Hallucinogen Persisting Perception Disorder: Etiology, Clinical Features, and Therapeutic Perspectives

Hallucinogen Persisting Perception Disorder (HPPD) is a rare, and therefore, poorly understood condition linked to hallucinogenic drugs consumption. The prevalence of this disorder is low; the condition is more often diagnosed in individuals with a history of previous psychological issues or substance misuse, but it can arise in anyone, even after a single exposure to triggering drugs. The aims of the present study are to review all the original studies about HPPD in order to evaluate the following: (1) the possible suggested etiologies; (2) the possible hallucinogens involved in HPPD induction; (3) the clinical features of both HPPD I and II; (4) the possible psychiatric comorbidities; and (5) the available and potential therapeutic strategies. We searched PubMed to identify original studies about psychedelics and Hallucinogen Persisting Perception Disorder (HPPD). Our research yielded a total of 45 papers, which have been analyzed and tabled to provide readers with the most updated and comprehensive literature review about the clinical features and treatment options for HPPD.


Introduction
Hallucinogens represent an enormous group of natural and synthetic agents [1,2]. The core features of hallucinogens include their being empathogenic and being able to induce alterations of consciousness, cognition, emotions, and perception. Their main characteristic is to profoundly affect a person's inner processes and the perception of the surrounding world. The perceptual distortions are mainly visual, as in the vast majority of induced psychoses [3][4][5]. The hallucinogenic properties of many natural products were known for thousands of years: popular healers, "brujos", and shamans used these substances in ancient times for medical, religious, spiritual, ritual, divination, and magical purposes. Nevertheless, the attention of western culture was drawn to psychedelics only at the beginning of the 20th century, but the turning point is considered to be 1938, the year in which the lysergic acid diethylamide, better known as LSD, was synthesized by Albert Hofmann. In the

Materials and Methods
We searched PubMed to identify original studies about psychedelics and Hallucinogen Persisting Perception Disorder (HPPD). The following search terms were used: "Hallucinogen Persisting Perception Disorder" OR "Hallucinogen Persisting Perceptual Disorder". The search was conducted on 15 September 2017 and yielded 46 records. We included all original articles (open-label or double-blind trials, prospective or retrospective observational studies, and case reports) written in English. We included all studies describing perceptual distortions in patients with a previous history of substance consumption. Reviews, commentaries, letters to the editor, and studies enrolling adolescents were excluded. All the authors agreed on the inclusion and exclusion criteria. We excluded 17 records by reading the titles and abstracts. By reading the full texts of the 29 remaining articles, we found 25 papers meeting our inclusion/exclusion criteria, and we, therefore, included them in the qualitative synthesis ( Figure 1).

Materials and Methods
We searched PubMed to identify original studies about psychedelics and Hallucinogen Persisting Perception Disorder (HPPD). The following search terms were used: "Hallucinogen Persisting Perception Disorder" OR "Hallucinogen Persisting Perceptual Disorder". The search was conducted on 15 September 2017 and yielded 46 records. We included all original articles (open-label or double-blind trials, prospective or retrospective observational studies, and case reports) written in English. We included all studies describing perceptual distortions in patients with a previous history of substance consumption. Reviews, commentaries, letters to the editor, and studies enrolling adolescents were excluded. All the authors agreed on the inclusion and exclusion criteria. We excluded 17 records by reading the titles and abstracts. By reading the full texts of the 29 remaining articles, we found 25 papers meeting our inclusion/exclusion criteria, and we, therefore, included them in the qualitative synthesis ( Figure 1). Full-text articles excluded, with reasons (n = 4): -Full-text unavailable (n = 2) -In vitro (n = 1) -Others (n = 1) Full-text articles included in qualitative synthesis (n = 25) Abstracts excluded (n = 17): -Review and Expert Opinion (n = 10) -Letters to Editor (n = 3) -Adolescents (n = 1) -Language (not in English) (n = 3)

