|
1 | 2
| 3 | 4
Neuroimaging-Assessed Brain Abnormalities
Numerous studies have demonstrated structural brain abnormalities
in schizophrenia (for example, 36). Subjects with schizophrenia
spectrum disorders such as SPD, but without psychosis, show similar,
though often milder, abnormalities (37,38). Structural magnetic
resonance imaging (sMRI) and magnetic resonance spectroscopy studies
demonstrate that first-degree relatives of patients with schizophrenia
who are not selected for SPD (that is, well over 90%) also differ
from control subjects in various ways. For example, such studies
demonstrate enlarged third ventricle (39) and pallidal (40) volumes
and reduced left amygdala (39), right amygdala, hippocampus, putamen,
left thalamus, brainstem (41), cerebellum (40), and overall brain
(39) volumes.
Recently, a few studies have focused on the degree of genetic loading
for schizophrenia. Lawrie and others compared high-risk subjects
(defined as having at least 2 affected first- or second-degree relatives)
with first-episode patients and healthy control subjects and found
abnormalities that were similar but not identical to those in patients
(42). Among these, the relatives showed amygdalo-hippocampal and
thalamic regions (bilaterally) that were small, compared with control
subjects, but large, compared with patients. Seidman and others
compared adult, first-degree relatives without psychosis who had
1 (that is, ‘simplex’) or 2 (that is, ‘multiplex’)
affected relatives with relatives who had schizophrenia and with
normal control subjects (43). Generally consistent with the findings
of Lawrie and others, the results showed that within families, nonpsychotic
relatives—and particularly those from multiplex families—showed
significantly smaller left hippocampi that did not differ from those
of patients. Moreover, measures of verbal memory were significantly
correlated with left hippocampal volumes.
Another variable related to the degree of risk for schizophrenia
involves pregnancy and obstetric complications (44,45). Interestingly,
documentation of fetal hypoxia predicted reduced gray matter and
increased cerebral spinal fluid in patients and in their nonpsychotic
relatives, but not in control subjects (46). These findings underscore
the importance of environmental factors in producing not only schizophrenia
but also the predisposition to schizophrenia.
In addition to sMRI studies, Seidman and others demonstrated in
a functional MRI study that adult nonpsychotic relatives of schizophrenia
patients were significantly impaired on working memory tasks with
interference, compared with normal control subjects (47). In both
groups, the tasks produced activation in the lateral and medial
frontal cortex, posterior parietal and prefrontal cortex, and thalamus.
Compared with control subjects, however, the relatives showed a
greater number of extraneous, bilateral activations on tasks in
which they performed poorly.
Neuropsychological Deficits
As in the domains described above, nonpsychotic relatives of schizophrenia
patients demonstrate multiple deficits in cognition that are similar
to those seen in schizophrenia sufferers (2,29,48–50). Individuals
with SPD, for example, show deficits in nonverbal learning, and
especially in verbal learning (51), in working memory (52), and
in several executive functions, including concept formation, abstraction,
and mental flexibility (53). Similarly, relatives of patients who
do not have SPD show dysfunction in several cognitive domains, including
motor and perceptual-motor ability, short-term memory, working memory,
learning and recall, verbal and language skills, sustained attention,
and executive function (2). Several of these deficits coexist in
relatives but not in control subjects (54), are stable over a 4-year
follow-up period (5), are more prominent in multiplex relatives
than in simplex relatives (55), and, when present in childhood (for
example, dysfunction in motor skills, verbal memory, and sustained
attention), predict schizophrenia-related psychosis in adulthood
(56).
Psychosocial Functioning
Poor social functioning is a common finding in child, adolescent,
and adult relatives of schizophrenia patients. Compared with control
subjects, child relatives demonstrate poor social functioning and
restricted interests (57), social incompetence and aggression (58),
and shyness, withdrawal, and antisocial behaviour (59,60). In the
Danish high-risk study, child relatives were described by teachers
as socially isolated, passive, less socially competent, and aggressive;
they were described by mothers as both passive and aggressive and
by peers as more aggressive, withdrawn, and unlikable (58).
