Verbal learning in schizophrenia in remission, first-degree relatives, and correlation to symptoms

Background: Cognitive impairments are fundamental in schizophrenia with verbal memory impairments commonly occurring not only in patients but also in unaffected genetically susceptible individuals. Deficits in verbal memory produce difficulty in problem-solving, emotional distress, and worsening of daily life skills resulting in a poor quality of life. Objectives: This study aims to evaluate the verbal memory in unaffected first-degree relatives of patients of schizophrenia in comparison to healthy controls as well as patients of schizophrenia in remission and to find correlation to symptom domains. Methods: It was a hospital-based, descriptive, cross-sectional case-control study. Three groups (n=40, each group) of patients, first-degree relatives, and controls were taken. Subjects were screened for mental retardation and remission was ascertained in the patient group by the Positive and Negative Syndrome Scale (PANSS). Rye’s Auditory Verbal Learning (RAVL) test was used to assess verbal leaning. Results: Verbal memory impairments were significant both in patients and unaffected first-degree relatives. Further, these impairments showed a strong correlation to negative symptoms. Conclusion: Significant (p<0.05) verbal learning impairments were noted in patients and first-degree relatives which showed a correlation to negative symptoms.


INTRODUCTION
Eugene Bleuler first described cognitive changes. Recent estimates suggest cognitive impairments to be present in up to 98% of the patients. [1] Clinical impairments arising as a result of cognitive deficits are noted in up to 80% of the patients and the remaining ones also fail to return to premorbid level of functioning. [2][3][4] The appearance and persistence of these deficits are unrelated to positive symptoms. They appear to predict the onset of psychosis. Deficits in verbal memory have been noted in healthy relatives also. [5][6][7][8][9] Medial temporal lobe and prefrontal cortex mediate verbal learning. It is the ability to retain, store, and use information heard. [10] Verbal memory is one of the most affected cognitive domains in schizophrenia. It has been regarded at times as a predictor of antipsychotic response and a determinant of functional outcome. [11,12] Deficits in verbal memory produce difficulty in problem-solving, emotional distress, and worsening of daily life skills. [13][14][15] In schizophrenia, deficits in verbal memory appear before the symptoms start and persist even after symptoms have recovered.
Studies on patients of first-episode psychosis showed that verbal memory continued to deteriorate even on treatment though symptoms improve. [16] This suggests verbal memory to be more of a core with a trait-like dimension. Bilder and colleagues [17] showed a large number of first-episode patients to be having a performance of 1.5 standard deviation (SD) lower than controls. Patients of chronic schizophrenia performed similarly to patients of Alzheimer's, showing impairments in delayed recall as well as delayed recognition. [18,19] Longitudinal studies on verbal memory had varying results. In a large outpatient-based study by Heaton and colleagues, [20] it was noted that visual and verbal memory remained stable at 1.5 and five-year follow-up. However, Gold and colleagues [21] had reported in a similar study that immediate recall improved with time, but deficient in delayed recall failed to improve. Verbal memory impairments were noted to correlate with positive symptoms. [22] Hoff and Kremen [2] followed-up patients annually from the first episode and found patients to be deteriorating progressively, with a performance one to two SD below normal.
Verbal memory impairments show an association with negative symptoms. [5] In a longitudinal study, [22] it was noted that deficits of verbal memory correlated more to symptoms at the onset of illness than after six months of treatment. Chronic patients did not show any benefit in verbal memory with treatment. This sets the need to explore the significance of verbal memory deficits in schizophrenia.

Aims and objectives
On reviewing the literature, the present study intends:

MATERIAL AND METHODS
It was a descriptive cross-sectional, case-control study. The study was conducted at Nil Ratan Sircar Medical College and Hospital, Kolkata in India in 2019. Ethical clearance was provided by the institutional ethics clearance committee by letter No/NMC/3691. Samples were obtained by purposive sampling from the patients, their relatives, and informants of patients coming to the psychiatry department.

Sampling and definition
It was a purposive sampling where all cases who fulfilled inclusion and exclusion criteria were included. Inclusion criteria were subjects between 18-50 years, who were educated in English up to at least till tenth standard and consented to participate in the study. Exclusion criteria were subjects who had comorbid conditions such as drug abuse, except nicotine dependence, affective disorders, dementia, mental retardation, history of head injury and seizure within a one-year duration. The first group was stable patients in remission, the stable schizophrenics. First-degree relatives' group was parents, children, and siblings of those patients of schizophrenia who were admitted to the psychiatry ward or presenting to the outpatient department. Those who fulfilled inclusion and exclusion criteria but did not belong to either of the above groups were controls or third group.

