Neurocognition in first-degree healthy relatives ... - Wiley Online Library

Psychiatry and Clinical Neurosciences 2008; 62: 190–196

doi:10.1111/j.1440-1819.2008.01754.x

Regular Article

Neurocognition in first-degree healthy relatives (siblings) of bipolar affective disorder patients Jitendra K. Trivedi, MD,* Dishanter Goel, MD, Mohan Dhyani, MBBS, Sachin Sharma, Anand P. Singh, PhD, Pramod K. Sinha, MSc, DSQC and Rajul Tandon, MD

MD,

Department of Psychiatry, King George’s Medical University, Lucknow, India

Aim: Cognitive deficits have been presupposed to be endophenotypic markers in bipolar disorder, but few studies have ascertained the cognitive deficits in healthy relatives of bipolar disorder patients. The aim of the present study was to assess the cognitive functions of first-degree relatives of patients with bipolar disorder and compare them with healthy controls. Methods: Ten first-degree apparently healthy relatives of patients with bipolar disorder were compared with 10 age- and education-matched control subjects on computer-based cognitive tests.

IPOLAR DISORDER IS highly heritable; family members of patients with bipolar disorder are at high risk of bipolar disorder and other affective disorders.1 Therefore, a useful endophenotype could be expressed in this population. Healthy first-degree relatives of bipolar disorder patients provide a powerful design to investigate whether cognitive deficits in bipolar disorder reflect state or trait markers.2 Cognitive deficits have been shown to be present in apparently healthy relatives of patients with bipolar disorder and thus they could be potential markers of familial vulnerability to bipolar disorder.3,4 Cognitive deficits seem to be more prevalent in relatives of patients with bipolar disorder type I compared to relatives of bipolar disorder type II patients.5 Research in this area is still lacking.

B

*Correspondence: Jitendra K. Trivedi, MD, Department of Psychiatry, King George’s Medical University, Lucknow, India. Email: [email protected]; [email protected]; drjktrivedi@sancharnet Received 12 June 2007; revised 29 October 2007; accepted 5 November 2007.

190

Results: As compared to the control group, the relatives group performed significantly poorly on tests for executive function and vigilance, while on the test for working memory the performance was not significantly different on most of the parameters. Conclusions: Executive functioning and vigilance could be potential markers of the endophenotype in bipolar patients. Key words: bipolar disorder, endophenotype, familial vulnerability, neurocognition.

A neuropsychological endophenotype for a disorder should be associated with trait dysfunction in affected probands and be present in unaffected firstdegree relatives of probands.6 Cognitive functions as endophenotypes in bipolar disorder have been further established by recent research assessing cognitive functions in unaffected twins discordant for affective disorder.7 Healthy twins discordant for unipolar disorder had lower performance on almost all measures of cognitive function: selective and sustained attention, executive function, language processing and working and declarative memory, and also after adjustment for demographic variables, subclinical symptoms and minor psychopathology. Healthy twins discordant for bipolar disorder had lower performance on tests measuring episodic and working memory, also after adjustment for the above-mentioned covariables.7 Various domains of cognition can easily be measured and are the key factors affecting the subject’s ability to function occupationally, socially and interpersonally. The three cognitive domains of executive

© 2008 The Authors Journal compilation © 2008 Japanese Society of Psychiatry and Neurology


function, working memory and attentional abilities included in the present study are considered to be the most important cognitive domains for daily functioning. The importance of cognition in the daily life of a person and also the large number of psychiatric patients catered for by general practitioners, underscore the importance of this knowledge to all medical personnel and not just to psychiatrists. Many a time such persons are not able to communicate their medical problems to the physicians because of their cognitive limitations, making it frustrating for the doctor. A sympathetic ear and a more humane approach can be rewarding. Furthermore, a cognitively compromised person has limited capacity to bear the stress of any medical or surgical ailment and may be difficult to deal with if the clinician does not understand their limitations. The study of cognitive functions in relatives of bipolar disorder patients has seldom been attempted in India, and therefore the present paper will contribute to the literature in this area. We based the present study on the hypothesis that the healthy relatives of bipolar disorder patients will also have some deficit in neurocognitive functions. The present study aims to assess neurocognitive functions in first-degree healthy relatives of patients with bipolar disorder and to compare them with healthy controls with no family history of any neuropsychiatric disorder, on parameters of executive functioning, working memory and vigilance, which are still under-researched neurocognitive domains in this context.

METHODS ‘The study was a time-limited, single-center study with a case–control design and was carried from March 2005 to December 2005. Subjects were screened consecutively from the adult psychiatry Outpatient Department (OPD) of King George Medical University, Lucknow, on specified OPD days.

