Psychedelic treatment of functional neurological disorder: a systematic review

Functional neurological disorder (FND) comprises motor or sensory symptoms that are incompatible with recognised neurological disease, and it is a common cause of neurological disability and clinic referral. Existing evidence-based treatments are limited: psychological therapy (notably cognitive–behavioural approaches) and physiotherapy have some support, while other interventions such as brief group psychoeducation, repetitive transcranial magnetic stimulation and multidisciplinary rehabilitation have emerging evidence. Pharmacological management is generally considered inappropriate, and prognosis can be poor, which together underline the need for novel therapeutic options. Interest in psychedelic compounds (for example LSD and psilocybin) has re-emerged over the past two decades because of their documented effects on neural networks involved in self-related processing and promising early findings in a range of neuropsychiatric conditions. Given contemporary theories that FND may arise from aberrant somatic self-representation, Butler and colleagues set out to systematically review the historical clinical literature on psychedelic therapy in patients diagnosed with ‘‘conversion’’ or ‘‘hysteria’’ syndromes to determine whether further clinical research in FND could be justified. Because legal restrictions after the late 1960s curtailed psychedelic research, the review anticipated finding only pre‑prohibition studies.

Butler and colleagues searched PsycINFO and MEDLINE for the period 1940–2000 using search terms encompassing classic psychedelics (LSD, psilocybin, mescaline, ayahuasca) and diagnostic labels in the historical FND spectrum (hysterical, dissociative, conversion, psychogenic, functional). The Multidisciplinary Association for Psychedelic Studies (MAPS) bibliography was also hand-searched, and reference lists of identified articles were examined. Initial screening of titles and abstracts was performed by two reviewers, with discrepancies resolved by consensus, followed by full-text review to confirm eligibility. The authors registered their review protocol on PROSPERO. Inclusion required medically supervised administration of a psychedelic compound, with or without psychotherapy, in patients with FND and presentation of outcome data; papers unavailable in English or French, reviews and duplicate reports were excluded. The extracted studies were almost exclusively case reports or case series, so methodological quality was assessed using a recent tool devised for case reports/series (four domains: selection, ascertainment, causality and reporting). Two reviewers rated studies and reached consensus on overall quality (low, moderate, high). Outcome assessment in the review was pragmatic and categorical: each patient was classified from study descriptions as ‘‘not improved’’, ‘‘improved’’ or ‘‘recovered’’. Adverse events were recorded qualitatively as a second primary outcome. Because the primary literature was heterogeneous and largely uncontrolled with inconsistent outcome metrics, no pooled effect size or formal meta-analysis was undertaken.

The search yielded 735 records (705 after duplicate removal); after screening, nine studies published between 1954 and 1967 met inclusion criteria and were analysed for outcome. These nine reports encompassed an aggregate of 26 patients with diagnoses described at the time as ‘‘conversion’’ or ‘‘hysteria’’. Sample sizes per report ranged from single case reports to small series (1–6 patients). In most reports psychedelic administration was combined with psychotherapy in what was then termed psycholytic therapy (low, repeated doses accompanying psychotherapeutic work), whereas some studies used larger, less frequent doses in a ‘‘psychedelic’’ therapy paradigm. LSD was the most commonly used compound; psilocybin, mescaline and LSD analogues were also reported. Dosing and route varied widely: reported oral or intramuscular LSD doses ranged from 25 µg to 2000 µg, and psilocybin doses ranged from 3 to 15 mg where specified. The number of therapeutic sessions reported varied from a single session to 26 sessions, and descriptions of psychotherapeutic technique were inconsistent and often sparse. Across the 26 patients, 69% (n = 18) were judged by the original study authors to have made at least some recovery and 23% (n = 6) were described as having recovered completely, based on heterogeneous and subjective clinician-rated criteria. Adverse events were variably reported; most described effects were mild and transient (for example nausea, transient anxiety, somatic sensations), but more serious reactions occurred in some cases, including depersonalisation and a nonepileptic seizure following intramuscular psilocybin in one report. There was at least one account of a patient terminating treatment because of a distressing experience. Several individual reports of note were summarised by the reviewers: Duché (1961) described a 17-year-old girl who experienced a nonepileptic seizure after psilocybin but subsequently showed symptom resolution sustained at 6 months; Heyder (1963) reported a patient who regained functional use of an immobile arm after three sessions of 300 µg LSD; Whitaker (1964) reported a study including a control group and five ‘‘conversion’’ cases, with mixed outcomes among those five; Baker (1967) described three refractory ‘‘conversion’’ patients within a larger series, two of whom were judged ‘‘some better’’ and one ‘‘much better’’. A placebo‑controlled study by Soskin (1973) was identified but not included in the main analysis because diagnostic and patient‑level details were not reported; that study showed a pre–post improvement in conversion scores in the LSD arm but no significant difference versus control. The methodological quality of all included studies was rated poor by the reviewers, with frequent absence of control groups, subjective outcome ascertainment and incomplete reporting.

