Classic psychedelic coadministration with lithium, but not lamotrigine, is associated with seizures: an analysis of online psychedelic experience reports

S. and colleagues situate their study in the context of renewed clinical and public interest in classic psychedelics (drugs acting primarily at the 5-HT2A receptor such as LSD and psilocybin) as potential treatments for psychiatric conditions. They note that people with bipolar disorder have been excluded from most modern psychedelic trials because of safety concerns (for example, precipitating mania or psychosis), but that the combination of (a) limited treatment options for bipolar depression, (b) growing public enthusiasm for psilocybin, and (c) the availability of psychedelics outside clinical settings raises the possibility that people prescribed mood stabilizers may self-administer psychedelics. The authors emphasise that little systematic information exists about interactions between classic psychedelics and commonly prescribed mood stabilizers such as lithium and lamotrigine. The paper therefore aims to identify signals of risk by analysing first- and second-person anecdotal ‘‘experience reports’’ posted on public online forums. The objective is not to establish causality but to assess whether reports available in public forums contain evidence of adverse interactions—particularly seizures—when classic psychedelics are coadministered with lithium or lamotrigine, to inform clinicians and guide future research and trial design.

The researchers performed a retrospective, descriptive analysis of publicly posted psychedelic experience reports from three online sources: Erowid.org (data accessed via Erowid's API on 7 August 2020), Shroomery.org (scraped on 21 August 2020 using R tools), and selected Reddit subreddits (accessed via the PRAW API on 22 August 2020). Experience reports were identified by searching for terms referring to classic psychedelics (LSD, psilocybin mushrooms, DMT, mescaline, peyote, San Pedro, ayahuasca) together with mood-stabiliser search terms (lithium, lamotrigine, valproic acid, carbamazepine, oxcarbazepine, including trade and alternative chemical names). Antipsychotics were not part of the initial search. Reports were filtered to reduce duplicate postings (duplicate usernames where identifiable) and to exclude accounts that did not contain first- or second-person descriptions of an actual psychedelic consumption event (this step was particularly important for Reddit posts). Demographic and clinical history data were generally not available in a systematic way and therefore were not collected. The authors excluded reports of psychedelic microdoses and reports indicating very small lithium dosages (e.g., <30 mg lithium orotate). Reports that mentioned both lithium and lamotrigine were set aside for separate reporting. All reports that met these criteria were manually coded by a research assistant (SMN) and reviewed by an author (NG). Coding captured whether other psychoactive medications were mentioned (e.g., antidepressants, benzodiazepines, stimulants), the psychedelic and mood stabiliser involved, and the reported clinical outcome. Outcomes were classified into six mutually exclusive categories: (1) seizure; (2) bad trip (negative/noctious experience excluding seizure); (3) intensified psychedelic effects (excluding seizures or bad trips); (4) decreased psychedelic effect; (5) no psychedelic effect; and (6) neutral (no reported effect of the mood stabiliser on the psychedelic experience). Discrepancies in coding were resolved by discussion. The analysis is descriptive (counts and percentages) of the final set of qualifying reports.

