Hippocampal volume in subjects at clinical high-risk for psychosis: A systematic review and meta-analysis

Accepted Manuscript Title: Hippocampal volume in subjects at clinical high-risk for psychosis: A systematic review and meta-analysis Author: Anna Walter Claudia Suenderhauf Harrisberger Fabienne Claudia Lenz Renata Smieskova Yoonho Chung Tyrone D. Cannon Carrie E. Bearden Charlotte Rapp Kerstin Bendfeldt Stefan Borgwardt Tobias Vogel PII: DOI: Reference:

S0149-7634(16)30422-5 http://dx.doi.org/doi:10.1016/j.neubiorev.2016.10.007 NBR 2625

To appear in: Received date: Revised date: Accepted date:

11-7-2016 9-10-2016 11-10-2016

Please cite this article as: Walter, Anna, Suenderhauf, Claudia, Fabienne, Harrisberger, Lenz, Claudia, Smieskova, Renata, Chung, Yoonho, Cannon, Tyrone D., Bearden, Carrie E., Rapp, Charlotte, Bendfeldt, Kerstin, Borgwardt, Stefan, Vogel, Tobias, Hippocampal volume in subjects at clinical high-risk for psychosis: A systematic review and meta-analysis.Neuroscience and Biobehavioral Reviews http://dx.doi.org/10.1016/j.neubiorev.2016.10.007 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Hippocampal volume in subjects at clinical high-risk for psychosis: A systematic review and meta-analysis

Anna Waltera*, Claudia Suenderhaufb, Harrisberger Fabiennea, Claudia Lenza, Renata Smieskovaa, Yoonho Chungc, Tyrone D. Cannonc, Carrie E. Beardend, Charlotte Rappe, Kerstin Bendfeldtf, Stefan Borgwardta, Tobias Vogela

a

Psychiatric University Clinics (UPK) Basel, Basel, Switzerland

b

Clinical Pharmacology and Toxicology, University Hospital Basel, Basel, Switzerland

c

Department of Psychology, Yale University, New Haven, Connecticut

d

Departments of Psychiatry, Psychology and Brain Research Institute, Semel Institute for Neuroscience and Human Behavior, University of California, Los Angeles, California

e

Center for Psychosis Treatment, Psychiatry Clinics Solothurn, Solothurn, Switzerland

f

Medical Image Analysis Center, University of Basel, Basel, Switzerland

Corresponding author: Anna Walter, Department of Psychiatry, University of Basel, Wilhelm Klein-Strasse 27, 4012 Basel, Switzerland, Phone: +41 61 325 51 11, Fax: + 41 61 325 55 12, e-mail: [email protected]

Highlights 

Magnetic resonance imaging studies of hippocampal volume in clinical high risk patients compared to healthy controls.



Clinical high risk patients showed no hippocampal volume reduction.



Hippocampal volume cannot be used as a biomarker.



This meta-analysis fosters the neurodegenerative hypothesis of hippocampal volume reduction around transition to psychosis.

Abstract Several magnetic resonance imaging studies have reported reductions in hippocampal volume in patients with psychosis. It is unclear whether structural abnormalities predate illness onset. We conducted a detailed, systematic literature search for studies reporting hippocampal volume in subjects with clinical high-risk, compared to healthy controls. The overall sample size comprised 1429 subjects. Meta-analysis revealed no difference for left, but a small, albeit significant, difference for right hippocampal volume, such that clinical high-risk patients had slightly smaller hippocampal volume than healthy controls (g = 0.24, p = 0.0418). Meta-regression indicated a moderating effect of manual tracing approach, due to one outlying site. The small difference on the right side did not remain significant (g = 0.14, 95%CI = [-0.03–0.32], p = 0.11) after removal of this outlier.

2

This meta-analysis suggests that there is no reduction in hippocampal volume before transition to psychosis and hippocampal volume cannot be used as a biomarker in clinical high-risk individuals.

Keywords Hippocampus; Clinical high-risk; Biomarker; Magnetic resonance imaging; At-risk mental state, Ultra high-risk; Psychosis; Early detection

1. Introduction The hippocampus is a key brain region in the pathophysiology of psychosis and structural and functional changes in the hippocampus have been consistently reported in psychotic patients (Adriano et al., 2012; Shenton et al., 2001; Tamminga et al., 2010; van Erp et al., 2015). The hippocampus is especially involved in the constructive process of spatial and declarative memory - the ability to learn, store, and retrieve information (Loureiro et al., 2012). The latter is of particular interest in psychosis and deficits are widely recognised as a consistent and critical component of schizophrenia (Heckers, 2001; Ragland et al., 2015; Rasetti et al., 2014). Several subregions have been differentially associated with deficits associated with psychosis. The anterior or ventral parts of the hippocampus seem to be involved in affective and cognitive deficits (Gothelf et al., 2000), as well as other complex

behaviours

such

as

stress,

emotion

(mediating

anxiety-related

behaviours), sensory–motor integration and goal-directed activity (Small et al., 2011; Strange et al., 2014). Several different studies have demonstrated that the posterior or dorsal hippocampus is more likely to be involved in memory and cognitive processing (Small et al., 2011). There is considerable evidence for abnormalities in the posterior hippocampus in schizophrenia (Lipska et al., 1993). 3

It has also been hypothesised that dopaminergic psychotic features may be driven by abnormally heightened hippocampal activity (Grace, 2012; Tamminga and Zukin, 2015). Meta-analysis in patients with diagnosed schizophrenia found significant bilateral reduction in hippocampal volume compared to healthy controls (Adriano et al., 2012; van Erp et al., 2015). However, in other neuropsychiatric disorders too, including major depressive disorder, anxiety disorder, bipolar disorder, and schizophrenia, meta-analysis of the relation between hippocampal volume and the BDNF rs6265 genotype revealed that patients had smaller hippocampal volume than healthy controls, regardless of the genotype (Harrisberger et al., 2015). Many factors are likely to underlie hippocampal abnormalities in patients with psychosis (including dose and duration of antipsychotic treatment, and numerous other confounding factors associated with the experience of the disorder). This is the reason why it is crucial to define whether these abnormalities are already present in patients with clinical high-risk for psychosis. Multivariate models incorporating risk factors from clinical, demographic, neurocognitive, and psychosocial assessments achieve high levels of predictive accuracy in help-seeking high-risk individuals (Cannon, 2015). An individualised risk calculator that could improve clinical decision making is available (Cannon, 2015). Unfortunately, these assessments are sophisticated and time-consuming. There is therefore a strong imperative to develop a biomarker for clinical high-risk that can be easily and objectively obtained as part of a routine screening exam and with minimum discomfort or risk to the patient. Measurement of hippocampal volume meets all these criteria. A review of hippocampal volume considering all types of high-risk subjects (Ganzola et al., 2014), and using three different classes of subjects, based on (1) 4

the sole presence of psychotic symptoms, (2) all kinds of risk factors (including low IQ, and high scores in either the Structured Interview for Schizotypy and Child Behavior Checklist or both), and (3) presence of combined risk symptoms, did not systematically assess patients with established criteria of clinical high-risk for psychosis (Fusar-Poli et al., 2015), and furthermore did not consider overlapping samples (Ganzola et al., 2014). Thus, no conclusive results could be achieved, aside from some vague evidence for hippocampal abnormalities preceding schizophrenia onset. In this meta-analysis, we wanted to focus on clinical high-risk individuals categorised by established criteria (Häfner et al., 1992; Miller et al., 2003; Riecher-Rössler et al., 2007; Schultze-Lutter, 2009; Yung et al., 2005, 1996), as these patients have about a 18%-36% risk of developing psychosis after 6 months to 3 years (Fusar-Poli et al., 2012a). Furthermore, research on the clinical high-risk for psychosis has progressed exponentially and preventive therapy is now possible (Addington et al., 2013). A range of neuroimaging techniques showed alterations in brain structure (DeLisi, 2008; Kempton et al., 2010; Mechelli et al., 2011), function (Fusar-Poli et al., 2007), and neurochemistry (Howes et al., 2007) in patients with clinical high-risk for psychosis (Borgwardt et al., 2011; DeLisi, 2011; Fusar-Poli and Borgwardt, 2012). These neuroimaging studies have shown that alterations in brain anatomy found in established psychosis are also present in people with a clinical high-risk (Smieskova et al., 2010). Overall, these patients exhibit qualitatively similar, but less pronounced, structural brain abnormalities than patients with established psychosis. Individual MRI studies in clinical high-risk individuals may be biased because they are typically obtained in small samples, are heterogeneous and may contain contradictory results.