Suggested Etiologies
HPPDs are poorly understood due to the enormous range and variability of recurrent sensory disturbances, and the multiple distinct subtypes [17,18].
The main neurobiological hypothesis is that LSD consumers might develop chronic disinhibition of visual processors and dysfunction in the function of the central nervous system (CNS) [4,[34][35][36]. This disinhibition may be linked to an LSD-generated intense current [37] that may determine the destruction or dysfunction [18] of cortical serotonergic inhibitory interneurons with gamma-Aminobutyric acid (GABAergic) outputs, implicated in sensory filtering mechanisms of unnecessary stimuli [34][35][36]38]. The efficacy of some treatment options in a subject with HPPD, such as pre-synaptic α 2 adrenergic agonists, selective serotonin reuptake inhibitor (SSRIs), benzodiazepines, and mood stabilizers would confirm this neurobiological hypothesis (see Section 3.2). Reverse tolerance or sensitization that emerges after LSD exposure may explain recurrent occurrences after the stimulus has been withdrawn [39]. Nonetheless, HPPD-like experiences, such as flashbacks, moments of derealization, and hyper-intense perceptions are reported in healthy populations and non-LSD exposed subjects [40]. Moving from biochemical receptor interactions towards macroscopic areas, a temporary or permanent impairment in the Lateral Geniculate Nucleus (LGN) has been hypothesized [4,[41][42][43]. The LGN, which is located in the thalamus, is associated with visual perception pathways [41][42][43]. Recent research highlighted a brain dysfunction in patients with visual snow, located mainly in the right lingual gyrus [44], perhaps implying LSD involvement. Halpern et al. [40] suggested that HPPD can be due to a subtle over-activation of predominantly neural visual pathways that worsens anxiety in predisposed subjects after ingestion of arousal-altering drugs, including non-hallucinogenic substances. According to Holland and Passie, environmental triggering by specific situations or stimuli or other elements related to the original experience may be involved in flashback experiences [45].
The use of cannabis has been associated with the development of perceptual distortions in seven patients [29,46,48,49,61]. In one case, it was associated with 3,4-Methylenedioxymethamphetamine (MDMA) and in another case with PCP [48,49]. In two patients, visual distortion followed the consumption of synthetic cannabinoids [61].
Lauterbach et al. reported the unique case of HPPD induced by the atypical antipsychotic Risperidone [60].

Clinical Features
According to DSM-5, Hallucinogen Persisting Perception Disorder is the recurrence of perceptive disturbances that firstly develop during intoxication. The contents of the perception and visual imagery range extensively [17,19]. DSM-5 and previous DSM editions report a list of the most common symptoms experienced by HPPD patients, but only a few symptoms have been described in the professional literature. The main group of symptoms reported by Criterion A of the DSM-5 are visual disturbances. In fact, as in the vast majority of induced psychoses, visual hallucinations are notably more common than auditory [3]. Regardless, every perceptual symptom that was experienced during intoxication may re-occur following hallucinogen withdrawal. We report a list of the main literature-reported visual disturbances in Table 2.
A latent period may antecede the onset of returning visual occurrences. This latent period may last from minutes, hours, or days up to years, and re-emerge as either HPPD I or II with or without any recognized or perceived precipitator [17,19]. Episodes of HPPD I and II may appear spontaneously or they may be triggered by identified and non-identified precipitators [18]. Episodes may be continuous, intermittent, or sudden. With regards to this point, neither HPPD I nor HPPD II can really be considered as persisting in a narrow sense of the word. Additionally, their differential diagnosis can only be proposed in terms of prognosis rather than clinical presentation.
The frequency of recurrence of perceptual distortions is lower for HPPD I than HPPD II [18]. Prior substance users can voluntarily elicit or produce visual disturbances with or without known triggers [4,17,18]. After HPPD II onset, hallucinogenic events tend to occur more frequently, and their duration and intensity increase. Subjects might perceive a partial or total loss of control.

