Brain morphometry of individuals with schizophrenia with and without antipsychotic medication – The Northern Finland Birth Cohort 1966 Study

European Psychiatry 30 (2015) 598–605

Contents lists available at ScienceDirect

European Psychiatry journal homepage: http://www.europsy-journal.com

Original article

Brain morphometry of individuals with schizophrenia with and without antipsychotic medication – The Northern Finland Birth Cohort 1966 Study J. Moilanen a,b,*, S. Huhtaniska a,b, M. Haapea a,b,c, E. Ja¨a¨skela¨inen b,d, J. Veijola a,b, M. Isohanni a, H. Koponen e, J. Miettunen a,b,d a

Institute of Clinical Medicine, Department of Psychiatry, University of Oulu and Oulu University Hospital, P.O. Box 5000, 90014 Oulu, Finland Medical Research Center Oulu, University of Oulu and Oulu University Hospital, P.O. Box 5000, 90014 Oulu, Finland c Department of Diagnostic Radiology, Oulu University Hospital, P.O. Box 5000, 90014 Oulu, Finland d Institute of Health Sciences, University of Oulu, P.O. Box 5000, 90014 Oulu, Finland e University of Helsinki and Helsinki University Hospital, Psychiatry, P.O. Box 22, 00014 Helsinki, Finland b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 10 December 2014 Received in revised form 10 February 2015 Accepted 11 February 2015 Available online 16 March 2015

Background: In schizophrenia, brain morphometric changes may be associated with antipsychotic medication. Only limited data is available concerning individuals with schizophrenia without antipsychotic medication. We aimed to study the associations of: use versus no use of antipsychotic medication; length of continuous time without antipsychotic medication; cumulative dose of lifetime antipsychotic medication; and type of antipsychotic medication; with brain morphometry in schizophrenia after an average of 10 years of illness. Methods: Data of 63 individuals with schizophrenia (mean duration of illness 10.4 years) from the Northern Finland Birth Cohort 1966 were gathered by interview and from hospital and outpatient records. Structural MRI data at age 34 years were acquired and grey matter volume maps with voxelbased morphometry were analyzed using FSL tools. Results: Of the individuals studied, 15 (24%) had taken no antipsychotic medication during the previous year. Individuals with antipsychotic medication had lower total grey matter (TGM) volume compared with non-medicated subjects, although this association was not statistically significant (Cohen’s d = – 0.51, P = 0.078). Time without antipsychotic medication associated with increased TGM (P = 0.028). Longer time without antipsychotic medication associated with increased regional volume in right precentral gyrus and right middle frontal gyrus. There were no associations between cumulative dose of lifetime antipsychotic medication or type of antipsychotic medication and brain morphometry. Conclusions: Unlike some previous investigators, we found no association between cumulative dose of lifetime antipsychotic medication and brain morphological changes in this population-based sample. However, longer continuous time without antipsychotic medication preceding the MRI scan associated with increased gray matter volume. ß 2015 Elsevier Masson SAS. All rights reserved.

Keywords: Antipsychotics Medicated Non-medicated MRI Schizophrenia

1. Introduction Morphological changes in the brain structure of individuals with schizophrenia have been widely reported. Honea et al. [18] reported in their meta-analysis, that the most consistent findings were of relative deficits in the left superior temporal

gyrus and the left medial temporal lobe [18]. In the longitudinal studies, there is evidence for volume change over time in the grey matter of anterior cingulate, frontal and temporal lobes, hippocampus/amygdala, thalamus, and insula [37,42], and that changes are found in both grey and white matter [31].

* Corresponding author. Institute of Clinical Medicine, Department of Psychiatry, University of Oulu and Oulu University Hospital, P.O. BOX 5000, 90014 Oulu, Finland. Tel.: +358 440690706; fax: +358 8 336169. E-mail address: [email protected].fi (J. Moilanen). http://dx.doi.org/10.1016/j.eurpsy.2015.02.009 0924-9338/ß 2015 Elsevier Masson SAS. All rights reserved.

