Psychosis

Cannabinoids are associated with significantly increased paranoid symptoms, with consistent effects in experimental and population-based studies.

Bayesian meta-analysis across k=13 studies (n=13.559) on cannabis and paranoid symptoms; experimental studies (k=5): cannabinoid participants developed more severe paranoid symptoms than placebo (SMD=0.47, 95% CI 0.13–0.48, PPI=94%); cross-sectional studies (k=4): cannabis users higher odds for paranoia (OR=1.75, 95% CI 1.43–2.07, PPI=99%); prospective studies (k=3) show association with later paranoia onset.

Moderate evidence for chronic pain and spasticity, low evidence for nausea/vomiting, weight gain and sleep disorders; increased risk of side effects.

n=79 RCTs (n=6.462), comprehensive SR on medical cannabis; identifies psychotic symptoms as a common side effect (moderate evidence), no therapeutic evidence for treatment of psychosis; relevant safety data.

Cannabis use was associated with significant memory impairments in healthy individuals, but with improved memory in patients with psychotic disorders.

Meta-analysis across k=88 studies (n=10.958: 7.697 healthy, 3.261 psychosis patients) on cannabis effects on memory. In healthy individuals: cannabis associated with impairments in global memory (d=0.27), prospective memory (d=0.61), verbal immediate (d=0.40) and delayed recall (d=0.36), visual recognition (d=0.41). In psychosis patients paradoxically better global memory (d=-0.11), visual immediate recall (d=-0.73) and recognition (d=-0.42). Cannabis-using patients were younger and had lower depression scores.

Systematic review shows that cannabis use is an important modifiable risk factor for psychosis and poorer prognosis; however, the exact neurobiological mechanisms remain unclear.

Systematic review (k=70 studies, >3000 patients with psychotic disorders or elevated psychosis risk) on neurobiological and neurochemical mechanisms of cannabis effects in psychosis. Longitudinal studies show: cannabis use increases risk of psychosis development, hospitalization, more frequent/longer hospital stays, treatment failure. Cannabis-using and non-using psychosis patients show distinct neurocognitive and neurodevelopmental impairment patterns; biological basis not yet fully clarified.

Cannabinoids showed limited efficacy for cannabis withdrawal symptoms, sleep disorders, tic disorders and autistic traits, but no significant effects on anxiety disorders, psychoses, PTSD and opioid disorders; increased risk of adverse effects overall.

Systematic review + meta-analysis on cannabinoids in mental disorders and substance use disorders; k=54 RCTs (n=2.477, 69% male, median age 33,3 years). 44% of studies with high risk of bias, quality of evidence mostly low. CBD+THC combination reduced cannabis withdrawal symptoms (SMD=-0.29, 95% CI -0.57 to -0.02) and weekly cannabis use (-1.00g, 95% CI -1.69 to [data incomplete in abstract]).

Contrary to expectations, cannabis use in schizophrenia is associated with small to medium advantages in neurocognitive performance.

Meta-analysis, k=8 studies, n=942 (356 cannabis users with schizophrenia, 586 non-users); effect sizes in the small to medium range; cannabis-using patients showed superior cognitive performance in all 7 neuropsychological domains (including general cognitive abilities, attention, working memory, executive functions) compared to non-using schizophrenia patients — no cognitive harm from cannabis use demonstrable in this population.

CBD showed mixed efficacy depending on indication, with a positive signal for anxiety disorders and schizophrenia, but no effect on cognitive function in psychosis.

Systematic review of k=25 studies (n=927, of which 22 RCTs) on CBD dosing, efficacy and safety in adults; short-term therapeutic effects for psychotic disorders (symptomatic improvement in schizophrenia), social anxiety disorder and substance use disorders; no effect on cognitive function in psychosis; overall good tolerability.

The evidence for efficacy and safety of CBD in psychiatry is limited; most studies showed no statistical significance and further well-designed RCTs are required.

Systematic review on CBD in psychiatric disorders (schizophrenia, anxiety disorders, addiction, among others); k=13 studies (6 case reports + 7 trials), n=201. WFSBP level of evidence C1 for reduction of positive symptoms in schizophrenia and in social anxiety disorder; C2 for cannabis dependence; B for cannabis withdrawal. Majority of individual studies without statistical significance. Most common side effects: sedation and dizziness.

