Introduction
Bipolar disorder I (BD-I) is a severe neuropsychiatric illness that causes extensive individual morbidity and socio-economic burden (Vos et al., Reference Vos, Flaxman, Naghavi, Lozano, Michaud, Ezzati, Shibuya, Salomon, Abdalla, Aboyans, Abraham, Ackerman, Aggarwal, Ahn, Ali, Alvarado, Anderson, Anderson, Andrews, Atkinson, Baddour, Bahalim, Barker-Collo, Barrero, Bartels, Basanez, Baxter, Bell, Benjamin, Bennett, Bernabe, Bhalla, Bhandari, Bikbov, Bin Abdulhak, Birbeck, Black, Blencowe, Blore, Blyth, Bolliger, Bonaventure, Boufous, Bourne, Boussinesq, Braithwaite, Brayne, Bridgett, Brooker, Brooks, Brugha, Bryan-Hancock, Bucello, Buchbinder, Buckle, Budke, Burch, Burney, Burstein, Calabria, Campbell, Canter, Carabin, Carapetis, Carmona, Cella, Charlson, Chen, Cheng, Chou, Chugh, Coffeng, Colan, Colquhoun, Colson, Condon, Connor, Cooper, Corriere, Cortinovis, de Vaccaro, Couser, Cowie, Criqui, Cross, Dabhadkar, Dahiya, Dahodwala, Damsere-Derry, Danaei, Davis, De Leo, Degenhardt, Dellavalle, Delossantos, Denenberg, Derrett, Des Jarlais, Dharmaratne, Dherani, Diaz-Torne, Dolk, Dorsey, Driscoll, Duber, Ebel, Edmond, Elbaz, Ali, Erskine, Erwin, Espindola, Ewoigbokhan, Farzadfar, Feigin, Felson, Ferrari, Ferri, Fevre, Finucane, Flaxman, Flood, Foreman, Forouzanfar, Fowkes, Franklin, Fransen, Freeman, Gabbe, Gabriel, Gakidou, Ganatra, Garcia, Gaspari, Gillum, Gmel, Gosselin, Grainger, Groeger, Guillemin, Gunnell, Gupta, Haagsma, Hagan, Halasa, Hall, Haring, Haro, Harrison, Havmoeller, Hay, Higashi, Hill, Hoen, Hoffman, Hotez, Hoy, Huang, Ibeanusi, Jacobsen, James, Jarvis, Jasrasaria, Jayaraman, Johns, Jonas, Karthikeyan, Kassebaum, Kawakami, Keren, Khoo, King, Knowlton, Kobusingye, Koranteng, Krishnamurthi, Lalloo, Laslett, Lathlean, Leasher, Lee, Leigh, Lim, Limb, Lin, Lipnick, Lipshultz, Liu, Loane, Ohno, Lyons, Ma, Mabweijano, MacIntyre, Malekzadeh, Mallinger, Manivannan, Marcenes, March, Margolis, Marks, Marks, Matsumori, Matzopoulos, Mayosi, McAnulty, McDermott, McGill, McGrath, Medina-Mora, Meltzer, Mensah, Merriman, Meyer, Miglioli, Miller, Miller, Mitchell, Mocumbi, Moffitt, Mokdad, Monasta, Montico, Moradi-Lakeh, Moran, Morawska, Mori, Murdoch, Mwaniki, Naidoo, Nair, Naldi, Narayan, Nelson, Nelson, Nevitt, Newton, Nolte, Norman, Norman, O'Donnell, O'Hanlon, Olives, Omer, Ortblad, Osborne, Ozgediz, Page, Pahari, Pandian, Rivero, Patten, Pearce, Padilla, Perez-Ruiz, Perico, Pesudovs, Phillips, Phillips, Pierce, Pion, Polanczyk, Polinder, Pope, Popova, Porrini, Pourmalek, Prince, Pullan, Ramaiah, Ranganathan, Razavi, Regan, Rehm, Rein, Remuzzi, Richardson, Rivara, Roberts, Robinson, De Leon, Ronfani, Room, Rosenfeld, Rushton, Sacco, Saha, Sampson, Sanchez-Riera, Sanman, Schwebel, Scott, Segui-Gomez, Shahraz, Shepard, Shin, Shivakoti, Singh, Singh, Singh, Singleton, Sleet, Sliwa, Smith, Smith, Stapelberg, Steer, Steiner, Stolk, Stovner, Sudfeld, Syed, Tamburlini, Tavakkoli, Taylor, Taylor, Taylor, Thomas, Thomson, Thurston, Tleyjeh, Tonelli, Towbin, Truelsen, Tsilimbaris, Ubeda, Undurraga, van der Werf, van Os, Vavilala, Venketasubramanian, Wang, Wang, Watt, Weatherall, Weinstock, Weintraub, Weisskopf, Weissman, White, Whiteford, Wiersma, Wilkinson, Williams, Williams, Witt, Wolfe, Woolf, Wulf, Yeh, Zaidi, Zheng, Zonies, Lopez, Murray, AlMazroa and Memish2012). Clinically, the disorder is defined by the occurrence of at least one manic episode and often includes recurrent states of depressive, manic, mixed and euthymic episodes (American Psychiatric Association, 2013). It is increasingly becoming evident that processing of motivation and reward are affected in all of these states (Ashok et al., Reference Ashok, Marques, Jauhar, Nour, Goodwin, Young and Howes2017). This is supported by recent findings showing a transdiagnostic dopamine dysfunction spanning from BP to schizophrenia (SZ) (Jauhar et al., Reference Jauhar, Nour, Veronese, Rogdaki, Bonoldi, Azis, Turkheimer, McGuire, Young and Howes2017). In order to understand the relationship between dysfunctional motivational processes and clinical disease manifestations within but also between different diagnostic entities, valid dimensional psychopathological constructs are essential (Cuthbert and Insel, Reference Cuthbert and Insel2013).