J. Moilanen et al. / European Psychiatry 30 (2015) 598–605

Recent studies have suggested possible effects of antipsychotic medication on brain morphometry [42,30,29,26,16,12]. It has been suggested that antipsychotic treatments act regionally rather than globally on the brain [42,30], and that the main part of brain volume reduction in schizophrenia is present before treatment is initiated [16]. According to one cross-sectional meta-analysis, major deficits have been detected in the frontal, superior temporal, insular, and parahippocampal regions of neuroleptic-treated compared with neuroleptic-naı¨ve firstepisode schizophrenia patients [26]. In a recent cross-sectional meta-analysis the results showed decreased volumes in intracranial, total brain and total grey matter volumes in antipsychotic-naive subjects, although the decrease was somewhat less pronounced than that observed in medicated subjects [16]. On the basis of a systematic review of longitudinal studies, antipsychotics may reduce the grey matter volume and increase lateral ventricles [12]. Previous reviews [30,45] have suggested that typical antipsychotic drugs may have a greater effect on brain morphometry than atypical drugs. In the study by Scherk and Falkai [35], switching from typical to atypical antipsychotics decreased the pathologically increased basal ganglia volume to the same level as in healthy controls. In the longitudinal study of Ho et al. [17] brain morphometrical changes were in different regions depending on type of antipsychotic in use (typical, atypical, clozapine). In all cases, the differences between the use of typical and atypical antipsychotics have been inconclusive [37]. To the best of our knowledge, there are no previous studies of the association between the length of continuous time without antipsychotics and brain morphometry. In our previous study, we investigated the characteristics and clinical course of individuals with schizophrenia with and without current antipsychotics [28]. In this study, we aimed to study brain morphometry in the same study sample. 2. Aims of the present work We aimed to study whether non-medicated subjects with schizophrenia have different brain morphometry at age 34 compared to medicated subjects, and to evaluate the effect of continuous time without antipsychotics on the brain morphometry of individuals with schizophrenia after on average 10 years of illness. We also studied whether type of medication or cumulative dose of lifetime antipsychotics are associated with brain morphometry. 3. Methods 3.1. The Northern Finland Birth Cohort 1966 The Northern Finland Birth Cohort 1966 (NFBC 1966) is an unselected, general population birth cohort identified during midpregnancy. The present study is based on 10,934 cohort members living in Finland in 1982 and 1997. Altogether 83 individuals forbade the use of their data and were excluded. The Ethics Committee of the Faculty of Medicine at the University of Oulu initially approved the NFBC 1966 project and keeps its study design under continuous review. The principal source of hospitalization data used in this study, the Finnish Hospital Discharge Register (FHDR), covers all general and mental hospitals, in-patient wards of local health centres, and private hospitals nationwide. Until recent years, most patients in Finland who experienced an episode of schizophrenic psychosis were hospitalized [19] and would therefore appear in the FHDR. The proportion of schizophrenia patients who do not receive

599

hospital treatment is still rather low, approximately 15% [32]. All the cohort members over 16 years of age appearing in the FHDR with any mental disorder (i.e., ICD-8 diagnoses 290-309, ICD-9 290-316, and ICD-10 F00-F69, F99) by the end of 1997 were identified. Their case records were scrutinized and diagnoses were assessed for DSM-III-R criteria, after which the diagnoses were rereviewed by a professional panel. The reliability of this procedure with respect to the diagnosis of schizophrenia was good (kappa = 0.85) [19,27]. 3.2. Sample The present study is based on a field study conducted at an age of 34 years, during 1999–2001. A total of 146 individuals (84 men) with a history of one or more known psychotic episodes by the end of 1997 were invited to participate, of whom 92 (63%; 52 men) participated [15]. Non-participants (n = 54) were more often individuals with schizophrenia, had more positive psychotic symptoms, were more often on disability pension, and had experienced more psychiatric hospitalisations [15]. Average duration of illness of the participants was 10.4 (SD 3.7) years. Written informed consent was obtained from the subjects after they had been given a complete description of the study. After the follow-up diagnostic interview, 22 individuals with other psychoses than schizophrenia spectrum disorder were excluded, which left 70 people with a schizophrenia spectrum disorder according to DSM-III-R. Of these 70 individuals, 66 had an adequate MRI scan and for 63 good quality data on life-time antipsychotic use was available. These 63 individuals (53 schizophrenia, two schizophreniform disorder, six schizoaffective disorder, and two delusional disorder) formed our study group for the current study. 3.3. Data on medication We collected data on the subjects’ life-time antipsychotic medication using all the available medical records (hospital and out-patient care case notes), an interview conducted during the field study, and the register of the Finnish Social Insurance Institution on psychoactive medications consumed during 1997. The medical records were obtained on the basis of information concerning the subjects’ treatment facilities, which we received from the FHDR. If the subject had no information in the FHDR, we requested the medical records from the outpatient facilities of the subjects’ area of residence. Individuals in this study had given their permission to collect medical records by signing the written informed consent. We had permission to collect the data from the Ministry of Social Affairs and Health. All medical records were reviewed to record the antipsychotic agent, dose and time period during which the medication had been used. This information was used to calculate the cumulative dose of lifetime antipsychotics, expressed as dose-years of a daily dose of 100 mg chlorpromazine. See Supplementary Appendix for a more exact description of the antipsychotic data [25,8,1,20,36]. The sample was divided into two groups (non-medicated, n = 15; and medicated, n = 48) based on the individuals’ antipsychotics usage during the previous year, because we wanted to study long-term effect of medication on brain morphometry. Longer time period than a year would have decreased the sample size too much. Those without antipsychotics had been nonmedicated on average for 5.7 years (standard deviation, SD 3.5). Current other psychoactive medication use in the non-medicated group was minor (anxiolytic for one subject), whereas in the medicated group 19 subjects used currently anxiolytics, 11 subjects antidepressants, and 2 subjects mood stabilizer (one lithium and one sodium valproate). The information on current use of other

