Cognitive changes in alcohol-induced psychotic disorder

The understanding of alcohol-induced psychotic disorder (AIPD) as an illness phenomenon is becoming is increasingly important for the clinician. AIPD is often described as a “rare complication” [1] of alcohol use disorders. Despite this description, it was noted that the number of persons’ diagnosed with alcohol psychosis has escalated by four times in certain countries [2] and for patients diagnosed with AIPD there is 68% risk of re-admission [3]. There is a 37% co-morbidity of AIPD with other mental disorders [4] and a 5–30% risk that patients with AIPD will develop a chronic schizophrenia-like syndrome [5]. Patients who develop AIPD are at risk of becoming suicidal and need to be carefully monitored [3].

Recent research [1, 6] concurs with earlier findings [7] that AIPD is a distinct diagnostic entity that can be distinguished from schizophrenia. Most commonly, AIPD mostly occurs with auditory hallucinations but also delusions, often of a paranoid nature. Whilst AIPD resembles paranoid schizophrenia, it generally occurs without negative symptoms [1, 7]. The psychotic episode reportedly often lasts no more than a month [8] to 6 months [6, 9], but during that time, reality testing is impaired. The person suffering from AIPD related auditory hallucinations has no insight into the fact that hallucinations are substance-induced [10] especially during the episode.

There are inconsistent reports regarding the outcome of this disorder [2, 5]. Some authors [2, 5] noted that AIPD may become prolonged and schizophrenia-like symptoms may develop. Other authors [9, 11] noted that the outcome of AIPD is usually good if abstinence is achieved. There is some indication that AIPD is a severe psychiatric disorder with poor prognosis [4]. This study [4] however included in their sample persons who suffered from alcohol-withdrawal delirium. A fifty year follow-up study [12] concurred with the previous study [4] and found that alcohol-induced psychotic syndrome (AIPS) is associated with increased risk for premature death. This study included sample persons with delirium tremens as well as AIPD, hence the use of the term AIPS. The authors however conclude that AIPD is a severe mental illness with poor prognosis [4].

The age of onset of alcoholism reported in AIPD varied between 21.4 [2] and 29.1 years [13] with the latter study showing a significantly younger age of onset of alcoholism for AIPD patients than their non-psychotic alcohol-dependent male counterparts. There were no significant differences [13] with reference to age, education, marital status and employment between male alcoholic patients with and without a history of psychosis. In this study [13] however there was a significant difference between the number of Caucasian men diagnosed with AIPD and the number of non-Caucasian men. AIPD was diagnosed more in the Caucasian men. A previous study [4] reported that low socio-economic status, paternal mental health or alcohol related difficulties, early onset of alcohol dependence and numerous hospitalizations increase the risk for developing AIPD.

Although a number of reports [2, 14, 15] addressed the phenomenological expression of AIPD, the nuanced clinical presentation received very little scientific attention [16]. This paucity of interest is deemed unjustified as the correct diagnosis of this condition is key to accurate treatment and improved prognosis [16]. This paucity of literature highlights the lack of scientific evidence addressing the cognitive deficits of this disorder. There is some evidence suggesting that cognitive deficits as measured by the mini-mental state examination (MMSE) [17] may be greater in AIPD compared to uncomplicated AD [1].

The present introductory study compared the neuropsychological functioning of patients who were diagnosed with AIPD with the neuropsychological functioning of alcohol dependent patients in order to differentiate between these two conditions from a neuropsychological perspective. Additionally, this paper aims to expose the associated brain region which is potentially at the genesis of AIPD. Understanding the underlying pathophysiology may lead to more effective treatment of AIPD and prevent relapse, including recurrent re-admissions and possibly the development of more severe pathology.

Upon initial ANOVA, the AD group had a statistically significant higher level of education (HLOE) than the AIPD group. Upon further comparative analysis with HLOE as a covariate, HLOE did not influence the findings. Even though the AD group reported more severe alcohol dependence than the AIPD group (SADQ), the difference did not reach statistical significance.

Patients with AIPD presented mainly with auditory hallucinations, delusions (with predominant persecutory themes) and anxiety in the presence of a clear consciousness. The mean scores on the PANSS scales were: 20.1 (positive scale), 16.7 (negative scale), 37.5 (general scale) and 74.3 (total scale).