Mental Illnesses Comorbid with HPPD
Recent observations reported a co-occurrence with depressive [20] and anxiety traits [51] and severe mental illnesses such as Major Depressive Disorder [23], Bipolar Disorder [23,62], and Schizophrenia Spectrum Disorders [17,58]. However, HPPD I and HPPD II onsets are not necessarily accompanied by any prominent additional psychiatric disorder, thus representing an independent condition [17,18]. In particular, the onset of HPPD II is often linked to a clear negative mood and affect. Anxiety and depressive features might aggravate new episodes. Anxiety might also evolve into a panic attack. Anticipatory anxiety may antecede future visual aberration events, and avoidant behavior may limit and restrict regular normal functioning [17,18]. Recently, in a study carried out by Halpern, a comprehensive survey of 20 subjects reporting Type-2 HPPD-like symptoms was presented and evaluated. The dissociative symptoms were consistently associated with HPPD, suggesting that HPPD is in most cases due to a subtle over-activation of predominantly neural visual pathways that worsens anxiety in predisposed subjects after the ingestion of arousal-altering drugs, including non-hallucinogenic substances. The authors report that many perceptual symptoms reported were not first experienced while intoxicated, and are partially associated with pre-existing psychiatric comorbidity, tempering the direct role of hallucinogens in the etiopathology of the disorder [40].
Only two observational studies and one case report evaluated psychotic patients with comorbid HPPD [57,58,60] (Table 3). Two observational, cross-sectional studies compared schizophrenic patients with prior use of LSD who developed HPPD (SCZ+HPPD, 49 patients) with those who did not (SCZ, 57 patients), for a total of 106 patients [57,58]. No differences between the two groups have been found with respect to demographic characteristics, age of psychotic onset, age of drug use onset, and type of substances abused [57,58]. As expected, SCZ+HPPD patients reported more distressing and horrific LSD experience ("bad trips") (p < 0.05) [57]. Interestingly, the positive subscale of the Positive and Negative Syndrome Scale (PANSS) did not differ between the two groups. On the contrary, SCZ+HPPD patients showed lower scores in the PANSS negative subscale, the PANSS General Psychopathology Subscale, and the PANSS total scores (p < 0.05) [57]. Moreover, 67% of the schizophrenic patients comorbid with HPPD were able to distinguish between perceptual distortion and psychotic hallucinations [58], and 9 out of 12 patients could identify precursory cues for perceptual distortion (substance-induced cues, situational cues, and mental cues) [58]. Lauterbach et al. [60] reported a case of HPPD comorbid with psychosis, in which visual distortions were induced by antipsychotic treatment. Interestingly, the patient did not report any history of previous substance abuse [60]. The patient was treated with Risperidone, Clonazepam and Trazodone, and she reported visual disturbances resembling HPPD, in particular, illusions, after three subsequent Risperidone dosage increases [60]. Table 3. Observational studies and case reports comparing schizophrenic patients with HPPD (SCZ+HPPD) and schizophrenic patients without HPPD (SCZ) (* p > 0.05, ** p < 0.05).