600

J. Moilanen et al. / European Psychiatry 30 (2015) 598–605

psychoactive medication was from interview at age 34 years. The length of continuous time without antipsychotics was calculated using exact dates from the medical records. Altogether 18 of the 63 individuals had been without antipsychotics preceding the MRI scan and the time without antipsychotics varied from 0.3 to 11.4 years (mean 4.9; SD 3.7). The medicated individuals were also categorized on the basis of the type of antipsychotics used during their life-time (mostly typical vs. atypical) and currently (typical, atypical, or both). 3.4. Covariates We studied differences between the medicated and nonmedicated individuals in the following variables:  sex: males versus females;  onset age: in this birth cohort design onset age corresponds with duration of illness. Onset age was ascertained from the medical records, utilizing additional information from health registers, and defined as the age when the first evident psychotic symptoms appeared;  service utilization: cumulative hospital treatment days and episodes due to any psychiatric disorder until the year 2000 were evaluated using the FHDR;  symptomatology and remission: symptoms during the preceding week were measured at the interview using the Positive and Negative Syndrome Scale (PANSS) [23], which measures the number and extent of psychopathological symptoms. Subscales of positive, negative, disorganization, excitement and emotional scores were used [43]. The non-medicated and medicated individuals were divided into remission groups based on the criteria of Andreasen et al. [4] with the exception that, as the symptoms had been assessed only once, the duration of remission criterion of six months was not applied;  functioning: the Social and Occupational Functional Assessment Scale (SOFAS) [40] measures individuals’ social and occupational functioning on a scale of 0–100, with higher score reflecting better functioning;  clinical global impression (CGI) [14] describes the severity of illness on a scale of 1–7, where 1 = healthy and 7 = very ill;  previous or current alcohol use disorder: the information on alcohol use disorder was gathered from the case records and from the diagnostic interview.

3.5. Image acquisition and analysis Structural MRI data were acquired from all participants on a GE Signa system (General Electric, Milwaukee, WI) operating at 1.5 T. T1-weighted SPGR images of the whole brain were collected with a slice thickness of 1.5 mm and in plane voxel size 0.94 mm  0.94 mm, TR = 35 ms, TE = 5 ms, Flip angle = 35o. The images were quality controlled by radiological screening. The MRI methods are described more precisely in previous publications [33,41]. Grey matter volume maps were constructed for each subject’s image using FSL-VBM (http://www.fmrib.ox.ac.uk/fsl/fslvbm/ index.html), a voxel-based morphometry style analysis [7,13] carried out with the FSL tools [39]. First, structural images were brain-extracted using the BET [38]. Next, tissue-type segmentation was carried out using the FAST4 [46]. The resulting greymatter partial volume images were then aligned to the MNI152 standard space using the affine registration tool FLIRT [21,22], followed by nonlinear registration using the FNIRT [2,3], which uses a b-spline representation of the registration warp field

[34]. The resulting images were averaged to create a studyspecific template, to which the native grey matter images were then non-linearly re-registered. The registered partial volume images were then modulated (to correct for local expansion or contraction) by dividing by the Jacobian of the warp field. The modulated segmentated images were then smoothed with an isotropic Gaussian kernel with a full-width-half-maximum of 4 mm to minimize slight mis-registration errors, resulting in smoothed voxelwise maps of grey matter volume. Finally, voxelwise GLM was applied using permutation-based nonparametric testing, correcting for multiple comparisons across space. 3.6. Statistical analyses The illness-related background variables were compared between medicated and non-medicated individuals using independent samples t-test or Mann-Whitney’s U test. Analysis of covariance was used to study the effect of:  antipsychotic medication use on TGM;  type of current and lifetime antipsychotics on TGM. Cohen’s d [9] was used to describe the magnitude of differences in TGM between medicated and non-medicated individuals.

Linear regression analysis was used to study the association between dose years and TGM, and time without antipsychotics and TGM. We present the results unadjusted and adjusted in different models including gender, onset age, remission, and psychiatric treatment days. Beta coefficients can be interpreted as small 0.10, moderate 0.30 and large 0.50 effect, whereas in Cohen’s d 0.2 indicates small, 0.5 moderate, and 0.8 large effect [9]. IBM SPSS Statistics 21.0 was used to perform the analyses. Associations between dichotomized antipsychotic use and grey matter volume were analyzed using a linear regression model at each voxel using permutation-based methods implemented in the FSL software package (http://www.fmrib.ox.ac.uk/ fsl/). Statistical inference using permutation-based statistics within FSL was based on threshold-free cluster enhancement, a method for finding ‘‘clusters’’ within MRI data without having to specify a single peak threshold above which groups of voxels are defined as a cluster. We used the default parameters in threshold-free cluster enhancement: height (default = 2), extent (default = 0.5), and connectivity (default = 6). All analyses were adjusted for TGM volume and gender, and corrected for multiple comparisons. We also conducted additional analyses to determine whether there was an effect of onset age or remission separately and together. 4. Results 4.1. Characteristics of the non-medicated and medicated groups Gender distributions (men/women) were 10 (67%) and 5 (33%) in the non-medicated and 24 (50%) and 24 (50%) in the medicated group. Nine (60%) individuals in the non-medicated and 12 (25%) in the medicated group were in remission. Onset age did not differ between the non-medicated and medicated groups (mean 24.3 vs. 23.4 years, P = 0.48), indicating approximately ten years duration of illness (Table 1). The non-medicated individuals had been in psychiatric hospital care less than medicated individuals before the follow-up interview (median 38 vs. 206 days, P < 0.001). Medicated individuals had more severe illness, lower social functioning score, and more symptoms, compared

J. Moilanen et al. / European Psychiatry 30 (2015) 598–605

601

Table 1 Illness related information on the medicated and non-medicated subjects at age 34 years with their correlation with total grey matter.

