- Email: [email protected]
Cognitive dysfunctions in parents of schizophrenic patients parallel the deficits found in patients
Schizophrenia Research 63 (2003) 285 – 293 www.elsevier.com/locate/schres
Cognitive dysfunctions in parents of schizophrenic patients parallel the deficits found in patients Melanie C.M. Appels a,*, Margriet M. Sitskoorn a, Paul Westers b, Elleke Lems a, Rene´ S. Kahn a a
Department of Psychiatry, University Medical Center, PO Box 85500, 3508 GA, Utrecht, The Netherlands b Center for Biostatistics, Utrecht University, Utrecht, The Netherlands Received 8 August 2001; accepted 15 June 2002
Abstract Schizophrenia is characterized by a global cognitive impairment, with varying degrees of deficit in all ability domains. Since genetic factors are important in the etiology of schizophrenia we investigated whether parents of schizophrenic patients also show cognitive deficits, particularly on those cognitive ability domains that are most severely affected in patients. Both biological parents of 37 patients with schizophrenia (N = 74 subjects) and 28 comparable healthy married control couples (N = 56 subjects) were included. A comprehensive and standardized cognitive battery was used including tests measuring verbal memory, executive functioning, language, attention, and psychomotor functioning. Parents of patients differed from control couples on those cognitive constructs that are generally considered to be most impaired in schizophrenic patients, i.e. global verbal memory, bilateral motor skill, continuous performance, and word fluency. In addition, parents differed significantly from control couples on some other cognitive constructs on which patients show a smaller but also significant difference compared to healthy controls, i.e. unilateral motor skill and digit span. Results suggest that the cognitive constructs on which patients show relatively most severe impairment may prove suitable as endophenotypic markers in schizophrenia. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Schizophrenia; Parents; Cognition; Genotypical marker
1. Introduction Family, twin and adoption studies suggest an important genetic role in the etiology of schizophrenia (Cardno et al., 1999; McGuffin et al., 1995). However, the mode of transmission of the disorder is still unknown (Tsuang et al., 1999). Identification of *
Corresponding author. Tel.: +31-30-2506025; fax: +31-302505443. E-mail address: [email protected] (M.C.M. Appels).
relatives of patients who carry a predisposing genotype is therefore not yet possible. One attempt to resolve this issue is to search for endophenotypic markers: characteristics that mark the presence of a genetic predisposition to schizophrenia (Faraone et al., 1999a). Apart from identifying individuals at risk, endophenotypic markers might also facilitate linkage analyses aimed at identifying the genetic loci involved in schizophrenia. Since cognitive dysfunctions are prominent features of schizophrenia, they might be suitable candi-
0920-9964/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. doi:10.1016/S0920-9964(02)00342-0
286
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293
dates as endophenotypic markers (Egan et al., 2000). Results from a quantitative review, which includes 7420 schizophrenic patients and 5865 healthy controls divided over 204 studies, show that schizophrenia is characterized by a general cognitive impairment, with varying degrees of deficit in all ability domains (Heinrichs and Zakzanis, 1998). In 7 of the 22 cognitive ability domains, the magnitude of the difference between schizophrenic patients and healthy controls as expressed in mean effect size (Md), was greater than 1. The greatest difference was found on global verbal memory (Md = 1.41). Global verbal memory was followed in effect size by bilateral motor skill (Md = 1.30), performance IQ (Md = 1.26), continuous performance (Md = 1.16), word fluency (Md = 1.15), attention as measured with the STROOP (Md = 1.11), and WAIS-R IQ (Md = 1.10). The cognitive ability domains that are most abnormal in schizophrenic patients might be the best candidates for endophenotypic markers. For these deficits to be considered suitable as endophenotypic markers for schizophrenia, they should also be present in relatives of patients. Most studies that investigated cognitive differences between relatives of patients and controls reported more cognitive dysfunctions in relatives of patients than in healthy controls (see, for reviews, Kremen et al., 1994; Faraone et al., 1995; Green, 1998; Thaker, 2000). However, it is not clear if the relatives of patients show deficits in specifically those cognitive ability domains that are most affected in schizophrenic patients. No quantitative review of the literature on cognitive test performance in relatives of schizophrenic patients exists, and results of individual studies are contradictory (see Kremen et al., 1994; Faraone et al., 1995). Furthermore, not all types of relatives of schizophrenic patients have been extensively investigated. So far, relatively few studies concentrated on the difference between parents of patients and controls (see Docherty, 1994; Harris et al., 1996; Dollfus et al., 2002). As far as we know, the difference in general cognitive functioning between both parents of schizophrenic patients and healthy control couples has not been investigated. Studying the parents of schizophrenic patients has an important advantage over studying other types of first-degree relatives of patients. Parents of patients have passed the age of risk for developing schizo-
phrenia. Therefore, the status with regard to schizophrenia can be determined with more confidence in parents than in other types of first-degree relatives. It is important to study both parents of patients since it is not clear whether schizophrenia is due to unilineal inheritance (only one parent is expected to carry the schizophrenia genotype) or to bilineal inheritance (both parents are expected to carry the schizophrenia genotype). By including both parents of a proband, one can ensure the inclusion of the obligate carrier(s) and one limits the risk of assessing only the healthiest parent. Furthermore, by comparing the parents of patients with healthy control couples, one can control for assortative mating. Assortative mating denotes a tendency for mated pairs to be more similar for some phenotypic traits than would be the case if the choice of a partner occurred at random. Assortative mating occurs for a variety of physical and psychological traits, including mental illness (Parnas, 1988; Mathews and Reus, 2001). In the present study, we investigate the difference between both biological parents of schizophrenic patients and healthy control couples on several cognitive ability domains. We hypothesize that parents of schizophrenic patients have a general cognitive impairment. Furthermore, we hypothesize that the difference between groups will be most pronounced on those cognitive ability domains that in a comprehensive meta-analysis (Heinrichs and Zakzanis, 1998) have been found to be most affected in schizophrenic patients themselves.
2. Methods 2.1. Subjects Thirty-seven biological parent couples (N = 74 subjects) of patients with schizophrenia and 28 healthy married control couples (N = 56 subjects) were included. The diagnosis of DSM-IV schizophrenia in the children of the parents was determined by using the Comprehensive Assessment of Symptoms and History (CASH) (Andreasen et al., 1992) or a Structural Clinical Interview DSM axis I interview (SCID interview) (First et al., 1997). Healthy controls and parents of patients were screened using the CASH interview, the ‘‘Schizophre-
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293
nia and Affective Disorder Schedule—Lifetime interview’’ (SADS-L interview) (Endicott and Spitzer, 1978), the Structured Interview for DSM-IV Personality Disorders (SIDP-IV) (Pfohl et al., 1995), and the Family Interview for Genetic Studies (FIGS) (National Institute of Mental Health, 1991). A history of neurological illness resulted in the exclusion of parents of patients and healthy controls, as did a history of a psychotic disorder and substance abuse. In addition, controls were excluded if they had a history of a depressive/manic syndrome, or a personality disorder. Controls were also excluded if a depressive, manic or psychotic illness was present in their first-degree relatives or if a history of psychotic illness was present in their second-degree relatives. Written informed consent was obtained from all subjects after full explanation of the study aims and procedures.
Table 1 Cognitive constructs and tests used
2.2. Demographic measures
Unilateral motor skill
The current IQ of the subjects was estimated with the aid of a short form of the Groningen Intelligence Test (GIT) (Luteyn and Van der Ploeg, 1983). The handedness of the subjects was assessed by means of the Edinburgh Handedness Inventory (Oldfield, 1971). The scoring system of Verhage (Verhage, 1983) was used to define a subject’s level of education. 2.3. Neuropsychological evaluation The cognitive constructs and cognitive tests used are described in Table 1. In order to compare our results with the results of Heinrichs and Zakzanis (1998), we tried to organize our cognitive tests as much as possible into the same cognitive constructs as they did. Trained predoctoral and postdoctoral examiners administered the cognitive tests. The examiners were supervised by a neuropsychologist. 2.4. Statistical analyses Data were analyzed using the Statistical Package for Social Sciences (SPSS), version 9.0. Demographic data were compared by ANOVA (age) or a Mann – Whitney U-test (level of education, handedness, IQ). In most cases, only one test score represented the score on the cognitive construct. In three cases (Word
287
Test or construct
Recorded test variablesa
Global verbal memory Bilateral motor skill
CVLT: trials I – V recall Purdue Pegboard test: bilateral score CPT-IP test: dprime of numbers (4 digits) fast CPT-IP test: dprime of shapes fast Category Fluency test (animals and professions), Letter Fluency (letters N and A) CVLT: long-delay free recall, consolidation, number of intrusions, discrimination index WCST: perseverative errors, perseverative responses, categories completed, percent perseverative errors Finger Tapping test: preferred hand score, non-preferred hand score Purdue Pegboard test: preferred hand score, non-preferred hand score Trail Making Test: part A Trail Making Test: part B Trail Making Test: part B – part A WAIS: Digit forward WAIS: Digit backward
Continuous performance, verbal version Continuous performance, spatial version Word fluency
Selective verbal memory
Wisconsin card sorting
Trail making—part A Trail making—part B Trail making—part B – A Digit forward Digit backward
a CVLT = California Verbal Learning Test (Mulder et al., 1996); Purdue Pegboard test (Tiffin, 1987); CPT-IP (Cornblatt et al., 1988); Category Fluency (Luteyn and Van der Ploeg, 1983); Letter Fluency (Benton and Hamsher, 1976); WCST = Wisconsin Card Sorting Test (Heaton et al., 1993); Finger Tapping test (Reitan and Wolfson, 1985); Trail Making Test (Reitan, 1992); WAIS Digit span (Stinissen et al., 1970).