Current cannabis users (CANN+) show poorer performance in several cognitive domains (premorbid IQ, current IQ, verbal learning, verbal working memory, motor inhibition) compared to never-users (CANN-), with deterioration in older patients.

Meta-analysis across k=14 studies on neurocognition in young psychosis patients (mean age 15–45 years) with current cannabis use (CANN+) vs. never-users (CANN-). CANN+ showed poorer performance in several domains: premorbid IQ, current IQ, verbal learning, verbal working memory, motor inhibition. Higher age predicted poorer processing speed, sustained attention, verbal memory. CANN+ outperformed CANN- only in conceptual set-shifting. Random-effects modeling, moderator analyses.

None of the five included studies showed clear efficacy of preventive interventions in reducing cannabis use in people with psychosis.

Systematic review of preventive interventions for cannabis use in psychosis patients; k=5 controlled studies (from 11.460 screened). No study showed clear efficacy in reducing cannabis use; no study measured cannabis-associated harms. All included studies had a high risk of bias.

CBD showed significant improvements in positive psychotic symptoms (PANSS) and clinical global impression of improvement (CGI-I) compared to placebo, with good tolerability.

Multicentre double-blind RCT, n=88 (CBD 1000 mg/day n=43 vs. placebo n=45) as add-on to existing antipsychotic medication in schizophrenia. After 6 weeks: CBD group showed lower positive psychosis symptoms (PANSS: treatment difference=-1.4, 95% CI=-2.5, -0.2), higher improvement rate (CGI-I: -0.5, 95% CI=-0.8, -0.1) and less severe illness (CGI-S: -0.3, 95% CI=-0.5, 0.0). Cognitive performance (BACS: +1.31, 95% CI=-0.10, 2.72) and overall functioning (GAF: +3.0, 95% CI=-0.4, 6.4) showed trends toward improvement without statistical significance. CBD was well tolerated, with adverse event rates similar to placebo.

Cannabidiol led to significant clinical improvement in acute schizophrenia with a superior side-effect profile compared to amisulpride.

Double-blind RCT, CBD vs. amisulpride in acute schizophrenia. Both treatments led to significant clinical improvement; CBD showed a markedly superior side-effect profile. CBD treatment was associated with a significant increase in serum anandamide levels, which correlated significantly with clinical improvement (p-values not given in the abstract, but 'significant' mentioned several times).

CBD induced significant changes in functional connectivity in the default mode network, but showed no effects on prefrontal metabolites or brain activity during reward processing.

n=31 patients with first-episode psychosis (≤5 years after diagnosis); adjunctive CBD 600 mg/day vs. placebo over 28 days. CBD treatment significantly altered functional connectivity in the default mode network (DMN; time×treatment interaction p=0.037): increase in the CBD arm (0.59±0.39 → 0.80±0.32) vs. decrease under placebo (0.77±0.37 → 0.62±0.33). Decrease in positive symptomatology correlated with decreasing glutamate (p=0.029) and NAA levels (p=0.019) only in the CBD group. No effect on prefrontal metabolite concentrations or reward processing.

CBD pretreatment worsened rather than improved the cognitive and psychotic effects of cannabis: verbal recall was worse (-1,3 words) and PANSS positive symptoms were higher (+2,2 points) compared to placebo.

RCT (n=30) in schizophrenia patients with cannabis use disorder; randomized, double-blind, cross-over design. CBD premedication 1000 mg vs. placebo before vaporized cannabis (THC 20-60 mg). Delayed verbal recall after CBD premedication worse than after placebo (3.5 vs. 4.8 words, MD=-1.3, 95% CI: -2.0 to -0.6, p=0.001). PANSS-P increase after CBD greater than after placebo (+5.0 vs. +2.9, MD=2.2, 95% CI: 0.6-3.7, p=0.01). CBD did not attenuate cannabis effects, but paradoxically enhanced them.

