With the aforementioned as preamble, it is noteworthy that a number of long-term sequelae have been attributed to the use of LSD. They include a thought disorder, subtle changes in the EEG, a decrease in higher cortical functions, abnormal Minnesota Multiphasic Personality Inventory results, an abnormality in color perception, a decrease of the serotonin catabolite 5-hydroxyindoleacetic acid (5-HIAA) in the CSF of psychotic subjects with prior LSD use, and psychosis. A ten-year follow-up study of medical LSD use in psychiatric outpatients and volunteers, however, failed to find evidence of lasting psychologic effects. The reported deviant response to the Reitan trail-making test among collegiate LSD users was not found among healthy adults.

In 1955 Elkes et al were among the first to suggest that LSD might have long-term effects in man. In 1958 Eisner and Cohen reported the spontaneous appearance of LSD phenomena at varying intervals following exposure, which were described by their patients "as relaxing and beneficial." Rosenthal presented evidence that visual hallucinations might last as long as five months from the time of last exposure. Cohen and Ditman reported a series of nine LSD users with prolonged effects from the drug lasting as long as two years.

Although to date there has been no systematic inventory of the types of visual disturbances experienced chronically in this population, Horowitz made one of the earliest efforts in this direction, reporting on 31 subjects with long-term LSD use who described three distinct visual types of flashbacks: perceptual distortions, heightened imagery, and recurrent unbidden images. In addition to perceptual changes, Shick and Smith described somatic and emotional flashbacks. Of special note is that although six reports proposed that such disturbances only appeared in subjects having multiple exposures, three studies failed to find statistically significant differences in such disturbances between heavy and light LSD users, and several observers found flashbacks after a single LSD exposure (D. X. Freedman, MD, written communication, Oct 13, 1981). One report suggested that visual disturbances and LSD use "seem to be nothing more than the association of two events bearing certain similarities." While the dose-response characteristics have been unclear in literature, reports on the incidence and severity of such visual disturbances also vary widely. Accordingly, the prevalence of flashbacks in populations of LSD users has varied from 15% to 77%.

Precipitants of such recurrences have not been studied systematically. A theory that such disturbances are linked to stress was not supported by a survey of LSD users before and after military combat. Though evidence shows that LSD interacts acutely with diverse regions of the CNS, and while the drug clearly appears in high concentration along visual pathways, no evidence has been presented that clarifies whether the drug operates at any CNS site on a long-term basis. Of special note is the association between LSD-type flashbacks and the use of marijuana. Six separate reports linked flashbacks to marijuana use, but in each report the subjects described an antecedent use of LSD that also explained the flashback. Further data from the survey of 720 servicemen in western Europe failed to find a single case of flashbacks in any subjects for whom hashish was the sole drug of consumption. In the same sample, however, 15 subjects were found for whom LSD flashbacks were precipitated by marijuana use. Another military sample of 2001 discovered that marijuana use had the highest, and only, significant association with the precipitation of LSD flashbacks among five classes of abused substances.

Finally, treatment data are contradictory. Moskowitz treated eight subjects for flashbacks successfully with haloperidol, Anderson and O'Malley found trifluoperazine hydrochloride useful in one case, but Schwarz reported that chlorpromazine given to two subjects during acute LSD experiences exacerbated visual symptoms. A report by Thurlow and Girvin offered evidence that an anticonvulsant, phenytoin, was helpful in a single case. The consensus of a number of clinicians, however, has been that neuroleptics are not the treatment of choice.

But what are the types of visual disturbances chronically experienced by these individuals? How long after the last use of LSD do such disturbances appear? What precipitates them? Do the disturbances correlate with other drugs of abuse? If so, which ones? And what about treatment? Which is effective? Can the phenomenon of the flashback teach us anything about which regions of the brain are involved in these hallucinatory phenomena? And most importantly, what is the basis of the great variety of responses to LSD in the medical literature? Some of these questions are approached in the following study, which focuses on the phenomenology of the LSD flashback, its precipitants, treatments, and clinical course.