Negative symptoms (NS) have been proposed as such a construct (Strauss et al., Reference Strauss, Vertinski, Vogel, Ringdahl and Allen2016). Originally considered as a hallmark symptom complex of SZ (Kraepelin, Reference Kraepelin1921), it has become evident that NS also occur outside of the SZ spectrum (Strauss and Cohen, Reference Strauss and Cohen2017). A recent study reported that NS can indeed be measured in BD-I patients (Strauss et al., Reference Strauss, Vertinski, Vogel, Ringdahl and Allen2016). NS can be grouped into a motivational dimension (hereafter referred to as apathy) containing anhedonia, avolition and asociality, and a diminished expression dimension, which includes blunted affect and alogia (Blanchard and Cohen, Reference Blanchard and Cohen2006; Strauss et al., Reference Strauss, Horan, Kirkpatrick, Fischer, Keller, Miski, Buchanan, Green and Carpenter2013). Critically, increasing evidence from several fields suggests that the two dimensions are caused by different neurobiological and behavioural mechanisms (Wolf et al., Reference Wolf, Satterthwaite, Kantrowitz, Katchmar, Vandekar, Elliott and Ruparel2014; Cathomas et al., Reference Cathomas, Hartmann, Seifritz, Pryce and Kaiser2015; Hager et al., Reference Hager, Kirschner, Bischof, Hartmann-Riemer, Kluge, Seifritz, Tobler and Kaiser2015; Hartmann et al., Reference Hartmann, Hager, Reimann, Chumbley, Kirschner, Seifritz, Tobler and Kaiser2015; Caravaggio et al., Reference Caravaggio, Fervaha, Menon, Remington, Graff-Guerrero and Gerretsen2018). This differentiation might therefore be critical for the development of effective treatment (Kaiser et al., Reference Kaiser, Lyne, Agartz, Clarke, Morch-Johnsen and Faerden2017; Galderisi et al., Reference Galderisi, Mucci, Buchanan and Arango2018). With respect to the relevance of NS in BD, several studies have reported elevated NS scores even in euthymic patients (Hawkins et al., Reference Hawkins, Hoffman, Quinlan, Rakfeldt, Docherty and Sledge1997; Mancuso et al., Reference Mancuso, Morgan, Mitchell, Berk, Young and Castle2015; Strauss et al., Reference Strauss, Vertinski, Vogel, Ringdahl and Allen2016), which are strongly associated with functional impairments (Atre-Vaidya et al., Reference Atre-Vaidya, Taylor, Seidenberg, Reed, Perrine and Glick-Oberwise1998; Samalin et al., Reference Samalin, Reinares, de Chazeron, Torrent, Bonnin, Hidalgo-Mazzei, Murru, Pacchiarotti, Geoffroy, Bellivier, Llorca and Vieta2016; Serafini et al., Reference Serafini, Vazquez, Gonda, Pompili, Rihmer and Amore2018). Studies comparing the severity of NS between SZ and BD yield mixed results (Tso et al., Reference Tso, Grove and Taylor2014; Strauss and Cohen, Reference Strauss and Cohen2017) but suggest that NS might be more stable and trait like in SZ. With respect to the two NS dimensions a recent study found that BD patients only differed from SZ patients on the diminished expression dimension but showed similar severity of apathy (Strauss et al., Reference Strauss, Vertinski, Vogel, Ringdahl and Allen2016). Furthermore, several studies have reported residual anhedonic symptoms in euthymic BD patients, which were comparable to SZ patients (Di Nicola et al., Reference Di Nicola, De Risio, Battaglia, Camardese, Tedeschi, Mazza, Martinotti, Pozzi, Niolu, Di Giannantonio, Siracusano and Janiri2013; Tso et al., Reference Tso, Grove and Taylor2014) and relevant for impaired reward learning (Pizzagalli et al., Reference Pizzagalli, Goetz, Ostacher, Iosifescu and Perlis2008). Although comparable in trait anhedonia, SZ patients showed more severe total NS, lower experience of pleasure and lower behavioural activation (Tso et al., Reference Tso, Grove and Taylor2014).
It is well established that striatal dysfunction during reward anticipation plays an important role in the pathophysiology of NS in patients with SZ (Heinz and Schlagenhauf, Reference Heinz and Schlagenhauf2010; Radua et al., Reference Radua, Schmidt, Borgwardt, Heinz, Schlagenhauf, McGuire and Fusar-Poli2015; Nielsen et al., Reference Nielsen, Rostrup, Broberg, Wulff and Glenthøj2018). Recent studies suggest a specific link between reduced ventral (VS) and dorsal striatal (DS) activity and apathy (Simon et al., Reference Simon, Biller, Walther, Roesch-Ely, Stippich, Weisbrod and Kaiser2010; Wolf et al., Reference Wolf, Satterthwaite, Kantrowitz, Katchmar, Vandekar, Elliott and Ruparel2014; Mucci et al., Reference Mucci, Dima, Soricelli, Volpe, Bucci, Frangou, Prinster, Salvatore, Galderisi and Maj2015; Kirschner et al., Reference Kirschner, Hager, Bischof, Hartmann, Kluge, Seifritz, Tobler and Kaiser2016b; Stepien et al., Reference Stepien, Manoliu, Kubli, Schneider, Tobler, Seifritz, Herdener, Kaiser and Kirschner2018). In contrast, results regarding reward anticipation in euthymic BD patients are conflicting, with studies showing both increased (Nusslock et al., Reference Nusslock, Almeida, Forbes, Versace, Frank, Labarbara, Klein and Phillips2012; Mason et al., Reference Mason, O'Sullivan, Montaldi, Bentall and El-Deredy2014) and unaltered (Caseras et al., Reference Caseras, Lawrence, Murphy, Wise and Phillips2013; Dutra et al., Reference Dutra, Cunningham, Kober and Gruber2015) VS activity compared with healthy controls (HC). To our knowledge, the association between NS and striatal dysfunction has not been investigated in patients with BD-I. A transdiagnostic approach taking advantage of the current findings from SZ research provides a unique opportunity to elucidate the neural substrates of motivational deficits in BD. Furthermore, given that only recently extra-striatal prefrontal correlates of apathy have been observed (Dowd et al., Reference Dowd, Frank, Collins, Gold and Barch2016; Wang et al., Reference Wang, Liu, Li, Wei, Jiang, Geng, Zou, Lui, Cheung, Pantelis and Chan2016), this transdiagnostic approach allows the comparison of potential contributions of extra-striatal dysfunction to the formation of apathy in BD and SZ.
The aims of the present study therefore were (1) to explore, on a psychopathological level, the association between apathy and other symptom dimensions in BD-I patients currently in an euthymic state compared with patients with SZ (Kirschner et al., Reference Kirschner, Hager, Bischof, Hartmann, Kluge, Seifritz, Tobler and Kaiser2016b), (2) to investigate whether euthymic BD-I patients show impaired VS or DS activity during reward anticipation during a variant of the Monetary Incentive Delay (MID) Task, (3) to compare BD-I patients to patients with SZ regarding the association between VS and DS activity during reward anticipation and the two NS domains apathy and diminished expression and (4) to study extra-striatal differences between BD-I patients and HC using an explorative whole-brain analysis approach.
Material and methods
Participants
Twenty-five participants with a diagnosis of BD-I and 27 patients with a diagnosis of SZ (Kirschner et al., Reference Kirschner, Hager, Bischof, Hartmann, Kluge, Seifritz, Tobler and Kaiser2016b) were recruited from in- and outpatient units of the Psychiatric Hospital of the University of Zurich, Switzerland and affiliated institutions. Twenty-five HC, all currently employed, were recruited from the general community. For all participants, the inclusion criterion for age was 18–55 years. Regarding the recruitment of euthymic BD-I patients, we included 18 patients from outpatient units and seven patients from inpatient units, which were at the end of their hospitalization. The data from SZ patients (11 outpatients and 16 inpatients) and HC rely on our previously published study (Kirschner et al., Reference Kirschner, Hager, Bischof, Hartmann, Kluge, Seifritz, Tobler and Kaiser2016b).