Males, n (%) Onset age, mean (SD) Psychiatric hospitalisation Treatment times, median (min–max) Treatment days, median (min–max) In remission, n (%) Current or earlier alcohol use disorder, n (%) Clinical global impression, mean (SD) SOFAS, mean (SD) PANSS Positive, mean (SD)g Negative, mean (SD)g Disorganization, mean (SD)g Excitement, mean (SD)g Emotional, mean (SD)g

Medicated n = 48

Non-medicated n = 15

P

24 (50%) 23.4 (4.3)

10 (67%) 24.3 (3.5)

0.38a 0.48c

7 (0–31) 206 (0–4703) 12 (25%) 10 (21%) 4.9 (1.3) 45 (15)

2 (1–7) 38 (1–478) 9 (60%) 2 (13%) 3.5 (1.6) 65 (13)

<0.001e <0.001e 0.025a 0.71a 0.001c <0.001c

0.04f 0.06f <0.01b 0.13b 0.10d 0.23d

12.4 16.1 18.3 11.9 13.2

11.9 (6.9) 8.9 (3.9) 14.1 (6.7) 10.1 (3.2) 10.1 (4.3)

0.78c <0.001c 0.12c 0.16c 0.025c

0.11d 0.03d 0.05d 0.07d 0.01d

(5.4) (10.0) (9.5) (4.5) (4.6)

Correlation with TGM 0.55b 0.04d

SD: standard deviation; SOFAS: Social and Occupational Functioning Assessment Scale; PANSS: Positive and Negative Syndrome Scale; TGM: total grey matter. a Significance from Chi2 test. b Biserial correlation coefficient. c Significance from independent samples t-test. d Pearson’s correlation coefficient. e Significance from Mann-Whitney’s test. f Spearman’s rank correlation coefficient. g n = 46 for medicated.

to non-medicated individuals. There were no differences between non-medicated and medicated individuals when comparing previous or current alcohol use disorder (13.3% vs. 20.8%, P = 0.71) (Table 1). 4.2. Differences between the non-medicated and medicated groups in brain morphometry Non-medicated individuals had in trend-level greater TGM compared to medicated individuals (649 vs. 617 cm3, d = –0.51, P = 0.078; Table 2), which was no longer close to statistical significance when adjusted for gender, onset age and psychiatric treatment days (637 vs. 618, d = –0.34, P = 0.24). The medicated and non-medicated individuals did not differ in the voxel-based analyses.

continuous time without medication was associated with increased regional volume in right precentral gyrus (P = 0.024; Talairach coordinates (x,y,z): 46,6,26; cluster size: 127) and right middle frontal gyrus (P = 0.026; Talairach coordinates (x,y,z): 34,22,46; cluster size: 119) when adjusted for TGM, gender and onset age (Fig. 2). When also adjusted for remission these associations did not persist. 4.4. Cumulative dose of lifetime antipsychotic medication We found no association between cumulative dose of lifetime antipsychotic medication and TGM (beta = –0.148; P = 0.25; Table 3) or voxel-based grey matter volumes. 4.5. Type of antipsychotic medication

4.3. Continuous time without antipsychotic medication Continuous time without antipsychotic medication preceding the MRI scan was associated statistically significantly with greater volume of TGM (Beta = 0.278; P = 0.028; Table 3) (Fig. 1). When adjusted for gender, onset age and remission this association was no longer significant, although the effect remained similar (Beta = 0.189) (Table 3). In the voxel-based analyses longer

There were no differences in TGM between those who currently used typical (n = 24), atypical (n = 11), or both (n = 10) types of antipsychotics (mean [SD]: 620 [67], 611 [55], and 602 [54], respectively; P = 0.72) or who had used mostly typical (n = 50) or atypical (n = 12) antipsychotics during their lifetime (624 [63] vs. 628 [61], P = 0.85). There were also no significant differences in the voxel-based analyses.

Table 2 Total grey matter of the medicated and non-medicated subjects. Estimated TGM (mm3) means (standard errors)

Crude Adjusted for Gender Gender and remission Gender and onset age Gender and psychiatric treatment days

Medicateda n = 48

Non-medicatedb n = 15

617 (8.8)

649 (15.6)

0.53

0.078

617 619 617 616

638 638 638 642

0.41 0.33 0.40 0.48

0.17 0.26 0.19 0.14

(7.5) (8.4) (7.5) (7.8)

(13.6) (13.8) (13.7) (14.9)

TGM: total grey matter. a Subjects who have used antipsychotic medication during the previous year. b Subjects who have not used antipsychotic medication during the previous year. c Significance from analysis of covariance.