Thus, a meta-analytic study should be useful to clarify 5

whether hippocampal volume is reduced in patients with a clinical high-risk for psychosis. This could support the neurodevelopmental theory of psychosis (Lewis and Levitt, 2002; Rapoport et al., 2012) and would be a putative biomarker. On the other hand, if there is no reduction in hippocampal volume, this would tend to support the neurodegenerative theory of psychosis (Lieberman, 1999), according to which, hippocampal volume is only reduced around or after the transition to psychosis. One important issue is the MRI technique used to measure hippocampal volume. We wished to focus on hippocampal volume, which is obtained using the region-of-interest (ROI) method. The volume of the structure (in mm3) is considered by accurately identifying its boundaries (Spoletini et al., 2011; Velakoulis et al., 1999). Investigators commonly use the measurement of manually delineated, anatomically defined ROI to assess and localize grey matter disruptions. The main advantage of manually selecting ROI is that the method is not susceptible to artefacts deriving from interfaces between bone, brain and air in the orbitofrontal areas. Although manually delineate anatomical regions might bear some inaccuracy when very small areas are investigated, ROI analyses were in general praised to enjoy substantial anatomic validity (Perlini et al., 2012). As the manual segmentation is highly time consuming, automated methods introduce substantial gains. On the other hand, the implementation of those methods is a challenge because of the low contrast of this structure in relation to the neighbouring areas of the brain (Dill et al., 2015). ROI analysis gives as an output the absolute the number of voxels in the investigated brain region, while voxel-based morphometry approaches return relative changes in grey matter within each voxel. This makes inter-study comparisons more feasible (Giuliani et al., 2005). In ROI analysis, one has to keep in mind that discrepancies in hippocampal subfield analysis could be due to different methods used to delineate the hippocampus (Ho et al., 2016).

6

We performed meta-analyses of hippocampal volume in patients meeting criteria for a clinical high-risk for psychosis, in order to clarify: (1) whether hippocampal volume reduction characterises clinical high-risk patients compared to healthy controls, and (2) whether the right and left hippocampus are affected in the same manner.

2. Materials and methods 2.1.Search strategy We applied a systematic search strategy, as recommended by the PRISMA group (Moher et al., 2015). The PRISMA Statement consists of an item checklist and a flow diagram that is essential for transparent reporting in systematic reviews and was adopted in the present manuscript. A systematic literature search was conducted in the public databases Medline (Ovid MEDLINE 1946 to November Week 3 2015, Ovid MEDLINE In-Process & Other Non-Indexed Citations December 10, 2015) and EMBASE (Embase 1996 to 2015 Week 50), using the key words MRI OR magnetic resonance imaging OR neuroimaging AND psychos* OR schizophren* AND high-risk OR at risk mental state OR prodrom* AND hippocamp*. We applied no restrictions on based on language or date of publication. We performed all searches on 17 December 2015. We re-ran searches on 15 March 2016, and updated the results. An overview of the applied search procedure is given in Figure 1.

2.2. Selection of studies Titles and abstracts of the articles that emerged were screened to filter for studies that reported hippocampal volume in individuals diagnosed with clinical high-risk 7

state for psychosis and healthy controls. Two members of our group (AW, TV) independently screened all abstracts to identify potential studies for meta-analysis and compared their search results to create the final database. For all of the studies in this compilation, full article copies were obtained and completely reviewed by the two group members for their suitability for inclusion and for metaanalysis. We resolved disagreements by applying the inclusion and exclusion criteria listed below and by referral to a third group member (CS).

Inclusion criteria for the studies were the following: 1. Presence of a healthy control comparison group. 2. Clinical high-risk state was diagnosed according to established criteria (CAARMS, PACE, SIPS/SOPS, BSIP, basic symptoms) (Häfner et al., 1992; Miller et al., 2003; Riecher-Rössler et al., 2007; Schultze-Lutter, 2009; Yung et al., 2005, 1996). 3. Hippocampal volume measured by the ROI method, which involved manually tracing hippocampal borders or employing a fully automated segmentation method. 4. Availability of mean (± SD) values of right and left hippocampal volume for each group. 5. Observational study design. 6. Publication in a peer-reviewed journal.

Exclusion criteria were the following: 1. Comorbidity with medical or neurological illnesses in patients or healthy controls. 2. History of psychiatric disorders in healthy control participants. 8

3. Lack of volumetric data (e.g., VBM studies). 4. Any post-mortem assessments. 5. Study groups of fewer than 10 (for patient or the healthy control group). 6. Studies on the chromosome 22q11.2 deletion syndrome.

The initial search, conducted in December 2015, produced 255 articles. After removing duplicates, 184 articles remained. 11 studies fulfilled the above mentioned inclusion criteria (a complete list of the 11 studies included is presented in Supplement, Table 1). As 4 studies were performed in Melbourne, and 2 studies in Boulder, we contacted the authors to ascertain whether there was any overlap between the samples. As a result, we had to exclude 3 studies due to overlap (Mittal et al., 2013; Phillips et al., 2002; Wood et al., 2005). Finally, our dataset used for meta-analysis consisted of 7 independent studies (Bühlmann et al., 2010; Dean et al., 2015; Hurlemann et al., 2008; Klauser et al., 2015; Velakoulis et al., 2006; Witthaus, 2010; Wood et al., 2010) and one multisite study (Cannon et al., 2015) with a total of 1429 subjects, 490 healthy controls and 939 individuals diagnosed with clinical high-risk for psychosis (Table 1).

2.3. Data extraction The following variables were extracted: First author name, publication year, sample size (healthy controls and clinical high-risk), gender distribution, mean age and standard deviation, clinical high-risk criteria, structural MRI measurement technique, field strength of scanner, inter- and intra-rater reliability, mean right and left uncorrected hippocampal volume and standard deviation, whole brain volume and intracranial volume. If necessary, units were transformed. When data were 9

missing but computation based on the original publication was possible, the missing values were calculated (i.e. we created one high-risk group out of two, if clinical high-risk patients were further differentiated into early and late high-risk. Otherwise, the authors of the original studies were contacted to provide missing variables.

