Journal of Affective Disorders 135 (2011) 241–250

Contents lists available at ScienceDirect

Journal of Affective Disorders

j ourna l homepage: www.e lsev ie r.com/ locate / j ad

Research report

Phenomenology of psychotic mood disorders: Lifetime and major depressiveepisode features

Daniel Souery a, Leonardo Zaninotto b, Raffaella Calati b,⁎, Sylvie Linotte c, Othman Sentissi d,Daniela Amital e, Ulrike Moser f, Siegfried Kasper f, Joseph Zohar g,Julien Mendlewicz h, Alessandro Serretti b

a Laboratoire de Psychologie Medicale, Université Libre de Bruxelles and “Psy Pluriel”, Centre Européen de Psychologie Medicale, Brussels, Belgiumb Institute of Psychiatry, University of Bologna, Bologna, Italyc Fonds de la Recherche Scientifique (FNRS), Laboratoire de Neurologie Expérimentale, Université Libre de Bruxelles, Bruxelles, Belgiumd Département de Psychiatrie Hôpitaux Universitaires de Genève, Faculté de Médecine de Genève, Geneva, Switzerlande Ness-Ziona Mental Health Center, Ness-Ziona, Israelf Department of Psychiatry and Psychotherapy, Medical University of Vienna, Vienna, Austriag Chaim Sheba Medical Center, Tel-Hashomer, Israelh Université Libre de Bruxelles, Brussels, Belgium

a r t i c l e i n f o

⁎ Corresponding author at: Institute of Psychiatry,Viale Carlo Pepoli 5, 40123 Bologna, Italy. Tel.: +39 05051 521030.

E-mail address: raffaella.calati@unibo.it (R. Calati)

0165-0327/$ – see front matter © 2011 Elsevier B.V.doi:10.1016/j.jad.2011.07.027

a b s t r a c t

Article history:Received 23 June 2011Received in revised form 29 July 2011Accepted 29 July 2011Available online 1 September 2011

Background: The nosological and clinical implications of psychotic features in the course ofmood disorders have been widely debated. Currently, no specification exists for defining asubgroup of lifetime Psychotic Mood Disorder (PMD) patients.Methods: A total of 2178 patients were examined, including subjects with Bipolar Disorder (BP)type I (n=519) and II (n=207) and Major Depressive Disorder (n=1452). Patients weredivided between PMD (n=645) and non-psychotic Mood Disorders (MD) (n=1533) by thelifetime presence of at least one mood episode with psychotic features. Subjects having adepressive episode at the time of assessment were also examined: HAM-D and YMRS scoreswere compared between MD and PMD subjects, both with and without current psychoticfeatures.Results: A diagnosis of BP-I, a higher familial load for BP, a higher number of mood episodeslifetime, and a higher prevalence of OCD and somatic comorbidities were all associated to PMD.A diagnosis of BP (OR=4.48) was the only significant predictor for psychosis. PMD with non-psychotic depression were apparently less severe than MD patients and had a lower rate of“non-responders” to AD treatment. Sub-threshold manic symptoms and suicidal risk were alsomore pronounced among PMD.Limitations: The lack of information about number and polarity of previous psychotic moodepisodes may be the major limitations of our study.Conclusions: BP diagnosis is the most significant predictor for psychosis in mood disorders.Non-psychotic mood episodes in PMD patients may be characterized by a distinctive symptomprofile and, possibly, a different response to treatment.

© 2011 Elsevier B.V. All rights reserved.

Keywords:PsychosisBipolarUnipolarMajor depressionMood disorders

University of Bologna,1 6584233; fax: +39

.

All rights reserved.

1. Introduction

In the late 19th century, Emil Kraepelin described manic–depressive insanity (now called bipolar disorder) anddementia praecox (the modern schizophrenia) as distinctdiagnostic entities (Kraepelin, 1921). His dichotomy between

242 D. Souery et al. / Journal of Affective Disorders 135 (2011) 241–250

mood disorders and schizophrenia was mainly based on theirdifferent illness course and prognosis.

Still now, course and outcome are crucial variables indetermining the validity of a diagnostic concept (Robins andGuze, 1970; Feinstein, 1977). However, a lack of convincingsupport for either categorical or dimensional diagnostic modelshas progressively moved attention to new domains of psycho-pathology, where particular patho-physiological disturbances orgenetic factorsmightplayamajor role (BuchananandCarpenter,1994; Flaum et al., 1995; Serretti et al., 2000; Henry and Etain,2010).