Fluency, Wisconsin Card Sorting and selective verbal memory), more than one test score belonged to the same cognitive construct. In order to calculate scores on these cognitive constructs, test scores were rescaled to obtain z-scores using the means and standard deviation of the controls as norm scores. The z-scores were adjusted for sex by using separate means and standard deviations for males and females. We did not cluster the Purdue Pegboard test with the Finger Tapping test because the preferred as well as the non-preferred hand score of these tests do not corre-
288
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293
late significantly (respectively r = 0.066; p = 0.473 and r = 0.138; p = 0.137). Cognitive differences between groups (parents of patients versus control couples) were calculated with a repeated measurement analysis if tests or z-scores were normally distributed and the Levene’s test of equality of error variances was not significant. Group was treated as a between factor. Sex was treated as a within-factor to control for the dependency of mother and father within each couple. If tests or z-scores were not normally distributed, the mean value of the score of each couple was calculated and used in a Mann – Whitney U-test. The mean value was used to take the dependency between husband and wife into account. In case of a missing value, the whole couple was excluded from the analysis. If a significant difference between groups was found on a cognitive construct, additional analyses (an ANOVA or a Mann – Whitney U-test) were done to investigate if parents of patients who carry a psychiatric diagnosis differ from the parents of patients who do not carry such a diagnosis. Two-tailed tests were used for all analyses and significance levels were set at p < 0.05.
Table 2 Demographic characteristics of parents of patients and healthy controls
3. Results
3.2. Cognitive differences between groups
3.1. Demographic characteristics of the sample
On global verbal memory, the bilateral motor test, the verbal version of the continuous performance test, identical pairs version (CPT-IP), Word Fluency, the dominant and non-dominant hand score of the Purdue Pegboard test and the digit forward test, parents of patients had a significantly lower score than healthy control couples. On the spatial version of the attention test, the selective verbal memory test, the Wisconsin Card Sorting test, the Trailmaking test, the Finger Tapping test (dominant and dominant score) and the digit backward test, parents of schizophrenic patients did not differ significantly from healthy controls (see Table 3). Additional analyses revealed no significant differences between the parents of patients who carried psychiatric diagnosis and the parents of patients who did not carry such a diagnosis on the dprime variable of the verbal attention test ( F = 0.272; df = 1; p = 0.60), on the Purdue Pegboard test preferred hand ( F = 0.128;
The groups were comparable with respect to age, handedness, IQ and level of education (see Table 2). Of the 74 parents of patients included, 26 carried a psychiatric diagnosis according to DSM-IV criteria and/or Research Diagnostic Criteria (RDC). Most of them had a diagnosis of depression (15 out of 26). Three persons with an axis 1 depressive disorder had an additional disorder (one a depressive disorder including a labile personality disorder, one a depression including a hypomanic disorder, a labile personality disorder and a generalized anxiety disorder, and one a depression including a panic disorder). Two parents of patients had a dysthymic disorder (one a dysthymic disorder including a labile personality disorder, a panic disorder and a generalized anxiety disorder). One parent had a hypomanic disorder. Two parents a manic –depressive disorder. One parent had a post-traumatic stress disorder, including a panic
Biological parents Healthy Between-group of schizophrenic controls differences patients (N = 56 persons) (N = 74 persons) Agea
M = 53.59 (S.D. = 4.06) Current IQb M = 115.38 (S.D. = 14.69) Handednessb M = 0.81 (S.D. = 0.36) Level of Median = 5.00 educationb (14 years)
M = 53.71 (S.D. = 4.59) M = 119.68 (S.D. = 11.91) M = 0.76 (S.D. = 0.37) Median = 6.00 (15years)
df = 1, p = 0.87 Z = 1.57, p = 0.12 Z = 1.40, p = 0.16 Z = 1.14, p = 0.25
M = mean. a ANOVA. b Mann – Whitney U-test.
disorder and a generalized anxiety disorder. One parent had an avoidant personality disorder and a generalized anxiety disorder. Two parents had a simple phobia, and one a generalized anxiety disorder. One parent had a gambling disorder. Only one control person carried a psychiatric diagnosis, this was a simple phobia according to the RDC criteria.
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293
289
Table 3 Differences between parents of schizophrenic patients and healthy control couples Test or construct
Controls (mean and S.D.)
Parents of patients (mean and SD)
Difference between groups
Global verbal memory Bilateral motor skill Continuous Performance, verbal version Continuous Performance, spatial version Word Fluency (z-score) Selective Verbal Memory (z-score) Wisconsin Card Sorting (z-score)
M = 51.75, S.D. = 9.47 M = 24.07, S.D. = 3.04 M = 1.67, S.D. = 0.74 M = 1.59, S.D. = 0.69 M = 0.00, S.D. = 0.79 M = 0.01, S.D. = 0.76 M = 0.000, S.D. = 0.93
M = 47.75, S.D. = 9.66 M = 22.64, S.D. = 3.34 M = 1.35, S.D. = 0.65 M = 1.51, S.D. = 0.66 M = 0.275, S.D. = 0.771 M = 0.228, S.D. = 0.779 M = 0.08, S.D. = 1.06
( F = 6.168; df = 1,60; p = 0.016) ( F = 6.405; df = 1,61; p = 0.014) ( F = 7.19; df = 1,58; p = 0.010) ( F = 0.316; df = 1,57; p = 0.576) (Z = 2.053; p = 0.040) ( F = 2.713; df = 1,58; p = 0.105) ( F = 0.202; df = 1,62; p = 0.655)
Unilateral motor skill Fingertapping test, preferred hand Purdue Pegboard test, preferred hand Fingertapping test, nonpreferred hand Purdue Pegboard test, nonpreferred hand
M = 46.51, M = 14.65, M = 41.16, M = 14.50,
M = 44.42, M = 13.75, M = 38.88, M = 13.18,
( F = 0.137; df = 1,61; p = 0.137) ( F = 9.652, df = 1,61; p = 0.003) ( F = 4.00; df = 1,59; p = 0.050) ( F = 13.935; df = 1,61; p = 0.000)
Trail Making—Part A Trail Making—Part B Trail Making—Part B – A Digit forward Digit backward
M = 32.63, S.D. = 9.99 M = 63.58, S.D. = 15.68 M = 30.96, S.D. = 15.24 M = 6.79, S.D. = 2.09 M = 6.63, S.D. = 1.79
S.D. = 8.69 S.D. = 1.51 S.D. = 7.53 S.D. = 1.70
df = 1; p = 0.722), Purdue Pegboard test both hands ( F = 1.242; df = 1; p = 0.269) word fluency ( F = 0.129; df = 1; p = 0.720) digit forward (Z = 0.798, p = 0.425), and global verbal memory ( F = 0.094; df = 1; p = 0.760). There was a significant difference between the parents of patients who carried psychiatric diagnosis and the parents of patients who did not carry such a diagnosis on the Purdue Pegboard test non-preferred hand ( F = 4.219, df = 1; p = 0.044).
4. Discussion In the present study, we compared the performance between both parents of schizophrenic patients and healthy control couples on several cognitive function domains. Parents of schizophrenic patients were more impaired than healthy control couples on global verbal memory, bilateral motor skill, the verbal version of the CPT test, word fluency, digit forward and unilateral motor skill measured with the Purdue Pegboard test. Interestingly, four of these cognitive ability domains (global verbal memory, bilateral motor skill, CPT performance, and word fluency) are reported to be most affected in schizophrenic patients (Heinrichs and Zakzanis, 1998).