2.4. Quality assessment The quality of the studies was assessed using an item-checklist constructed specifically for our review process and similar to the previously published quality assessment (Fusar-Poli et al., 2013). The recorded variables were assessed in terms of precision, directness and consistency of the data. The categories scored in the quality assessment are listed in the Supplement (Table 2), with the range of 0 to 2 points for minimum to maximum quality, respectively. The code of the range was developed a priori and modified after the first run of quality assessment. The disagreements were discussed between the authors (AW and CR) and the consensus was put in Table 2. Quality assessment was conducted in the following categories: (1) role of the funding (2) sample size (3) clearly reported inclusion criteria (4) exclusion criteria, substance abuse (5) gender distribution (6) race distribution (7) IQ, educational level (8) previous antipsychotic medication (9) psychopathological ratings 10

and in case of manual tracing: (10) inter-rater reliability (11) intra-rater reliability (12) blindness of rater The included studies were rated according to the sum of the points and characterised as high quality (above 80% of the maximal sum of points), moderate-high (60–79%), moderate (40–59%), moderate-low (20–39%), and low quality studies (below 19%) (see more details in Table 2).

2.5. Statistical analysis Quantitative meta-analysis and meta-regression was performed using the metaphor package for R 3.2.2. statistical software (R core Team). The extracted data were converted to Hedges’s g effect sizes, which provides an unbiased standardised mean difference and incorporates a correction for small sample size (Lipsey and Wilson, 2001). Hedge’s g was calculated from mean right and left hippocampal volume, standard deviations and sample sizes. Random effects model were applied with the DerSimonion-Laird estimator, using the metafor package version 1.9-8 in R (DerSimonian and Laird, 1986). The random effects model shows more flexibility with respect to variable effect size in different studies and study populations (Cooper et al., 2009), as it incorporates the between-study variance theta 2. With high between-study heterogeneity, the random effects model is the model of choice, rather than the fixed effect model (Ioannidis et al., 2007).

2.6. Study heterogeneity

11

Cochran’s Q test was then used to calculate between-group heterogeneity; the magnitude of heterogeneity was assessed by I2 (Higgins and Thompson, 2002). I2 is an estimate of variability across studies based on heterogeneity rather than chance, ranging from 0 to 100%. Values above 25%, 50% and 75% correspond to low, moderate and high heterogeneity, respectively. We investigated potential publication bias by funnel plot asymmetry and Egger’s regression test (Egger et al., 1997). In case of a bias, we planned to perform the "trim and fill" method (Duval and Tweedie, 2000). Moreover, meta-regression analyses using the metafor package version 1.9-8 were carried out to assess the impact of possible moderating factors such as publication year, age of participants, gender ratio, sample size, quality rating, magnetic field strength, and applied hippocampal measuring techniques. For sensitivity analysis, each side was checked for potential outliers (Viechtbauer and Cheung, 2010). Power analysis was performed by using G*Power (Faul et al., 2007).

3. Results 3.1. Description of studies All included studies were published between 2006 and 2015. A total of 1429 subjects (939 clinical high-risk, 490 healthy controls from 8 independent studies (including one multisite study)) were selected for this random effects meta-analysis (mean age: 21.1 years, 57.8 % males) (Bühlmann et al., 2010; Cannon et al., 2015; Dean et al., 2015; Hurlemann et al., 2008; Klauser et al., 2015; Velakoulis et al., 2006; Witthaus, 2010; Wood et al., 2010).

12

3 studies reported ethnicity (Supplement 1). Quality assessment revealed that all included studies were rated as being of moderate-high (60-79%) or high (above 80% of the maximal sum of points) quality (Table 2).

3.2. Meta-analysis of right and left hippocampal volume The meta-analysis of all datasets (k = 8) for right hippocampal volume indicated slightly smaller hippocampal volume for high-risk individuals than for healthy controls (g = 0.24, 95%CI = [0.01–0.48], p = 0.0418, see Fig. 2 and Table 3) and no difference for left hippocampal volume (g = 0.25, 95%CI = [-0.04–0.54], p = 0.0860, see Fig. 3 and Table 3) with indications of significant between-study heterogeneity for the right (I2 = 69.30%, Q(df = 7) = 22.80, p < 0.0018) and left (I2 = 79.99%, Q(df = 7) = 34.99, p < 0.0001) sides. Visual inspection of the funnel plot (se Fig. 4 & 5) as well as the Egger's regression test (right p = 0.38, left p = 0.43) indicated no potential publication bias.

3.3. Meta-Regression Meta-regression analysis did not reveal any effect of year of publication (p = 0.08 left, 0.09 right), sample size (p = 0.64 left, 0.58 right), gender ratio (p = 0.16 left, 0.59 right), age of participants (p = 0.42 left, 0.09 right), quality rating (p = 0.39 left, 0.65 right), or magnetic field strength (p = 0.28 left, 0.11). However, the analysis of the meta-regressions indicated a potential moderating effect of the measurement technique (p = 0.0317 left, 0.0163 right, Table 4), which seemed to be driven by the moderating effect of the manual tracing technique (p = 0.009 left, 0.0041 right), but not of the automated measurement technique (p = 0.77 left, 0.94 right). 13

3.4. Sensitivity Analysis Sensitivity analysis indicated that, on the left side, Wood’s study (Wood et al., 2010) might be a potential outlier, and on the right side Hurleman’s study (Hurlemann et al., 2008) (Table 5). Removal of the respective study on the corresponding side still revealed no difference between high-risk individuals and healthy controls on the left side (g = 0.13, 95%CI = [-0.10–0.37], p = 0.27, see Fig. 3), whereas the small difference on the right side was no longer significant (g = 0.14, 95%CI = [-0.03–0.32], p = 0.11, see Fig. 2). Additionally, after removing the respective study, meta-regression indicated that there was no longer a potential moderating effect of the measurement technique on the left side (p = 0.16) and only a trend on the right side (p=0.06).

3.5. Power Analysis Power analysis suggested that 253 individuals (506 in total) of each group for left, and 274 individuals (548 in total) for right hippocampal volume would be necessary to achieve a power of 80% at α-level of 0.05 (two-sided). Power analysis using the effect sizes without outliers raises the number to 930 individuals per group (1960 in total) on the left side, and 802 (1604 in total) on the right side.

4. Discussion In this meta-analytic study and after outlier removal, we found no significant reduction in left or right hippocampal volume in individuals with a clinical high-risk for psychosis compared to healthy controls. These results are in line with a longitudinal study in clinical high-risk individuals, which revealed a significant 14

decrease in hippocampal volume after the transition to psychosis, and which implies that reduction in hippocampal volume is a marker of illness rather than a marker of risk (Walter et al., 2012). At the first glance, this seems to be contradictory to a meta-analysis in high-risk individuals (n = 198), using the VBM approach, and which detected significant reduction in GM compared to healthy controls (n = 254) in regions including the right hippocampus (p < 0.001) (Fusar-Poli et al., 2012b) and a recent VBM study that found relative reduction in grey matter in the right middle/superior temporal cortex of clinical high-risk individuals (Nenadic et al., 2015). It is interesting that studies in genetic high-risk individuals (i.e. two or more first- or second-degree relatives with a clinical diagnosis of schizophrenia) found no significant difference in hippocampal volume compared to healthy controls (Bois et al., 2015), whereas other studies in non-psychotic, first-degree relatives of individuals diagnosed with schizophrenia revealed significantly smaller left hippocampi compared to healthy controls (Seidman et al., 2014). On the other hand, measures of hippocampal volume alone may miss subtle but important changes in hippocampal shape and size that occur during adolescence, which is the typical onset age for the psychosis high-risk state. Using a volumetric approach alone, changes in one area of a structure or a specific subfield may be missed entirely or volumetric increases in one area may cancel out decreases in another area (Dean et al., 2015) and converging evidence suggests early atrophy of CA1 in schizophrenia, with later extension to other hippocampal subfields (Ho et al., 2016; Narr et al., 2004). The differing hippocampal delineation methods have also to be taken into account, as it has been discussed that divergent measuring protocols could have an important impact on hippocampal volume (Geuze et al., 2005).