In this sense, the nosological and clinical implications ofpsychotic features in the course of mood disorders have beenwidely debated. Currently, psychotic symptoms are consideredas a criterion for severity of mood episodes in both BipolarDisorder (BP) and Major Depressive Disorder (MDD) (A.P.A.,2000).No specification exists for defininga subgroupof lifetimePsychotic Mood Disorder (PMD) patients, though some studiessuggest that these patients might share with schizophrenia notonly symptom presentation, but also epidemiology (Murrayet al., 2004), and, probably, genetic susceptibility (Craddock etal., 2006; Goes et al., 2008; Ivleva et al., 2010). Secondly, wheresome authors have argued that psychotic features have novalidity in terms of prognosis, treatment or family history formood disorders (Pope and Lipinski, 1978; Maj et al., 1990; Jageret al., 2005), others have found a poorer response tomedications(Coryell et al., 1984), a worse short and long-term outcome(Coryell et al., 1982; Coryell et al., 1996; Coryell et al., 2001) andamore severe cognitive impairment (Grant et al., 2001; Hill et al.,2004; Jeste et al., 1996; Glahn et al., 2006). Moreover, while theclinical and prognostic implications of psychotic mania havebeen extensively explored (Pope and Lipinski, 1978; Coryellet al., 2001;Haro et al., 2006; Canuso et al., 2008), themeaningofpsychotic features for depressionmight have been biased by theseparation between unipolar (Coryell et al., 1996; Leyton et al.,1995; Harrow et al., 2000) and bipolar (Guze et al., 1975;Brockington et al., 1982;Mitchell et al., 2001; Parker et al., 2000;Endicott et al., 1985; Tafalla et al., 2009) patients. Indeed,although there is a growing interest in the study of thedepressive syndrome by a spectrum-based dimensional ap-proach – e.g. see the concept of “mixed depression” (Akiskal andBenazzi, 2003; Benazzi, 2000; Benazzi, 2003; Koukopoulos andKoukopoulos, 1999) – most studies on delusional or psychoticdepression seem to be implicitly inspired by a polarity-basedcategorical model, where unipolar and bipolar psychoticdepression are referred to as separate constructs. Finally, sincemost studies on the clinical and severity pattern of depressionusually make a comparison between psychotic and non-psychotic patients (Parker et al., 1991; Gaudiano et al., 2008),symptom profile and psychopathological presentation of non-psychotic mood episodes in PMD subjects have not beenexplored yet.

To address these issues we examined a largemulti-centersample in order to detect any demographic and/or clinicalfeature that might be associated with PMD. PMDs weredefined by the presence of at least one mood episode withpsychotic features during lifetime, independently frommood polarity. Secondly, we analyzed the phenomenologyof a Major Depressive Episode (MDE) with and withoutpsychotic features in both PMD and mood disorder patientswith no history of psychotic symptoms (MD). The purpose

was to determine whether a lifetime history of psychosismay affect symptom presentation of non-psychotic moodepisodes.

2. Methods

2.1. Study sample

Patients were recruited within the context of two largestudy projects: 1) the Group for the Study of ResistantDepression (GSRD) (started in January 2000) and 2) the COPEBipolar program (started in December 2003).

The GSRD is a large multicenter, multinational study withthe objective of defining some key issues in TreatmentResistant Depression (TRD), such as diagnosis, clinicalfeatures and treatment adequacy (for a description ofinclusion and exclusion criteria see (Souery et al., 2007,Souery et al., 2011a)). Seven centers have been involved inthis project: 1) Department of Psychiatry and Psychotherapy,Medical University of Vienna, Vienna, Austria; 2) Departmentof Psychiatry, Chaim Sheba Medical Center, Tel-Hashomer,Israel; 3) Department of Psychiatry, Erasme Hospital, Uni-versité Libre de Bruxelles, Brussels, Belgium; 4) Departmentof Psychiatry, University Hospital Gasthuisberg, Leuven,Belgium; 5) Hôpital la Salpetriere, INSERM U302, Paris,France; 6) Sint-Truiden Psychiatric center, Sint-Truiden,Belgium; 7) Ness-Ziona Mental Health Center, Israel.

The definitions of treatment resistance vary from nonresponse to a single antidepressant given at an adequate dosefor sufficient duration (Souery et al., 1999; Thase, 2001; Fava,2003) to the more complicated failure to respond to twoseparate courses of treatment with antidepressants from twodifferent classes for adequate duration and dose (Thase, 2001;Committee for Medicinal Products for Human Use CHMP,2002). For the present study “non-responders”were defined ashaving a HAM-D score higher than 17 after at least 4 weeks ofone antidepressant treatment given at an adequate dose.