Table 1. – Characteristics of the Study Sample

Drug histories were taken on each subject's life experience regarding hallucinogens, marijuana, narcotics, hypnosedatives, amphetamines, cocaine, and alcohol. A summary of this these data is shown in Table 2.

Table 2. – Lifetime Drug Abuse Other Than LSD

An 83-item questionnaire was constructed that incorporated the 16 visual disturbances most commonly reported in the first phase. The information fell into six categories: (1) psychiatric and social history; (2) psychiatric diagnoses and treatments; (3) visual disturbances, their causes, and treatments; (4) other flashback symptoms and treatments; (5) other drug history scored as continuous variables; and (6) a history of the subject's lifetime use of LSD in an effort to calculate a lifetime dose. Since "mescaline" had been previously found to be chemically assayed, such trips were included in the final tally for each subject. Life doses from each subject then were segregated into cells arranged at the following intervals: one to ten trips, 11 to 20, 21 to 30, 31 to 40, 41 to 50, 51 to 75, 76 to 100, 101 to 300, 301 to 500, and greater than 500 LSD exposures over a lifetime. This life dose served as an abscissa against which the dependent variables of visual disturbances and flashbacks could be graphed as ordinates. In addition, 22 subjects were referred for a battery of psychophysical tests that included kinetic periphery, color vision, flicker perimetry, dark adaptation, perception under glare, and a test of simultaneous color contrast, results of which are reported separately. Following the questionnaire, subjects also were interviewed in an unstructured fashion for approximately one hour, usually in the setting of the Acute Psychiatric Service.

Completed data were recorded for 89% of the sample. Incompletions were due largely to the clinical needs of the patients who required hospitalization, crisis intervention, or whose answers were excluded by virtue of prepossessing delusions or hallucinations. This biased the sample toward the healthier subjects, but also toward the more reliable ones.

Table 3 ranks the kinds of visual disturbances encountered by the sample. A core of visual disturbances, most commonly including geometric and imagistic pseudohallucinations, illusions of movement, trails, flashes of color, and the prolonged intensification of color, emerged from the data. Pearson's correlation between the presence of these visual disturbances and a clinical description of flashbacks by these subjects was .586 (P<.001).

Table 3. – Ranked Frequency of Visual Disturbances Associated With Prior LSD Use

Figure 1 illustrates the relationship between the total number of reported LSD exposure at ten dose levels and the percentage of flashbacks of subjects in each level with a clinical appearance of flashbacks. The hypothesis that there is a linear relationship between LSD life dose and flashbacks was rejected. What appears, instead, is a relationship between dose and response that has a peak at 15 exposures, a second peak at approximately 40 exposures, and a plateau thereafter. Table 4 illustrates Pearson's correlation coefficients against the total number of reported visual disturbances using data from LSD users and control subjects. The greatest correlation is for LSD, and the least for narcotics. Also, there was a weak but significant trend between visual disturbances and marijuana use. It is noteworthy that while LSD has the highest correlation with visual disturbances in this sample, coefficients for other drugs of abuse fall in the general order that one would expect regarding their respective abilities to stimulate visual perceptions.

Table 4. – Pearson's Correlation Coefficients Between Drugs and Visual Disturbances

* A two-tailed test was used to calculate P values.

Figure 2 illustrates the relative stability of visual disturbances in the user population over a five-year span. The hypothesis that all such disturbances disappear in time was rejected, since approximately half the user in each time interval from last drug exposure continued to report flashbacks at the time of the interview.

Table 5 itemizes 19 separately reported precipitants of visual disturbances following the use of LSD. The most common precipitant is the emergence into a dark environment in 16% of the users. Characteristically, a subject described entering a dwelling after being outdoors for a number of minutes or longer. Once indoors the subject described a variety of disturbances, the most common being geometric pseudohallucinations. Another common precipitant was intention in 14%. Subjects intentionally induced visual aberrations by staring at a blank wall or blackboard.