Subjects A total of 103 patients, on specified days, from the outpatient section of Department of Psychiatry, King George Medical University, Lucknow, were screened and those patients (n = 73) who were suffering from bipolar affective disorder, without any family history of any other psychiatric disorder including a history of bipolar affective disorder in any other member of

Neurocognition in siblings of bipolar

191

the family, were selected and the diagnosis was reviewed by two independent psychiatrists. The siblings of these patients were screened for selection criteria and informed consent was obtained. A total of 56 siblings were screened, of whom 46 first-degree relatives (siblings) were excluded, either for not fulfilling the selection criteria or for unavailability for assessment. Ten first-degree relatives were chosen, if they were in age group 18–45 years and had received formal education for a minimum of 10 years. If, for any patient, more than one sibling was available then the eldest one within the required age group was chosen for assessment. Only one sibling was taken from each family. They were excluded if they had ever had a history of any psychiatric illness, concurrent major illness or systemic dysfunction (including seizures), history of head injury severe enough to cause unconsciousness, history of substance abuse or dependence, benzodiazepine use or any other medication known to impair cognitive functions. All the subjects were clinically assessed for intellectual capacity; in some cases patients were subject to formal IQ assessment, and those with IQ < 70 were excluded from the study. Subjects were also excluded if they scored >3 on General Health Questionnaire (GHQ). Ten age-, sex- and education-matched controls were selected from the caregivers of the patients who were not the relatives of the patient, so as to provide a control sample as close to the patient group as possible, in their demographic profile. The controls were excluded if they had ever had a history of treatment for any psychiatric illness, and history of any substance abuse or dependence. The presence of psychiatric illness in any of their first-degree relatives was assessed clinically on the basis of history. The presence of any current psychiatric illness was ruled out by clinical assessment and GHQ > 3.

Protocol and procedure Informed consent was obtained from all the subjects and the study was done in accordance with the Indian Council of Medical Research’s Ethical Guidelines for Biomedical Research on Human Subjects (2000).8 All the computer based cognitive tests were used and standardized for the Indian population during a multicenter pilot study (CPRG001, 2002–2004) and a study based on these tests has been published earlier.9 The psychologist who administered the tests


192

J. K. Trivedi et al.

in the present study was formally trained in their usage during the previous multicenter study. The tests were explained to the patients at the start of each one and all possible queries were answered promptly. They were asked to relax and to concentrate on the tests, as much as possible. To accomplish this the tests were administered in a quiet room, with minimal disturbance. Each test was preceded by a mock test series to check the understanding and involvement of the patient in the test. During the tests the patients were not required to do any computer-based operation (except for the Spatial Working Memory Test [SWMT], where the patients had to press a predecided keyboard button in response to a target stimulus). The patients were not required to have any knowledge of computers for the tests. Any queries regarding the tests were addressed without giving them any clue about the possible solutions. In both the groups cognitive assessments were done between 12.00 hours and 16.00 hours for all the subjects; to minimize the confounding factors of drowsiness or withdrawal symptoms the patients were instructed not to take benzodiazepines in the preceding 8 h, other medications, if any, were continued as usual. The tests were conducted in a single sitting for all patients, sequentially with an interval of 15 min between each test. Assessment was done on a semi-structured sociodemographic proforma that listed demographic and personal details, personality assessment, details of physical examination, and mental status examination. The diagnosis of the patients (bipolar disorder type I) was confirmed by two independent psychiatrists using DSM-IV.10 Structured Clinical Interviews for DSM-IV (SCID-I, SCID-II) were applied to all subjects to rule out any subthreshold psychiatric illness or personality disorder. The 17-item Hamilton Depression Rating Scale11 and the Young Mania Rating Scale (YMRS)12 were used to verify the absence of depressive and manic symptomatology, respectively. Physical health was assessed by means of a health questionnaire, a standard physical examination by a physician, and urine screening. The subjects who were included were assessed on computer-based neurocognitive tests;13 these included the Wisconsin Card Sorting test (WCST), the Continuous Performance Test (CPT) and the SWMT. The controls were assessed using the 12-item GHQ14 and computer-based neurocognitive tests (WCST, CPT and SWMT).


All three tests have been used for research purposes and have been well authenticated for the study of cognitive functions. Although they are not used in routine clinical practice, they adequately represent the cognitive functions used in everyday functioning by an individual. In both groups all cognitive assessments were done between 12 pm and 4 pm.