Regarding the COPE Bipolar program, patients have beenrecruited within the “Psy Pluriel” center, Centre Européen dePsychologieMédicale (outpatients only), and thedepartmentofPsychiatry of ErasmeHospital (in- and outpatients), in Brussels(for a description of the sample see (Souery et al., 2011b)).Briefly, in this program the assessment of mood disorderpatients is completed using “COPE-Bipolar.COM” (ClinicalOutcome Measures for Bipolar Disorder), a software programconsisting on a structured examination tool and immediatedata capture. It is composed of ten “modules”, each of themdedicated to essential elements of mood disorders, such associo-demographic characteristics, familiarity, diagnosis, treat-ment, and quality of life and functioning.

Lifetime and current diagnosis, course of illness andcomorbidities were assessed by specifically trained psychiatristson the basis of the M.I.N.I. interview (Sheehan et al., 1998).Patientswere included if theymetDSM-IV criteria for adiagnosisof mood disorders (BP-I, BP-II and MDD). Individuals withmental retardation, dementia, neurological disorders or severeorganic diseases were excluded. Written informed consent wasobtained from all participants in the studies.

Psychiatric familial antecedents were screened by clinicalinvestigation. Positive or negative family history was used asa dichotomous indicator for familial loading using all possible

243D. Souery et al. / Journal of Affective Disorders 135 (2011) 241–250

sources of information including previous charts, familymembers and previous treating clinicians.

A total of 2178 patients were examined for demographicand clinical variables. One thousand six hundred twenty two(74.5%) subjects came from the GSRD, including: 265 (16.3%)BP-I, 106 (6.6%) BP-II and 1251(77.1%) MDD. Five hundred andfifty-six (25.5%) subjects came from the COPE Bipolar Program,including: 254 (45.7%) BP-I, 101 (18.2%) BP-II and 201 (36.1%)MDD. The overall sample included BP-I (n=519), BP-II(n=207) and MDD (n=1452) patients, who were dividedbetween Psychotic Mood Disorders (PMD) (n=645, 29.6%)and non-psychotic Mood Disorders (MD) (n=1533; 70.4%).Finally, we examined all those subjects who had a MajorDepressive Episode (MDE) according to the M.I.N.I. interview(Sheehan et al., 1998) at the time of assessment. PMD patientshaving a MDE with (n=256) and without (n=174) currentpsychotic features were compared to depressed MD patients(n=1215) (Fig. 1).

2.2. Assessment scales

Severity of mood symptoms for the current mood episodewas assessed using the Hamilton Depression Rating Scale, 21items (HAM-D) (Hamilton, 1960) and the Young Mania RatingScale (YMRS) (Young et al., 1978). For each scale, all the itemswereavailable.HAM-D itemswerealsopooled into the followingfactors:Core (items1, 2, 7, 8, 10, 13), Sleep(items4, 5, 6), Activity(items 7, 8), Psychic anxiety (items 9, 10), Somatic anxiety(items 11, 12, 13), andDelusion (items 2, 15, 20) (Lattuada et al.,1999; Serretti et al., 1998; Serretti et al., 1999; Sobin andSackeim, 1997).

Though the use of a rating scale for mania in depressedpatients may be questionable, we also decided to compareYMRS items among groups in order to detect any potentialdifference in the presence of sub-threshold manic symptoms.Although there is some evidence on a factor structure of theYMRS (Double, 1990), itemswere not pooled into factors. Theaccuracy and completeness of the collected clinical informa-tion have been independently tested by two researchers. Allraters were trained for the use of instruments with goodinter-rater reliability (kN0.8).

Non Psychotic(n=174)

Total S(

PMD with MDE at inclusion (n=430)

PMD (n=645)

Psychotic MDE (n=256)

Fig. 1. Flow chart of study sample. Legend: PMD=Psychotic Mood Disorders; M

2.3. Statistical analysis

Demographic and illness characteristics were compared byusing Chi-square test for categorical data and T-test andanalysis of variance (ANOVA) for continuous measures.Whenever indicated, post-hoc analyses (LSD test) wereperformed to discriminate individual effects. Finally, a non-linear logistic regression model was performed to determinethe best predictors for a diagnosis of PMD. Variables included inthe model were those which were significant on univariateanalyses.