All patients had a stable medication for at least 14 days and a dose of lorazepam not exceeding 1 mg daily. Using the structured Mini-International Neuropsychiatric Interview for DSM-IV (Ackenheil et al., Reference Ackenheil, Stotz and Dietz-Bauer1999), we (A) confirmed the diagnosis of BD-I or SZ, (B) excluded patients with a current manic, hypomanic or major depressive episode, (C) excluded patients with a history of schizoaffective disorder, and (D) excluded patients with any other DSM-IV Axis I disorder. HC were screened for neuropsychiatric disorders using the structured Mini-International Neuropsychiatric Interview to ensure that they had no previous or present psychiatric illnesses. All participants were required to have a normal physical and neurologic status (including the assessment of extrapyramidal side effects, i.e. a total score ⩽2 on the Modified Simpson–Angus Scale (MSAS)] (Simpson and Angus, Reference Simpson and Angus1970) and no history of major head injury or neurological disorder. Patients with BD-I were clinically euthymic and on a stable medication for at least 2 weeks before the study. None of the patients with BD-I showed more than sub-syndromal depressive symptoms (Hamilton Depression Rating Scale, HAMD 17 Score <17), as defined by the International Society for Bipolar Disorder (ISBD) Task Force (Tohen et al., Reference Tohen, Frank, Bowden, Colom, Ghaemi, Yatham, Malhi, Calabrese, Nolen, Vieta, Kapczinski, Goodwin, Suppes, Sachs, Chengappa, Grunze, Mitchell, Kanba and Berk2009). The absence of manic symptoms was confirmed with the Young Mania Rating Scale (Young et al., Reference Young, Biggs, Ziegler and Meyer1978) (mean = 0.4, s.d. = 0.91, min = 0, max = 4). The study was approved by the local ethics committee of the canton of Zurich and all participants provided written informed consent.
Clinical and neuropsychological assessment
NS were assessed using the Brief Negative Symptom Scale (Kirkpatrick et al., Reference Kirkpatrick, Strauss, Nguyen, Fischer, Daniel, Cienfuegos and Marder2011) to answer all research questions in the present study. For comparison with earlier studies, additional assessment of NS using the Scale for the Assessment of Negative Symptoms (SANS) [Andreasen NC. Scale for the Assessment of Negative Symptoms (SANS). Iowa City, IA: University of Iowa; 1983] can be found in the online Supplementary Results. Apathy was defined based on the BNSS subscales anhedonia (items 1–3), asociality (items 5, 6) and avolition (items 7, 8) and diminished expression was defined as the sum of the BNSS subscales blunted effect (items 9–11) and alogia (items 12, 13) (Kirkpatrick et al., Reference Kirkpatrick, Strauss, Nguyen, Fischer, Daniel, Cienfuegos and Marder2011; Mucci et al., Reference Mucci, Dima, Soricelli, Volpe, Bucci, Frangou, Prinster, Salvatore, Galderisi and Maj2015). Please note, that the BNSS total score also includes the BNSS distress item, which was not included in one of the two factors (Kirkpatrick et al., Reference Kirkpatrick, Strauss, Nguyen, Fischer, Daniel, Cienfuegos and Marder2011; Mucci et al., Reference Mucci, Dima, Soricelli, Volpe, Bucci, Frangou, Prinster, Salvatore, Galderisi and Maj2015).
Additional psychopathological assessment included the Hamilton Depression Rating Scale 21 (Hamilton, Reference Hamilton1967) and the Calgary Depression Scale for Schizophrenia (CDSS) (Addington et al., Reference Addington, Addington and Schissel1990) to assess depressive symptoms. The CDSS has also been validated for patients with major depressive disorders (Micoulaud-Franchi et al., Reference Micoulaud-Franchi, Faugere, Weibel, Faget, Lancon, Richieri and Cermolacce2018) and has been suggested as a transdiagnostic tool to assess subclinical depressive symptoms across BD-I patients and SZ patients. The Global Assessment of Functioning Scale (GAF) (Frances et al., Reference Frances, Pincus and First1994) and Personal and Social Performance Scale (PSP) (Juckel et al., Reference Juckel, Schaub, Fuchs, Naumann, Uhl, Witthaus, Hargarter, Bierhoff and Brune2008) were used to assess the global level of functioning, and the Positive and Negative Syndrome Scale (PANSS) (Kay et al., Reference Kay, Opler and Lindenmayer1989) was used to assess all psychotic symptom dimensions. The PANSS factor scores were calculated according to the five-factor model of Wallwork et al. (Reference Wallwork, Fortgang, Hashimoto, Weinberger and Dickinson2012). Since the different factor scores do not include all the PANSS items, sums of the subscales do not add up to the total PANSS score. All participants performed a comprehensive neuropsychological test battery, which has been used in previous studies (Hager et al., Reference Hager, Kirschner, Bischof, Hartmann-Riemer, Kluge, Seifritz, Tobler and Kaiser2015; Hartmann et al., Reference Hartmann, Hager, Reimann, Chumbley, Kirschner, Seifritz, Tobler and Kaiser2015) (online Supplementary Methods).
MID task
We employed a variant of the MID (Knutson et al., Reference Knutson, Westdorp, Kaiser and Hommer2000) with stimuli based on the Cued-Reinforcement Reaction Time Task (Cools et al., Reference Cools, Blackwell, Clark, Menzies, Cox and Robbins2005). This modified version was originally developed by Simon et al. (Reference Simon, Cordeiro, Weber, Friederich, Wolf, Weisbrod and Kaiser2015) and used in previous studies to investigate reward anticipation (Simon et al., Reference Simon, Cordeiro, Weber, Friederich, Wolf, Weisbrod and Kaiser2015; Kirschner et al., Reference Kirschner, Hager, Bischof, Hartmann, Kluge, Seifritz, Tobler and Kaiser2016b, Reference Kirschner, Hager, Muff, Bischof, Hartmann-Riemer, Kluge, Habermeyer, Seifritz, Tobler and Kaiser2018). In this variant, reward amount was directly determined by the individual response time (RT) of each participant (online Supplementary Fig. S2). This adaptation allowed us to investigate the motivational properties of reward anticipation in the presence of strong action–outcome contingencies. Briefly, before starting the experiment, participants were informed that they would receive all the money won during the two experimental sessions. At the beginning of each trial, one of three different cues was presented for 0.75 s. The cue indicated the maximum amount participants could gain [i.e. 2 Swiss francs (CHF), 0.40 CHF or 0 CHF; 1 CHF = US$1.03). After a delay of 2.5–3 s, participants had to identify an outlier from three presented circles and press a button as fast as possible. Immediately after the button press, participants were notified of the money they had won (duration of feedback 2 s) (online Supplementary Fig. S1). Error trials were defined as trials with an incorrect or late response (after 1 s). In all other trials, we calculated the pay-out structure for each trial on the basis of the RTs of the previous 15 individual trials (online Supplementary Fig. S2). Therefore, the amount of money won depended on the RT in the current trial in relation to the RT in the previous 15 trials. The maximum amount of money to be won was 50 CHF. Every participant performed two training runs, one outside and one inside the scanner. Excluding the training sessions, the experiment included two runs with 36 trials of about 10 s each. The inter-trial interval was jittered from 1 to 9 s with a mean of 3.5 s to enhance statistical power. In total, one run lasted about 6 min.