Table 5. – Ranked Frequency of Reported Precipitants of Visual Disturbances

Fourteen percent of the users also reported that marijuana could precipitate LSD-type flashbacks. A further examination of 15 control subjects with a history of heavy marijuana use ("more than once a day") revealed that none of the subjects ever reported experiencing any flashback phenomena, while seven of the 12 LSD users with the same marijuana history did report flashbacks (P=.003). This finding strengthens a reportedly unique form of drug-drug interaction between LSD and marijuana, though the evidence to date remains solely based on clinical histories.

It appears that LSD can be associated with a chronic, irreversible, or slowly reversible collection of visual disturbances in a sample of heavy LSD users that largely included psychiatric outpatients but also included a number of otherwise normal volunteers. Such disturbances most commonly reported were geometric pseudohallucinations, illusions of movement in the peripheral visual field, trailing phenomena, flashes of color, imagistic pseudohallucinations, and intensified colors for brief periods of time. Other visual symptoms may be present, but these occurred in 28% or less of the study sample.

Of note is that while 12.9% of the sample sought clinical help for flashbacks, 32.9% reported considerable numbers of visual symptoms (more than an SD beyond the control mean), usually of a mild nature, suggesting that the LSD flashback in this sample may represent a more pronounced species of a broader range of drug-related perceptual disorder. With this in mind, one many construct a working definition of an LSD flashback as the recapitulation of an LSD-like subjective experience, often but not always in the form of altered visual perceptions occurring weeks to years following an exposure to the drug. In the present sample such disturbances were invariably pseudohallucinations rather than perceived and believed hallucinated imagery.

While such visual disturbances can be striking, no generalization can be made about the population of LSD users at large. The sample was drawn in large measure from an acute psychiatric outpatient population and, though steps were taken to minimize a bias toward the selection of LSD casualties, selection could not utterly escape that bias. In addition, a retrospective study of flashbacks faces formidable problems in the reliability of self-reports; the kind and dose of drug abused; and the confounding of variables by such factors as intelligence, personality, and preexisting psychopathology. It is, however, the unique, consensual reportage that makes the foregoing observations appear to be valid.

The observation that emergence into a dark environment is one of the most common precipitants of these disturbances suggests that the neurophysiologic basis of flashbacks may be related to a failure of an inhibitory system of the visual pathway to function properly. Such a locking of visual circuitry into an "on" position following perception of a visual stimulus would explain such diverse complaints as trailing, color intensification, positive afterimages, phosphenes, and color confusions, each of which may represent a failure of the respective visual function to turn off the brain's response to the stimulus once the stimulus is gone. This would also explain the disparate reportage between positive afterimages (28.6%) and negative ones (15.7%). The latter are induced commonly by a depletion of visual pigment in retinal color receptors, whereas one mechanism for inducing the former may be mediated within the lateral geniculate nucleus (LGN) of the thalamus. If the LGN were a target organ for LSD toxicity, one would expect to find LSD-sensitive neurons there, which, in fact, has been reported.

It has been suggested previously that LSD has the ability to disrupt defensive and perceptual constancies by which we order our lives. Such habits of perception have been classified into perceptual "types" reflecting character traits. While this concept remains controversial, LSD flashbacks may be modulated by perceptual style and psyhobiologic maturity. Flashback phenomena, then, appear to operate at multiple levels of organization of the CNS and may be conceptualized productively by models of sensitization and habituation, neurobehavioral and cellular processes, state-dependent learning, cortical arousal, information processing, and psychodynamics.

The observations of a variegate sensitivity to LSD are confirmed by these data. Figure 1 suggests that the study group is composed of three distinct populations of users: a group that reports flashbacks at 15 exposures and appears to be extremely sensitive to the drug; a second group with a peak at 40 to 50 lifetime exposures; and a third group that appears relatively resistant to LSD even at high exposures. The most intriguing hypothesis this observation generates is that in the study sample there exists a genetic basis for variable sensitivities to LSD. Such a sensitivity to the aftereffects of LSD has been proposed previously, and a familial history of major mental disorders reportedly is linked to the frequency of chronic psychotic disturbances after LSD use.