15

Patients in our meta-analysis were defined by several different high-risk criteria (i.e. PACE, SIPS, ERIraos, and CAARMS), including three heterogeneous subgroups (attenuated psychotic symptoms, brief limited intermittent psychotic symptoms, and trait vulnerability plus a marked decline in psychosocial functioning). Moreover, lack of volumetric abnormalities in clinical high-risk individuals may be due to the fact that mild deficits are too subtle or more of a functional nature at this stage, and are not (yet) reflected in macroscopic brain structure. Indeed, clinical high-risk was associated with increased hippocampal resting state activity (Allen et al., 2015), and subsequent resolution of the high-risk state was linked to normalisation of activity. One could hypothesise that functional changes in high-risk individuals are even less prominent than in patients with full blown psychosis, where several functional abnormalities have been discussed. For example, recent studies indicated that activation of the hippocampus does lead to a hyperdopaminergic state (Grace, 2012). In juvenile rats, it was demonstrated that activation of low-affinity D2-like dopamine receptors simulating a hyperdopaminergic tone leads to lasting depression of NMDA receptors at the hippocampal–prefrontal projection, thus resulting in cognitive deficits, a hallmark in psychosis (Banks et al., 2015). An alternative theory postulates hippocampal hyperactivity due to subfield-specific hippocampal molecular pathology, including increased NMDA receptors in hippocampal CA3, along with postsynaptic changes. These changes in CA3 in psychosis would lead to increased vulnerability of the hippocampus to pathologically increased neuronal activity, feed-forward excitation and false memory formation, which would be a potential explanation for psychotic symptoms (Tamminga and Zukin, 2015). 16

Heckers et al. explored GABAergic mechanisms of this hippocampal hyperactivity, which could contribute to some psychotic symptoms (Heckers and Konradi, 2015). A model integrating early stress exposure and hippocampal pathology is intriguing, since the hippocampus is known to be a stress-sensitive brain region (Magarinos and McEwen, 1995; McEwen, 2000; Sapolsky, 2000) and high levels of the stress hormone cortisol are associated with a smaller hippocampal volume in first episode schizophrenia patients (Mondelli et al., 2010). In addition, Samudra et al. (2015) suggested decreased hippocampal connectivity with limbic and frontal contributions in psychosis. Bernard et al. (2015) provided evidence for decreasing fractional anisotropy (FA) values in hippocampal-thalamic connections in clinical high-risk individuals. Integration of glutamatergic and mesolimbic dopaminergic inputs to the Nucleus accumbens (Loureiro et al., 2015) is thought to be a crucial neural mechanism by which appropriate attention to environmentally salient versus non-salient cues is regulated (Cahill et al., 2014; Pascoli et al., 2014).

5. Conclusion In summary, this review and meta-analysis helps to elucidate the question as to whether hippocampal volume reduction observed in patients with schizophrenia is of neurodevelopmental origin or neurodegenerative origin. As hippocampal volume reduction seemed to characterise first- episode and chronic patients equally, but not individuals with a clinical high-risk for psychosis, hippocampal volume cannot be considered as an indicator or biomarker for the clinical high-risk state.

Conflict of interest

17

The authors certify that they have no commercial associations that may pose a conflict of interests in connection with the article.

Acknowledgements Special thanks go to Helen Juan Zhou, Vijay Mittal, Carrie Bearden, Tyrone Cannon, Yoonho Chung, and Stephen Wood who provided additional information and volumetric data.

References Addington, F.-P.P.B.S.B.A., Fusar-Poli, P., Borgwardt, S., Bechdolf, A., Addington, J., RiecherRössler, A., Schultze-Lutter, F., Keshavan, M., Wood, S., Ruhrmann, S., Seidman, L.J., Valmaggia, L., Cannon, T., Velthorst, E., De Haan, L., Cornblatt, B., Bonoldi, I., Birchwood, M., McGlashan, T., Carpenter, W., McGorry, P., Klosterkötter, J., McGuire, P., Yung, A., FusarPoli, P., B.S.B.A.A., 2013. The Psychosis High-Risk State: A Comprehensive State-of-the-Art Review. JAMA psychiatry 70, 107–20. doi:10.1001/jamapsychiatry.2013.269.The Adriano, F., Caltagirone, C., Spalletta, G., 2012. Hippocampal volume reduction in first-episode and chronic schizophrenia: a review and meta-analysis. Hippocampal Vol. Reduct. FirstEpisode Chronic Schizophr. A Rev. Meta-Analysis 18, 180–200. doi:10.1177/1073858410395147 Allen, P., Chaddock, C.A., Egerton, A., Howes, O.D., Bonoldi, I., Zelaya, F., Bhattacharyya, S., Murray, R., McGuire, P., 2015. Resting Hyperperfusion of the Hippocampus, Midbrain, and Basal Ganglia in People at High Risk for Psychosis. Am. J. Psychiatry appiajp201515040485. 18

doi:10.1176/appi.ajp.2015.15040485 Banks, P.J., Burroughs, A.C., Barker, G.R.I., Brown, J.T., Warburton, E.C., Bashir, Z.I., 2015. Disruption of hippocampal-prefrontal cortex activity by dopamine D2R-dependent LTD of NMDAR transmission. Proc. Natl. Acad. Sci. U. S. A. 112, 11096–101. doi:10.1073/pnas.1512064112 Bernard, J.A., Orr, J.M., Mittal, V.A., 2015. Abnormal hippocampal–thalamic white matter tract development and positive symptom course in individuals at ultra-high risk for psychosis. npj Schizophr. 1, 15009. doi:10.1038/npjschz.2015.9 Bois, C., Levita, L., Ripp, I., Owens, D.C.G., Johnstone, E.C., Whalley, H.C., Lawrie, S.S., 2015. Hippocampal, amygdala and nucleus accumbens volume in first-episode schizophrenia patients and individuals at high familial risk: A cross-sectional comparison. Schizophr. Res. doi:10.1016/j.schres.2015.03.024 Borgwardt, S., McGuire, P., Fusar-Poli, P., 2011. Gray matters!--mapping the transition to psychosis. Schizophr. Res. 133, 63–7. doi:10.1016/j.schres.2011.08.021 Bühlmann, E., Berger, G.E., Aston, J., Gschwandtner, U., Pflueger, M.O., Borgwardt, S.J., Radue, E.W., Riecher-Rössler, A., 2010. Hippocampus abnormalities in at risk mental states for psychosis? A cross-sectional high resolution region of interest magnetic resonance imaging study. J. Psychiatr. Res. 44, 447–53. doi:10.1016/j.jpsychires.2009.10.008 Cahill, E., Pascoli, V., Trifilieff, P., Savoldi, D., Kappès, V., Lüscher, C., Caboche, J., Vanhoutte, P., 2014. D1R/GluN1 complexes in the striatum integrate dopamine and glutamate signalling to control synaptic plasticity and cocaine-induced responses. Mol. Psychiatry 19, 1295–1304. doi:10.1038/mp.2014.73 Cannon, T.D., 2015. Brain Biomarkers of Vulnerability and Progression to Psychosis. Schizophr. Bull. sbv173. doi:10.1093/schbul/sbv173 Cannon, T.D., Chung, Y., He, G., Sun, D., Jacobson, A., Van Erp, T.G.M., McEwen, S., Addington, J., Bearden, C.E., Cadenhead, K., Cornblatt, B., Mathalon, D.H., McGlashan, T., Perkins, D., Jeffries, C., Seidman, L.J., Tsuang, M., Walker, E., Woods, S.W., Heinssen, R., 2015. 19