Then, HAM-D factors and items and YMRS items werecompared among diagnostic groups (PMD and MD with non-psychotic MDE and PMDwith psychotic MDE) by using analysisof covariance (ANCOVA),with the total scoreof the rating scale ascovariant. All p values were 2-tailed, and statistical significancewas conservatively set at the 0.001 level since this sample hasbeen repeatedly investigated (Souery et al., 2007; Souery et al.,2011a, 2011b). Traditional statistical analyses were performedusing “Statistica” package (StatSoft, 1995). Power analysis hasbeen performed using G*Power 3.0.10 (Faul et al., 2007). Withtheseparameters (p=0.001)wehada sufficient power (0.80) todetect a small effect size (d=0.28)between the twomaingroups(PMDwithpsychoticMDE vsMDwithnon-psychoticMDE), that,as an example, corresponds to a difference in the HAM-D Corecluster of 0.98 points (Cohen, 1988).

3. Results

3.1. Demographic and clinical features of PMD

In our sample, PMD patients (n=645) resulted to be morefrequently single orwidowed, andwith a higher level education,when compared to MD patients (n=1533). They were alsomore represented among inpatients, supporting the view of amore severe illness course in this group (Table 1). Indeed, thesepatients exhibited also a higher rate of BP-I diagnosis, a higherincidenceof early onset, a younger ageat thefirst hospitalization,and a higher number of mood episodes lifetime (Table 2). In oursample, early age at onset was defined by the cut-off of 21 years(Fisfalen et al., 2005; Potash et al., 2007; Benazzi, 2009).We also

MDE

tudy Sample n=2178)

MD (n=1533)

MD with MDE at inclusion (n=1215)

D=non-psychotic Mood Disorders; MDE=Major Depressive Episode.

Table 1Demographic features of study sample. Significant variables (pb0.001) are indicated by bold font.

Total 2178 patients PMD MD χ2/T d.f. p

N=645 (29.6%) N=1533 (70.4%)

N (%)/Mean (SD) N (%)/Mean (SD)

Age 47.8 (14.0) 49.5 (14.3) −2.57 2173 0.0103Males 236 (36.6%) 470 (30.7%) 7.18 1 0.0074Civil status

Single 182 (31.2%) 356 (25.5%) 89.43 3 b0.0001Married/Living with someone 230 (39.5%) 743 (53.3%)Separated/Divorced 64 (11.0%) 210 (15.1%)Widowed 107 (18.4%) 85 (6.1%)

Children (Nr.) 1.53 (1.57) 1.66 (1.45) −1.87 2107 0.0618Education

None/Elementary 123 (19.9%) 395 (26.9%)Intermediate 275 (44.4%) 659 (44.9%) 16.76 2 0.0002High level/University 221 (35.7%) 414 (28.2%)

ProfessionStudent 22 (3.8%) 27 (1.9%)Worker/Employee 367 (63.1%) 824 (56.8%)Self employed/Management 127 (21.8%) 412 (28.4%) 16.78 4 0.0021Unemployed 41 (7.0%) 116 (8.0%)Retired 25 (4.3%) 72 (5.0%)

OriginInpatients 394 (62.5%) 824 (54.6%) 11.41 1 0.0007

Legend: PMD=Psychotic Mood Disorders; MD=non-psychotic Mood Disorders.

244 D. Souery et al. / Journal of Affective Disorders 135 (2011) 241–250

found that PMD patients had a significantly higher incidence ofpost-partum onset and of seasonal recurrence of depressiveepisodes, though this informationwas available for less than halfof the sample.

Regarding comorbidities, PMD patients seemed to have ahigher prevalence of comorbid obsessive–compulsive disor-der (OCD) together with a marginally higher prevalence ofagoraphobia. No difference was found in the rate of panicdisorder, generalized anxiety disorder (GAD), post-traumaticstress disorder (PTSD), substance abuse or dependence, andeating disorders (Table 2).

In general, the PMD group also showed a higher rate ofsomatic comorbidities (including, in our sample: arthritis orany other form of rheumatism, asthma, bronchitis, any heartdisease, hypertension, epilepsy, a history of stroke or TIA,migraine, gastric ulcer or gastritis) and thyroid disorders.However, when excluding subjects who had a lifetime historyof lithium use (n=212), no difference was found in the rateof thyroid dysfunction between groups (χ2=0.57; d.f.=1;p=0.4484).

Considering both first and second degree relatives, apositive family history for mood disorders, especially for BP,was significantly more frequent in the PMD group (Table 3).Conversely, we found no difference in terms of familiarity forsubstance use disorders, schizophrenia or schizoaffectivedisorders, anxiety disorders, or other psychiatric diagnoses(including ADHD and eating disorders), though this findingmay be biased by the limited availability of this information(n=788).

Significant demographic and clinical variables wereincorporated as predictors in a non-linear logistic regressionmodel with psychosis (PMD vs MD Group) as the outcomevariable. In our model, only a diagnosis of BP (OR=4.74;pb0.0001) was found to be significantly associated with anincreased risk of having a PMD.