S.D. = 8.35 S.D. = 1.86 S.D. = 6.77 S.D. = 1.80
M = 33.79, S.D. = 10.38 M = 76.44, S.D. = 38.10 M = 42.65, S.D. = 33.36 M = 5.91, S.D. = 1.78 M = 6.04, S.D. = 1.82
(Z = (Z = (Z = (Z = (Z =
0.662; 1.603; 1.325; 2.374; 1.756;
p = 0.508) p = 0.109) p = 0.185) p = 0.018) p = 0.079)
Contrary to our expectation, parents of patients did not show a general cognitive impairment. They did not differ significantly from healthy control couples on the spatial version of the CPT test, on selective verbal memory, Wisconsin Card Sorting, Trail Making, unilateral motor skill measured with the Finger Tapping test and the Digit Backward test. However, although groups did not significantly differ on these cognitive ability domains, the differences between groups are all in the expected direction (see Table 3). Many studies investigated one or more aspects of verbal memory in relatives of patients, however in general studies do not split verbal memory into ‘‘general verbal memory’’ and ‘‘selective verbal memory.’’ Most of the studies that investigated verbal memory report verbal memory abnormalities in relatives compared to controls (see, for review, Kremen et al., 1994; see, e.g. Cannon et al., 1994; Blackwood et al., 1999; Faraone et al., 1999b; Faraone et al., 2000). However, some studies did not find a significantly lower score on any of the studied aspects of verbal memory performance in relatives of patients as compared to controls (Goldberg et al., 1990; Faraone et al., 1996; Harris et al., 1996; Staal et al., 2000). The absence of a significant difference between relatives of patients and controls in these studies might be
290
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293
explained in terms of insufficient power. Staal et al. (2000) and Goldberg et al. (1990) included a relatively small group of relatives of patients compared to studies that did find a difference in verbal memory between relatives and controls. Staal et al. (2000) and Goldberg et al. (1990) included 15 siblings and 16 discordant twins, respectively. Staal et al. (2000) also did not find a difference in verbal memory between schizophrenic patients and controls. Faraone et al. (1996) included siblings and parents of patients, although not necessarily both parents of patients. Thus, the failure to find a difference between relatives and controls in that study might have been due to the possibility that in some cases only the non-obligate carrier decided to participate. Harris et al. (1996) included both parents of 14 patients, but compared them to a small group of much younger controls (n = 18). A difference between relatives of patients and controls in psychomotor functioning as measured with the Purdue Pegboard test has, as far as we know, not been investigated previously. Of the 204 studies that compared cognitive functioning between schizophrenic patients and healthy controls, only 5 used the Purdue Pegboard test (Heinrichs and Zakzanis, 1998). Since psychomotor functioning as measured with the Purdue Pegboard test is one of the most pronounced cognitive deficits in patients (Heinrichs and Zakzanis, 1998) and since this deficit appears, as is reported here, also to be present in parents of patients, future studies may consider including this test in their cognitive battery. A difference between parents of patients and controls in word fluency is in concordance with the results of Dollfus et al. (2002), who also report that word fluency is significantly impaired in parents of patients compared to controls. Abnormalities in the word fluency test have also quite consistently been found in other studies that compared relatives of patients (not necessarily parents) to healthy controls (e.g., Keefe et al., 1994; Chen et al., 2000). Again, Goldberg et al. (1990) did not found word fluency abnormalities in relatives of patients. A study in relatives that reports separate performance data for the forward and the backward tasks (Conklin et al., 2000) found relatives to perform significantly worse than controls on the backward, but not on the forward digit span task. The difference
between relatives and controls was, however, in the same direction on the forward digit span test as we found; both Conklin et al. (2000) and our results report a lower score on the digit forward test in patients’ relatives than in controls. Four other studies compared the relatives of patients with healthy controls on the verbal version of the CPT-IP test (Franke et al., 1994; Laurant et al., 1999, 2000; Cosway et al., 2002). Our finding of verbal attentional deficits in relatives is partly consistent with results of the study by Franke et al. (1994), who reported a lower ability to sustain adequate levels of attention in verbal as well as in spatial stimuli in siblings of patients compared to controls. Our finding is not consistent with the result of Laurent et al. (1999, 2000) who reported a lower ability to sustain adequate levels of attention when it comes to spatial stimuli, but not when it comes to verbal stimuli in relatives of patients compared to controls. Our results likewise did not agree with the results of Cosway et al. (2002), who reported no difference between relatives and controls. Methodological differences between all studies that investigated CPT-IP performance in relatives of patients and controls might explain the inconsistency in study results. Firstly, Franke et al. (1994), Laurent et al. (1999, 2000) and Cosway et al. (2002) always presented the verbal and spatial attention test in the same order to their subjects (spatial after verbal). We used a randomized order because this minimizes order effects. Secondly, the discrepancy in test results might be caused by the fact that we compared the parents of patients with controls whereas the other studies did not concentrate on the parents of patients. The fact that we included parents of patients might also be the reason why we did not find differences on some other cognitive tests whereas other studies did. For example, differences between relatives of patients and controls have been found on the Trail Making test (e.g., Keefe et al., 1994), the backward digit span test (e.g., Conklin et al., 2000), the Wisconsin Card Sorting Test (e.g., Saoud et al., 2000) and on the spatial version of the CPT-IP test (Franke et al., 1994; Laurent et al., 1999, 2000). Parents of patients might be the healthiest group of first-degree relatives of schizophrenic patients since they passed the age of risk to develop schizophrenia. Obviously, all parents included in our study were capable of reproduction and parenting. The process
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293
of marriage and reproduction leads to a ‘‘positive’’ selection for mental health as compared to those who remain unmarried (Gottesman and Shields, 1982). Secondly, it is not known yet whether schizophrenia is due to unilineal or bilineal inheritance. If schizophrenia is unilineally inheritable, then only one parent of a proband is expected to carry the schizophrenia genotype. We included both parents in order to ensure the inclusion of the obligate carrier(s). However, the consequence of this approach is that possibly half of the parents included might not carry the genotype, which might have reduced the power of the study. The fact that a lower ability on card sorting is found in schizophrenic patients (Heinrichs and Zakzanis, 1998) and their siblings (Franke et al., 1992; Franke et al., 1993; Saoud et al., 2000) but not in their parents (Dollfus et al., 2002 and our study) supports the concept that parents of patients have fewer or less obvious phenotypic characteristics than patients themselves or their siblings. Of the 74 parents of patients included, 26 carried a psychiatric diagnosis. Therefore, a potential lower score on any of the cognitive tests in parents of schizophrenic patients may have been secondary to their current or past psychiatric diagnosis instead of their carrying the schizophrenic genotype. However, only on the non-preferred hand score of the Purdue Pegboard test has our analyses revealed a difference between parents of patients with a psychiatric diagnosis and parents without a psychiatric diagnosis. This means that the group differences found on other cognitive tests can not be explained by the fact that more parents of patients than healthy controls carried a psychiatric diagnosis. In order to compare our results with the results of Heinrichs and Zakzanis (1998), we tried to organize our cognitive tests into the same cognitive constructs as they did (see Table 1). Our cognitive constructs deviate, however, on some aspects from the constructs of Heinrichs and Zakzanis (1998). This is not only due to the fact that Heinrichs and Zakzanis (1998) included more tests in the cognitive constructs than we did. We had also some theoretical reasons to define some constructs differently. We did not cluster the digit forward and the digit backward span since Conklin et al. (2000) revealed that these tests tap different cognitive abilities. We analyzed the verbal (4 digits) and shape version of the CPT-IP test separately
291