Functional image acquisition
We used the same protocol as in our previous studies (Kirschner et al., Reference Kirschner, Hager, Bischof, Hartmann-Riemer, Kluge, Seifritz, Tobler and Kaiser2016a, Reference Kirschner, Hager, Bischof, Hartmann, Kluge, Seifritz, Tobler and Kaiser2016b), which is described in the online Supplementary Methods.
Data analyses
Data (demographic, clinical, neuropsychological and behavioural) were analysed using SPSS (version 23, SPSS Inc.), and the test for normal distribution was applied for all data. Group differences in socio-demographic and neuropsychological characteristics were investigated using two-sample t tests for continuous and χ2 tests for categorical data. For non-normally distributed data, Mann–Whitney U tests were applied. The association symptom dimensions were investigated using bivariate Spearman correlations (r s). In particular, apathy was correlated with potential secondary sources for NS including depressive symptoms, positive symptoms and antipsychotic dose equivalents. FMRI data were analysed using SPM8 (revision 5236, 4 February 2013) (Statistical Parametric Mapping, Wellcome Trust Centre for Neuroimaging, London, UK). The statistical tests were selected for separate comparisons of BD-I patients with HC or SZ, excluding the previously reported comparison of HC with SZ (Kirschner et al., Reference Kirschner, Hager, Bischof, Hartmann, Kluge, Seifritz, Tobler and Kaiser2016b).
Behavioural data analyses
The main behavioural outcome measure was RT, defined as the time between target presentation and pressing the correct button. We conducted a two-way repeated-measures analysis of variance (ANOVA) with RT as a dependent variable, group as a between-subject factor and reward condition (neutral, low, high) as a within-subject factor. We performed the Mauchly test for the assumption of sphericity and in case of violations report Greenhouse–Geisser-corrected degrees of freedom. Bonferroni-corrected pairwise comparisons were calculated as post-hoc tests for significant main effects. We performed a correlation analysis between NS factors and reward-related speeding. Reward-related speeding was calculated by subtracting the RT during the neutral condition (CHF 0) from the RT during the high reward condition (CHF 2.0). One BD-I subject was excluded only from the behavioural analysis due to corrupted data files. To account for multiple comparisons (group comparison of BD-I patients with HC and patients with SZ separately), we considered findings at p < 0.025 as significant.
Image pre-processing
The Image pre-processing followed the same protocol as in our previous studies (Kirschner et al., Reference Kirschner, Hager, Bischof, Hartmann-Riemer, Kluge, Seifritz, Tobler and Kaiser2016a, Reference Kirschner, Hager, Bischof, Hartmann, Kluge, Seifritz, Tobler and Kaiser2016b) and is outlined in the online Supplementary Methods.
First- and second-level image analyses
A general linear model approach was applied to assess data in an event-related design at the first level. For the three different reward anticipation phases, separate regressors were included: anticipation of no reward (CHF 0), anticipation of low reward (CHF 0.40) and anticipation of high reward (CHF 2.0). For the outcome phases, regressor for each condition (three basic regressors) was included. Additionally, for the low and high reward conditions the two outcome regressors were parametrically modulated by the actual outcome amount of each trial. Target presentation (one regressor) and anticipation, target and outcome phase in error trials (three regressors) were modelled as regressors of no interest. In total, the first-level model included 12 regressors. The canonical haemodynamic response function was used for convolving all explanatory variables. For reward anticipation, the contrast anticipation of high reward (CHF 2.0) v. anticipation of no reward (CHF 0) was calculated. At the second-level analysis, individual contrast images of all participants were included in a random-effects model. Within-group activation was calculated using a one-sample t test and between-group activation using a two-sample t test. For all whole-brain analysis, the statistical threshold was set to p < 0.05, whole-brain cluster-level family-wise error (FWE) rate corrected for multiple comparisons with a cluster-defining voxel-level threshold of p < 0.001 uncorrected.
Region of interest image analysis
The VS and DS were defined as regions of interest (ROIs) during anticipation of reward. We derived VS coordinates [Montreal Neurological Institute (MNI)] from a meta-analysis (Knutson and Greer, Reference Knutson and Greer2008) (left: x = −12, y = 10, z = −2; right: x = 10, y = 8, z = 0; 9 mm sphere in one single VS ROI). DS coordinates were derived from a previously published fMRI data, indicating DS activation in response to the MID task (left: x = −9, y = 3, z = 15; right: x = 9, y = 3, z = 15; 9 mm sphere in one single DS ROI) (Yip et al., Reference Yip, Worhunsky, Rogers and Goodwin2015). This ROI approach was adapted from Yip et al. (Reference Yip, Worhunsky, Rogers and Goodwin2015) and has been used in our previous studies (Kirschner et al., Reference Kirschner, Hager, Bischof, Hartmann-Riemer, Kluge, Seifritz, Tobler and Kaiser2016a, Reference Kirschner, Hager, Bischof, Hartmann, Kluge, Seifritz, Tobler and Kaiser2016b). The ROIs were constructed with the Wake Forest University Toolbox implemented in SPM8 (Maldjian et al., Reference Maldjian, Laurienti, Kraft and Burdette2003). The statistical thresholds at the voxel level were set to a FWE-corrected threshold for multiple comparison of p = 0.05 in each ROI. Mean per cent signal changes were extracted for all voxels in the VS and DS using the REX toolbox (http://web.mit.edu/swg/software.htm). In other words, extraction and average calculation were not restricted to the significant voxels but to all voxels of the ROIs. In addition, we performed an extra-striatal ROI analysis in the prefrontal cortex, which was based on previous findings showing an association between a behavioural measure of apathy and reduced inferior frontal gyrus (IFG) activation (Kluge et al., Reference Kluge, Kirschner, Hager, Bischof, Habermeyer, Seifritz, Walther and Kaiser2018). Structural masks of the left and right IFG were derived from the IBASPM 71 atlas implemented in the Wake Forest University Toolbox in SPM8 (Maldjian et al., Reference Maldjian, Laurienti, Kraft and Burdette2003).
