Elsevier

Biological Psychiatry

Volume 45, Issue 5, 1 March 1999, Pages 639-646
Biological Psychiatry

Original Articles
Neuropsychological performance of monozygotic twins discordant for bipolar disorder

https://doi.org/10.1016/S0006-3223(98)00148-6Get rights and content

Abstract

Background: A paradigm that involves cognitive assessment of monozygotic (MZ) twins discordant for a neuropsychiatric disorder (here bipolar illness) allows for the examination of both disease-specific impairments (in the comparison of affected to unaffected twins) and risk factors (in the comparison of unaffected twins to normal twins).

Methods: Neuropsychological functions were evaluated in seven MZ twin pairs discordant for bipolar illness and seven pairs of normal MZ twins in an attempt to highlight cognitive abilities associated with manifestations of disease and genetic risk factors. At the time of testing, 3 of the affected twins were euthymic, 2 had depressive symptoms, and 2 had manic symptoms; all were receiving medication. All twins receive neuropsychological tests to evaluate intelligence, attention, visuospatial skills, language, learning and memory, and problem solving.

Results: Statistical analyses revealed that the affected twins were significantly impaired as compared to the unaffected (and normal) twins on some measures of visuospatial functioning and some verbal memory measures. In contrast to a sample of MZ twins discordant for schizophrenia studied previously, the cognitive impairments we observed in bipolar twins were mild in nature and fairly circumscribed. The unaffected twins performed significantly worse than normal controls on a Brown–Petersen memory task, verbal list learning, and overall Wechsler Memory Quotient.

Conclusions: These data suggest that while some visuospatial deficits and verbal memory deficits may be features of bipolar disorder related to disease parameters, mild attenuations in overall memory or retrieval function may be related to genetic factors associated with the illness.

Introduction

Numerous investigators have described various types of cognitive deficits in bipolar illness, but the data have varied in consistency. For example, findings of global intellectual impairment in this population have varied qualitatively from one study to another Goldberg et al 1993, Donnely et al 1982. Attentional dysfunction, as demonstrated by impaired performance on measures such as digit span and continuous performance tasks, has been more consistently reported Coffman et al 1990, Goldberg et al 1993, Goodwin and Jamison 1989. Additionally, some studies have demonstrated deficits in “right hemisphere” functioning, such as impairments of visual perception, spatial relations, and complex nonverbal tasks Coffman et al 1990, Robertson and Taylor 1985. Other investigators have shown deficits in verbal learning and memory that are more profound in bipolar than either unipolar subjects or normal control subjects Dupont et al 1990, Gruzelier et al 1988, Wolfe et al 1987. Finally, some studies have demonstrated that bipolar subjects have mild impairments on measures of abstract and logical reasoning Goldberg et al 1993, Coffman et al 1990, Morice 1990, Savard et al 1980. In general motor functions, praxis, and language abilities have been found to be intact. Several studies also suggest that cognitive dysfunction in bipolar patients persist after improvement in psychiatric symptoms Altshuler 1993, Coffman et al 1990, Sapin et al 1987, Savard et al 1980, Friedman et al 1977. It is unclear whether this finding reflects a disease residual or is a stable trait marker.

In the present study we evaluated cognitive dysfunction associated with bipolar illness by comparing the neuropsychological performances of seven monozygotic (MZ) twin pairs discordant for the disorder. This is a unique sample in that genetic and psychosocial variances are controlled; thus, differences between the affected from the unaffected twin will likely represent cognitive dysfunction related to the intercession of disease.

It is also meaningful to compare the unaffected twin of the discordant pairs to normal MZ twins to evaluate specific cognitive processing deficits that may be genetically linked to the disease. From this perspective, there are three possible outcomes. First, the unaffected twins may perform at a level significantly above the affected group and equivalent to the normal group. With this result, abnormalities in the affected twin would be considered to be associated with the intercession of the disease per se. Second, the performance of the unaffected and affected twins may be equivalent and significantly different from the normal twins. This pattern would suggest abnormalities associated with markers of genetic risk and not the disease per se. Finally, the performance of the unaffected twin may fall between the affected and normal MZ groups. Neuropsychological measures consistent with this pattern would then be considered to reflect genetic risk as well as an additional factor that is associated with the actual development of disease symptomatology.