because verbal and spatial attention are independent and each heritable to some degree (Cornblatt et al., 1988). In addition, the Dutch version of the CVLT differs from other CVLT versions (Mulder et al., 1996). Therefore, the CVLT subtests included in the ‘‘selective verbal memory’’ cluster differ from the CVLT subtests Heinrichs and Zakzanis included in this cluster. Finally, we did not cluster the Purdue Pegboard test and the Finger Tapping test since our correlation analyses revealed that these tests do not correlate (see statistics). In conclusion, we investigated differences in cognitive functioning between the parents of schizophrenic patients and healthy control couples. Several cognitive constructs were found to be impaired in parents as compared to controls. Interestingly, most of these constructs are among the constructs that are most impaired in schizophrenic patients themselves. Most deficits found in parents cannot be explained by a psychiatric illness of the participants, nor to a difference in general intellectual abilities between groups. This suggests that these cognitive constructs may be suitable as endophenotypic markers in schizophrenia. Future meta-analytic studies must point out if the same pattern of deficits found in schizophrenic patients can also be found in siblings and offspring of patients and if it can be replicated in parents of patients. References Andreasen, N.C., Flaum, M., Arndt, S., 1992. The Comprehensive Assessment of Symptoms and History (CASH). An instrument for assessing diagnosis and psychopathology. Arch. Gen. Psychiatry 49, 615 – 623. Benton, A.L., Hamsher, K., 1976. Multilingual Aphasia Examination. University of Iowa, Iowa City. Blackwood, D.H., Glabus, M.F., Dunan, J., O’Caroll, R.E., Muir, W.J., Ebmeier, K.P., 1999. Altered cerebral perfusion measured by SPECT in relatives of patients with schizophrenia. Correlations with memory and P300. Br. J. Psychiatry 175, 357 – 366. Cannon, T.D., Zorilla, L.E., Shtasel, D., Gur, R.E., Gur, R.C., Marco, E.J., Moberg, P., Price, R.A., 1994. Neuropsychological functioning in siblings discordant for schizophrenia and healthy volunteers. Arch. Gen. Psychiatry 51, 651 – 661. Cardno, A.G., Marshall, E.J., Coid, B., Macdonald, A.M., Ribchester, T.R., Davies, N.J., Venturi, P., Jones, L.A., Lewis, S.W., Sham, P.C., Gottesman, I.I., Farmer, A.E., McGuffin, P., Reveley, A.M., Murray, R.M., 1999. Heritability estimates for psychotic disorders: the Maudsley twin psychosis series. Arch. Gen. Psychiatry 56, 162 – 168. Chen, Y.L., Chen, Y.H., Lieh, M.F., 2000. Semantic verbal fluency
292
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293
deficit as a familial trait marker in schizophrenia. Psychiatry Res. 95, 133 – 148. Conklin, H.M., Curtis, C.E., Katsanis, J., Iacono, W.G., 2000. Verbal working memory impairment in schizophrenia patients and their first-degree relatives: evidence from the digit span task. Am. J. Psychiatry 157, 275 – 277. Cornblatt, B.A., Risch, N.J., Faris, G., Friedman, D., ErlenmeyerKimling, L., 1988. The Continuous Performance Test, Identical Pairs version (CPT-IP): I. New findings about sustained attention in normal families. Psychiatry Res. 26, 223 – 238. Cosway, R., Byrne, M., Clafferty, R., Hodges, A., Grant, E., Morris, J., Abukmeil, S.S., Lawrie, S.M., Miller, P., Owens, D.G.C., Johnstone, E.C., 2002. Sustained attention in young people at risk for schizophrenia. Psychol. Med. 32, 277 – 286. Docherty, N.M., 1994. Cognitive characteristics of the parents of schizophrenic patients. J. Nerv. Ment. Dis. 182, 443 – 451. Dollfus, S., Lombardo, C., Be´nali, K., Halbecq, I., Abadie, P., Marie´, R.M., Brazo, P., 2002. Executive/attentional cognitive functions in schizophrenic patients and their parents: a preliminary study. Schizophr. Res. 53, 93 – 99. Egan, M.F., Goldberg, T.E., Gscheidle, T., Weirich, M., Bigelow, L.B., Weinberger, D.R., 2000. Relative risk of attention deficits in siblings of patients with schizophrenia. Am. J. Psychiatry 157, 1309 – 1316. Endicott, J., Spitzer, R.L., 1978. A diagnostic interview: the schedule for affective disorders and schizophrenia. Arch. Gen. Psychiatry 35, 837 – 844. Faraone, S.V., Kremen, W.S., Lyons, M.J., Pepple, J.R., Seidman, L.J., Tsuamg, M.T., 1995. Diagnostic accuracy and linkage analysis: how useful are schizophrenia spectrum phenotypes? Am. J. Psychiatry 152, 1286 – 1290. Faraone, S.V., Seidman, L.J., Kremen, W.S., Toomey, R., Lyons, M.J., Tsuang, M.T., 1996. Neuropsychological functioning among the elderly nonpsychotic relatives of schizophrenic patients. Schizophr. Res. 21, 27 – 31. Faraone, S.V., Tsuang, M.T., Tsuang, D.W., 1999a. The future of psychiatric genetics. Genetics of Mental Disorders. A Guide for Students, Clinicians, and Researchers. The Guilford Press, New York, pp. 196 – 234. Faraone, S.V., Seidman, L.J., Toomey, R., Kremen, W.S., Pepple, J.R., Tsuang, M.T., 1999b. Neuropsychological functioning among the nonpsychotic relatives of schizophrenic patients: a 4-year follow-up study. J. Abnorm. Psychol. 108, 176 – 181. Faraone, S.V., Seidman, L.J., Kremen, W.S., Toomey, R., Pepple, J.R., Tsuang, M.T., 2000. Neuropsychologic functioning among the nonpsychotic relatives of schizophrenic patients: the effect of genetic loading. Biol. Psychiatry 48, 120 – 126. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 1997. Structured Clinical Interview for DSM-IV Axis 1 Disorders— Patient Edition (SCID-I/P) Version 2.0, Dutch Version. Swets & Zeitlinger, Lisse. Franke, P., Maier, W., Hain, C., Klinger, T., 1992. Wisconsin Card Sorting Test: an indicator of vulnerability to schizophrenia? Schizophr. Res. 6, 243 – 249. Franke, P., Maier, W., Hardt, J., Hain, C., 1993. Cognitive functioning and anhedonia in subjects at risk for schizophrenia. Schizophr. Res. 10, 77 – 84.
Franke, P., Maier, W., Hardt, J., Hain, C., Cornblatt, B.A., 1994. Attentional abilities and measures of schizotypy: their variation and covariation in schizophrenic patients, their siblings, and normal control subjects. Psychiatry Res. 54, 259 – 272. Goldberg, T.E., Ragland, J.D., Torrey, E.F., Gold, J.M., Bigelow, L.B., Weinberger, D.R., 1990. Neuropsychological assessment of monozygotic twins discordant for schizophrenia. Arch. Gen. Psychiatry 47, 1066 – 1072. Gottesman, I.I., Shields, J., 1982. Genetical puzzle pieces: family studies. Schizophrenia: The Epigenetic Study. Cambridge Univ. Press, Cambridge, pp. 83 – 98. Green, M.F., 1998. Neurocognitive indicators of vulnerability to schizophrenia. Schizophrenia from a Neurocognitive Perspective. Probing the Impenetrable Darkness. Allan & Bacon, Needham Heights. Harris, J.G., Adler, L.E., Young, D.A., Cullum, C.M., Rilling, L.M., Cicerello, A., Intemann, P.M., Freedman, R., 1996. Neuropsychological dysfunction in parents of schizophrenics. Schizophr. Res. 20, 253 – 260. Heaton, R.K., Chelune, G.J., Talley, J.L., Kay, G.G., Curtiss, G., 1993. Wisconsin Card Sorting Test Manual—Revised and Expanded. PARPsychological Assessment Resources, Inc., USA. Heinrichs, R.W., Zakzanis, K.K., 1998. Neurocognitive deficit in schizophrenia: a quantitative review of the evidence. Neuropsychology 12, 426 – 445. Keefe, R.S., Silverman, J.M., Roitman, S.E., Harvey, P.D., Duncan, M.A., Alroy, D., Siever, L.J., Davis, K.L., Mohs, R.C., 1994. Performance of nonpsychotic relatives of schizophrenic patients on cognitive tests. Psychiatry Res. 53, 1 – 12. Kremen, W.S., Seidman, L.J., Pepple, J.R., Lyons, M.J., Tsuang, M.T., Faraone, S.V., 1994. Neuropsychological risk indicators for schizophrenia: a review of family studies. Schizophr. Bull. 20, 103 – 119. Laurent, A., Saoud, M., Bougerol, T., D’Amato, T., Anchisi, A.M., Biloa-Tang, M., Dalery, J., Rochet, T., 1999. Attentional deficits in patients with schizophrenia and in their non-psychotic firstdegree relatives. Psychiatry Res. 89, 147 – 159. Laurent, A., Rochet, T., D’Amato, T., Anchisi, A.M., Daumal, M., Favre, P., Bougerol, T., Dalery, J., 2000. Vulnerability to schizophrenia: III. Importance and limits of the identical pairs continuous performance test. Encephale 26, 48 – 55. Luteyn, F., Van der Ploeg, F.A.E., 1983. The Groningen Intelligence Test. Swets & Zeitlinger, Lisse, in Dutch. Mathews, C.A., Reus, V.I., 2001. Assortative mating in the affective disorders: a systematic review and meta-analysis. Comp. Psychiatry 42, 257 – 262. McGuffin, P., Owen, M.J., Farmer, A.E., 1995. Genetic basis of schizophrenia. Lancet 346, 678 – 682. Mulder, J.L., Dekker, R., Dekker, P.H., 1996. California Verbal Learning Test. Swets & Zeitlinger, Lisse, Dutch version. National Institute of Mental Health, 1991. Molecular Genetics Initiative: Family Interview for Genetic Studies. National Institute of Mental Health, Bethesda, MD. Oldfield, R.C., 1971. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 9, 97 – 113. Parnas, J., 1988. Assortative mating in schizophrenia: results from the Copenhagen High-Risk Study. Psychiatry 51, 58 – 64.
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293 Pfohl, B., Blum, N., Zimmerman, M., 1995. Structured interview for DSM-IV personality SIDP-IV. University of Iowa, Department of Psychiatry, Iowa. Reitan, R.M., 1992. Trailmaking test. Manual for administration and Scoring. Reitan Neuropsychology Laboratory, Tucson. Reitan, R.M., Wolfson, D., 1985. The Halstead – Reitan Neuropsychological Test Battery; Theory and Clinical Interpretation. Neuropsychology Press, Tuscon, AZ. Saoud, M., D’Amato, T., Gutknecht, C., Triboulet, P., Bertaud, J.P., Marie-Cardine, M., Dalery, J., Rochet, T., 2000. Neuropsychological deficit in siblings discordant for schizophrenia. Schizophr. Bull. 26, 893 – 902. Staal, W.G., Hijman, R., Hulshoff Pol, H.E., Kahn, R.S., 2000. Neuropsychological dysfunctions in siblings discordant for schizophrenia. Psychiatry Res. 95, 227 – 235.
293
Stinissen, J., Willems, P.J., Coetsier, P., Hulsman, W.L.L., 1970. Dutch Adaption of the Wechsler Adult Intelligent Scale. Swets & Zeitlinger, Amsterdam. Thaker, G.K., 2000. Defining the schizophrenia phenotype. Curr. Psychiatry Rep. 2, 398 – 403. Tiffin, J., 1987. Purdue Pegboard. MacMillan/McGraw-Hill, Park Ridge, IL, USA. Tsuang, M.T., Stone, W.S., Faraone, S.V., 1999. Schizophrenia: a review of genetic studies. Harv. Rev. Psychiatry 7, 185 – 207. Verhage, F., 1983. Revised Scoring Method. University Hospital Groningen. Department of Neuropsychology, Groningen, The Netherlands.