Spatial Working Memory Test In the SWMT, which is a test of memory for spatial locations,13,15 the subject views a brief presentation of black circle on computer screen and then is asked to point to the location of circle after a delay of 0 s and 20 s, randomly. The patient is given a choice of eight possible locations. During the 20-s delay the subject is engaged in a distraction task, which involves continuously repeating in reverse order the 3-digits of the number that appears on the screen. There are a total of 48 responses, 24 at 0 s and 24 at 20 s delay. The result in SWMT is obtained as the number of correct responses and number of non-adjacent errors at 0 s and at 20 s delay, respectively.

Continuous Performance Test Sustained attention, vigilance and impulse control is assessed on the CPT.16–18 The test requires a participant to respond to a specified target when it is presented spontaneously within a stream of interfering visual stimuli. The task involves monitoring a random series of geometrical figures. It requires the subject to distinguish targets from non-targets, an ability known as sensitivity. The results obtained are in terms of correct responses, wrong responses, missed responses and the reaction or response time. In the test, target stimulus is rare in frequency and presentation latency is brief. A total of 328 stimuli are presented, out of which 28 are for practice. Stimulus duration and interstimulus latency are 50 ms and 1000 ms, respectively.

Wisconsin Card Sorting Test The WCST is a measure of executive function requiring the ability to develop and maintain an appropriate problem-solving strategy across changing stimulus conditions in order to achieve a future goal.13,19,20 It is the classical test for dorsolateral prefrontal cortex function.21 There are three sets of stimulus, each having four variations. These stimulus cards reflect three stimulus



Table 1. Sample characteristics Variable Age (years) Sex (%) Male Female Years of education IQ

Neurocognition in siblings of bipolar

193

RESULTS

First-degree relatives (n = 10)

Control group (n = 10)

30.10 ⫾ 11.19

29.30 ⫾ 5.77

90 10 14.40 ⫾ 2.63 102.5 ⫾ 5.40

80 20 16.30 ⫾ 0.82 103.5 ⫾ 5.79

parameters, color, form and number, and are displayed in four different ways: for forms they are crosses, circles, triangles or stars; for colors: red, blue, yellow or green; and numbers: one, two, three or four. These cards are numbered from 1 to 64 on the lower left corner of the reverse side to ensure a standard order of presentation. It can be applied on subjects aged 61/2-89 years.

Data analysis The scores for the three tests in the two groups were statistically analyzed for significance using Mann– Whitney U-test because the sample size of the two groups was small and also because normalcy of the observations cannot be presumed. Mean and standard deviation of the various parameters are given in Table 1. P < 0.05 was considered significant.

Subject characteristics The subject group and the control group were matched for age, sex, education and place of domicile (urban/rural). There was no significant difference between the two groups on age, sex and education or place of domicile (i.e. urban). The control group (n = 10) had a mean age of 29.30 ⫾ 5.77 years, 80% of were male and had mean years of education of 16.30 ⫾ 0.82 years. The first-degree relatives (n = 10) had a mean age of 30.10 ⫾ 11.19 years, 90% of them were male, and had mean years of education of 14.40 ⫾ 2.63 years. Table 2 shows the results for the CPT. The responses obtained between the control group and the first-degree relatives were significantly different on wrong responses and missed responses. The mean differences show that the first-degree relatives committed more wrong responses (mean, 2.00 ⫾ 1.49) and had more missed responses (mean, 1.80 ⫾ 0.92) as compared to the control group. On WCST it was observed that the two groups differed significantly on completed and perseverative errors, suggesting that control subjects completed more categories (mean, 6.00 ⫾ 0.00) and committed less perseverative errors (mean, 5.50 ⫾ 1.90) on WCST, than the first-degree relatives.

Table 2. Results on cognition tests (mean ⫾ SD) Mann–Whitney U-test Variable Spatial Working Memory Test Correct responses at 0 s delay Non adjacent errors at 0 s delay Correct responses at 20 s delay Non adjacent errors at 20 s delay Continuous Performance Test Wrong responses Missed responses Response time Wisconsin Card-Sorting Test Categories completed Perseverative errors

Control group

First-degree relatives

U

P

24.00 ⫾ 0.00 0.00 ⫾ 0.00 23.00 ⫾ 0.67 0.00 ⫾ 0.00

23.60 ⫾ 0.52 0.00 ⫾ 0.00 21.90 ⫾ 1.66 0.20 ⫾ 0.42

30 50 26 40

NS NS NS NS

0.50 ⫾ 0.53 0.40 ⫾ 0.51 0.38 ⫾ 0.07

2.00 ⫾ 1.49 1.80 ⫾ 0.92 0.48 ⫾ 0.14

20 11 33