A final impetus for this study was as a comparison to earlier studies of MZ twins discordant for schizophrenia (Goldberg et al 1990, in press; Torrey et al 1994). Goldberg and colleagues found that the affected schizophrenic twins performed worse than their unaffected cotwins on cognitive measures of attention, response speed, memory, and executive function. Furthermore, the cognitive performance of their unaffected discordant group fell between the affected and normal groups for certain measures; differences for selected variables reached significant or near significant levels between the unaffected and normal group. In the present study, we evaluated seven MZ twin pairs discordant for bipolar illness and seven normal MZ twin pairs on a wide range of neuropsychological tests that measured general intelligence, attention, language, visuospatial functioning, learning and memory, and abstract problem solving. To determine the specific effects associated with the intercession of the disease, we directly compared the performances of the affected to the unaffected cotwins in the sample discordant for bipolar disorder. Additionally, to begin to examine genetic risk factors associated with bipolar illness, we compared the discordant twins to a sample of healthy normal MZ twins. Given our small sample of seven MZ twin pairs, our analyses are exploratory and our results heuristic rather than definitive.

Section snippets

Subjects

Seven pairs of MZ twins discordant for bipolar disorder were studied. Mean years discordant for illness was 14.9. One pair was male and six pairs were female. DSM-III-R diagnoses of all twins were based on psychiatric interviews using the SCID-R (Spitzer et al 1988) and a review of history and clinical records. Affected twins were in various symptomatic states of their disease at the time of testing; 3 were euthymic, 2 were depressed, and 2 were manic. One twin had a sister who also had bipolar

Comparisons of discordant MZ twins

As can be seen in Table 2, the mean scores for the estimated Wechsler Adult Intelligence Scale–Revised (WAIS-R) and Wide Range Achievement Test–Revised (WRAT-R) reading subtest, measures of current and premorbid intellectual abilities, did not differ significantly between the MZ twins discordant for bipolar disorder. Hence, there was no evidence of intellectual deterioration in the affected twins.

As listed in Table 2 and illustrated in Figure 1, matched pair t tests and effect sizes indicated

Discussion

In this study of cognitive function in bipolar disorder we used a paradigm that attempted to reduce genetic and psychosocial variance between comparison groups by testing pairs of MZ twins discordant for the disorder. We found that the affected twins showed specific deficits in select visual processing measures and short- and long-term verbal learning and memory as compared to their unaffected cotwins and a control group of normal MZ twins. It is important to note that all groups had equivalent

References (27)

  • R.M Dupont et al.

    Subcortical abnormalities detected in bipolar affective disorder using magnetic resonance imaging

    Arch Gen Psychiatry

    (1990)
  • M Friedman et al.

    On the safety of long-term treatment with lithium

    Am J Psychiatry

    (1977)
  • T.E Goldberg et al.

    Neuropsychological assessment of monozygotic twins discordant for schizophrenia

    Arch Gen Psychiatry

    (1990)
  • Cited by (156)

    • Deficits in generalized cognitive ability, visual sensorimotor function, and inhibitory control represent discrete domains of neurobehavioral deficit in psychotic disorders

      2021, Schizophrenia Research
      Citation Excerpt :

      Snitz et al. (2006) reviewed 58 studies of cognitive performance in unaffected relatives of schizophrenia probands and found cognitive impairments compared to healthy controls over a diverse array of cognitive tasks in the small to medium effect size range, consistent with findings in the B-SNIP study (Hill et al., 2013). This was consistent with literature showing mild cognitive dysfunction across several domains in unaffected relatives of schizophrenia probands (Hill et al., 2008a) and bipolar disorder with psychosis probands (Gourovitch et al., 1999; Bora et al., 2009). Sensorimotor disturbance and cognitive deficits have also been well established in psychotic disorders using eye movement tasks.

    • Neurocognitive findings in youth at high risk for bipolar disorder: Potential endophenotypes?

      2018, Bipolar Disorder Vulnerability: Perspectives from Pediatric and High-Risk Populations
    View all citing articles on Scopus
    View full text