This study demonstrated statistically significant deficit in immediate verbal memory, increased effort when completing a memory task and impaired abstract verbal reasoning skills in patients with AIPD when compared to patients with uncomplicated AD. Memory, specifically explicit memory, short term memory and learning, may be impaired by the chronic use of alcohol. Our results do not support the observations of Bleuler [14] of intact memory function for persons suffering from AIPD. On the immediate verbal memory task the AIPD group’s performance was below average while the AD group’s performance fell within the average range. Immediate memory refers to the initial phase of short term memory and theoretically it holds information before the information is registered [40]. Immediate verbal memory impairment is associated with lesions in the dominant perisylvian cortex [47]. Of interest in this regard is that neuroimaging studies suggest that the development of auditory hallucinations in schizophrenia are also associated with altered structural and functional connectivity within the perisylvian language network [48]. It may be, therefore, that similar underlying neurobiological mechanisms are involved in the genesis of auditory hallucinations in AIPD and schizophrenia.

Both groups performed above the cut-off score on the task measuring effort when completing a memory task and the extent to which memory problems are exaggerated which lends further support for more severe memory difficulties in the AIPD group compared to the AD group. On delayed verbal recall, the AIPD group performed in the below average range, while the AD group performed within the average range but the difference did not reach statistical significance. Delayed verbal recall, as measured by the RAVLT, refers to recall of words after a 2–45 min delay [40]. Impaired performance on delayed verbal recall tasks was connected to left hippocampal volume/atrophy in an Alzheimer Dementia [49] for both normal control subjects and subjects suffering from Alzheimer’s Dementia.

With reference to impairment on the task assessing abstract verbal reasoning skills, in the AIPD group, the similarities subtest is reportedly sensitive to cognitive pathology that affect verbal abstraction which is connected to left temporal and frontal involvement [40]. This may partially explain the occurrence of psychotic symptoms and supports frontal and temporal lobe involvement in AIPD [15, 50]. In addition, both groups scored within the average range for word initiation even though the AIPD group’s performance was slightly poorer than that of the AD group.

The findings concur with previously reported research suggesting cognitive impairments such as difficulty with attention and concentration are found in AIPD [23, 24]. The AIPD group performed in the below average range in tasks measuring attention and concentration and the AD group in the low average range. The difference between the two groups was not found to be statistically significant.

Both groups demonstrated ability for new learning with a positive, but fluctuating learning curve. New learning for the AIPD group fell within the below average range while new learning for the AD group was marginally better and was in the average range but the difference also did not reach statistical significance.

On tasks assessing visual memory (immediate and delayed), both groups scored within the average range with the AD group’s score better but again no statistically significant difference recorded. In terms of visuospatial assessment, both groups demonstrated average ability.

It is interesting that the AIPD group’s average of reported drinking severity (SADQ) was lower than that of the AD group. However, this may be misleading as it could be explained by recall bias in the AIPD group due to greater memory impairment, associated impairment of insight or to deliberate under reporting [40].

The study has a number of limitations. It is a cross-sectional study with a small number of participants. The participants were selected from two different clinical populations within one institution. The AD group was selected from a program that requires a certain level of education which increased the possibility of selection bias. The scope of neuropsychological exploration needs to be expanded to include all the major neuropsychological domains, such as attention and concentration, all aspects of visual and verbal memory, speech and language functioning, motor control and all aspects of executive functions. Furthermore, a detailed analysis of predisposing factors needs to be performed to rule out their association with the cognitive deficits. It therefore remains to be determined to what extent the cognitive deficits in AIPD demonstrated in this study suggests pre-existing vulnerability rather than primary effect of alcohol. These limitations, together with the fact that we did not correct for multiple comparisons, mean that our results should be regarded as preliminary.

In conclusion, this study concurs with an earlier [1] that some cognitive deficits are greater in AIPD compared to uncomplicated AD. The AIPD group generally performed poorer on all tasks, but a statistically significant poorer performance was only recorded for tasks assessing immediate verbal memory, delayed verbal recall and abstract verbal reasoning abilities. These findings support the notion that several brain regions [15, 50] and possibly several neurotransmitter systems [6] are involved in the pathogenesis of AIPD. Of particular interest is the the perisylvian language network. Even though the findings of this study imply an association between AIPD and the emergence of cognitive pathology, one has to keep in mind the contribution of pre-existing factors in the development and severity of the cognitive deficits demonstrated in this study. This condition is relatively uncommon; it is often transient and generally has high levels of co-morbidity. Notwithstanding these challenges, further exploration of cognitive deficits possibly associated with other correlates should significantly improve the future understanding of AIPD and perhaps also the cognitive and neurobiological underpinnings of auditory hallucinations in other psychotic disorders.