Cognitive dysfunctions in parents of schizophrenic patients parallel the deficits found in patients Melanie C.M. Appels a,*, Margriet M. Sitskoorn a, Paul Westers b, Elleke Lems a, Rene´ S. Kahn a a
Department of Psychiatry, University Medical Center, PO Box 85500, 3508 GA, Utrecht, The Netherlands b Center for Biostatistics, Utrecht University, Utrecht, The Netherlands Received 8 August 2001; accepted 15 June 2002
Abstract Schizophrenia is characterized by a global cognitive impairment, with varying degrees of deficit in all ability domains. Since genetic factors are important in the etiology of schizophrenia we investigated whether parents of schizophrenic patients also show cognitive deficits, particularly on those cognitive ability domains that are most severely affected in patients. Both biological parents of 37 patients with schizophrenia (N = 74 subjects) and 28 comparable healthy married control couples (N = 56 subjects) were included. A comprehensive and standardized cognitive battery was used including tests measuring verbal memory, executive functioning, language, attention, and psychomotor functioning. Parents of patients differed from control couples on those cognitive constructs that are generally considered to be most impaired in schizophrenic patients, i.e. global verbal memory, bilateral motor skill, continuous performance, and word fluency. In addition, parents differed significantly from control couples on some other cognitive constructs on which patients show a smaller but also significant difference compared to healthy controls, i.e. unilateral motor skill and digit span. Results suggest that the cognitive constructs on which patients show relatively most severe impairment may prove suitable as endophenotypic markers in schizophrenia. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Schizophrenia; Parents; Cognition; Genotypical marker
1. Introduction Family, twin and adoption studies suggest an important genetic role in the etiology of schizophrenia (Cardno et al., 1999; McGuffin et al., 1995). However, the mode of transmission of the disorder is still unknown (Tsuang et al., 1999). Identification of *
Corresponding author. Tel.: +31-30-2506025; fax: +31-302505443. E-mail address: [email protected] (M.C.M. Appels).
relatives of patients who carry a predisposing genotype is therefore not yet possible. One attempt to resolve this issue is to search for endophenotypic markers: characteristics that mark the presence of a genetic predisposition to schizophrenia (Faraone et al., 1999a). Apart from identifying individuals at risk, endophenotypic markers might also facilitate linkage analyses aimed at identifying the genetic loci involved in schizophrenia. Since cognitive dysfunctions are prominent features of schizophrenia, they might be suitable candi-
0920-9964/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. doi:10.1016/S0920-9964(02)00342-0
286
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293
dates as endophenotypic markers (Egan et al., 2000). Results from a quantitative review, which includes 7420 schizophrenic patients and 5865 healthy controls divided over 204 studies, show that schizophrenia is characterized by a general cognitive impairment, with varying degrees of deficit in all ability domains (Heinrichs and Zakzanis, 1998). In 7 of the 22 cognitive ability domains, the magnitude of the difference between schizophrenic patients and healthy controls as expressed in mean effect size (Md), was greater than 1. The greatest difference was found on global verbal memory (Md = 1.41). Global verbal memory was followed in effect size by bilateral motor skill (Md = 1.30), performance IQ (Md = 1.26), continuous performance (Md = 1.16), word fluency (Md = 1.15), attention as measured with the STROOP (Md = 1.11), and WAIS-R IQ (Md = 1.10). The cognitive ability domains that are most abnormal in schizophrenic patients might be the best candidates for endophenotypic markers. For these deficits to be considered suitable as endophenotypic markers for schizophrenia, they should also be present in relatives of patients. Most studies that investigated cognitive differences between relatives of patients and controls reported more cognitive dysfunctions in relatives of patients than in healthy controls (see, for reviews, Kremen et al., 1994; Faraone et al., 1995; Green, 1998; Thaker, 2000). However, it is not clear if the relatives of patients show deficits in specifically those cognitive ability domains that are most affected in schizophrenic patients. No quantitative review of the literature on cognitive test performance in relatives of schizophrenic patients exists, and results of individual studies are contradictory (see Kremen et al., 1994; Faraone et al., 1995). Furthermore, not all types of relatives of schizophrenic patients have been extensively investigated. So far, relatively few studies concentrated on the difference between parents of patients and controls (see Docherty, 1994; Harris et al., 1996; Dollfus et al., 2002). As far as we know, the difference in general cognitive functioning between both parents of schizophrenic patients and healthy control couples has not been investigated. Studying the parents of schizophrenic patients has an important advantage over studying other types of first-degree relatives of patients. Parents of patients have passed the age of risk for developing schizo-
phrenia. Therefore, the status with regard to schizophrenia can be determined with more confidence in parents than in other types of first-degree relatives. It is important to study both parents of patients since it is not clear whether schizophrenia is due to unilineal inheritance (only one parent is expected to carry the schizophrenia genotype) or to bilineal inheritance (both parents are expected to carry the schizophrenia genotype). By including both parents of a proband, one can ensure the inclusion of the obligate carrier(s) and one limits the risk of assessing only the healthiest parent. Furthermore, by comparing the parents of patients with healthy control couples, one can control for assortative mating. Assortative mating denotes a tendency for mated pairs to be more similar for some phenotypic traits than would be the case if the choice of a partner occurred at random. Assortative mating occurs for a variety of physical and psychological traits, including mental illness (Parnas, 1988; Mathews and Reus, 2001). In the present study, we investigate the difference between both biological parents of schizophrenic patients and healthy control couples on several cognitive ability domains. We hypothesize that parents of schizophrenic patients have a general cognitive impairment. Furthermore, we hypothesize that the difference between groups will be most pronounced on those cognitive ability domains that in a comprehensive meta-analysis (Heinrichs and Zakzanis, 1998) have been found to be most affected in schizophrenic patients themselves.
2. Methods 2.1. Subjects Thirty-seven biological parent couples (N = 74 subjects) of patients with schizophrenia and 28 healthy married control couples (N = 56 subjects) were included. The diagnosis of DSM-IV schizophrenia in the children of the parents was determined by using the Comprehensive Assessment of Symptoms and History (CASH) (Andreasen et al., 1992) or a Structural Clinical Interview DSM axis I interview (SCID interview) (First et al., 1997). Healthy controls and parents of patients were screened using the CASH interview, the ‘‘Schizophre-
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293
nia and Affective Disorder Schedule—Lifetime interview’’ (SADS-L interview) (Endicott and Spitzer, 1978), the Structured Interview for DSM-IV Personality Disorders (SIDP-IV) (Pfohl et al., 1995), and the Family Interview for Genetic Studies (FIGS) (National Institute of Mental Health, 1991). A history of neurological illness resulted in the exclusion of parents of patients and healthy controls, as did a history of a psychotic disorder and substance abuse. In addition, controls were excluded if they had a history of a depressive/manic syndrome, or a personality disorder. Controls were also excluded if a depressive, manic or psychotic illness was present in their first-degree relatives or if a history of psychotic illness was present in their second-degree relatives. Written informed consent was obtained from all subjects after full explanation of the study aims and procedures.
Table 1 Cognitive constructs and tests used
2.2. Demographic measures
Unilateral motor skill
The current IQ of the subjects was estimated with the aid of a short form of the Groningen Intelligence Test (GIT) (Luteyn and Van der Ploeg, 1983). The handedness of the subjects was assessed by means of the Edinburgh Handedness Inventory (Oldfield, 1971). The scoring system of Verhage (Verhage, 1983) was used to define a subject’s level of education. 2.3. Neuropsychological evaluation The cognitive constructs and cognitive tests used are described in Table 1. In order to compare our results with the results of Heinrichs and Zakzanis (1998), we tried to organize our cognitive tests as much as possible into the same cognitive constructs as they did. Trained predoctoral and postdoctoral examiners administered the cognitive tests. The examiners were supervised by a neuropsychologist. 2.4. Statistical analyses Data were analyzed using the Statistical Package for Social Sciences (SPSS), version 9.0. Demographic data were compared by ANOVA (age) or a Mann – Whitney U-test (level of education, handedness, IQ). In most cases, only one test score represented the score on the cognitive construct. In three cases (Word
287
Test or construct
Recorded test variablesa
Global verbal memory Bilateral motor skill
CVLT: trials I – V recall Purdue Pegboard test: bilateral score CPT-IP test: dprime of numbers (4 digits) fast CPT-IP test: dprime of shapes fast Category Fluency test (animals and professions), Letter Fluency (letters N and A) CVLT: long-delay free recall, consolidation, number of intrusions, discrimination index WCST: perseverative errors, perseverative responses, categories completed, percent perseverative errors Finger Tapping test: preferred hand score, non-preferred hand score Purdue Pegboard test: preferred hand score, non-preferred hand score Trail Making Test: part A Trail Making Test: part B Trail Making Test: part B – part A WAIS: Digit forward WAIS: Digit backward
Continuous performance, verbal version Continuous performance, spatial version Word fluency
Selective verbal memory
Wisconsin card sorting
Trail making—part A Trail making—part B Trail making—part B – A Digit forward Digit backward
a CVLT = California Verbal Learning Test (Mulder et al., 1996); Purdue Pegboard test (Tiffin, 1987); CPT-IP (Cornblatt et al., 1988); Category Fluency (Luteyn and Van der Ploeg, 1983); Letter Fluency (Benton and Hamsher, 1976); WCST = Wisconsin Card Sorting Test (Heaton et al., 1993); Finger Tapping test (Reitan and Wolfson, 1985); Trail Making Test (Reitan, 1992); WAIS Digit span (Stinissen et al., 1970).