Progressive reduction in cortical thickness as psychosis develops: A multisite longitudinal neuroimaging study of youth at elevated clinical risk. Biol. Psychiatry 77, 147–157. doi:10.1016/j.biopsych.2014.05.023 Cooper, H.M., Hedges, L. V., Valentine, J.C., 2009. The Handbook of Research Synthesis and MetaAnalysis, Psychological Review April 2006. Dean, D.J., Orr, J.M., Bernard, J. a., Gupta, T., Pelletier-Baldelli, A., Carol, E.E., Mittal, V. a., 2015. Hippocampal Shape Abnormalities Predict Symptom Progression in Neuroleptic-Free Youth at Ultrahigh Risk for Psychosis. Schizophr. Bull. i, sbv086. doi:10.1093/schbul/sbv086 DeLisi, L.E., 2011. Moving on in Schizophrenia Research to the next decade. Schizophr. Res. doi:10.1016/j.schres.2011.01.012 DeLisi, L.E., 2008. The concept of progressive brain change in schizophrenia: Implications for understanding schizophrenia. Schizophr. Bull. doi:10.1093/schbul/sbm164 DerSimonian, R., Laird, N., 1986. Meta-analysis in clinical trials. Control. Clin. Trials 7, 177–188. doi:10.1016/0197-2456(86)90046-2 Dill, V., Franco, A.R., Pinho, M.S., 2015. Automated Methods for Hippocampus Segmentation: the Evolution and a Review of the State of the Art. Neuroinformatics. doi:10.1007/s12021-0149243-4 Duval, S., Tweedie, R., 2000. Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics 56, 455–463. doi:10.1111/j.0006341x.2000.00455.x Egger, M., Davey Smith, G., Schneider, M., Minder, C., 1997. Bias in meta-analysis detected by a simple, graphical test. BMJ 315, 629–34. doi:10.1136/bmj.316.7129.469 Faul, F., Erdfelder, E., Lang, A.-G., Buchner, A., 2007. G * Power 3 : A flexible statistical power analysis program for the social , behavioral , and biomedical sciences. Behav. Res. Methods 39, 175–191. doi:10.3758/bf03193146 Fusar-Poli, P., Bonoldi, I., Yung, A.R., Borgwardt, S., Kempton, M.J., Valmaggia, L., Barale, F., Caverzasi, E., McGuire, P., 2012a. Predicting psychosis: meta-analysis of transition outcomes 20

in individuals at high clinical risk. Arch. Gen. Psychiatry 69, 220–9. doi:10.1001/archgenpsychiatry.2011.1472 Fusar-Poli, P., Borgwardt, S., 2012. Predictive power of attenuated psychosis syndrome: Is it really low? The case of mild cognitive impairment. Schizophr. Res. doi:10.1016/j.schres.2011.11.023 Fusar-Poli, P., Cappucciati, M., Borgwardt, S., Woods, S.W., Addington, J., Nelson, B., Nieman, D.H., Stahl, D.R., Rutigliano, G., Riecher-Rössler, A., Simon, A.E., Mizuno, M., Lee, T.Y., Kwon, J.S., Lam, M.M.L., Perez, J., Keri, S., Amminger, P., Metzler, S., Kawohl, W., Rössler, W., Lee, J., Labad, J., Ziermans, T., An, S.K., Liu, C.-C., Woodberry, K.A., Braham, A., Corcoran, C., McGorry, P., Yung, A.R., McGuire, P.K., 2015. Heterogeneity of Psychosis Risk Within Individuals at Clinical High Risk. JAMA Psychiatry 1. doi:10.1001/jamapsychiatry.2015.2324 Fusar-Poli, P., Perez, J., Broome, M., Borgwardt, S., Placentino, A., Caverzasi, E., Cortesi, M., Veggiotti, P., Politi, P., Barale, F., McGuire, P., 2007. Neurofunctional correlates of vulnerability to psychosis: a systematic review and meta-analysis. Neurosci. Biobehav. Rev. 31, 465–84. doi:10.1016/j.neubiorev.2006.11.006 Fusar-Poli, P., Radua, J., McGuire, P., Borgwardt, S., 2012b. Neuroanatomical maps of psychosis onset: Voxel-wise meta-analysis of antipsychotic-naive vbm studies. Schizophr. Bull. 38, 1297–1307. doi:10.1093/schbul/sbr134 Fusar-Poli, P., Smieskova, R., Kempton, M.J., Ho, B.C., Andreasen, N.C., Borgwardt, S., 2013. Progressive brain changes in schizophrenia related to antipsychotic treatment? A metaanalysis of longitudinal MRI studies. Neurosci. Biobehav. Rev. doi:10.1016/j.neubiorev.2013.06.001 Ganzola, R., Maziade, M., Duchesne, S., 2014. Hippocampus and amygdala volumes in children and young adults at high-risk of schizophrenia: research synthesis. Schizophr. Res. 156, 76– 86. doi:10.1016/j.schres.2014.03.030 Geuze, E., Vermetten, E., Bremner, J., 2005. MR-based in vivo hippocampal volumetrics: 1. Review of methodologies currently employed. Mol. Psychiatry 10, 147–159. 21

doi:10.1038/sj.mp.4001580 Giuliani, N.R., Calhoun, V.D., Pearlson, G.D., Francis, A., Buchanan, R.W., 2005. Voxel-based morphometry versus region of interest: A comparison of two methods for analyzing gray matter differences in schizophrenia. Schizophr. Res. 74, 135–147. doi:10.1016/j.schres.2004.08.019 Gothelf, D., Soreni, N., Nachman, R.P., Tyano, S., Hiss, Y., Reiner, O., Weizman, a, 2000. Evidence for the involvement of the hippocampus in the pathophysiology of schizophrenia. Eur. Neuropsychopharmacol. 10, 389–395. doi:10.1016/S0924-977X(00)00097-3 Grace, A.A., 2012. Dopamine system dysregulation by the hippocampus: Implications for the pathophysiology and treatment of schizophrenia. Neuropharmacology. doi:10.1016/j.neuropharm.2011.05.011 Häfner, H., Riecher-Rössler, A., Hambrecht, M., Maurer, K., Meissner, S., Schmidtke, A., Fätkenheuer, B., Löffler, W., van der Heiden, W., 1992. IRAOS: an instrument for the assessment of onset and early course of schizophrenia. Schizophr. Res. 6, 209–223. doi:10.1016/0920-9964(92)90004-O Harrisberger, F., Smieskova, R., Schmidt, A., Lenz, C., Walter, A., Wittfeld, K., Grabe, H.J., Lang, U.E., Fusar-Poli, P., Borgwardt, S., 2015. BDNF Val66Met polymorphism and hippocampal volume in neuropsychiatric disorders: A systematic review and meta-analysis. Neurosci. Biobehav. Rev. 55, 107–118. doi:10.1016/j.neubiorev.2015.04.017 Heckers, S., 2001. Neuroimaging studies of the hippocampus in schizophrenia. Hippocampus 11, 520–528. doi:10.1002/hipo.1068 Heckers, S., Konradi, C., 2015. GABAergic mechanisms of hippocampal hyperactivity in schizophrenia. Schizophr. Res. 167, 4–11. doi:10.1016/j.schres.2014.09.041 Higgins, J.P.T., Thompson, S.G., 2002. Quantifying heterogeneity in a meta-analysis. Stat. Med. 21, 1539–1558. doi:10.1002/sim.1186 Ho, N.F., Iglesias, J.E., Sum, M.Y., Kuswanto, C.N., Sitoh, Y.Y., De Souza, J., Hong, Z., Fischl, B., Roffman, J.L., Zhou, J., Sim, K., Holt, D.J., 2016. Progression from selective to general 22