JWH-018 induced psychotomimetic symptoms including psychedelic effects (altered perception), dissociative effects (amnesia, derealisation, depersonalisation) and feelings of confusion.

n=24 healthy participants (10m, 14f), placebo-controlled crossover design with synthetic cannabinoid JWH-018 (average 5.52mg inhaled, 75µg/kg body weight + booster). JWH-018 induced pronounced psychotomimetic symptoms (dissociation, derealisation, depersonalisation, amnesia, confusion, altered perception) in healthy subjects without psychiatric history. Time window 4.5h, CADSS/Bowdle/POMS/SCRQ scales.

CBD augmentation showed no improvement in cognitive or psychotic symptoms compared to placebo in chronic schizophrenia.

n=36 stable antipsychotically treated patients with chronic schizophrenia; 6-week CBD 600 mg/d vs. placebo as augmentation. No main effect on MCCB total score; significant drug×time effect (p=0,02), but only the placebo group improved (p=0,03). PANSS total decreased over time (p<0,0001), no significant difference between groups (p=0,18). Negative RCT — CBD augmentation did not improve cognition or symptoms.

THC worsened inhibitory control and correlated with the extent of acute psychotic symptoms.

n=36 healthy men; 10 mg oral THC vs. placebo (double-blind crossover, fMRI); THC induced transient psychotic symptoms; symptom severity correlated directly with inhibition error frequency and inversely with inhibition efficiency; attenuated left inferior frontal activation as neural mechanism of THC-induced psychotic symptoms.

Delta9-THC increased anxiety and intoxication, while CBD showed a trend toward anxiety reduction with differing effects on brain activation and autonomic responses.

n=15 healthy subjects, fMRI study of THC vs. CBD during emotional processing; THC induced activation in limbic regions (amygdala, anterior cingulate), CBD showed opposite effects with attenuation of subcortical activation.

Δ9-THC increased psychotic symptoms, anxiety and sedation, modified brain activation in the parahippocampus and ventrostriatum, but did not significantly impair learning performance; cannabidiol showed no such effects.

n=15 healthy subjects, THC-induced modulation of mediotemporal and ventrostriatal function using fMRI; THC modulated activity in the hippocampus and ventral striatum, regions relevant to psychosis pathophysiology.

Continuous cannabis use was associated with persistently higher psychotic symptomatology and worse functioning; cessation of use significantly reduced symptomatology.

n=314 first-episode psychosis patients (OPUS), 5-year course. Continued cannabis use vs. abstinence: higher psychosis symptoms (difference 0,97; p<0,001) and lower functioning (GAF difference 8,26; p=0,01). Cessation after disease onset protective: adjusted symptom difference -1,04 (p=0,006), also after controlling for antipsychotic compliance.

Continued daily use of high-potency cannabis after first psychotic event significantly increases relapse risk compared to abstinence.

n=256 first-episode psychosis patients: daily use of high-potency cannabis over 24 months associated with increased relapse risk (OR=3,28; 95% CI 1,22–9,18) and more frequent relapses (IRR=1,77; 95% CI 0,96–3,25) vs. abstinent patients; discontinuation after disease onset was associated with the most favorable disease course.

Continued cannabis use after first-episode psychosis increases relapse risk, partially mediated by poorer medication adherence.

n=245 first-episode psychosis patients; 37% relapse rate over 2 years; cannabis use after onset of psychosis predicted relapse, number of relapses, intensity of care; structural equation models: 20–36% of the effect mediated via medication adherence (β=0,06–0,08, all 95% CI >0).

Cannabis use increases the risk of psychosis incidence and considerably worsens the prognosis in existing vulnerability to psychosis.

3-year follow-up, subgroup n=59 individuals with existing psychosis diagnosis; cannabis use was associated with a risk difference of 54.7% for clinically relevant worsening (vs. 2.2% without pre-existing psychosis; p for interaction=0.001). Result: cannabis use considerably increases the risk of psychosis persistence and poor course in already affected patients.

Decrease in cannabis use is accompanied by fewer psychotic experiences, increase by more positive symptoms, but not by negative or depressive symptoms.