Cohen’s d

Pc

P-valuesc of adjusting variables

Gender < 0.001 Gender < 0.001, remission 0.68 Gender < 0.001, onset age 0.74 Gender < 0.001, treatment days 0.53

602

J. Moilanen et al. / European Psychiatry 30 (2015) 598–605

Table 3 Associations between cumulative dose of lifetime antipsychotic medication and total grey matter, and length of continuous time without antipsychotic medication and total grey matter. Coefficients from the linear regression models B (SE) Dose years Crude Adjusted for Gender Gender and onset age Gender and remission Gender and psychiatric treatment days Time without antipsychotic medication Crude Adjusted for Gender Gender and onset age Gender and remission Gender and psychiatric treatment days

Beta

Pa

7.2 (6.2)

0.148

0.25

6.3 7.1 5.5 9.7

0.129 0.145 0.111 0.198

0.23 0.27 0.39 0.15

(5.2) (6.4) (6.2) (6.6)

5.3 (2.3)

0.278

0.028

3.6 3.5 3.6 4.4

0.193 0.189 0.189 0.232

0.075 0.087 0.12 0.054

(2.0) (2.1) (2.3) (2.2)

P-valuesa of adjusting variables

Gender < 0.001 Gender < 0.001, onset age 0.82 Gender < 0.001, remission 0.80 Gender < 0.001, treatment days 0.41

Gender < 0.001 Gender < 0.001, onset age 0.82 Gender < 0.001, remission 0.95 Gender < 0.001, treatment days 0.44

SE: standard error. a Significance from analysis of covariance.

5. Discussion 5.1. Main results Longer continuous time without antipsychotics preceding the MRI scan was associated statistically significantly with greater volume of TGM. In the voxel-based analyses, longer continuous time without medication was associated with increased regional volume in the right precentral gyrus and right middle frontal gyrus. When remission was taken into account, these associations were no longer statistically significant. Non-medicated individuals had in trend-level greater TGM compared to medicated individuals. We found no association between cumulative dose of lifetime antipsychotics or type of medication and brain morphometric changes. 5.2. Antipsychotic medication and brain morphometry Morphological changes in brain structures have been widely reported in schizophrenia, and several studies have reported the

Fig. 1. Association between continuous time without antipsychotic medication preceding MRI scan and total grey matter.

effect of antipsychotics on these changes [42,30,29,26,16,12]. In a meta-analysis of longitudinal studies, Fusar-Poli et al. [12] reported that antipsychotics may reduce the grey matter volume and increase lateral ventricles. In our sample, non-medicated individuals had in trend-level greater TGM compared to medicated individuals. The systematic reviews on antipsychotics and brain morphometry have not specifically studied time without antipsychotics, which was one our main focuses. We found that time without antipsychotics associated with increased TGM volume and in the voxel-based analyses longer time without medication was associated with increased regional volume in the right precentral gyrus and right middle frontal gyrus. Although the associations were statistically non-significant after adjustment for other factors (e.g. remission), these adjustments did not affect the magnitude of the effect. In our sample, all individuals except one had a history of previous antipsychotic use. Previous antipsychotic use and the positive correlation between the time without antipsychotics and TGM raise the question whether the possible effect of antipsychotics on brain morphometry is reversible. We did not have a longitudinal study, and therefore it is not possible to make this type of interpretation on the basis of our results. Veijola et al. [44] followed a sub-sample (n = 33) of the current study sample with a second MRI scan after nine years. They found that higher cumulative antipsychotic dose between scans was associated with volume reductions in the parietal lobe and periventricular area. They did not study the possible effect of time without antipsychotics, or the period before age 34, which were the focuses of our study. Earlier studies have reported contradictory findings regarding differences in effect of different types of antipsychotics [37]. In the study of Ho et al. [17], type of medication (typical, non-clozapine atypical, or clozapine) was associated with brain morphometrical changes in different regions. In our sample, there was no association between type of medication and brain morphometry. Our findings regarding a positive correlation between time without antipsychotics and grey matter volumes are in line with previous findings of deficits in the frontal regions in neuroleptictreated individuals with schizophrenia [26]. The clinical meaning and mechanisms behind these alterations are still somewhat unclear. Andreasen et al. [5] found that frontal tissue loss was associated with poorer performance in tests of verbal learning, attention and working memory. Artigas [6] wrote a clinical overview concerning the mechanism of action of antipsychotics,

J. Moilanen et al. / European Psychiatry 30 (2015) 598–605

603

Fig. 2. Association of length of continuous time without antipsychotic medication preceding the MRI scans with voxel-based morphometry. Longer continuous time without medication associated with increased regional volume in right precentral gyrus (P = 0.024) and right middle frontal gyrus (P = 0.026) when adjusted for total grey matter, gender and onset age.