involvement of hippocampal subfields in schizophrenia. Mol. Psychiatry 1–11. doi:10.1038/mp.2016.4 Howes, O.D., Montgomery, A.J., Asselin, M.C., Murray, R.M., Grasby, P.M., Mcguire, P.K., 2007. Molecular imaging studies of the striatal dopaminergic system in psychosis and predictions for the prodromal phase of psychosis. Br. J. Psychiatry. doi:10.1192/bjp.191.51.s13 Hurlemann, R., Jessen, F., Wagner, M., Frommann, I., Ruhrmann, S., Brockhaus, A., Picker, H., Scheef, L., Block, W., Schild, H.H., Moller-Hartmann, W., Krug, B., Falkai, P., Klosterkotter, J., Maier, W., 2008. Interrelated neuropsychological and anatomical evidence of hippocampal pathology in the at-risk mental state. Psychol. Med. 38, 843–51. doi:10.1017/S0033291708003279 Ioannidis, J.P.A., Patsopoulos, N.A., Evangelou, E., 2007. Heterogeneity in meta-analyses of genome-wide association investigations. PLoS One 2. doi:10.1371/journal.pone.0000841 Kempton, M.J., Stahl, D., Williams, S.C.R., DeLisi, L.E., 2010. Progressive lateral ventricular enlargement in schizophrenia: A meta-analysis of longitudinal MRI studies. Schizophr. Res. 120, 54–62. doi:10.1016/j.schres.2010.03.036 Klauser, P., Zhou, J., Lim, J.K.W., Poh, J.S., Zheng, H., Tng, H.Y., Krishnan, R., Lee, J., Keefe, R.S.E., Adcock, R.A., Wood, S.J., Fornito, A., Chee, M.W.L., 2015. Lack of Evidence for Regional Brain Volume or Cortical Thickness Abnormalities in Youths at Clinical High Risk for Psychosis: Findings From the Longitudinal Youth at Risk Study. Schizophr. Bull. doi:10.1093/schbul/sbv012 Lewis, D.A., Levitt, P., 2002. SCHIZOPHRENIA AS A DISORDER OF NEURODEVELOPMENT - Annual Review of Neuroscience, 25(1):409. Annu. Rev. Neurosci. doi:10.1146/annurev.neuro.25.112701.142754 Lieberman, J.A., 1999. Is schizophrenia a neurodegenerative disorder? A clinical and neurobiological perspective. Biol. Psychiatry. doi:10.1016/S0006-3223(99)00147-X Lipsey, M.W., Wilson, D.B., 2001. Practical meta-analysis Lipsey, M W Wilson, D B, October. Lipska, B.K., Jaskiw, G.E., Weinberger, D.R., 1993. Postpubertal emergence of 23

hyperresponsiveness to stress and to amphetamine after neonatal excitotoxic hippocampal damage: a potential animal model of schizophrenia. Neuropsychopharmacology 9, 67–75. doi:10.1038/npp.1993.44 Loureiro, M., Kramar, C., Renard, J., Rosen, L.G., Laviolette, S.R., 2015. Cannabinoid Transmission in the Hippocampus Activates Nucleus Accumbens Neurons and Modulates Reward and Aversion-Related Emotional Salience. Biol. Psychiatry. doi:10.1016/j.biopsych.2015.10.016 Loureiro, M., Lecourtier, L., Engeln, M., Lopez, J., Cosquer, B., Geiger, K., Kelche, C., Cassel, J.C., Pereira De Vasconcelos, A., 2012. The ventral hippocampus is necessary for expressing a spatial memory. Brain Struct. Funct. 217, 93–106. doi:10.1007/s00429-011-0332-y Magarinos, A.M., McEwen, B.S., 1995. Stress-induced atrophy of apical dendrites of hippocampal CA3c neurons: Comparison of stressors. Neuroscience 69, 83–88. doi:10.1016/03064522(95)00256-I McEwen, B.S., 2000. The neurobiology of stress: From serendipity to clinical relevance. Brain Res. doi:10.1016/S0006-8993(00)02950-4 Mechelli, A., Riecher-Rössler, A., Meisenzahl, E.M., Tognin, S., Wood, S.J., Borgwardt, S.J., Koutsouleris, N., Yung, A.R., Stone, J.M., Phillips, L.J., McGorry, P.D., Valli, I., Velakoulis, D., Woolley, J., Pantelis, C., McGuire, P., 2011. Neuroanatomical abnormalities that predate the onset of psychosis: a multicenter study. Arch. Gen. Psychiatry 68, 489–495. doi:10.1016/j.ypsy.2011.08.098 Miller, T.J., McGlashan, T.H., Rosen, J.L., Cadenhead, K., Cannon, T., Ventura, J., McFarlane, W., Perkins, D.O., Pearlson, G.D., Woods, S.W., 2003. Prodromal assessment with the structured interview for prodromal syndromes and the scale of prodromal symptoms: predictive validity, interrater reliability, and training to reliability. Schizophr. Bull. 29, 703–715. doi:10.1093/oxfordjournals.schbul.a007040 Mittal, V.A., Gupta, T., Orr, J.M., Pelletier-Baldelli, A., Dean, D.J., Lunsford-Avery, J.R., Smith, A.K., Robustelli, B.L., Leopold, D.R., Millman, Z.B., 2013. Physical activity level and medial temporal health in youth at ultra high-risk for psychosis. J. Abnorm. Psychol. 122, 1101–10. 24

doi:10.1037/a0034085 Moher, D., Shamseer, L., Clarke, M., Ghersi, D., Liberati, A., Petticrew, M., Shekelle, P., Stewart, L.A., 2015. Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) 2015 statement. Syst. Rev. 4, 1. doi:10.1186/2046-4053-4-1 Mondelli, V., Pariante, C.M., Navari, S., Aas, M., D’Albenzio, A., Di Forti, M., Handley, R., Hepgul, N., Marques, T.R., Taylor, H., Papadopoulos, A.S., Aitchison, K.J., Murray, R.M., Dazzan, P., 2010. Higher cortisol levels are associated with smaller left hippocampal volume in firstepisode psychosis. Schizophr. Res. 119, 75–78. doi:10.1016/j.schres.2009.12.021 Narr, K.L., Thompson, P.M., Szeszko, P., Robinson, D., Jang, S., Woods, R.P., Kim, S., Hayashi, K.M., Asunction, D., Toga, A.W., Bilder, R.M., 2004. Regional specificity of hippocampal volume reductions in first-episode schizophrenia. Neuroimage 21, 1563–1575. doi:10.1016/j.neuroimage.2003.11.011 Nenadic, I., Dietzek, M., Schönfeld, N., Lorenz, C., Gussew, A., Reichenbach, J.R., Sauer, H., Gaser, C., Smesny, S., 2015. Brain structure in people at ultra-high risk of psychosis, patients with first-episode schizophrenia, and healthy controls: a VBM study. Schizophr. Res. 161, 169–76. doi:10.1016/j.schres.2014.10.041 Pascoli, V., Terrier, J., Espallergues, J., Valjent, E., O’Connor, E.C., Lüscher, C., 2014. Contrasting forms of cocaine-evoked plasticity control components of relapse. Nature 509, 459–64. doi:10.1038/nature13257 Perlini, C., Bellani, M., Brambilla, P., 2012. Structural imaging techniques in schizophrenia. Acta Psychiatr. Scand. 126, 235–242. doi:10.1111/j.1600-0447.2012.01868.x Phillips, L.J., Velakoulis, D., Pantelis, C., Wood, S., Yuen, H.P., Yung, A.R., Desmond, P., Brewer, W., McGorry, P.D., 2002. Non-reduction in hippocampal volume is associated with higher risk of psychosis. Schizophr. Res. 58, 145–58. Ragland, J., Ranganath, C., MP, H., al, et, 2015. Functional and neuroanatomic specificity of episodic memory dysfunction in schizophrenia: A functional magnetic resonance imaging study of the relational and item-specific encoding task. JAMA Psychiatry 72, 909–916. 25