Butler and colleagues interpret the historical literature as indicating that, in the pre‑prohibition era, adjunctive psychedelic psychotherapy was judged by treating clinicians to produce clinician‑rated improvements in a majority of reported FND cases, but that the evidence is of very low quality and cannot establish efficacy. Small sample sizes, heterogeneous and often poorly described psychotherapeutic approaches, wide variation in dosing and schedules, and subjective outcome measures make it impossible to disentangle which components of the multimodal interventions (drug, psychotherapy, therapist attention) accounted for any observed gains. The reviewers situate these findings within contemporary mechanistic models of FND. Modern theories propose that symptoms may arise from aberrant somatic self‑representation and excessive top‑down prediction relative to sensory evidence, and neuroimaging studies implicate altered functional connectivity within the default mode network (DMN), a system linked to self‑related processing. Psychedelics act as 5‑HT2A receptor agonists and have been shown in imaging studies to reduce functional connectivity in networks such as the DMN, an effect that correlates with subjective ‘‘ego‑dissolution’’. The authors therefore note a plausible mechanistic rationale for investigating psychedelics in FND, given their capacity to perturb networks implicated in self‑representation. Safety considerations are discussed cautiously. The authors report that psychedelics are physiologically non‑toxic, largely non‑addictive and have high estimated toxicity ratios relative to many other psychoactive drugs, with low rates of serious adverse events in modern controlled trials. Nonetheless, historical reports document psychological distress, depersonalisation and at least one nonepileptic seizure, and the reviewers stress that FND patients may be particularly sensitive to drug‑induced somatic and motor phenomena that could theoretically worsen symptoms. They advise that safety protocols used in contemporary Phase I/II psychedelic trials would be important to apply in any future work with FND populations. Key limitations of the review are acknowledged: all included studies predate modern diagnostic systems and standardised outcome measures, raising concerns about diagnostic heterogeneity and misclassification; reporting was often anecdotal and incomplete; selection bias was probable, and follow‑up data were generally lacking. The heterogeneity of compounds, doses and psychotherapeutic approaches, together with the absence of control groups, prevented meta‑analysis and precluded conclusions about efficacy. Despite these limitations, the authors conclude that the historical data are sufficiently suggestive to warrant carefully designed feasibility, acceptability and safety studies—preferably using psilocybin with modern safeguards—before any efficacy trials in FND are attempted. They recommend preliminary patient liaison and acceptability work to inform future trial design.

The authors summarise that pre‑prohibition literature describing psychedelic psychotherapy in ‘‘conversion’’/FND populations indicates that such treatment was generally tolerated and, by subjective clinician judgement, associated with improvements in some patients. Given the poor methodological quality of historical studies, no firm conclusions about efficacy can be drawn. Butler and colleagues suggest that further investigations of feasibility and safety—particularly using psilocybin in contemporary, well‑controlled protocols—represent a potential future research direction, and they recommend preliminary work on patient acceptability before clinical trials in FND are undertaken.