First-Line Medications
Pre-synaptic α 2 adrenergic agonists are a treatment with a low side-effect profile for patients with a previous history of substance-related disorders. Symptoms alleviation has been reported in some patients treated with these drugs [17,18,52,63]. The effectiveness may be based on the evidence that clonidine may elevate plasma GABA levels in humans, having a benzodiazepine-like effect. Clonidine may also decrease locus coeruleus activity, leading to a reduction of adrenergic activity [64], which can be effective in the management of PTSD [65]. Therefore, as in PTSD-related recurring flashbacks, some visual disturbances could be associated with excessive sympathetic nervous activity. Thus, these visual distortions could be ameliorated by Clonidine [52,63].
A dosage of 0.75 mg/die of Clonidine has been evaluated as a treatment option for nine HPPD patients [51,59] (Table 4). The total remission has been reported in a single patient with flashbacks and anxiety treated with 0.25 mg of Clonidine three times a day for two months [59]. In the 2 months open study on eight HPPD patients, the Clinical Global Impression (CGI) and Patient's Severity Perception significantly decreased between entry and endpoint scores [51], although two patients dropped out at week 3 and week 5, respectively [52]. Lofexidine (0.2-0.8 mg/day) is a sympatholytic centrally acting α 2 presynaptic adrenergic agonist that showed similar efficacy in some cases [23,65,66].
Benzodiazepines may be useful and effective in eliminating benign HPPD I and ameliorating, but not completely eradicating, pervasive HPPD II symptoms [18,67]. The effectiveness of Benzodiazepines may be related to their activity on the cortical serotonergic-inhibitory inter-neurons with GABAergic outputs [2,4]. Alprazolam (0.25-0.75 mg/day) has been prescribed with some success and Clonazepam (0.5-1.5 mg/day) appears to be the most reliable and effective benzodiazepine even at low doses [17,18,51,67]. Higher doses (4 mg/day) have also been used with good outcomes [68]. Clonazepam may act on serotonergic systems, improving, enhancing, and augmenting transmission [17,18,51,67], thus promoting alleviation and a marked improvement [51,67]. Clonazepam has been evaluated in three case reports and one open-label trial by Lerner [19,50,51]. In the clinical trial, 16 HPPD patients were treated with a Clonazepam dosage of 2 mg/day [51]. Their symptoms improved significantly after treatment initiation and the improvement persisted during a 6-month follow-up after treatment discontinuation [51]. The same author reported two cases of cannabis-induced visual disturbances and correlated anxiety features. In both cases, Clonazepam (2 mg/day) was effective in improving symptoms, but focal visual disturbances without anxiety (trailing phenomena in one case, and black moving spots in the second case) persisted during and after therapy [19]. More recently, Clonazepam (6 mg/day) has been proved to be effective in improving cannabis-induced HPPD symptoms [50]. On the other hand, the intrinsic abuse potential of benzodiazepines might be inconvenient in certain individuals with a past history of substance use [17,18]. Given the benign nature of HPPD I, the use of benzodiazepines should be proposed only for severe cases, in the acute phase, and for the short term.
HPPD patients appear to be sensitive to first-generation antipsychotics at low doses, requiring monitoring of extrapyramidal side effects. Haloperidol [69] and Trifluoperazine [70] were reported to be helpful. Perphenazine (4-8 mg/day) [17,23], Sulpiride (50-100 mg/day) [23], and Zuclopenthixol (2-10 mg/day) [17,23], at very low doses, are well tolerated and may be an effective treatment. Some of the long-acting first-generation antipsychotics may still be useful in co-occurring Psychotic Spectrum Disorders and HPPD II [58]. In one study, haloperidol was noted to reduce hallucinations, but an exacerbation of flashbacks in the early phases of treatment was highlighted as well [1,69].