doi:10.1001/jamapsychiatry.2015.0276 Rapoport, J.L., Giedd, J.N., Gogtay, N., 2012. Neurodevelopmental model of schizophrenia: update 2012. Mol. Psychiatry 17, 1228–38. doi:10.1038/mp.2012.23 Rasetti, R., Mattay, V.S., White, M.G., Sambataro, F., Podell, J.E., Zoltick, B., Chen, Q., Berman, K.F., Callicott, J.H., Weinberger, D.R., 2014. Altered Hippocampal-Parahippocampal Function During Stimulus Encoding: A Potential Indicator of Genetic Liability for Schizophrenia. JAMA psychiatry 71, 236–247. doi:10.1001/jamapsychiatry.2013.3911 Riecher-Rössler, a, Gschwandtner, U., Aston, J., Borgwardt, S., Drewe, M., Fuhr, P., Pflüger, M., Radü, W., Schindler, C., Stieglitz, R.-D., 2007. The Basel early-detection-of-psychosis (FEPSY)study--design and preliminary results. Acta Psychiatr. Scand. 115, 114–25. doi:10.1111/j.1600-0447.2006.00854.x Samudra, N., Ivleva, E.I., Hubbard, N.A., Rypma, B., Sweeney, J.A., Clementz, B.A., Keshavan, M.S., Pearlson, G.D., Tamminga, C.A., 2015. Alterations in hippocampal connectivity across the psychosis dimension. Psychiatry Res. - Neuroimaging 233, 148–157. doi:10.1016/j.pscychresns.2015.06.004 Sapolsky, R.M., 2000. Glucocorticoids and hippocampal atrophy in neuropsychiatric disorders. Arch. Gen. Psychiatry 57, 925–935. doi:10.1001/archpsyc.57.10.925 Schultze-Lutter, F., 2009. Subjective symptoms of schizophrenia in research and the clinic: The basic symptom concept. Schizophr. Bull. doi:10.1093/schbul/sbn139 Seidman, L.J., Rosso, I.M., Thermenos, H.W., Makris, N., Juelich, R., Gabrieli, J.D.E., Faraone, S. V., Tsuang, M.T., Whitfield-Gabrieli, S., 2014. Medial temporal lobe default mode functioning and hippocampal structure as vulnerability indicators for schizophrenia: A MRI study of nonpsychotic adolescent first-degree relatives. Schizophr. Res. 159, 426–434. doi:10.1016/j.schres.2014.09.011 Shenton, M.E., Dickey, C.C., Frumin, M., McCarley, R.W., 2001. A review of MRI findings in schizophrenia. Schizophr. Res. 49, 1–52. doi:10.1016/S0920-9964(01)00163-3 Small, S.A., Schobel, S.A., Buxton, R.B., Witter, M.P., Barnes, C.A., 2011. A pathophysiological 26

framework of hippocampal dysfunction in ageing and disease. Nat. Rev. Neurosci. 12, 585– 601. doi:10.1038/nrn3085 Smieskova, R., Fusar-Poli, P., Allen, P., Bendfeldt, K., Stieglitz, R.D., Drewe, J., Radue, E.W., McGuire, P.K., Riecher-Rössler, A., Borgwardt, S.J., 2010. Neuroimaging predictors of transition to psychosis-A systematic review and meta-analysis. Neurosci. Biobehav. Rev. doi:10.1016/j.neubiorev.2010.01.016 Spoletini, I., Piras, F., Fagioli, S., Rubino, I.A., Martinotti, G., Siracusano, A., Caltagirone, C., Spalletta, G., 2011. Suicidal attempts and increased right amygdala volume in schizophrenia. Schizophr. Res. 125, 30–40. doi:10.1016/j.schres.2010.08.023 Strange, B. a, Witter, M.P., Lein, E.S., Moser, E.I., 2014. Functional organization of the hippocampal longitudinal axis. Nat. Publ. Gr. 15, 655–669. doi:10.1038/nrn3785 Tamminga, C.A., Stan, A.D., Wagner, A.D., 2010. The hippocampal formation in schizophrenia. Am. J. Psychiatry. doi:10.1176/appi.ajp.2010.09081187 Tamminga, C.A., Zukin, R.S., 2015. Schizophrenia: Evidence implicating hippocampal GluN2B protein and REST epigenetics in psychosis pathophysiology. Neuroscience. doi:10.1016/j.neuroscience.2015.07.038 van Erp, T.G.M., Hibar, D.P., Rasmussen, J.M., Glahn, D.C., Pearlson, G.D., Andreassen, O. a, Agartz, I., Westlye, L.T., Haukvik, U.K., Dale, a M., Melle, I., Hartberg, C.B., Gruber, O., Kraemer, B., Zilles, D., Donohoe, G., Kelly, S., McDonald, C., Morris, D.W., Cannon, D.M., Corvin, A., Machielsen, M.W.J., Koenders, L., de Haan, L., Veltman, D.J., Satterthwaite, T.D., Wolf, D.H., Gur, R.C., Gur, R.E., Potkin, S.G., Mathalon, D.H., Mueller, B. a, Preda, A., Macciardi, F., Ehrlich, S., Walton, E., Hass, J., Calhoun, V.D., Bockholt, H.J., Sponheim, S.R., Shoemaker, J.M., van Haren, N.E.M., Pol, H.E.H., Ophoff, R. a, Kahn, R.S., Roiz-Santiañez, R., Crespo-Facorro, B., Wang, L., Alpert, K.I., Jönsson, E.G., Dimitrova, R., Bois, C., Whalley, H.C., McIntosh, a M., Lawrie, S.M., Hashimoto, R., Thompson, P.M., Turner, J. a, 2015. Subcortical brain volume abnormalities in 2028 individuals with schizophrenia and 2540 healthy controls via the ENIGMA consortium. Mol. Psychiatry 1–7. doi:10.1038/mp.2015.63 27