Childhood social deficits increase between childhood and early
adolescence and continue through adolescence (61). Results from
the Jerusalem Infant Development Study (JIDS) indicate that adolescents
who have a parent with schizophrenia show poor social adjustment
not related to concurrent onset of schizophrenia (or another disorder)
(62). Social difficulties observed in adolescents include increasingly
poor functioning at work and at school, a decreased number of friends,
immaturity, a lack of popularity with peers, poor peer engagement
(especially with members of the opposite sex), and a lack of dating.
Adult relatives without psychosis also demonstrate difficulties
in social functioning. Toomey and others showed that relatives of
schizophrenia patients exhibit deficits in perception of nonverbal
social cues when assessed with the Profile of Nonverbal Sensitivity
test (PONS), compared with control subjects, (63).
Taken together, these lines of evidence support the hypothesis
that some relatives in schizophrenia families have a clinically
meaningful, familiarly transmitted syndrome or set of traits—schizotaxia—that
includes negative symptoms, psychophysiological abnormalities, neuroimaging-assessed
brain abnormalities, neuropsychological deficits, and psychosocial
impairments. While some of these deficits may be present in other
schizophrenia-related disorders such as SPD, they also exist independently
(see [2] for further discussion of similarities and differences
between SPD and schizotaxia).
Definition and Validation of Schizotaxia
The identification of schizotaxic features raises the issue of
whether they can be used to select preschizophrenia children for
primary prevention protocols. At present, the answer is no. Although,
as noted above, studies of children at risk for schizophrenia show
that schizotaxia symptoms predict schizophrenia and related disorders
(50,56), more work is needed to determine which individuals will
or will not develop psychiatric problems. Even less is known about
moderating variables that might serve as protective factors to mitigate
the effects of vulnerability factors (64). The notion of protective
factors is interesting and refers to variables that actively reduce
risk, as opposed to the simple absence of variables that confer
it. Eventually, an understanding of how risks and protective factors
combine to produce an overall level of risk, consistent with a diathesis
stress model (1), will be used to tailor individual treatment plans.
Until then, we can pursue at least 2 productive lines of inquiry.
One of these involves the validation of schizotaxia as a syndrome,
and the other involves the schizotaxia intervention protocol to
evaluate treatments.
Defining Schizotaxia
The first step is to define a syndrome of schizotaxia. Tsuang and
others developed preliminary research criteria for schizotaxia based
on a combination of negative symptoms and neuropsychological deficits
(4). These criteria reflect only a subset of the symptoms described
above. They were selected because evidence for abnormalities in
these areas is well established in relatives of schizophrenia patients
and because they establish clearly different dimensions to the syndrome.
This is particularly important to minimize the likelihood of falsely
classifying individuals as schizotaxic (that is, as phenocopies).
Nevertheless, the initial criteria are tentative, because schizotaxia
is an evolving concept.
In our initial study, subjects who met preinclusion criteria were
first-degree relatives of patients with schizophrenia, spoke English
as a first language, had estimated IQ scores of at least 70, were
aged 19 to 50 years (the age range was partly related to treatment
administration), and provided informed consent. Exclusion criteria
were designed to minimize the influence of comorbid neurological,
psychiatric, or other medical conditions that could mimic schizotaxia
symptoms (for example, head injuries, current substance abuse, or
history of electroconvulsive treatments). We excluded individuals
with any lifetime history of psychosis.
Criteria for schizotaxia were met if a subject was shown to have
at least moderately severe negative symptoms and neuropsychological
deficits. We used fairly stringent criteria as another hedge against
false-positive classification. We assessed negative symptoms using
the Scale for the Assessment of Negative Symptoms (SANS) (65). The
neuropsychological assessment focused on 3 cognitive domains, including
vigilance–working memory, long-term verbal declarative memory,
and executive functions. Moderate or greater deficits (defined as
approximately 2 or more standard deviations below appropriate norms
in 1 domain and at least 1 standard deviation below average in a
second domain) were required in at least 2 of the 3 cognitive domains
to meet the neuropsychological criteria. In each domain, specific
cut-off scores on particular tests were used to assess whether cognitive
criteria were met.