Correlation analysis
The main hypothesis was tested by calculating bivariate Spearman correlations (r s) between NS (apathy and diminished expression) and per cent signal change in the VS and DS. Finally, the Steiger test for dependent correlation coefficients was performed to test for potential differences between these correlations (Steiger, Reference Steiger1980).
Results
Sample characteristics
Participant characteristics, clinical data and group comparisons are summarized in Table 1. Patients with BD-I had a significant higher average number of education years compared with HC and SZ patients. Patients in the BD-I group were older than SZ patients and were characterized by a longer duration of illness compared with the SZ group. On both the GAF (t = −2.613, p = 0.012) and the PSP (t = −2.163, p = 0.036), BD-I patients scored higher, indicating a higher level of functioning compared with patients with SZ. In the BD group, 18 out of 25 patients were treated with atypical antipsychotics, 18 patients received mood stabilizer and seven patients antidepressants. All patients with SZ received atypical antipsychotics. Mean chlorpromazine equivalents were higher in patients with SZ compared with BD-I patients (BD-I: 185.99 ± 259.8, SZ: 508.01 ± 369.2; U = 133, p < 0.001). None of the patients received typical antipsychotics.
Table 1. Summary of the socio-demographic and neuropsychological characteristics of the study participants
BD-I, bipolar disorder I; SZ, schizophrenia; HC, healthy controls; BNSS, Brief Negative Symptom Scale. BNSS Apathy was defined based on the BNSS subscales anhedonia (items 1–3), asociality (items 5, 6) and avolition (items 7, 8) and BNSS Diminished expression was defined as the sum of the BNSS subscales blunted affect (items 9–11) and alogia (items 12, 13); PANSS, Positive and Negative Syndrome Scale; HAMD, Hamilton Depression Rating Scale; CDSS, Calgary Depression Scale for Schizophrenia; GAF, Global Assessment of Functioning; PSP, Personal and Social Performance Scale; MWT IQ, Multiple Word Test Intelligence Quotient.
Data are presented as average ± standard deviation.
Psychopathological data
To address our first aim, we compared the severity of NS in BD-I with SZ patients (Fig. 1 and Table 1 for statistics). Whereas the two groups did not differ in BNSS apathy (Fig. 1a), BD-I patients scored significantly lower on the BNSS expression subscale (Fig. 1b). Moreover, we observed a trend that BD-I patients have less total BNSS scores compared with SZ patients.
Fig. 1. (a) Comparison of BNSS apathy between BD-I patients and patients with schizophrenia revealed no differences, whereas (b) BD-I patients showed lower BNSS expression and (c) higher CDSS total scores than patients with schizophrenia. There was a significant correlation between BNSS apathy and CDSS in (d) BD-I but not in (e) patients with schizophrenia. *p < 0.05.
BD-I patients showed higher CDSS scores than SZ patients (Fig. 1c). There was a significant correlation between BNSS apathy and CDSS in BD-I (Fig. 1d) but not in patients with SZ (Fig. 1e). There was no significant correlation between BNSS diminished expression and CDSS in either group (BD-I: r s = 0.268, p = 0.195; SZ: r s = 0.174, p = 0.385). Given the significant correlation between apathy and subclinical depressive symptoms, we performed additional correlation analyses with the three subdomains anhedonia, avolition and asociality of the BNSS apathy factor. In BD, all three subdomains were correlated with depressive symptoms (anhedonia: r s = 0.599, p = 0.002; avolition: r s = 0.418, p = 0.038; asociality: r s = 0.397, p = 0.049). Although these findings suggest a stronger association between anhedonia and depression, we did not observe a significant difference compared to the correlation between avolition and depression (Z = 1.60, p = 0.109) and the correlation between asociality and depression (Z = 1.464, p = 0.143). Additionally, the BNSS items comprising blunted affect did not correlate with the CDSS score in BD I (r s = 0.294, p = 0.154) and SZ patients (r s = 0.163, p = 0.418). In sum, these analyses suggest that in BD-I patients, subclinical depressive symptoms were associated with all three subdomains of the apathy factor.
Taken together, these findings revealed that BD-I patients have similar levels of apathy but lower levels of diminished expression compared with SZ patients. Interestingly, apathy was correlated with sub-syndromal depressive symptoms in BD-I patients but not in patients with SZ.
BD-I patients had lower scores on the PANSS-positive symptom subscale than SZ patients. However, there was no correlation of positive symptoms with BNSS apathy in either of the groups (BD-I: r s = −0.019, p = 0.928; SZ: r s = −0.136, p = 0.500). Regarding antipsychotic medication, chlorpromazine equivalents were higher in patients with SZ compared with BD-I patients (BD-I: 185.99 ± 259.8, SZ: 508.01 ± 369.2; U = 133, p < 0.001). However, there was no correlation with BNSS apathy in either of the groups (BD-I: r s = 0.207, p = 0.320; SZ: r s = −0.047, p = 0.816). Thus, no evidence for a contribution of positive symptoms or antipsychotic medication to the observed apathy symptoms was found.
Behavioural data
Comparison between BD-I patients and HC
The repeated-measures ANOVA with reward condition as within-subject factor and group affiliation as between-subject (BD-I patients and HC) factor revealed a significant main effect of reward [F (1.5,70) = 67.661, p < 0.001] but no significant effect of group [F (1,47) = 2.196, p = 0.15] or group × reward interaction [F (1.5,70) = 2.117, p = 0.14]. Bonferroni post-hoc pairwise comparison of RTs revealed significant differences between all reward conditions (all p < 0.001; no reward < low reward < high reward). These results indicate intact reward-related speeding in HC and BD-I. Furthermore, neither BNSS apathy nor BNSS expression correlated significantly with reward-related speeding (BNSS apathy: r = −0.062, p = 0.774; BNSS expression: r = 0.018, p = 0.933). Finally, we did not find group differences in total error rates (BD-I: 5.4 ± 3.6, HC: 5.7 ± 4.0; U = 293.5, p = 0.896) or total money gain (BD-I: 39.5 ± 4.5, HC: 38.9 ± 5.2; U = 274, p = 0.603) (online Supplementary Table S1).
Comparison between BD-I patients and SZ patients
We repeated the same analysis to compare the BD-I patients with the previously published sample of chronic SZ patients (Kirschner et al., Reference Kirschner, Hager, Bischof, Hartmann, Kluge, Seifritz, Tobler and Kaiser2016b). The repeated-measures ANOVA with reward condition as within-subject factor and group affiliation as between-subject factor (BD-I and SZ patients) revealed a significant main effect of reward [F (1.4,67) = 39.376, p < 0.001] but no significant effect of group [F (1,49) = 0.049, p = 0.83] or group × reward interaction effect was observed [F (1.4,67) = 2.907, p = 0.08] (online Supplementary Table S1). None of the measures for task performance (RT, error rate, total gain) showed a significant correlation with IQ (estimated with the MWT) (online Supplementary Table S2).
fMRI data
Group comparison of VS and DS activation between BD-I patients and HC
A voxel-wise whole-brain analysis of the reward anticipation contrast in the combined sample of BD-I patients and HC revealed strong task-related activation in reward coding regions including the VS and DS (online Supplementary Table S6, Fig. 2).