The use of second-generation antipsychotics in HPPD patients without comorbid psychotic disorders is debated. Anderson recently reported the case of a young woman presenting prolonged and distressing multimodal pseudo-hallucinations, depressive symptoms, and anxiety, who was treated with Risperidone for three months without any significant improvement [48]. At the same time, conflicting evidence exists on the antipsychotics effects in psychotic HPPD patients. One study did not report differences in antipsychotic treatment response between SCZ and SCZ+HPPD patients [58]. On the other hand, a more recent study has shown the ineffectiveness of antipsychotic medications in an SCZ+HPPD population [57].  Risperidone was usually prescribed due to its proven efficiency in the treatment of perceptual disturbances in Psychotic Spectrum Disorders, mainly in Schizophrenic Disorders. LSD seems to work as a partial agonist of postsynaptic serotonin receptors. Therefore, Risperidone, which is a strong antagonist of both postsynaptic 5-HT 2 and D 2 receptors, was expected to be convenient. In contrast with this supposition, Risperidone at recommended [71] and lower doses [72] worsens visual disturbances and accompanying anxiety, or does not show any effect [54]. This was presumably due to Risperidone's α 2 presynaptic antagonism and noradrenaline release [34]. In addition, Risperidone was associated to the re-experiencing of visual disturbances in some patients suffering from schizophrenia with a past history of LSD use [73]. One psychotic patient treated with Risperidone, Clonazepam, and Trazodone reported visual disturbances resembling HPPD after three subsequent Risperidone dosage increases [60]. At the same time, Risperidone has been shown to be effective in improving PCP-induced HPPD with anxiety in one patient, while in the same patient Olanzapine produced symptoms exacerbation [21].
Evidence not included in our systematic review suggested that low dosages of atypical antipsychotics may be useful, specifically Aripiprazole (5-10 mg/day) [23], also because of its efficacy in substance and alcohol use disorders [74].
Visual oddities and disturbances with sudden paroxysmal onset have been interpreted as visual seizures and prompted the use of antiepileptic drugs in HPPD. This consideration helped to explicate the efficacy of benzodiazepines and led to the prescription of Phenytoin [75,76]. Today, Phenytoin is not used for HPPD treatment due to its troubled side effect profile. Medications such as Valproic Acid (200-600 mg/day), Carbamazepine (200-600 mg/day), Oxcarbamazepine (300 mg/day), Gabapentin (300-900 mg/day), Topiramate (25-100 mg/day), and Lamotrigine (50-100) may be useful [23], also because of their efficacy in substance and alcohol use disorders [77][78][79]. In a single case of HPPD symptoms and electroencephalographic (EEG) abnormalities, compatible with toxic encephalopathy, the visual hallucinations that recurred at any alcohol ingestion improved, but did not disappear with the use of Valproic Acid (1500 mg/day) [46]. Levetiracetam has shown to reduce some visual symptoms as well as HPPD related-depersonalization and derealization [80]. Lamotrigine has shown to be efficacious in a recent severe case of HPPD with some EEG abnormalities (Anderson et al., 2018). These medications may also be helpful when visual disturbances are accompanied by co-occurring mood swings and mood disorders.
Antidepressant medications could help in the management of co-occurring HPPD II with anxiety and depressive disorders [17,18,20,51,67]. HPPD II alone does not appear to be an appropriate target. There are questionable and controversial results regarding Sertraline, which has been reported to worsen [81] as well as to improve visual disturbances. Amelioration following long-term administration of SSRIs was attributed to the down-regulation of 5-HT 2 receptors, providing more evidence to corroborate the serotonergic mechanisms underlying this condition. Other prescribed SSRIs did not show any benefits. Norepinephrine reuptake inhibitors (NRIs) such as Reboxetine have been tried with some success in LSD-induced HPPD symptoms comorbid with Major Depressive Disorder [20]. Agomelatine, given its peculiar function on neurotrophic factors [74], could have some benefits on the syndrome, although no data are available until now.