Fluency, Wisconsin Card Sorting and selective verbal memory), more than one test score belonged to the same cognitive construct. In order to calculate scores on these cognitive constructs, test scores were rescaled to obtain z-scores using the means and standard deviation of the controls as norm scores. The z-scores were adjusted for sex by using separate means and standard deviations for males and females. We did not cluster the Purdue Pegboard test with the Finger Tapping test because the preferred as well as the non-preferred hand score of these tests do not corre-
288
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293
late significantly (respectively r = 0.066; p = 0.473 and r = 0.138; p = 0.137). Cognitive differences between groups (parents of patients versus control couples) were calculated with a repeated measurement analysis if tests or z-scores were normally distributed and the Levene’s test of equality of error variances was not significant. Group was treated as a between factor. Sex was treated as a within-factor to control for the dependency of mother and father within each couple. If tests or z-scores were not normally distributed, the mean value of the score of each couple was calculated and used in a Mann – Whitney U-test. The mean value was used to take the dependency between husband and wife into account. In case of a missing value, the whole couple was excluded from the analysis. If a significant difference between groups was found on a cognitive construct, additional analyses (an ANOVA or a Mann – Whitney U-test) were done to investigate if parents of patients who carry a psychiatric diagnosis differ from the parents of patients who do not carry such a diagnosis. Two-tailed tests were used for all analyses and significance levels were set at p < 0.05.
Table 2 Demographic characteristics of parents of patients and healthy controls
3. Results
3.2. Cognitive differences between groups
3.1. Demographic characteristics of the sample
On global verbal memory, the bilateral motor test, the verbal version of the continuous performance test, identical pairs version (CPT-IP), Word Fluency, the dominant and non-dominant hand score of the Purdue Pegboard test and the digit forward test, parents of patients had a significantly lower score than healthy control couples. On the spatial version of the attention test, the selective verbal memory test, the Wisconsin Card Sorting test, the Trailmaking test, the Finger Tapping test (dominant and dominant score) and the digit backward test, parents of schizophrenic patients did not differ significantly from healthy controls (see Table 3). Additional analyses revealed no significant differences between the parents of patients who carried psychiatric diagnosis and the parents of patients who did not carry such a diagnosis on the dprime variable of the verbal attention test ( F = 0.272; df = 1; p = 0.60), on the Purdue Pegboard test preferred hand ( F = 0.128;
The groups were comparable with respect to age, handedness, IQ and level of education (see Table 2). Of the 74 parents of patients included, 26 carried a psychiatric diagnosis according to DSM-IV criteria and/or Research Diagnostic Criteria (RDC). Most of them had a diagnosis of depression (15 out of 26). Three persons with an axis 1 depressive disorder had an additional disorder (one a depressive disorder including a labile personality disorder, one a depression including a hypomanic disorder, a labile personality disorder and a generalized anxiety disorder, and one a depression including a panic disorder). Two parents of patients had a dysthymic disorder (one a dysthymic disorder including a labile personality disorder, a panic disorder and a generalized anxiety disorder). One parent had a hypomanic disorder. Two parents a manic –depressive disorder. One parent had a post-traumatic stress disorder, including a panic
Biological parents Healthy Between-group of schizophrenic controls differences patients (N = 56 persons) (N = 74 persons) Agea
M = 53.59 (S.D. = 4.06) Current IQb M = 115.38 (S.D. = 14.69) Handednessb M = 0.81 (S.D. = 0.36) Level of Median = 5.00 educationb (14 years)
M = 53.71 (S.D. = 4.59) M = 119.68 (S.D. = 11.91) M = 0.76 (S.D. = 0.37) Median = 6.00 (15years)
df = 1, p = 0.87 Z = 1.57, p = 0.12 Z = 1.40, p = 0.16 Z = 1.14, p = 0.25
M = mean. a ANOVA. b Mann – Whitney U-test.
disorder and a generalized anxiety disorder. One parent had an avoidant personality disorder and a generalized anxiety disorder. Two parents had a simple phobia, and one a generalized anxiety disorder. One parent had a gambling disorder. Only one control person carried a psychiatric diagnosis, this was a simple phobia according to the RDC criteria.
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293
289
Table 3 Differences between parents of schizophrenic patients and healthy control couples Test or construct
Controls (mean and S.D.)
Parents of patients (mean and SD)
Difference between groups
Global verbal memory Bilateral motor skill Continuous Performance, verbal version Continuous Performance, spatial version Word Fluency (z-score) Selective Verbal Memory (z-score) Wisconsin Card Sorting (z-score)
M = 51.75, S.D. = 9.47 M = 24.07, S.D. = 3.04 M = 1.67, S.D. = 0.74 M = 1.59, S.D. = 0.69 M = 0.00, S.D. = 0.79 M = 0.01, S.D. = 0.76 M = 0.000, S.D. = 0.93
M = 47.75, S.D. = 9.66 M = 22.64, S.D. = 3.34 M = 1.35, S.D. = 0.65 M = 1.51, S.D. = 0.66 M = 0.275, S.D. = 0.771 M = 0.228, S.D. = 0.779 M = 0.08, S.D. = 1.06
( F = 6.168; df = 1,60; p = 0.016) ( F = 6.405; df = 1,61; p = 0.014) ( F = 7.19; df = 1,58; p = 0.010) ( F = 0.316; df = 1,57; p = 0.576) (Z = 2.053; p = 0.040) ( F = 2.713; df = 1,58; p = 0.105) ( F = 0.202; df = 1,62; p = 0.655)
Unilateral motor skill Fingertapping test, preferred hand Purdue Pegboard test, preferred hand Fingertapping test, nonpreferred hand Purdue Pegboard test, nonpreferred hand
M = 46.51, M = 14.65, M = 41.16, M = 14.50,
M = 44.42, M = 13.75, M = 38.88, M = 13.18,
( F = 0.137; df = 1,61; p = 0.137) ( F = 9.652, df = 1,61; p = 0.003) ( F = 4.00; df = 1,59; p = 0.050) ( F = 13.935; df = 1,61; p = 0.000)
Trail Making—Part A Trail Making—Part B Trail Making—Part B – A Digit forward Digit backward
M = 32.63, S.D. = 9.99 M = 63.58, S.D. = 15.68 M = 30.96, S.D. = 15.24 M = 6.79, S.D. = 2.09 M = 6.63, S.D. = 1.79
S.D. = 8.69 S.D. = 1.51 S.D. = 7.53 S.D. = 1.70
df = 1; p = 0.722), Purdue Pegboard test both hands ( F = 1.242; df = 1; p = 0.269) word fluency ( F = 0.129; df = 1; p = 0.720) digit forward (Z = 0.798, p = 0.425), and global verbal memory ( F = 0.094; df = 1; p = 0.760). There was a significant difference between the parents of patients who carried psychiatric diagnosis and the parents of patients who did not carry such a diagnosis on the Purdue Pegboard test non-preferred hand ( F = 4.219, df = 1; p = 0.044).
4. Discussion In the present study, we compared the performance between both parents of schizophrenic patients and healthy control couples on several cognitive function domains. Parents of schizophrenic patients were more impaired than healthy control couples on global verbal memory, bilateral motor skill, the verbal version of the CPT test, word fluency, digit forward and unilateral motor skill measured with the Purdue Pegboard test. Interestingly, four of these cognitive ability domains (global verbal memory, bilateral motor skill, CPT performance, and word fluency) are reported to be most affected in schizophrenic patients (Heinrichs and Zakzanis, 1998).