with particular emphasis on its action in the prefrontal cortex, stating that histological studies indicate the presence of a large proportion of neurons expressing monoaminergic receptors sensitive to the action of antipsychotics. Konopaske et al. [24] found a 20.5% lower astrocyte and 12.9% lower oligodendrocyte number in the antipsychotic-exposed monkeys when compared to non-exposed. Dorph-Petersen et al. [10] found that chronic exposure to antipsychotics in the same monkeys was associated with similar volume decrements in both the parietal and frontal lobes. 6. Strengths and limitations The role of antipsychotics in the observed brain morphometric changes in schizophrenia is unclear. There are only a limited number of individuals with schizophrenia without antipsychotics and it is not possible to study these subjects in clinical trials [11]. In this population-based birth cohort sample we were able to study brain morphometry in individuals who were currently without antipsychotics and to compare them to those with antipsychotics, which is a remarkable strength of our study. To our knowledge, there are no previous population-based studies on association between medication status and brain morphometry in schizophrenia ten years after the onset of illness. Because our sample was population-based it included individuals who were less severely ill than in studies in clinical settings. Furthermore, there are no previous studies in general investigating time without antipsychotics and brain morphometry. Our data collection on medication use was extensive. We went through all the available hospital and outpatient care case notes, asked about the individuals’ antipsychotic medication history in the interview and also used register data. Using all this information we were able to gather detailed and reliable data about the individuals’ life-time antipsychotic use. The limitations of the study are that the sample was rather small and the individuals were in different phases of the illness. Our study is not necessarily generalizable to the whole original cohort, as the participants generally had less severe illness than the non-participants [15]. We also performed only one MRI scan, which makes it difficult to interpret these associations as causal. One potential confounding factor between medication and brain volume loss is illness severity, as it is conceivable that patients with the most severe illness lose more brain volume over time and also use antipsychotics with high doses and for long periods.

Another potential confounding factor is that current use of other psychoactive medication than antipsychotics was more common within medicated subjects. The lack of information on life-time use of other psychoactive medications is a limitation in our study. We used remission as a marker of severity of illness, which is not necessarily the best possible indicator of the severity of a life-time illness. It should be noted that in this birth cohort study onset age corresponds to duration of illness, which has often been used as a marker of severity of illness. The association between time without antipsychotics and brain morphometry remained when onset age was taken into account. However, when also adjusted for remission the associations were no longer statistically significant. 7. Conclusion We have earlier published a paper [28] reporting a study of the same sample, in which we found that non-medicated individuals had a favourable clinical and functional outcome compared to medicated ones. In this study, we studied the brain morphometry of the same sample, which resulted in a positive correlation between continuous time without antipsychotics preceding the MRI scan and the morphometric changes. Positive correlation between the length of continuous time without antipsychotics and TGM volume raises the question of whether the possible effect of antipsychotic medication on brain morphometry is reversible. Although our sample size was small and the morphometric changes were not extensive, these results indicate that the possible effect of medication discontinuation on brain morphometry and functioning in schizophrenia requires further investigation. Disclosure of interest The authors declare that they have no conflicts of interest concerning this article. Acknowledgements This work was supported by unrestricted grants from the Academy of Finland (grants 110 143, 125 853, 214 273, and 268 336), the Northern Finland Health Care Support Foundation, the Jalmari and Rauha Ahokas Foundation, the Oy H. Lundbeck Foundation, the Sigrid Juse´lius Foundation, the Stanley Medical Research Institute, the NARSAD: Brain and Behavior Research Fund, and The Finnish Medical Society Duodecim Oulu.

J. Moilanen et al. / European Psychiatry 30 (2015) 598–605

604

Appendix. Chlorpromazine equivalents and types for different antipsychotics Agent

Finnish trade name

Equivalenta Type

Asenapine Chlorpromazine

Asenapine Klorproman (t/i), largactil Truxal, cloxan Leponex, froidir Fluanxol Siqualone inj Pacinol Haloperin, serenase Haloperin inj Levozin, nozinan Melpax Moban Zyprexa Neulactil, neuperil Peratsin, pertriptyl Peratsin inj Orap Piportyl inj Sparine Roxiam Risperdal Risperdal consta inj Serdolect Suprium Majeptil Orsanil, tioridil Cisordinol Cisordinol inj, cisordinol acutard inj

5 100

Atypical Woods Typical Kroken

50 100 2 1.07 2 3 2 100 100 10 5 24 8 1.9 2 1.43 100 75 1.5 1 5.33 200 10 100 25 14

Typical Typical Typical Typical Typical Typical Typical Typical Typical Typical Atypical Typical Typical Typical Typical Typical Typical Typical Atypical Atypical Atypical Typical Typical Typical Typical Typical

Chlorprothixene Clozapine Flupentixol Fluphenazine inj. Fluphenazine Haloperidol Haloperidol inj. Levomepromazine Melperone Molindone Olanzapine Pericyazine Perphenazine Perphenazine inj. Pimozide Pipothiazine inj. Promazine Remoxipride Risperidone Risperidone inj. Sertindole Sulpiride Thioproperazine Thioridazine Zuclopenthixol Zuclopenthixol inj.

Reference

Kroken Kroken Kroken PDD PDD Kroken Kroken Kroken Janssen Ahuja Kroken PDD Kroken Kroken PDD Semple PDD PDD Kroken Kroken Kroken PDD Ahuja Kroken Kroken Kroken

t/i: tablet or injection. a Dose that equals 100 mg of chlorpromazine, references [1,8,20,25,36], Woods SW. Webpage (http://www.scottwilliamwoods.com).

