT. Additional measures included a repeated hourly VAS for subjective drug intensity on dosing days, vital signs (blood pressure, heart rate) at baseline, 2 h and 6 h on CPT days, and plasma LSD and O‑H‑LSD concentrations from blood sampled 2 h after dosing. Ultrahigh-performance liquid chromatography–tandem mass spectrometry (UHPLC-MS/MS) was used for drug assays. Sample size planning used G*Power for a repeated-measures ANOVA interaction (treatment × time) aiming for 80% power at alpha = 0.05 with effect size f = 0.25, yielding a required N = 24; however the investigators used Linear Mixed Models (LMM) for analysis to accommodate missing data without listwise deletion. Randomisation was block-based (blocks of four) and prepared by an experimenter not in contact with participants; blinding was maintained for participants and study staff. Statistical analysis employed LMMs with fixed effects for Treatment, Test day (baseline, treatment day 1, treatment day 4, follow-up), Time (1 h, 5 h), their interactions, and a random subject intercept with an unstructured covariance. Additional LMMs tested potential covariation by adding vital signs or subjective under-the-influence scores as covariates. Non-parametric Spearman correlations and canonical correlation analyses were used to probe relations between physiological variables and pain measures. An exploratory post-hoc analysis excluded participants exhibiting ceiling performance on the CPT (hand submerged for the full 180 s at baseline or in ≥3 of 4 CPTs) to mitigate possible masking of treatment effects.

Recruitment ran from January 2020 to February 2022. Of 53 recruited volunteers, three dropped out and two were excluded for missing CPT data, leaving 48 participants (24 female, 24 male; mean age 36.9, SD = 16.6) analysed. Lifetime psychedelic use was reported by 28 (58.3%) participants; 14 (29%) had used psychedelics in the past year. Other substance use frequencies were reported in the sample. Plasma assays confirmed the presence of LSD: mean (SD) plasma LSD at 2 h was 302 (105) pg/mL after dose 1 and 326 (117) pg/mL after dose 4; O‑H‑LSD means were 17 (7.2) pg/mL and 17 (4.9) pg/mL respectively. Primary analyses using LMM found no significant Treatment × Test day interaction on pain tolerance (primary outcome; p = .198) and no Treatment × Test day × Time interaction (p = .959). Graphical means were reported but did not indicate treatment effects in the whole sample. Vital-sign analyses showed changes within normal clinical ranges; LMM detected a significant Treatment × Test day interaction for systolic blood pressure (F(5, 234) = 2.7; p = .022). In the LSD group, pain tolerance correlated positively with systolic BP (Spearman ρ = .305, p = .000) and diastolic BP (ρ = .297, p = .000). Canonical correlation analysis between BP/HR and the set of pain measures was significant (F(44, 343) = 1.94, p = .001) with canonical r = .41, explaining 16% of variance, indicating that higher BP/HR associated with greater analgesia. Subjective "under the influence" ratings increased on dosing days: LMM showed a significant Treatment × Test day interaction (F(3, 95) = 3.43, p = .02). An LMM restricted to treatment days 1 and 4 yielded a main effect of treatment (F(1, 48) = 4.125, p = .048), interpreted by the investigators as indicating no tolerance across the four doses. These subjective scores did not correlate with pain measures, and adding them as covariates did not alter pain-model interactions. A substantial ceiling effect in the CPT was observed: 42% of participants reached the 180 s maximum. After excluding participants who were pain-tolerant at baseline (n = 18) or during ≥3 of 4 CPTs on treatment days (n = 2), a subsample of n = 28 remained (14 placebo, 14 LSD). In this subsample LMM showed a significant Treatment × Test day interaction on pain tolerance (F(3, 55) = 5.196, p = .003). Follow-up contrasts suggested the effect was mainly driven by a difference at follow-up (β = 29.47, p = .013). The extracted text indicates that other contrasts were generally nonsignificant and that no clear Treatment × Test day or Treatment × Test day × Time effects were found elsewhere; the sentence in the extraction is incomplete regarding those latter results. Overall, no serious adverse events were reported and vital-sign changes remained within normal limits.

Cavarra and colleagues interpret the primary findings as providing no evidence that a repeated 15 μg LSD microdosing regimen produces analgesia in a mixed healthy-volunteer sample tested with the Cold Pressor Task. The authors note that exploratory post-hoc analyses in a smaller subsample that excluded pain-tolerant participants suggested a marginal analgesic effect on pain tolerance and unpleasantness, but this effect appeared after the first dose and was not consistently present on the fourth dosing day. Several explanatory possibilities are advanced. First, the authors propose that 15 μg may be below a threshold dose required for consistent analgesia, citing prior work showing analgesia at 20 μg but not 10 μg. Second, a ceiling effect in the CPT limited sensitivity in the full sample; a high proportion of participants reached the 180 s cap, reducing the ability to detect group differences. Third, the lack of sustained effects by the fourth dose could reflect either development of tolerance to a mild analgesic action or fluctuation around a threshold effect in some individuals. Physiological findings showed modest, clinically normal increases in systolic blood pressure in the LSD group, and blood-pressure/heart-rate measures correlated positively with pain tolerance. However, inclusion of blood pressure as a covariate did not remove the treatment-by-day interaction in the subsample analysis, so the investigators conclude that cardiovascular changes do not fully mediate any observed marginal analgesic effect. Subjective drug-intensity ratings were slightly elevated on dosing days but did not correlate with pain outcomes and did not mediate the treatment effects in models. The authors acknowledge important limitations: the high rate of CPT ceiling performance impaired sensitivity and reduced effective sample size after exclusions; the exploratory subsample was small and underpowered for many contrasts; and the study examined healthy volunteers rather than clinical pain populations. They recommend future work using a wider range of pain paradigms to avoid ceiling effects, larger samples, higher doses (for example >20 μg) to establish dose–response, assessment of tolerance with repeated dosing, and ultimately trials in patient populations to evaluate clinical efficacy, neuroplasticity effects, specificity across pain conditions, and combination with psychological interventions such as mindfulness or psychotherapy.

In this randomised, placebo-controlled study, repeated administration of 15 μg LSD twice weekly for 2 weeks produced no evidence of analgesic effects on experimentally induced cold pain in the full healthy-volunteer sample. Post-hoc analyses excluding participants who showed ceiling performance on the Cold Pressor Task suggested a marginal, short-lived analgesic signal after the first dose in a small subsample, but this was not robust across dosing days. The investigators conclude that 15 μg is unlikely to produce a reliable analgesic effect and recommend further research with larger, better-selected samples, alternative pain paradigms, higher doses and clinical populations to clarify LSD's potential in pain treatment.