S.D. = 8.35 S.D. = 1.86 S.D. = 6.77 S.D. = 1.80
M = 33.79, S.D. = 10.38 M = 76.44, S.D. = 38.10 M = 42.65, S.D. = 33.36 M = 5.91, S.D. = 1.78 M = 6.04, S.D. = 1.82
(Z = (Z = (Z = (Z = (Z =
0.662; 1.603; 1.325; 2.374; 1.756;
p = 0.508) p = 0.109) p = 0.185) p = 0.018) p = 0.079)
Contrary to our expectation, parents of patients did not show a general cognitive impairment. They did not differ significantly from healthy control couples on the spatial version of the CPT test, on selective verbal memory, Wisconsin Card Sorting, Trail Making, unilateral motor skill measured with the Finger Tapping test and the Digit Backward test. However, although groups did not significantly differ on these cognitive ability domains, the differences between groups are all in the expected direction (see Table 3). Many studies investigated one or more aspects of verbal memory in relatives of patients, however in general studies do not split verbal memory into ‘‘general verbal memory’’ and ‘‘selective verbal memory.’’ Most of the studies that investigated verbal memory report verbal memory abnormalities in relatives compared to controls (see, for review, Kremen et al., 1994; see, e.g. Cannon et al., 1994; Blackwood et al., 1999; Faraone et al., 1999b; Faraone et al., 2000). However, some studies did not find a significantly lower score on any of the studied aspects of verbal memory performance in relatives of patients as compared to controls (Goldberg et al., 1990; Faraone et al., 1996; Harris et al., 1996; Staal et al., 2000). The absence of a significant difference between relatives of patients and controls in these studies might be
290
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293
explained in terms of insufficient power. Staal et al. (2000) and Goldberg et al. (1990) included a relatively small group of relatives of patients compared to studies that did find a difference in verbal memory between relatives and controls. Staal et al. (2000) and Goldberg et al. (1990) included 15 siblings and 16 discordant twins, respectively. Staal et al. (2000) also did not find a difference in verbal memory between schizophrenic patients and controls. Faraone et al. (1996) included siblings and parents of patients, although not necessarily both parents of patients. Thus, the failure to find a difference between relatives and controls in that study might have been due to the possibility that in some cases only the non-obligate carrier decided to participate. Harris et al. (1996) included both parents of 14 patients, but compared them to a small group of much younger controls (n = 18). A difference between relatives of patients and controls in psychomotor functioning as measured with the Purdue Pegboard test has, as far as we know, not been investigated previously. Of the 204 studies that compared cognitive functioning between schizophrenic patients and healthy controls, only 5 used the Purdue Pegboard test (Heinrichs and Zakzanis, 1998). Since psychomotor functioning as measured with the Purdue Pegboard test is one of the most pronounced cognitive deficits in patients (Heinrichs and Zakzanis, 1998) and since this deficit appears, as is reported here, also to be present in parents of patients, future studies may consider including this test in their cognitive battery. A difference between parents of patients and controls in word fluency is in concordance with the results of Dollfus et al. (2002), who also report that word fluency is significantly impaired in parents of patients compared to controls. Abnormalities in the word fluency test have also quite consistently been found in other studies that compared relatives of patients (not necessarily parents) to healthy controls (e.g., Keefe et al., 1994; Chen et al., 2000). Again, Goldberg et al. (1990) did not found word fluency abnormalities in relatives of patients. A study in relatives that reports separate performance data for the forward and the backward tasks (Conklin et al., 2000) found relatives to perform significantly worse than controls on the backward, but not on the forward digit span task. The difference
between relatives and controls was, however, in the same direction on the forward digit span test as we found; both Conklin et al. (2000) and our results report a lower score on the digit forward test in patients’ relatives than in controls. Four other studies compared the relatives of patients with healthy controls on the verbal version of the CPT-IP test (Franke et al., 1994; Laurant et al., 1999, 2000; Cosway et al., 2002). Our finding of verbal attentional deficits in relatives is partly consistent with results of the study by Franke et al. (1994), who reported a lower ability to sustain adequate levels of attention in verbal as well as in spatial stimuli in siblings of patients compared to controls. Our finding is not consistent with the result of Laurent et al. (1999, 2000) who reported a lower ability to sustain adequate levels of attention when it comes to spatial stimuli, but not when it comes to verbal stimuli in relatives of patients compared to controls. Our results likewise did not agree with the results of Cosway et al. (2002), who reported no difference between relatives and controls. Methodological differences between all studies that investigated CPT-IP performance in relatives of patients and controls might explain the inconsistency in study results. Firstly, Franke et al. (1994), Laurent et al. (1999, 2000) and Cosway et al. (2002) always presented the verbal and spatial attention test in the same order to their subjects (spatial after verbal). We used a randomized order because this minimizes order effects. Secondly, the discrepancy in test results might be caused by the fact that we compared the parents of patients with controls whereas the other studies did not concentrate on the parents of patients. The fact that we included parents of patients might also be the reason why we did not find differences on some other cognitive tests whereas other studies did. For example, differences between relatives of patients and controls have been found on the Trail Making test (e.g., Keefe et al., 1994), the backward digit span test (e.g., Conklin et al., 2000), the Wisconsin Card Sorting Test (e.g., Saoud et al., 2000) and on the spatial version of the CPT-IP test (Franke et al., 1994; Laurent et al., 1999, 2000). Parents of patients might be the healthiest group of first-degree relatives of schizophrenic patients since they passed the age of risk to develop schizophrenia. Obviously, all parents included in our study were capable of reproduction and parenting. The process
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293
of marriage and reproduction leads to a ‘‘positive’’ selection for mental health as compared to those who remain unmarried (Gottesman and Shields, 1982). Secondly, it is not known yet whether schizophrenia is due to unilineal or bilineal inheritance. If schizophrenia is unilineally inheritable, then only one parent of a proband is expected to carry the schizophrenia genotype. We included both parents in order to ensure the inclusion of the obligate carrier(s). However, the consequence of this approach is that possibly half of the parents included might not carry the genotype, which might have reduced the power of the study. The fact that a lower ability on card sorting is found in schizophrenic patients (Heinrichs and Zakzanis, 1998) and their siblings (Franke et al., 1992; Franke et al., 1993; Saoud et al., 2000) but not in their parents (Dollfus et al., 2002 and our study) supports the concept that parents of patients have fewer or less obvious phenotypic characteristics than patients themselves or their siblings. Of the 74 parents of patients included, 26 carried a psychiatric diagnosis. Therefore, a potential lower score on any of the cognitive tests in parents of schizophrenic patients may have been secondary to their current or past psychiatric diagnosis instead of their carrying the schizophrenic genotype. However, only on the non-preferred hand score of the Purdue Pegboard test has our analyses revealed a difference between parents of patients with a psychiatric diagnosis and parents without a psychiatric diagnosis. This means that the group differences found on other cognitive tests can not be explained by the fact that more parents of patients than healthy controls carried a psychiatric diagnosis. In order to compare our results with the results of Heinrichs and Zakzanis (1998), we tried to organize our cognitive tests into the same cognitive constructs as they did (see Table 1). Our cognitive constructs deviate, however, on some aspects from the constructs of Heinrichs and Zakzanis (1998). This is not only due to the fact that Heinrichs and Zakzanis (1998) included more tests in the cognitive constructs than we did. We had also some theoretical reasons to define some constructs differently. We did not cluster the digit forward and the digit backward span since Conklin et al. (2000) revealed that these tests tap different cognitive abilities. We analyzed the verbal (4 digits) and shape version of the CPT-IP test separately
291