Fig. 2. Activation map of the reward anticipation contrast (anticipation of high reward v. anticipation of no reward) across the complete sample. Illustrated in coronal orientation (upper line) and axial (lower line) orientation. The VS ROI (a) and DS ROI (b) were overlaid on the activation map. The threshold was set at p < 0.05 FWE corrected.
To address the second aim of the study (relation of group to VS and DS activity), we compared the mean contrast signal from our a priori defined VS and DS ROI between HC and patients with BD-I (Fig. 3a, c). There were no significant group differences in neural activation during reward anticipation in the VS (t = 0.412, p = 0.682) or DS (t = 1.09, p = 0.279). Thus, the VS and DS showed increased activity related to reward anticipation irrespective of group membership. Furthermore, in an explorative analysis we did not observe group differences in the low reward–no reward anticipation contrast and high reward–low reward anticipation contrast (online Supplementary Table S3).
Fig. 3. (a, d) Group comparison of activation during reward anticipation revealed no differences between HC and individuals with BD-I in VS (a) or DS (d). (b, c) VS activity during reward anticipation did not correlate with apathy or diminished expression. (e, f) DS activity during reward anticipation did not correlate with apathy or diminished expression.
Group comparison of VS and DS activation between BD-I patients and SZ
In the next step, we compared the mean contrast signal of the VS and DS ROI between BD-I patients and the previously published sample of chronic SZ patients (Kirschner et al., Reference Kirschner, Hager, Bischof, Hartmann, Kluge, Seifritz, Tobler and Kaiser2016b). There were no significant group differences in neural activation during reward anticipation in the VS (t = 0.93, p = 0.357) or DS (t = 0.267, p = 0.79). Thus, VS and DS activations during reward anticipation did not differ significantly between BD-I patients and chronic SZ patients.
Correlation analysis between reward anticipation and NS domains
The third aim of the study was to investigate the association between the two NS dimensions apathy and diminished expression and the mean contrast signal in our a priori ROIs of the VS and DS. Neither VS activity nor DS activity during reward anticipation was associated with one of the two NS dimensions apathy or diminished expression (VS: apathy r s = −0.002, diminished expression r s = −0.058; DS: apathy r s = 0.189, diminished expression r s = 0.236, all p > 0.05) (Fig. 3b, d). In sum, in patients with BD-I neither apathy nor diminished expression showed any association with reward anticipation in the striatum.
In further analysis, we investigated the potential correlation between depressive symptoms (CDSS Total score) and reward anticipation in the VS and DS. Again, we did not find any association between symptom expression and VS and DS activity (VS: r s = 0.136, p = 0.515; DS: r s = 0.145, p = 0.490). Furthermore, neither in BD-I patients nor in patients with SZ antipsychotic medication dose correlated with striatal activation during reward anticipation (SZ, VS: r s = 0.128, p = 0.525, DS: r s = 0.045, p = 0.824; BD-I, VS: r s = 0.086, p = 0.684, DS: r s = −0.052, p = 0.804). Additionally, VS and DS activation during reward anticipation were not associated with functioning or education (online Supplementary Table S5). In the next step, we compared the correlation between apathy and VS activity in BD-I patients with the previous published negative correlation between apathy and VS activity in SZ patients. Using Fisher's r-to-z transformation, we found a trend-level significance (Z = 1.72, p = 0.085 two-tailed) for the differences between the two correlation coefficients suggesting that the association between apathy and VS activation was stronger in patients with SZ than in patients with BD-I. Finally, we conducted explorative correlation analyses for the low reward–no reward anticipation contrast and high reward–low reward anticipation contrast, which revealed no significant correlation with apathy (online Supplementary Table S4).
Extra-striatal dysfunction during reward anticipation
The fourth aim of the study was to identify potential extra-striatal prefrontal group differences in brain activity during reward anticipation. In our a priori IFG ROI, no group differences were observed comparing BD-I patients with HC and SZ patients. Furthermore, SZ patients did not differ from HC. Additional explorative whole-brain analysis in BD-I patients revealed reduced activation during reward anticipation in the fusiform gyrus, lingual gyrus and precuneus compared with HC (online Supplementary Table S7).
In the next step, we aimed to identify potential associations between apathy and reduced activation in extra-striatal regions. Based on previous findings from our own group, we expected a negative association between apathy and IFG activation. Please note that the main effect of high reward anticipation v. reward anticipation across the complete sample revealed activity in the right and left IFG supporting the relevance of this region in reward anticipation. In BD-I patients, we found a significant negative association between right IFG activation and apathy (56, 27, 9, cluster size = 41, t = 5.05, p = 0.034, peak-level FWE <0.05 corrected) (Fig. 4). No association was observed in the left IFG. Please note that after Bonferroni correction for the number of ROIs, this result was only at trend-level significance (p = 0.068). A secondary linear regression model with the log-transformed CDSS scores revealed no significant association between sub-syndromal depressive symptoms and reduced IFG activation. To test whether BD-I and SZ patients share the same apathy-related IFG hypoactivation, we repeated the linear regression with apathy and IFG activation during reward anticipation in patients with SZ. In contrast to BD-I patients, we did not observe an association between apathy and reduced IFG activation in patients with SZ.
Fig. 4. Associations between apathy and reduced activation in the right inferior frontal gyrus (56, 27, 9, cluster size = 41, t = 5.05, p = 0.034, peak-level FWE <0.05 corrected) in euthymic BD-I patients.
Discussion
The present study investigated apathy and diminished expression in euthymic BD-I patients and its association with VS and DS activation during reward anticipation in a variant of the MID task. In euthymic BD-I patients, clinically measured severity of apathy was similar to patients with SZ (Kirschner et al., Reference Kirschner, Hager, Bischof, Hartmann, Kluge, Seifritz, Tobler and Kaiser2016b). However, BD-I patients showed less emotional expressivity deficits than patients with SZ. On the neural level, apathy was not associated with VS and DS activity during reward anticipation but showed an association with reduced prefrontal activation in the IFG.
