Hallucinogen persisting perception disorder: what do we know after 50 years?

Reports of 'flashbacks' after hallucinogen use have appeared in the scientific and popular literature for many decades. Standardised diagnostic criteria were not available until the publication of DSM-III-R in 1986 and were slightly revised in DSM-IV as Hallucinogen Persisting Perception Disorder (HPPD). DSM-IV requires re-experiencing perceptual symptoms that occurred during intoxication, clinically significant distress or impairment, and exclusion of medical or other psychiatric causes. Halpern and Pope set out to review quantitative studies of post-hallucinogen perceptual disturbances to determine what is reliably known about HPPD. Specifically, they aimed to locate studies reporting groups of cases, assess how well these reports map onto modern DSM-IV criteria, and summarise evidence on epidemiology, aetiology, and treatment.

The investigators conducted a literature survey using MEDLINE (1966 to present) and reference lists from retrieved papers. Inclusion required studies to present at least eight cases of individuals who had ingested hallucinogens and to include some quantitative assessment of perceptual phenomena resembling those of hallucinogen intoxication. For each eligible study they attempted to extract information on the hallucinogens used, phenomenology of the reported 'flashbacks', comorbid psychiatric or medical conditions, and any treatments given. In applying DSM-IV standards to older studies, the authors were somewhat more permissive than DSM-IV in one respect: they included cases even when explicit evidence of distress or impairment was not reported. Screening began with 85 potentially relevant publications; 11 were excluded as reviews/commentaries, 25 for having fewer than eight cases, ten for insufficient evidence of hallucinogen exposure, and 12 for describing visual or EEG changes not clearly occurring after intoxication. This left 20 qualifying studies reported across 29 publications for detailed analysis.

Twenty studies met the authors' selection criteria but showed marked heterogeneity in design, methods, and case definition. Most reports predated operational diagnostic criteria for HPPD, and many provided insufficient detail to judge whether cases would meet current DSM-IV standards. Definitions of 'flashback' varied widely across studies, with some including predominantly perceptual phenomena and others grouping together non-perceptual psychiatric symptoms such as panic, depersonalisation, mood change, or transcendental experiences. Reported prevalence of perceptual disturbances varied greatly across settings. Large investigator questionnaires and series of research or therapeutic LSD administrations often found very few persistent perceptual disturbances (for example, one survey of investigators covering some 5,000 subjects reported only four fleeting afterimages after mescaline and none after LSD). By contrast, some community and clinical samples reported much higher rates: examples include 32% of 34 LSD-related psychiatric admissions at one hospital, 26% of 31 Haight-Ashbury community subjects, 55% of 20 community hallucinogen users recruited by advertisement, 77% of 65 hallucinogen-using patients in a Norwegian hospital sample, and 28% of 235 LSD users in a chain-referral study. Methodological caveats are common: many subjects continued to use other drugs, some reported flashbacks despite no hallucinogen use, and several samples included people with pre-existing or concurrent psychiatric disorders. Longitudinal and better-screened samples tended to report lower rates consistent with strict HPPD. In a 10-year follow-up of 247 subjects who had received LSD in therapeutic or research contexts, 36 (15%) reported uncontrolled LSD-like experiences after use but only five (2%) described major perceptual changes suggestive of HPPD. In a rigorously selected series, one group reported 44 individuals who fulfilled DSM-III-R criteria for HPPD; these subjects reported a mean of 8.1 different visual disturbances, a mean symptom duration of about 9 years, and a median of 16 LSD exposures. Neurophysiological findings in that sample included EEG changes reported as alpha acceleration and shortened flash visual evoked potential latency (visual evoked potential latency refers to the time between a visual stimulus and the brain's electrical response). Triggers for recurrent perceptual phenomena were often reported: entering dark environments and intentional recall were common, and many subjects reported that use of other drugs—particularly cannabis, but also alcohol and stimulants—could precipitate or worsen episodes. Several case reports and small series suggested that some subjects experienced symptom exacerbation with phenothiazines, risperidone, or selective serotonin reuptake inhibitors, whereas open reports described apparent benefit from haloperidol, anticonvulsants such as carbamazepine, benzodiazepines, clonidine, sunglasses, psychotherapy, or behavioural approaches. One small open study of eight polysubstance users meeting DSM-IV criteria reported improvement on clonidine over two months, but detailed symptom data were not provided. No randomised controlled trials of pharmacological treatments for HPPD were identified. Across studies it was frequently difficult to exclude alternative explanations for persistent perceptual disturbances: active or prior use of other substances, neurological disorders (including later-diagnosed temporal lobe epilepsy in at least one subject), primary psychotic or mood disorders, malingering, and anxiety disorders were inconsistently assessed or reported. The extracted data therefore do not permit a reliable estimate of the population prevalence of strict HPPD; evidence suggests it is uncommon in well-screened research or therapeutic cohorts and appears most frequently in some illicit-user samples, but selection and reporting biases limit interpretation.

Halpern and Pope emphasise that the literature on 'flashbacks' and HPPD is sparse and methodologically heterogeneous, limiting firm conclusions. A central problem is the multiplicity of definitions for 'flashback', which has led to widely divergent prevalence estimates and conflation of perceptual phenomena with broader psychiatric events. When attention is restricted to the specific visual and perceptual symptoms enumerated in DSM-IV, the data indicate that a subset of individuals—most often after illicit LSD use—experience persistent perceptual disturbances that can last months or years and cause substantial morbidity. The authors outline three broad explanatory constructs drawn from the literature. First, some reported phenomena may reflect heightened awareness of normal visual experiences rather than a pathological syndrome. Second, a proportion of cases could represent intrusive, emotionally charged memories of intense drug experiences rather than a primary perceptual disorder. Third, a core of strict HPPD cases may reflect an underlying neurological or psychiatric diathesis; proposed mechanisms include disinhibition of visual processing related to serotonergic receptor changes, and interaction between hallucinogen exposure and other biological cofactors. The reviewers note puzzling inconsistencies, such as the lack of a clear relationship between cumulative hallucinogen exposure and risk of persistent symptoms. Treatment evidence is anecdotal and mixed. Some patients reportedly improve with non-pharmacological measures or various medications, but other agents have been associated with symptom worsening. Importantly, no randomised controlled trials address HPPD treatment. The authors therefore call for well-designed studies: large-scale screening of hallucinogen users, application of rigorous DSM-IV-based case definitions, careful exclusion of alternative medical or psychiatric causes, longitudinal follow-up, and controlled treatment trials to clarify prevalence, aetiology, and effective interventions.