After applying the search and coding procedure, 96 experience reports met inclusion criteria for analysis and involved either lithium or lamotrigine; 4 additional reports that mentioned both lithium and lamotrigine were analysed separately. The analysed sample comprised 62 lithium reports and 34 lamotrigine reports. Lithium: Among the 62 reports involving lithium and a classic psychedelic, 29 of 62 (46.8%) were coded as involving a seizure. An additional 11 of 62 (17.7%) were coded as ‘‘bad trips’’ (negative experiences excluding seizures), 12 of 62 (19.4%) were coded as intensified psychedelic effects, and 5 of 62 (8.1%) were coded as neutral (lithium was not described to affect the psychedelic experience). By psychedelic type, of 55 lithium-plus-LSD reports, 27 (49.1%) involved seizures; of 6 lithium-plus-psilocybin reports, 2 (33.3%) involved seizures; the single lithium-plus-DMT report did not report a seizure. Eighteen lithium reports included mention of another psychoactive drug (most commonly a selective serotonin reuptake inhibitor); excluding those 18 reports did not eliminate the signal—54.5% of the remaining lithium reports still involved seizures. The authors also note that 39% of lithium reports included mention of emergency medical treatment (ambulance or hospital). Two additional Erowid reports of 2C-class psychedelics with lithium (not included in the primary analysis due to low counts) both described seizures. Lamotrigine: Of the 34 lamotrigine-plus-psychedelic reports, the majority (22 of 34; 64.7%) described lamotrigine as not impacting the psychedelic experience. Six of 34 (17.6%) reported decreased psychedelic effects, 3 of 34 (8.8%) reported no psychedelic effect at all, and 3 of 34 (8.8%) reported intensified effects. No lamotrigine reports were coded as bad trips, and no seizures were reported in the analysed lamotrigine set. Removing 12 lamotrigine reports that mentioned another psychoactive drug did not substantially change the pattern: in that subset, 73.1% were still judged not to affect the psychedelic experience. Lithium plus lamotrigine: Four reports involved both mood stabilisers with a psychedelic; 1 of these 4 (25%) involved a seizure (that report also documented concurrent use of the MAOI Syrian Rue), and the remaining 3 were described as intensified experiences. The authors present these findings descriptively; no inferential statistical testing is reported in the extracted text.

The authors interpret their findings as a signal—albeit from low-quality, self-reported data—that coadministration of classic psychedelics with lithium may carry a substantial seizure risk. They highlight that nearly half of the lithium-plus-psychedelic reports described seizures (47%) and that lithium reports were more likely than lamotrigine reports to be described as bad trips (18% versus 0%). The authors caution that experience reports from public forums are subject to multiple biases (selection bias, lack of clinical verification, missing information about doses and medical history, and possible duplicate reporting) and therefore do not establish causality. They note that lithium doses were rarely reported, so dose–response relationships cannot be examined, and that microdoses were explicitly excluded from analysis. Despite these caveats, the authors argue that the severity and frequency of the reported seizures and the proportion of reports requiring emergency medical attention (reported in 39% of lithium accounts) support a conservative clinical stance: they recommend that individuals taking lithium avoid non-microdose classic psychedelics. They also suggest that, for future clinical trials involving patients who continue mood stabilisers, lamotrigine may present a lower probability of adverse interaction based on the relative lack of signal in these reports, although they emphasise that the benignity of lamotrigine in these anecdotes does not prove safety. The authors discuss several mechanistic possibilities. These include (1) a pre-existing greater seizure propensity in people taking lithium (unmeasured in these reports); (2) increased systemic lithium concentrations (noted to elevate seizure risk) or increased psychedelic concentrations, though a pharmacokinetic interaction is considered unlikely because lithium is renally excreted whereas classic psychedelics undergo hepatic metabolism; (3) the possibility that some events reflected serotonin syndrome (some seizure reports included accompanying features that could be compatible with serotonin syndrome, such as diaphoresis, tachycardia, vomiting, mydriasis, agitation and confusion), though these signs overlap with expected psychedelic effects and cannot be reliably interpreted from anecdotal reports; and (4) a plausible pharmacodynamic interaction in which psychedelics' 5-HT2A agonism (increasing cortical excitability) combined with lithium-related effects or with psychedelics' reduction of tonic locus coeruleus activity (a region that can be seizure-protective) could synergistically lower the seizure threshold. The authors underline the limitations of anonymous online data and call for further research, but they conclude that the available anecdotal evidence is sufficient to consider lithium a contraindication to classic psychedelic use pending better data.

S. and colleagues conclude that nearly half (47%) of online psychedelic experience reports involving lithium and a non-microdose classic psychedelic described seizures, 18% described otherwise negative experiences, and 39% mentioned emergency medical treatment. By contrast, most lamotrigine-plus-psychedelic reports (64.7%) did not describe an impact on the psychedelic experience. The authors state that, despite the substantial limitations of anonymous online reports, lithium should presently be considered a contraindication to non-microdose classic psychedelic use.