It has recently been suggested that NS constitute a valid transdiagnostic psychopathological construct (Strauss et al., Reference Strauss, Vertinski, Vogel, Ringdahl and Allen2016). In the present study, we confirm that NS can be assessed in euthymic BD-I patients and that the overall severity of both total BNSS and BNSS apathy is not different compared with a recently published study with SZ patients (Kirschner et al., Reference Kirschner, Hager, Bischof, Hartmann, Kluge, Seifritz, Tobler and Kaiser2016b). In contrast, BD-I patients scored less on the BNSS expression subscale than SZ patients. Findings of similar scores on the BNSS apathy subscale but differences in BNSS expression are in line with a recent study by Strauss et al. (Reference Strauss, Vertinski, Vogel, Ringdahl and Allen2016). Although not fulfilling the criteria for a depressive episode, BD-I patients showed more depressive symptoms compared with SZ patients and these correlated significantly with BNSS apathy in the BD-I but not in the SZ group. Interestingly in euthymic BD-I patients, subclinical depressive symptoms were related to all three subdomains of the BNSS apathy factor: anhedonia, avolition and asociality. These findings suggest that when exploring NS in BD, it might be difficult to differentiate symptoms such as anhedonia from other depressive symptoms (here assessed with the CDSS). This overlap and its potential impact on underlying neurobiological mechanism should be taken into account when investigating NS and depression in a transdiagnostic way. Furthermore, it is well known that BD-I patients in the euthymic state as determined by diagnostic criteria and structured interviews display a broad range of neuropsychological and psychopathological impairments, including sub-syndromal depressive symptoms (Bourne et al., Reference Bourne, Aydemir, Balanza-Martinez, Bora, Brissos, Cavanagh, Clark, Cubukcuoglu, Dias, Dittmann, Ferrier, Fleck, Frangou, Gallagher, Jones, Kieseppa, Martinez-Aran, Melle, Moore, Mur, Pfennig, Raust, Senturk, Simonsen, Smith, Bio, Soeiro-de-Souza, Stoddart, Sundet, Szoke, Thompson, Torrent, Zalla, Craddock, Andreassen, Leboyer, Vieta, Bauer, Worhunsky, Tzagarakis, Rogers, Geddes and Goodwin2013; Roux et al., Reference Roux, Raust, Cannavo, Aubin, Aouizerate, Azorin, Bellivier, Belzeaux, Bougerol, Cussac, Courtet, Etain, Gard, Job, Kahn, Leboyer, Olie, Henry and Passerieux2017). These residual depressive symptoms – in contrast to residual hypomanic symptoms – are an important predictor of functional outcome and quality of life (Dias et al., Reference Dias, Brissos, Frey and Kapczinski2008; Bonnin et al., Reference Bonnin, Martinez-Aran, Torrent, Pacchiarotti, Rosa, Franco, Murru, Sanchez-Moreno and Vieta2010).
When interpreting the results of the present study, it is therefore important to take into account the concept of secondary NS: whereas primary NS are defined by being inherent to the disorder, secondary NS are thought to have different underlying sources such as depressive symptoms, productive psychotic features or medication (Kirkpatrick, Reference Kirkpatrick2014; Kirschner et al., Reference Kirschner, Aleman and Kaiser2017). Our data clearly show that in euthymic patients with BP, motivational NS are at least in part secondary to sub-syndromal depressive symptoms. In contrast, we found no evidence for a contribution of other secondary sources of NS such as positive symptoms or medication side-effects. Taken together, the psychopathological findings of our study suggest that whereas NS can indeed be assessed as a transdiagnostic construct, the specific characteristics of the different subdomains are likely to differ between traditional categorical disease entities, emphasizing the importance of a detailed psychopathological assessment.
While the clinically measured severity of apathy in BD-1 patients was comparable to that of patients with SZ, the specific association of apathy to sub-syndromal depressive symptoms in BD-I patients but not SZ patients suggests different underlying neural mechanisms during reward anticipation: In contrast to the growing evidence that apathy is associated with reduced striatal activity in SZ (Wolf et al., Reference Wolf, Satterthwaite, Kantrowitz, Katchmar, Vandekar, Elliott and Ruparel2014; Mucci et al., Reference Mucci, Dima, Soricelli, Volpe, Bucci, Frangou, Prinster, Salvatore, Galderisi and Maj2015; Kirschner et al., Reference Kirschner, Hager, Bischof, Hartmann, Kluge, Seifritz, Tobler and Kaiser2016b), this association was not observed in patients with euthymic BD-I. In addition, in line with previous studies in euthymic BD patients (Caseras et al., Reference Caseras, Lawrence, Murphy, Wise and Phillips2013; Yip et al., Reference Yip, Worhunsky, Rogers and Goodwin2015) we did not observe an association between striatal activity and sub-syndromal depressive symptoms. Though speculative, one explanation for these divergent findings might be that primary and secondary NS have divergent neural correlates. While apathy, as a primary NS, is associated with striatal dysfunction, apathy secondary to sub-syndromal depressive symptoms may be related to extra-striatal mechanisms. Our findings suggest that apathy in patients with BD-I might be more strongly related to cognitive processes relevant for action initiation and execution as reflected by prefrontal cortex activation (Barch and Dowd, Reference Barch and Dowd2010). Critically, the observed association between apathy and reduced IFG activation during reward anticipation is in line with previous results showing reduced IFG activation in bipolar depression (Redlich et al., Reference Redlich, Dohm, Grotegerd, Opel, Zwitserlood, Heindel, Arolt, Kugel and Dannlowski2015). The relevance of prefrontal activity in euthymic and sub-syndromal depressive BD-I patients is underpinned by research reporting reduced top-down control of anteroventral prefrontal cortex on VS activity during reward processing (Trost et al., Reference Trost, Diekhof, Zvonik, Lewandowski, Usher, Keil, Zilles, Falkai, Dechent and Gruber2014). Nevertheless, our findings have to be interpreted with caution given that only apathy but not depression was associated with reduced prefrontal cortex activation in the present study. In addition, some studies in bipolar depression, unipolar depression and SZ have shown a correlation between depressive symptom scores and reduced striatal activity (Arrondo et al., Reference Arrondo, Segarra, Metastasio, Ziauddeen, Spencer, Reinders, Dudas, Robbins, Fletcher and Murray2015; Hagele et al., Reference Hagele, Schlagenhauf, Rapp, Sterzer, Beck, Bermpohl, Stoy, Strohle, Wittchen, Dolan and Heinz2015; Satterthwaite et al., Reference Satterthwaite, Kable, Vandekar, Katchmar, Bassett, Baldassano, Ruparel, Elliott, Sheline, Gur, Gur, Davatzikos, Leibenluft, Thase and Wolf2015). Taken together, the potentially shared and distinct neural substrates of primary apathy and secondary apathy due to depression still have to be elucidated.