Longitudinal study (N=705, age 18–27 years): Increase in cannabis use significantly associated with increased positive psychotic symptoms (β=0,07; p=0,02). Decrease in use associated with a decrease in the overall frequency of psychotic experiences (β=−0,096; p=0,01). First longitudinal study to directly link changes in cannabis use with changes in psychotic experiences in the general population.

Synthetic cannabinoid users showed more severe psychotic symptoms, longer hospitalisation duration and more previous hospitalisations than cannabis users, but this is a comparative observational study without an intervention effect.

n=223 psychiatrically hospitalised patients (synthetic cannabinoids n=60 vs. cannabis n=163); SC users younger (30,5±7,8 vs. 34,7±10,1 years, p=0.009), longer hospitalisation (43,5±54,0 vs. 22,9±31,4 days, p=0.005), more previous hospitalisations (3,7±5,1 vs. 2,0±5,1, p<0.001), higher PANSS scores (82,5±23,1 vs. 70,0±19,9, p=0.008), more frequently court-ordered admissions (36,7% vs. 19,9%, p=0.028).

CSF anandamide levels are 8-fold elevated in untreated first-episode schizophrenics and correlate negatively with psychotic symptoms; the clinical significance and causality remain unclear.

n=203 (47 antipsychotic-naive first-episode paranoid patients, 84 healthy, 13 dementia, 22 affective disorder, 37 typical antipsychotics, 34 atypical antipsychotics). CSF anandamide levels 8-fold elevated in untreated acute paranoid-schizophrenic patients vs. controls (p<0.001). Negative correlation between CSF anandamide and psychotic symptoms (rS=-0.452, p=0.001). The elevation was absent under typical antipsychotic treatment.

In 4 of 6 treatment-refractory patients an improvement of schizophrenia symptoms was reported, of whom 3 showed a reduction of core psychotic symptoms; no clinically significant side effects mentioned.

n=6 treatment-refractory chronic schizophrenia patients with self-reported cannabis-induced improvement received dronabinol (synthetic Δ9-THC). 4 of 6 showed symptom improvement; in 3 of the 4 responders reduction of core psychotic symptoms (not merely nonspecific sedation). No clinically significant side effects.

CBD showed promising effects in schizophrenia (particularly in the early stage of illness) and in cannabis withdrawal symptoms, but the evidence base is, in the authors' assessment, incomplete.

Narrative review on the effect of CBD in schizophrenia and substance use disorders. CBD as mono- or add-on therapy improved symptoms in schizophrenia patients, particularly promising in early stages of illness. Anandamide blood levels discussed as a potential biomarker. CBD-THC mixtures showed positive effects on cannabis withdrawal and craving.

Narrative review on the disturbance of the endocannabinoid system in schizophrenia; no intervention examined, only associative findings and pathophysiological mechanisms described.

Narrative overview of endocannabinoid dysfunction in schizophrenia; reports increased CB1 receptor density in the prefrontal cortex/hippocampus/basal ganglia in schizophrenia patients and elevated anandamide levels in CSF/serum (including the prodromal phase). Qualitative synthesis without numerical effect sizes; discusses a possible protective role of endocannabinoids in psychosis homeostasis.

Genome-wide association study on cannabis use disorder (CanUD), cannabis use and psychiatric disorders; CHRNA2 variant significantly shared between CanUD and schizophrenia (colocalization + local genetic correlation); Mendelian Randomization shows bidirectional causal relationship between CanUD and most psychiatric disorders, unlike pure cannabis use.

CB1 receptor density is increased in a region- and cause-specific manner: in the prefrontal cortex in schizophrenia, in the caudate-putamen with cannabis use.

Postmortem measurement of CB1 receptor density in the human brain: In the dorsolateral prefrontal cortex (BA9) significantly increased binding density in schizophrenia vs. controls (142±9,9 vs. 119±6,6 fmol/mg; p<0,05), independent of recent cannabis use. In the striatum (caudate nucleus-putamen) significantly increased CB1 density following recent cannabis use vs. non-users (151±9,0 vs. 123±7,2 fmol/mg; p<0,05), independent of diagnosis. Indication of distinct neurobiological substrates for schizophrenia-associated vs. cannabis-induced CB1 changes.