Velakoulis, D., Pantelis, C., McGorry, P.D., Dudgeon, P., Brewer, W., Cook, M., Desmond, P., Bridle, N., Tierney, P., Murrie, V., Singh, B., Copolov, D., 1999. Hippocampal Volume in First-Episode Psychoses and Chronic Schizophrenia. Arch. Gen. Psychiatry 56, 133. doi:10.1001/archpsyc.56.2.133 Velakoulis, D., Wood, S.J., Wong, M.T.H., McGorry, P.D., Yung, A., Phillips, L., Smith, D., Brewer, W., Proffitt, T., Desmond, P., Pantelis, C., 2006. Hippocampal and amygdala volumes according to psychosis stage and diagnosis: a magnetic resonance imaging study of chronic schizophrenia, first-episode psychosis, and ultra-high-risk individuals. Arch. Gen. Psychiatry 63, 139–49. doi:10.1001/archpsyc.63.2.139 Viechtbauer, W., Cheung, M.W.-L., 2010. Outlier and influence diagnostics for meta-analysis. Res. Synth. Methods 1, 112–125. doi:10.1002/jrsm.11 Walter, A., Studerus, E., Smieskova, R., Kuster, P., Aston, J., Lang, U.E., Radue, E.-W., RiecherRössler, A., Borgwardt, S., 2012. Hippocampal volume in subjects at high risk of psychosis: a longitudinal MRI study. Schizophr. Res. 142, 217–22. doi:10.1016/j.schres.2012.10.013 Witthaus, H. et al., 2010. Hippocampal subdivision and amygdalar volumes in patients in an at-risk mental state for schizophrenia. J. Psychiatry Neurosci. 35, 33–40. Wood, S.J., Kennedy, D., Phillips, L.J., Seal, M.L., Yücel, M., Nelson, B., Yung, A.R., Jackson, G., McGorry, P.D., Velakoulis, D., Pantelis, C., 2010. Hippocampal pathology in individuals at ultra-high risk for psychosis: a multi-modal magnetic resonance study. Neuroimage 52, 62–8. doi:10.1016/j.neuroimage.2010.04.012 Wood, S.J., Yücel, M., Velakoulis, D., Phillips, L.J., Yung, A.R., Brewer, W., McGorry, P.D., Pantelis, C., 2005. Hippocampal and anterior cingulate morphology in subjects at ultra-high-risk for psychosis: the role of family history of psychotic illness. Schizophr. Res. 75, 295–301. doi:10.1016/j.schres.2004.10.008 Yung, A.R., Mc Gorry, P.D., McFarlane, C.A., Jackson, H.J., Patton, G.C., Rakkar, A., 1996. Monitoring and care of young people at incipient risk of psychosis. Schizophr. Bull. 22, 283– 303. doi:10.1093/schbul/22.2.283 28

Yung, A.R., Yuen, H.P., McGorry, P.D., Phillips, L.J., Kelly, D., Dell’Olio, M., Francey, S.M., Cosgrave, E.M., Killackey, E., Stanford, C., Godfrey, K., Buckby, J., 2005. Mapping the onset of psychosis: The comprehensive assessment of at risk mental states. Schizophr. Res. 39, 964– 971. doi:10.1016/S0920-9964(03)80090-7

29

30

Figure 2: Forest plot for global random effects, meta-analysis of right hippocampal volume. Positive effect sizes indicate smaller hippocampi in clinical high-risk patients. Dashed lines indicate zero line. Square size proportional to sample size.

31

Figure 3: Forest plot for global random effects meta-analysis of left hippocampal volume. Positive effect sizes indicate smaller hippocampi in clinical high-risk patients. Dashed lines indicate zero line. Square size proportional to sample size.

32

Figure 4: Funnel plot of potential bias where trim and fill method revealed no missing studies to correct for potential publication bias in right hippocampal volume

*indicate studies with automated measurement technique.

33

Figure 5: Funnel plot of potential bias where trim and fill method revealed no missing studies to correct for potential publication bias in left hippocampal volume

*indicate studies with automated measurement technique.

34

HC

UHR

Author

Year

Cohort

n

n

Age, mean (SD)

% Male

n

Age, mean (SD)

% Male

UHR Criteria

Tesla

Tech

Bühlmann et al.

2010

Basel

59

22

23(4.3)

59

37

24.7(5.6)

59

PACE

1.5

Man

Dean et al.

2016

Boulder

80

42

18.7(1.9)

50

38

18.9(1.4)

58

SIPS

3.0

Auto FSL

Hurlemann et al.

2008

Köln, Bonn

66

30

28.2(6.4)

77

36

27.1(5.6)

56

ERIraos

1.5

Man

Klauser et al.

2015

Singapore

101

32

23(3.9)

53

69

21.5(3.5)

68

CAARMS

3.0

Auto Free

Velakoulis et al.

2006

Melbourne

222

87

26.9(10)

63

135

20.1(3.6)

58

PACE

1.5

Man

Witthaus et al.

2010

Berlin

58

29

25.7(5.2)

59

29

25.3(4.3)

66

SIPS

1.5

Man

Wood et al.

2010

Melbourne

94

29

21.1(4.7)

38

65

19.2(3.2)

38

CAARMS

3.0

Man

Cannon et al.

2015

NAPLS

777

224

20.5(4.6)

54

553

19.6(4.2)

62

SIPS

3.0

Auto Free

Table 1: Overview of all studies included in the final meta-analysis

35

Table 2: Quality assessment of included studies Author

Year

Funding

Sample Size

Inclusion Criteria

Exclusion Criteria Substance

Gender

Race

IQ

Antipsychotics

Psychopathology

Inter-rate reliability

Bühlmann et al.

2010

2

2

2

2

2

0

2

2

2

0

Cannon et al.

2015

2

2

2

2

2

2

2

2

2

-

Dean et al.

2016

2

2

2

2

2

2

1

2

2

-

Hurleman n et al.

2008

2

2

2

2

2

0

2

2

2

2

Klauser et al.

2015

2

2

2

2

2

2

2

2

2

-

Velakoulis et al.

2006

2

2

2

2

2

0

2

2

0

2

Witthaus et al.

2010

2

2

2

2

2

0

2

2

2

2

Wood et al.

2010

2

2

2

2

2

0

0

2

2

2

36

Table 3: Overview of the results from the meta-analysis of left and right hippocampal volume

Meta-analysis

Heter

Hedge’ sg

Standard error

Lower confidence interval

Upper confidence interval

Zvalue

p-value of z

Heter I2

Left HV (k=8, n=1429)

0.25

0.72

-0.04

0.54

1.72

0.09

79.99

Right HV (k=8, n=1429)

0.24

0.12

0.0091

0.48

2.04

0.04*

69.30

Left HV without outlier (Wood 2010, k=7, n=1335)

0.13

0.12

-0.10

0.37

1.11

0.27

66.13

Right HV without outlier (Hurlemann 2008, k=7, n=1363)

0.14

0.09

-0.03

0.32

1.61

0.11

42.43

*Significant, HV = hippocampal volume

37

Table 4: Publication bias, trim and fill, meta-regression

Publication bias

Trim & fill

Meta-regression: p-values

p value of Eggers regression test

Number of missing studies

Publication year

Sample size

Gender ratio

Left HV (k=8, n=1429)

0.43

0

0.08

0.64

0.16

Right HV (k=8, n=1429)

0.38

0

0.09

0.58

0.59

*Significant, HV = hippocampal volume

38

Table 5: Externally studentised residuals (outliers) Left

Right

Residuals

SE

Z

Residuals

SE

Z

Klauser

-0.3414

0.4698

-0.7267

-0.2617

0.3859

-0.6781

Dean

-0.5661

0.4563

-1.2405

-0.5397

0.3674

-1.4689

Cannon

-0.2408

0.4932

-0.4881

-0.1840

0.3878

-0.4744

Wood

0.9952

0.3626

2.7443*

0.5662

0.3453

1.6399

Witthaus

-0.2867

0.4894

-0.5858

-0.0409

0.4135

-0.0988

Velakoulis

0.0374

0.4807

0.0777

-0.0790

0.3843

-0.2056

Hurlemann

0.8630

0.4226

2.0424

0.9422

0.3175

2.9675*

Buehlmann

-0.3589

0.4901

-0.7322

-0.2106

0.4156

-0.5067

*Outlier = z > 2.5, SE = Standard Error

39