An additional explanation for these divergent findings might be the greater heterogeneity of neural activation during reward anticipation in different stages and subtypes of BD. While the meta-analysis of Radua et al. (Reference Radua, Schmidt, Borgwardt, Heinz, Schlagenhauf, McGuire and Fusar-Poli2015) demonstrated that there is consistent evidence for reduced striatal activation during reward anticipation in SZ, the findings in BD are still inconclusive (Ashok et al., Reference Ashok, Marques, Jauhar, Nour, Goodwin, Young and Howes2017). Recently, Caseras et al. reported increased bilateral VS activity in BD-II patients, but no differences in BD-I patients compared with HC (Caseras et al., Reference Caseras, Lawrence, Murphy, Wise and Phillips2013). This is in line with our observation of comparable activation during reward activation in euthymic BD-I patients and HC. However, a recent study by Yip et al. showed that during reward anticipation, unmedicated BD-II patients had decreased DS but not VS activity (Yip et al., Reference Yip, Worhunsky, Rogers and Goodwin2015). Here, similar to the literature on SZ, which reported opposing results in unmedicated patients and patients treated with atypical antipsychotics (Schlagenhauf et al., Reference Schlagenhauf, Juckel, Koslowski, Kahnt, Knutson, Dembler, Kienast, Gallinat, Wrase and Heinz2008; Nielsen et al., Reference Nielsen, Rostrup, Wulff, Bak, Lublin, Kapur and Glenthoj2012), the divergent findings might be due to differences in medication between the present study (medicated) and the previous report from Yip et al. (unmedicated) (Yip et al., Reference Yip, Worhunsky, Rogers and Goodwin2015). In sum, these findings point towards different mechanisms underlying motivational deficits between different stages and subtypes of BD. Of note, the explorative whole-brain analyses of our study revealed reduced extra-striatal activation in fusiform gyrus, lingual gyrus and precuneus in BD-I compared with HC. These results replicated previous findings in BD-II patients (Yip et al., Reference Yip, Worhunsky, Rogers and Goodwin2015) and suggest that temporal, occipital and parietal regions involved in reward processing (Sescousse et al., Reference Sescousse, Caldu, Segura and Dreher2013) are impaired in euthymic BD-I patients.
Limitations and future directions
There are several limitations to the present study: First, the sample size is only moderate and although we made an effort to recruit a homogeneous sample regarding the current disease stage, it is challenging to minimize variance in a number of past depressive/manic episodes and duration of illness in chronic BD-I patients. These characteristics could mask specific mechanisms underlying various stages of the disorder. In particular, we did not use a conservative definition of euthymia (Samalin et al., Reference Samalin, Reinares, de Chazeron, Torrent, Bonnin, Hidalgo-Mazzei, Murru, Pacchiarotti, Geoffroy, Bellivier, Llorca and Vieta2016), but allowed sub-syndromal depressive symptoms according to the definition of the ISBD Task Force (Tohen et al., Reference Tohen, Frank, Bowden, Colom, Ghaemi, Yatham, Malhi, Calabrese, Nolen, Vieta, Kapczinski, Goodwin, Suppes, Sachs, Chengappa, Grunze, Mitchell, Kanba and Berk2009). In order to elucidate specific neural correlates of different stages in BD, future neuroimaging studies should take advantage of precise recommendations to overcome the high variability in clinical criteria and duration for euthymia used in previous research (Robinson et al., Reference Robinson, Thompson, Gallagher, Goswami, Young, Ferrier and Moore2006; Tohen et al., Reference Tohen, Frank, Bowden, Colom, Ghaemi, Yatham, Malhi, Calabrese, Nolen, Vieta, Kapczinski, Goodwin, Suppes, Sachs, Chengappa, Grunze, Mitchell, Kanba and Berk2009; Samalin et al., Reference Samalin, Reinares, de Chazeron, Torrent, Bonnin, Hidalgo-Mazzei, Murru, Pacchiarotti, Geoffroy, Bellivier, Llorca and Vieta2016). Furthermore, the present patient sample consists of medicated patients and does not take into account possible confounding effects of psychotropic drugs (Phillips et al., Reference Phillips, Travis, Fagiolini and Kupfer2008) since subgroups were too small to conduct post hoc comparisons between different drug classes. Third, the lack of specific measures for anhedonia other than the BNSS may have hampered the ability to identify distinct correlates for subdomains of apathy. Therefore, further studies should investigate detailed psychopathological characteristics including specific measures for anhedonia and motivational deficits during different episodes of the disorder. Especially, within-subject designs investigating different disease stages on an individual level could reveal important insights into the variability of the underlying neural mechanisms. Though speculative, understanding these processes could be an important step towards the development of targeted treatment interventions that address the oftentimes so burdensome switching between the different episodes of BD-I.
Conclusion
The present study indicates that on a psychopathological level, apathy is present in euthymic BD-I patients comparable to patients with SZ but appear to be related to sub-syndromal depressive symptoms. This highlights the relevance of detailed clinical assessments of other symptom dimensions in transdiagnostic studies of apathy. The lack of convergent neural correlates of apathy in the striatum and the association of apathy with extra-striatal dysfunction in BD-I patients suggests divergent neurobiological mechanisms of apathy in BD-I patients compared with SZ patients. Thus, this transdiagnostic approach contributes to the empirical differentiation of shared and divergent pathophysiological mechanisms of apathy across psychiatric disorders. Taken together, these findings strengthen the notion to investigate motivational deficits in a dimensional approach as conceptualized in the Research Domain Criteria (Cuthbert and Insel, Reference Cuthbert and Insel2010) in order to foster progress for new treatments (Strauss and Cohen, Reference Strauss and Cohen2017; Husain and Roiser, Reference Husai and Roiser2018).
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0033291719000801.
Author ORCIDs
Matthias Kirschner, 0000-0002-9486-1439.
Acknowledgements
This study was supported by the Swiss National Science Foundation (Grant No. 105314_140351 to Stefan Kaiser). Matthias Kirschner was supported by the National Bank Fellowship Award (Montreal Neurological Institute, McGill University). Philippe Tobler was supported by the Swiss National Science Foundation (PP00P1_150739, and PP00P1_00014_165884). Andrei Manoliu received a grant from the swiss national science foundation (P2SKP3_178107). We are grateful to Dr Philipp Staempfli for his excellent technical support. Furthermore, we would like to thank all participants for their time and interest in our study.
Conflict of interest
Stefan Kaiser has received speaker honoraria from Roche, Lundbeck, Janssen and Takeda. He receives royalties for cognitive test and training software from Schuhfried. Philippe Tobler has received grant support from Pfizer. Erich Seifritz has received grant support from H. Lundbeck and has served as a consultant and/or speaker for AstraZeneca, Otsuka, Takeda, Eli Lilly, Janssen, Lundbeck, Novartis, Pfizer, Roche and Servier. None of these activities are related to the present study. All other authors declare no biomedical financial interests or potential conflicts of interest.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the ethics committee of the canton of Zurich, Switzerland and with the Helsinki Declaration of 1975, as revised in 2008.