because verbal and spatial attention are independent and each heritable to some degree (Cornblatt et al., 1988). In addition, the Dutch version of the CVLT differs from other CVLT versions (Mulder et al., 1996). Therefore, the CVLT subtests included in the ‘‘selective verbal memory’’ cluster differ from the CVLT subtests Heinrichs and Zakzanis included in this cluster. Finally, we did not cluster the Purdue Pegboard test and the Finger Tapping test since our correlation analyses revealed that these tests do not correlate (see statistics). In conclusion, we investigated differences in cognitive functioning between the parents of schizophrenic patients and healthy control couples. Several cognitive constructs were found to be impaired in parents as compared to controls. Interestingly, most of these constructs are among the constructs that are most impaired in schizophrenic patients themselves. Most deficits found in parents cannot be explained by a psychiatric illness of the participants, nor to a difference in general intellectual abilities between groups. This suggests that these cognitive constructs may be suitable as endophenotypic markers in schizophrenia. Future meta-analytic studies must point out if the same pattern of deficits found in schizophrenic patients can also be found in siblings and offspring of patients and if it can be replicated in parents of patients. References Andreasen, N.C., Flaum, M., Arndt, S., 1992. The Comprehensive Assessment of Symptoms and History (CASH). An instrument for assessing diagnosis and psychopathology. Arch. Gen. Psychiatry 49, 615 – 623. Benton, A.L., Hamsher, K., 1976. Multilingual Aphasia Examination. University of Iowa, Iowa City. Blackwood, D.H., Glabus, M.F., Dunan, J., O’Caroll, R.E., Muir, W.J., Ebmeier, K.P., 1999. Altered cerebral perfusion measured by SPECT in relatives of patients with schizophrenia. Correlations with memory and P300. Br. J. Psychiatry 175, 357 – 366. Cannon, T.D., Zorilla, L.E., Shtasel, D., Gur, R.E., Gur, R.C., Marco, E.J., Moberg, P., Price, R.A., 1994. Neuropsychological functioning in siblings discordant for schizophrenia and healthy volunteers. Arch. Gen. Psychiatry 51, 651 – 661. Cardno, A.G., Marshall, E.J., Coid, B., Macdonald, A.M., Ribchester, T.R., Davies, N.J., Venturi, P., Jones, L.A., Lewis, S.W., Sham, P.C., Gottesman, I.I., Farmer, A.E., McGuffin, P., Reveley, A.M., Murray, R.M., 1999. Heritability estimates for psychotic disorders: the Maudsley twin psychosis series. Arch. Gen. Psychiatry 56, 162 – 168. Chen, Y.L., Chen, Y.H., Lieh, M.F., 2000. Semantic verbal fluency
292
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293
deficit as a familial trait marker in schizophrenia. Psychiatry Res. 95, 133 – 148. Conklin, H.M., Curtis, C.E., Katsanis, J., Iacono, W.G., 2000. Verbal working memory impairment in schizophrenia patients and their first-degree relatives: evidence from the digit span task. Am. J. Psychiatry 157, 275 – 277. Cornblatt, B.A., Risch, N.J., Faris, G., Friedman, D., ErlenmeyerKimling, L., 1988. The Continuous Performance Test, Identical Pairs version (CPT-IP): I. New findings about sustained attention in normal families. Psychiatry Res. 26, 223 – 238. Cosway, R., Byrne, M., Clafferty, R., Hodges, A., Grant, E., Morris, J., Abukmeil, S.S., Lawrie, S.M., Miller, P., Owens, D.G.C., Johnstone, E.C., 2002. Sustained attention in young people at risk for schizophrenia. Psychol. Med. 32, 277 – 286. Docherty, N.M., 1994. Cognitive characteristics of the parents of schizophrenic patients. J. Nerv. Ment. Dis. 182, 443 – 451. Dollfus, S., Lombardo, C., Be´nali, K., Halbecq, I., Abadie, P., Marie´, R.M., Brazo, P., 2002. Executive/attentional cognitive functions in schizophrenic patients and their parents: a preliminary study. Schizophr. Res. 53, 93 – 99. Egan, M.F., Goldberg, T.E., Gscheidle, T., Weirich, M., Bigelow, L.B., Weinberger, D.R., 2000. Relative risk of attention deficits in siblings of patients with schizophrenia. Am. J. Psychiatry 157, 1309 – 1316. Endicott, J., Spitzer, R.L., 1978. A diagnostic interview: the schedule for affective disorders and schizophrenia. Arch. Gen. Psychiatry 35, 837 – 844. Faraone, S.V., Kremen, W.S., Lyons, M.J., Pepple, J.R., Seidman, L.J., Tsuamg, M.T., 1995. Diagnostic accuracy and linkage analysis: how useful are schizophrenia spectrum phenotypes? Am. J. Psychiatry 152, 1286 – 1290. Faraone, S.V., Seidman, L.J., Kremen, W.S., Toomey, R., Lyons, M.J., Tsuang, M.T., 1996. Neuropsychological functioning among the elderly nonpsychotic relatives of schizophrenic patients. Schizophr. Res. 21, 27 – 31. Faraone, S.V., Tsuang, M.T., Tsuang, D.W., 1999a. The future of psychiatric genetics. Genetics of Mental Disorders. A Guide for Students, Clinicians, and Researchers. The Guilford Press, New York, pp. 196 – 234. Faraone, S.V., Seidman, L.J., Toomey, R., Kremen, W.S., Pepple, J.R., Tsuang, M.T., 1999b. Neuropsychological functioning among the nonpsychotic relatives of schizophrenic patients: a 4-year follow-up study. J. Abnorm. Psychol. 108, 176 – 181. Faraone, S.V., Seidman, L.J., Kremen, W.S., Toomey, R., Pepple, J.R., Tsuang, M.T., 2000. Neuropsychologic functioning among the nonpsychotic relatives of schizophrenic patients: the effect of genetic loading. Biol. Psychiatry 48, 120 – 126. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 1997. Structured Clinical Interview for DSM-IV Axis 1 Disorders— Patient Edition (SCID-I/P) Version 2.0, Dutch Version. Swets & Zeitlinger, Lisse. Franke, P., Maier, W., Hain, C., Klinger, T., 1992. Wisconsin Card Sorting Test: an indicator of vulnerability to schizophrenia? Schizophr. Res. 6, 243 – 249. Franke, P., Maier, W., Hardt, J., Hain, C., 1993. Cognitive functioning and anhedonia in subjects at risk for schizophrenia. Schizophr. Res. 10, 77 – 84.
Franke, P., Maier, W., Hardt, J., Hain, C., Cornblatt, B.A., 1994. Attentional abilities and measures of schizotypy: their variation and covariation in schizophrenic patients, their siblings, and normal control subjects. Psychiatry Res. 54, 259 – 272. Goldberg, T.E., Ragland, J.D., Torrey, E.F., Gold, J.M., Bigelow, L.B., Weinberger, D.R., 1990. Neuropsychological assessment of monozygotic twins discordant for schizophrenia. Arch. Gen. Psychiatry 47, 1066 – 1072. Gottesman, I.I., Shields, J., 1982. Genetical puzzle pieces: family studies. Schizophrenia: The Epigenetic Study. Cambridge Univ. Press, Cambridge, pp. 83 – 98. Green, M.F., 1998. Neurocognitive indicators of vulnerability to schizophrenia. Schizophrenia from a Neurocognitive Perspective. Probing the Impenetrable Darkness. Allan & Bacon, Needham Heights. Harris, J.G., Adler, L.E., Young, D.A., Cullum, C.M., Rilling, L.M., Cicerello, A., Intemann, P.M., Freedman, R., 1996. Neuropsychological dysfunction in parents of schizophrenics. Schizophr. Res. 20, 253 – 260. Heaton, R.K., Chelune, G.J., Talley, J.L., Kay, G.G., Curtiss, G., 1993. Wisconsin Card Sorting Test Manual—Revised and Expanded. PARPsychological Assessment Resources, Inc., USA. Heinrichs, R.W., Zakzanis, K.K., 1998. Neurocognitive deficit in schizophrenia: a quantitative review of the evidence. Neuropsychology 12, 426 – 445. Keefe, R.S., Silverman, J.M., Roitman, S.E., Harvey, P.D., Duncan, M.A., Alroy, D., Siever, L.J., Davis, K.L., Mohs, R.C., 1994. Performance of nonpsychotic relatives of schizophrenic patients on cognitive tests. Psychiatry Res. 53, 1 – 12. Kremen, W.S., Seidman, L.J., Pepple, J.R., Lyons, M.J., Tsuang, M.T., Faraone, S.V., 1994. Neuropsychological risk indicators for schizophrenia: a review of family studies. Schizophr. Bull. 20, 103 – 119. Laurent, A., Saoud, M., Bougerol, T., D’Amato, T., Anchisi, A.M., Biloa-Tang, M., Dalery, J., Rochet, T., 1999. Attentional deficits in patients with schizophrenia and in their non-psychotic firstdegree relatives. Psychiatry Res. 89, 147 – 159. Laurent, A., Rochet, T., D’Amato, T., Anchisi, A.M., Daumal, M., Favre, P., Bougerol, T., Dalery, J., 2000. Vulnerability to schizophrenia: III. Importance and limits of the identical pairs continuous performance test. Encephale 26, 48 – 55. Luteyn, F., Van der Ploeg, F.A.E., 1983. The Groningen Intelligence Test. Swets & Zeitlinger, Lisse, in Dutch. Mathews, C.A., Reus, V.I., 2001. Assortative mating in the affective disorders: a systematic review and meta-analysis. Comp. Psychiatry 42, 257 – 262. McGuffin, P., Owen, M.J., Farmer, A.E., 1995. Genetic basis of schizophrenia. Lancet 346, 678 – 682. Mulder, J.L., Dekker, R., Dekker, P.H., 1996. California Verbal Learning Test. Swets & Zeitlinger, Lisse, Dutch version. National Institute of Mental Health, 1991. Molecular Genetics Initiative: Family Interview for Genetic Studies. National Institute of Mental Health, Bethesda, MD. Oldfield, R.C., 1971. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 9, 97 – 113. Parnas, J., 1988. Assortative mating in schizophrenia: results from the Copenhagen High-Risk Study. Psychiatry 51, 58 – 64.
M.C.M. Appels et al. / Schizophrenia Research 63 (2003) 285–293 Pfohl, B., Blum, N., Zimmerman, M., 1995. Structured interview for DSM-IV personality SIDP-IV. University of Iowa, Department of Psychiatry, Iowa. Reitan, R.M., 1992. Trailmaking test. Manual for administration and Scoring. Reitan Neuropsychology Laboratory, Tucson. Reitan, R.M., Wolfson, D., 1985. The Halstead – Reitan Neuropsychological Test Battery; Theory and Clinical Interpretation. Neuropsychology Press, Tuscon, AZ. Saoud, M., D’Amato, T., Gutknecht, C., Triboulet, P., Bertaud, J.P., Marie-Cardine, M., Dalery, J., Rochet, T., 2000. Neuropsychological deficit in siblings discordant for schizophrenia. Schizophr. Bull. 26, 893 – 902. Staal, W.G., Hijman, R., Hulshoff Pol, H.E., Kahn, R.S., 2000. Neuropsychological dysfunctions in siblings discordant for schizophrenia. Psychiatry Res. 95, 227 – 235.
293
Stinissen, J., Willems, P.J., Coetsier, P., Hulsman, W.L.L., 1970. Dutch Adaption of the Wechsler Adult Intelligent Scale. Swets & Zeitlinger, Amsterdam. Thaker, G.K., 2000. Defining the schizophrenia phenotype. Curr. Psychiatry Rep. 2, 398 – 403. Tiffin, J., 1987. Purdue Pegboard. MacMillan/McGraw-Hill, Park Ridge, IL, USA. Tsuang, M.T., Stone, W.S., Faraone, S.V., 1999. Schizophrenia: a review of genetic studies. Harv. Rev. Psychiatry 7, 185 – 207. Verhage, F., 1983. Revised Scoring Method. University Hospital Groningen. Department of Neuropsychology, Groningen, The Netherlands.