Validating Schizotaxic Criteria
The next step is to validate the proposed syndrome or set of traits.
Consistent with criteria proposed by Robins and Guze (66), it will
be necessary to assess the validity of schizotaxia with converging
evidence from multiple domains. In this context, 3 lines of evidence
support the validity of the syndrome. First, we recently obtained
concurrent validation of schizotaxia by comparing subjects who met
our criteria for schizotaxia with those who did not on independent
measures of clinical function (67). These measures included the
DSM-IV Global Assessment of Functioning Scale (GAF) (68), the Social
Adjustment Scale (SAS) (69), the Symptom Checklist-90-R (SCL-90-R)
(70), and Chapman’s Physical Anhedonia Scale (PAS) (71). On
each of these scales, the schizotaxia subjects showed poorer clinical
or social function, regardless of whether they were rated by the
subjects themselves or by the investigators (blindly). Differences
between groups were not attributable to age, IQ, education, parental
education, family genetic loading, sex, or comorbid psychiatric
disorders. Thus, subjects with schizotaxia as it is currently defined
are not impaired globally. Instead, they have clinically meaningful
symptoms in circumscribed areas of function.
The second line of support for the validity of schizotaxia involves
the response to intervention. If our conceptualization is correct,
then treatments that attenuate symptoms in schizophrenia might also
attenuate symptoms of schizotaxia. Of the 8 subjects who met our
criteria for schizotaxia, 6 agreed to receive a brief 6-week trial
of low-dose risperidone (up to 2.0 mg daily) (72). Medication side
effects were temporary and mainly mild, and all 6 subjects completed
the intervention protocol. Based on subjective assessments, 5 of
the 6 individuals reported increased cognitive abilities during
the risperidone trial, and 3 reported greater interest in and enjoyment
of social activities. Objective assessments demonstrated that 5
out of 6 subjects showed reduced SANS scores, particularly in the
Anhedonia Asociality section. Of those 5, 3 showed moderate reductions
in their scores (approximately 50%), and 2 showed milder reductions
(approximately 25%). Of the 6 subjects, 5 also showed substantial
improvements in attention and working memory. On one test of verbal
learning and memory, 5 out of 6 subjects showed better learning,
although this gain was not reflected on a similar test and was not
reflected by better recall over a longer period of time. Overall,
these findings are encouraging, although larger, double-blind studies
are required to determine whether the initial findings can be replicated.
A third line of support for the validity of schizotaxia involves
one component of the syndrome (negative symptoms), because the other
component (neuropsychological functioning) was not assessed. We
drew a sample from the National Institutes of Mental Health (NIMH)
Genetics Initiative for Schizophrenia study (73,74). This multisite
collaborative study of the genetics of schizophrenia involved 71
pedigrees that contained 218 nuclear families and 987 individuals.
Twenty-nine pedigrees (343 individuals) were of African-American
descent, and 42 pedigrees (644 individuals) were of European-American
descent. Families were recruited systematically, based upon the
DSM-III-R definition of schizophrenia. Each family had 1 member
with schizophrenia, and at least 1 other member with either schizophrenia
or schizoaffective disorder (depressed type). In our sample, we
found that if all subjects with schizophrenia, schizophrenia-related
diagnoses (for example, SPD), or other disorders that include psychosis
were excluded, at least 1 subgroup remained, characterized by negative
symptoms (Stone, Wilcox, Faraone, Tsuang, unpublished). The presence
of negative symptoms in a portion of unaffected relatives adds further
to the validity of the schizotaxia syndrome, as currently defined.
1 | 2
| 3 | 4
|