References [1] Ahuja N. Antipsychotic drugs. In: Vyas AJ, Ahuja N, editors. Textbook of Postgraduate Psychiatry. Jaypee Brothers Medical Publishers (P) Ltd; 1999 . p. 737–52. [2] Andersson JLR, Jenkinson M, Smith S. Non-linear optimisation; 2007, FMRIB technical report TR07JA1 from http://www.fmrib.ox.ac.uk/analysis/techrep. [3] Andersson JLR, Jenkinson M, Smith S. Non-linear registration, aka spatial normalisation; 2007, FMRIB technical report TR07JA2 from http://www. fmrib.ox.ac.uk/analysis/techrep. [4] Andreasen NC, Carpenter WT, Kane JM, Lasser RA, Marder SR, Weinberger DR. Remission in schizophrenia: proposed criteria and rationale for consensus. Am J Psychiatry 2005;162:441–9. [5] Andreasen NC, Nopoulos P, Magnotta V, Pierson R, Ziebell S, Ho BC. Progressive brain change in schizophrenia: a prospective longitudinal study of first-episode schizophrenia. Biol Psychiatry 2011;70(7):672–9. [6] Artigas F. The prefrontal cortex: a target for antipsychotic drugs. Acta Psychiatr Scand 2010;121(1):11–21. [7] Ashburner J, Friston KJ. Voxel-based morphometry – the methods. Neuroimage 2000;11:805–21. [8] Bazire S. In: Bazire S, editor. Psychotropic Drug Directory 2003/2004. Fivepin Publishing; 2003. p. 179–80. [9] Cohen J. A power primer. Psychol Bull 1992;112(1):155–9. [10] Dorph-Petersen KA, Pierri JN, Perel JM, Sun Z, Sampson AR, Lewis DA. The influence of chronic exposure to antipsychotic medications on brain size before and after tissue fixation: a comparison of haloperidol and olanzapine in macaque monkeys. Neuropsychopharmacology 2005;30:1649–61. [11] Emsley R, Fleischhacker WW. Is the ongoing use of placebo in relapseprevention clinical trials in schizophrenia justified? Schizophr Res 2013;150(2–3):427–33. [12] Fusar-Poli P, Smieskova R, Kempton MJ, Ho BC, Andreasen NC, Borgwardt S. Progressive brain changes in schizophrenia related to antipsychotic treatment? A meta-analysis of longitudinal MRI studies. Neurosci Biobehav Rev 2013;37(8):1680–91. [13] Good CD, Johnsrude IS, Ashburner J, Henson RNA, Friston KJ, Frackowiak RSJ. Voxel-based morphometric study of ageing in 465 normal adult human brains. Neuroimage 2001;14:21–36. [14] Guy W:. EDCEU Assessment Manual for Psychopharmacology – Revised (DHEW Pupl No ADM 76 338). Rockville, MD, U.S.: Department of Health, Education, and Welfare, Public Health Service, Alcohol, Drug Abuse, and Mental Health Administration, NIMH Psychopharmacology Research Branch, Division of Extramural Research Programs; 1976. p. 534–7.