Second Line Medications
Naltrexone has been usually used, alone or with other medications, in chronic patients with continuous visual imagery that previously did not respond to other medications [17,18].
Calcium Channel Blockers and Beta Blockers may be helpful in patients with co-occurring HPPD II and anxiety disorders [18]. Propanolol at low (20-60 mg/day) and high doses (240 mg/day), as well as Atenolol 25-50 mg/day, have been used to diminish accompanying anxiety of visual imagery [18,23]. Investigations of HPPD patients with EEG mapping showed that HPPD is represented by disinhibition [35] in the cerebral cortex [34]. The rationale behind this interesting and novel approach is that improving sensory gating by dopaminergic enhancers may cause an inhibition of catechol-O-methyl transferase (COMT), that may improve HPPD symptomatology.

Brain Stimulation Treatments
Currently, brain stimulation treatments have been proposed as a possible therapeutic option to enhance the recovery of refractory symptoms in several disorders [82,83]. Repetitive Transcranial Magnetic Stimulation (rTMS) is a non-invasive brain stimulation approach that acts by modulating specific brain circuits. While high-frequency (>5 Hz) stimulation determines a depolarization of nerve cells, with long-term potentiation (LTP) effects, low-frequency stimulation protocols (1 Hz) determine the long-term depression (LTD) of the targeted area, with the possibility to induce the localized inhibition of specific disordered networks. According to the cortical hyperexcitability hypothesis about its pathogenesis, several case reports propose that rTMS could be a promising therapeutic method for refractory visual hallucinations in schizophrenia [84,85].
To date, no studies have investigated the potential use of rTMS in HPPD. Interestingly, Kilpatrick and Ermentrout (2012) [86] studied the spatiotemporal dynamics of neuronal networks in HPPD, with spike frequency adaptation. This study reported that altering parameters controlling the strength of synaptic connections in the network can lead to spatially structured activity suggestive of symptoms of HPPD. Future research is necessary to test the possible effectiveness of the rTMS neuromodulatory effect on HPPD. Putative targets of stimulation could be hypothesized to be located in the visual cortical areas, as well as in the occipitotemporal sulcus [87]. Functional neuroimaging may be beneficial in localizing a specific target for stimulation and may prevent wasting time and money on targets which are not as likely to be involved in the pathogenesis.

Discussion
It has to be highlighted that a limitation of the study might be represented by the search method: in fact, we decided to limit the literature search to the DSM terminology in order to exclude simple "flashback phenomena" that are commonly reported in psychopathology, and that may not follow the use of hallucinogens. This could have narrowed the results, preventing the inclusion of other studies using the ICD terminology, which is less "technical" about the issue.
The main consideration that has to be done with respect to HPPD is its rare and unpredictable nature [16]: current prevalence estimates are unknown, but DSM-5 suggests 4.2% [88]. The condition is more often diagnosed in individuals with a history of previous psychological issues or substance misuse [56], but it can arise in anyone, even after a single exposure (mostly to LSD, but it has also been reported after use of other psychedelics) [89]. In many cases, HPPD may also be explained in terms of a heightened awareness of and concern about ordinary visual phenomena, which is supported by the high rates of anxiety, obsessive-compulsive disorder, hypochondria, and paranoia seen in many patients [90].
The crucial movement towards a comprehensive clinical understanding of Hallucinogen Persisting Perception Spectrum Disorders (HPPSD) [23] is the establishment of an accepted operative nomenclature. This wide spectrum of disorders encompasses different subtypes, ranging from HPPD I to HPPD II, according to our hypothetical distinction. Among the innumerable triggers able to precipitate HPPD, prospectively, the use of natural and synthetic cannabinoids appears to be the most frequent. This is consistent with the rapid and vast diffusion of these novel psychoactive compounds, nowadays easily available without specific cultural filters and references [91,92]. Distinct substances, with completely different mechanisms of action, might lead or precipitate the genesis of HPPD, therefore suggesting a multifaceted etiology. Thus, it is accordingly conceivable that different medications could be useful and helpful in the treatment of different subtypes of HPPD. Tracers and trailing phenomena appear to be the most resistant symptoms. Concomitant coexisting psychiatric disorders can represent a further clinical challenge, with the clinical construct of the lysergic psychoma as a possible heuristic model. According to this theory, the presence of induced psychopathological phenomena (the Psychoma) may trigger a specific reaction excepted by the not-affected part of the mind, trying to counteract the psychoma, which is perceived as a "foreign body in the mind". Of course, when the psychoma is strong and repeated in its nature, the possibility to determine a full-blown psychosis may become more concrete [93,94].
Regarding treatment options, a combination of medications may be needed according to the preceding or subsequent psychopathology. Given the limited literature about HPPD, a possible hypothesis about the pharmacotherapy of choice in relation to different etiologies has not been considered. However, the presence of psychiatric and neurological comorbidities could represent a valid criterion to address the choice. Clinical experience and an extensive and comprehensive knowledge of these phenomena are vital for successful treatment outcomes.
Controlled clinical investigations are mostly needed in order to better understand the etiology, mechanisms of action, clinical issues, and pharmacological treatment options for Hallucinogen Persisting Perception Spectrum Disorders (HPPSD).