Rating and assessment by tools
A specially prepared proforma was used to collect the patients' sociodemographic data which included age, gender, educational status, residential status, socioeconomic status, past history, family history, personal history, general physical examination, systemic examination, and mental status examination. The World Health Organization's tenth revision of the diagnostic criteria for research of the International Statistical Classification of Diseases and Related Health Problems (ICD-10 DCR) [23] was used to diagnose each patient. Mental retardation was ruled out by screening all subjects by Raven's Colored Progressive Matrices. [24] All subjects were screened by the Positive and Negative Syndrome Scale (PANSS) [25] for assessing the severity of illness in schizophrenia. The accepted criteria for remission as in literature was a PANSS score of less than three for all items in the last six months. [1] General Health Questionnaire-28 (GHQ-28) [26]was used for checking the presence of any psychiatric comorbidity.
Rey's Auditory Verbal Learning (RAVL) test [27] was used for the assessment of verbal learning. In this test, first, a list-A having 15 words are presented successively five times. After reading each time, the subject is asked to recall from memory in any order, though no hint is given. Next, a new list, list-B is presented and asked to recall immediately. This serves as interference list. Later, he has to recall list-A immediately for immediate recall and after a gap of 20 minutes, for delayed recall. Ultimately, a new list having new words mixed with those of list-A is presented and he is asked to identify words of list-A.
PANSS is a 30-item interval rating scale where each item can be rated up to seven. It rates separately positive, negative, and general symptom domains. [25]

Statistical analysis
The data was tabulated and descriptive values like mean and SD were calculated. Epi info7 and Microsoft Excel were used for analysis. Chi-square test was used for categorical variables and analysis of variance (ANOVA) for continuous variables. Pearson's coefficient of correlation was used to find a correlation to symptom domains.

RESULTS
There is no statistically significant difference existing among the groups with respect to age, educational status, occupation, and gender. A significant difference between groups was noted in the area of residence. Hence, groups were matched on age, gender, education, and occupation by design (Table 1).
Four set of scores were obtained (

. Delayed recognition (hits and errors) 4. Long-term retention percentage (LTRP)
Patients and first-degree relatives performed significantly worse than controls. Post-hoc (least significant difference [LDS]) showed first-degree relatives performed similar to patients in remission. However, in delayed recognition, patients performed worse than first-degree relatives. Symptom domains showed minimal symptoms as patients were in remission (Table 3). A strong negative correlation was found between negative symptoms with recall and retention (Table 4).

DISCUSSION
We aimed to study verbal memory in three groups, the remitted patients, first-degree relatives, and controls. The three groups were demographically matching in age and education. The version of RAVL test used is a test modified and standardised for Indian subjects. [27] It is designed to provide four sets of scores. It is a performance test where various tasks involving memory is allotted with fixed time and scoring criteria. Patients of schizophrenia even in remission showed significant impairments in verbal learning with respect to controls. Posthoc LSD showed significant difference existing between groups. Impairments include deficits in immediate recall, delayed recall as well as long-term recall. This indicates that all stages of memory formation, i.e. registration, retention, and recall are affected. Further, this points to the fact that these deficits supersede the symptoms of the illness. [5] Our findings are similar to those noted by Bilder and colleagues, [17] in which a large number of patients in the post-psychotic phase in remission performed 1.5 SD below normal. Similar to previous studies, we observed that treatment could remit symptoms but not verbal memory deficits. [2,16] Asymptomatic first-degree relatives have also been shown to harbour such impairments. [7] This raises the question that does verbal learning impairment possesses the potential of being a screening tool? Addington and Addington [28] compared the nature of verbal learning impairment in firstepisode patients to those with a chronic course and found the deficits to be similar.
We intended to study the correlation of the extent of impairments to the symptom domains. Though positive and general symptoms did not show any correlation, negative symptoms showed a high inverse correlation to verbal learning. Addington and Addington, [28] in their study had also noted a similar finding. However, McDermid Vaz and Heinrichs [22] had noted positive symptoms to be showing a correlation to verbal memory deficits. In their study, patients of schizoaffective disorders were also included in cases. Hence, results may not be exactly comparable. Negative symptoms are the most difficult to treat and are the ones left over resulting in residual deficits. As treatment options for targeting negative symptoms are far less than positive ones, a suitable trait marker for early intervention may be useful for better disability limitations.
These facts taken together point to learning impairments as a trait marker present in the genetically susceptible population more often than the general population; hence, may be regarded as endophenotype. [13] It appears to be the    strongest predictor of functional outcome, causes emotional discomfort, and difficulty in problem-solving. [15] Ultimately, these deficits limit the functional ability and skills needed for activities of daily living. People forget names of things they have been asked to fetch and show the inability to remember the sequence of events they are directed to perform. As a result, occupational and activities of daily living are affected. Screening for these verbal learning impairments could be a part of a routine assessment to aid in disability limitations in the future.

Strengths
1. Instruments standardised for the Indian population were used. 2. Groups were demographically matched by design.

Limitations
Being a hospital-based study, sample size remained a constraint.

Conclusion
Deficits in verbal learning persist in patients of schizophrenia even in remission. Genetically high-risk asymptomatic subjects also harbour such deficits, significantly more often than others. Further, deficits in verbal learning show a strong correlation to negative symptoms.