[15] Haapea M, Miettunen J, Veijola J, Lauronen E, Tanskanen P, Isohanni M. Non-participation may bias the results of a psychiatric survey: an analysis from the survey including magnetic resonance imaging within the Northern Finland 1966 Birth Cohort. Soc Psychiatry Psychiatric Epidemiol 2007;42: 403–9. [16] Haijma SV, Van Haren N, Cahn W, Koolschijn PC, Hulshoff Pol HE, Kahn RS. Brain volumes in schizophrenia: a meta-analysis in over 18 000 subjects. Schizophr Bull 2013;39(5):1129–38. [17] Ho BC, Andreasen NC, Ziebell S, Pierson R, Magnotta V. Long-term antipsychotic treatment and brain volumes: a longitudinal study of first-episode schizophrenia. Arch Gen Psychiatry 2011;68(2):128–37. [18] Honea R, Crow TJ, Passingham D, Mackay CE. Regional deficits in brain volume in schizophrenia: a meta-analysis of voxel-based morphometry studies. Am J Psychiatry 2005;162(12):2233–45. [19] Isohanni M, Ma¨kikyro¨ T, Moring J, Ra¨sa¨nen P, Hakko H, Partanen U, et al. A comparison of clinical and research DSM-III-R diagnoses of schizophrenia in a Finnish national birth cohort. Soc Psychiatry Psychiatr Epidemiol 1997;32: 303–8. [20] Janssen B, Weinnmann S, Berger M, Gaebel W. Validation of polypharmacy process measures in inpatient schizophrenia care. Schizophr Bull 2004;30: 1023–33. [21] Jenkinson M, Smith S. A global optimisation method for robust affine registration of brain images. Med Image Anal 2001;5(2):143–56. [22] Jenkinson M, Bannister PR, Brady JM, Smith SM. Improved optimisation for the robust and accurate linear registration and motion correction of brain images. Neuroimage 2002;17(2):825–41. [23] Kay SR, Fiszbein A, Opler LA. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull 1987;13:261–76. [24] Konopaske GT, Dorph-Petersen KA, Sweet RA, Pierri JN, Zhang W, Sampson AR, et al. Effect of chronic antipsychotic exposure on astrocyte and oligodendrocyte numbers in macaque monkeys. Biol Psychiatry 2008;63(8):759–65. [25] Kroken RA, Johnsen E, Ruud T, Wentzel-Larsen T, Jørgensen HA. Treatment of schizophrenia with antipsychotics in Norwegian emergency wards, a crosssectional national study. BMC Psychiatry 2009;9:24. [26] Leung M, Cheung C, Yu K, Yip B, Sham P, Li Q, et al. Gray matter in first-episode schizophrenia before and after antipsychotic drug treatment. Anatomical likelihood estimation meta-analyses with sample size weighting. Schizophr Bull 2011;37(1):199–211. [27] Moilanen K, Veijola J, La¨ksy K, Ma¨kikyro¨ T, Miettunen J, Kantoja¨rvi L, et al. Reasons for the diagnostic discordance between clinicians and researchers in schizophrenia in the Northern Finland 1966 Birth Cohort. Soc Psychiatry Psychiatric Epidemiol 2003;38:305–10. [28] Moilanen J, Haapea M, Miettunen J, Ja¨a¨skela¨inen E, Veijola J, Isohanni M, et al. Characteristics of subjects with schizophrenia spectrum disorder with and without antipsychotic medication – a 10-year follow-up of the Northern Finland 1966 Birth Cohort study. Eur Psychiatry 2013;28(1):53–8. [29] Moncrieff J, Leo J. A systematic review of the effects of antipsychotic drugs on brain volume. Psychol Med 2010;40(9):1409–22. [30] Navari S, Dazzan P. Do antipsychotic drugs affect brain structure? A systematic and critical review of MRI findings. Psychol Med 2009;39(11):1763–77. [31] Olabi B, Ellison-Wright I, McIntosh AM, Woods SJ, Bullmore E, Lawrie SM. Are there progressive brain changes in schizophrenia? A meta-analysis of structural magnetic resonance imaging studies. Biol Psychiatry 2011;70(1):88–96. [32] Pera¨la¨ J, Suvisaari J, Saarni SI, Kuoppasalmi K, Isometsa¨ E, Pirkola S, et al. Lifetime prevalence of psychotic and bipolar I disorders in a general population. Arch Gen Psychiatry 2007;64(1):19–28. [33] Ridler K, Veijola JM, Tanskanen P, Miettunen J, Chitnis X, Suckling J, et al. Frontocerebellar systems are associated with infant motor and adult executive functions in healthy but not in schizophrenia. Proc Natl Acad Sci U S A 2006;103(42):15651–6. [34] Rueckert D, Sonoda LI, Hayes C, Hill DLG, Leach MO, Hawkes DJ. Non-rigid registration using free-form deformations: application to breast MR images. IEEE Trans Med Imaging 1999;18(8):712–21. [35] Scherk H, Falkai P. Effects of antipsychotics on brain structure. Curr Opin Psychiatry 2006;19(2):145–50. [36] Semple D, Smyth R. In: Semple D, Smyth R, editors. Oxford Handbook of Psychiatry. 2nd ed., Oxford University Press; 2009 [Table 6.3]. [37] Shepherd AM, Laurens KR, Matheson SL, Carr VJ, Green MJ. Systematic metareview and quality assessment of the structural brain alterations in schizophrenia. Neurosci Biobehav Rev 2012;36(4):1342–56. [38] Smith S. Fast robust automated brain extraction. Hum Brain Mapp 2002;17(3): 143–55. [39] Smith SM, Jenkinson M, Woolrich M, Beckmann CF, Behrens TEJ, JohansenBerg H, et al. Advances in functional and structural MR image analysis and implementation as FSL. Neuroimage 2004;23:208–19. [40] Spitzer RL, Gibbon M, Endicott J. Global Assessment Scale (GAS), Global Assessment of Functioning (GAF) scale, Social and Occupational Functional Assessment Scale (SOFAS). In: Rush JA, et al., editors. Handbook of psychiatric measures. Washington: American Psychiatric Association; 2000. p. 96–100. [41] Tanskanen P, Ridler K, Murray G, Haapea M, Veijola J, Ja¨a¨skela¨inen E, et al. Morphometric brain abnormalities in schizophrenia in a population-based sample: relationship to duration of illness. Schizophr Bull 2010;36(4):766–77. [42] Torres US, Portela-Oliveira E, Borgwardt S, Busatto GF. Structural brain changes associated with antipsychotic treatment in schizophrenia as revealed by voxel-based morphometric MRI: an activation likelihood estimation metaanalysis. BMC Psychiatry 2013;13:342.

J. Moilanen et al. / European Psychiatry 30 (2015) 598–605 [43] van der Gaag M, Hoffman T, Remijsen M, Hijman R, de Haan L, van Meijel B, et al. The five-factor model of the Positive and Negative Syndrome Scale II: a ten-fold cross-validation of a revised model. Schicophr Res 2006;85(1–3):280–7. [44] Veijola J, Guo JY, Moilanen JS, Ja¨a¨skela¨inen E, Miettunen J, Kyllo¨nen M, et al. Longitudinal changes in total brain volume in schizophrenia: relation to symptom severity, cognition and antipsychotic medication. PLoS One 2014;9:e101689.

605

[45] Vita A, De Peri L, Deste G, Sacchetti E. Progressive loss of cortical gray matter in schizophrenia: a meta-analysis and meta-regression of longitudinal MRI studies. Transl Psychiatry 2012;2:e190. [46] Zhang Y, Brady M, Smith S. Segmentation of brain MR images through a hidden Markov random field model and the expectation maximization algorithm. IEEE Trans Med Imaging 2001;20:45–57.