3.2. Psychopathological features of a major depressive episodein PMD and MD subjects

One thousand six hundred forty-five subjects had a MDEat the time of assessment: 430 (26.1%) in the PMD group and1215 (73.9%) in the MD group. Regarding diagnosis, 1201(73.0%) subjects had a MDD, 135 (8.2%) were BP-II, and 309(18.8%) were BP-I. More than half of PMD patients (n=256)had a MDE with psychotic features at the time of assessment(Fig. 1). We analyzed symptom presentation in the threegroups of patients: PMD with psychotic depression (n=256;15.5%), PMD with non-psychotic depression (n=174; 10.6%)and MD (n=1215; 73.9%).

A higher prevalence of patients with a severe suicidal risklevel was found among PMDs, both with (n=86; 37.6%) andwithout (n=64; 40.5%) current psychotic features, whencompared to MD subjects (n=260; 28.6%) (χ2=28.18;d.f.=6; p=0.0001).

As expected, PMD patients with psychotic depressionexhibited significantly higher HAM-D total scores (F=40.91;d.f.=2.1443;pb0.0001).Adjusting foroverall symptomseverity,these patients scored significantly higher on “Core” (F=34.43;d.f.=2.1441; pb0.0001) and “Delusion” HAM-D factors(F=194.05; d.f.=2.1441; pb0.0001). However, PMD patientswith no current psychotic features exhibited the lowest scores interms of “Core” symptoms, “Sleep” disturbances (F=9.93;d.f.=2.1441; pb0.001) and “Somatic Anxiety” (F=8.47;d.f.=2.1441; p=0.0002). A similar trend was found by post-hoc analyses on “Activity” and “Psychic Anxiety” factors, wherePMDwith psychotic depression had the highest scores, followedbyMDpatients and then by PMDwith non-psychotic depression(all pb0.0001). Similarly, subjects considered to be “non-responders” to antidepressant treatment were significantlymore represented among psychotic depressed (n=155; 66.8%)and MD (n=672; 63.1%) patients, than among PMD with no

Table 2Clinical features of study sample. Significant variables (pb0.001) are indicated by bold font.

Total 2178 patients PMD MD χ2/F d.f. p

N=645 (29.6%) N=1533 (70.4%)

N (%)/Mean (SD) N (%)/Mean (SD)

DiagnosisMDD 264 (40.9%) 1188 (77.5%)BP-I 313 (48.5%) 206 (13.4%) 326.67 2 b0.0001BP-II 68 (10.5%) 139 (9.1%)

Course of illnessEarly onset (at or below 20) 248 (40.4%) 358 (24.2%) 59.98 1 b0.0001Onset (years) 27.6 (14.3) 33.9 (15.4) −8.63 2089 b0.0001Nr. of episodes (tot) 12.1 (7.4) 6.7 (6.0) 10.84 742 b0.0001Age 1st MDE (years) 28.4 (14.2) 34.2 (15.5) −8.01 2076 b0.0001Nr. of MDE 5.9 (5.0) 4.5 (5.0) 5.76 1987 b0.0001Age 1st (hypo)manic ep. (years) 29.7 (14.0) 34.4 (14.8) −3.94 574 b0.0001Nr. of (hypo)manic ep. 5.1 (4.7) 1.6 (3.4) 12.14 788 b0.0001Age 1st hosp. (years) 30.8 (11.2) 37.1 (13.3) −4.24 279 b0.0001Hosp. lifetime (weeks) 31.8 (25.7) 31.3 (26.6) 0.28 1469 0.7824Age 1st hosp. for MDE (years) 30.8 (11.0) 37.0 (13.5) −3.9 252 0.0001Hosp. for MDE (weeks) 31.1 (24.9) 29.8 (23.8) 0.98 1496 0.3290Age 1st hosp. for (hypo)manic ep. (years) 32.1 (13.8) 39.1 (12.0) −1.56 65 0.1228Hosp. for (hypo)manic ep. (weeks) 9.1 (28.7) 2.3 (8.2) 2.59 336 0.0099Pre-menstrual depressive disorder (n=853)a 69 (35.9%) 170 (25.7%) 7.70 1 0.0056Post-partum onset (n=722)a 64 (34.8%) 117 (21.8%) 12.40 1 0.0004Seasonal recurrence (n=962)a 138 (52.5%) 213 (30.5%) 39.91 1 b0.0001

Medical comorbiditiesThyroid disorder 135 (21.2%) 211 (14.0%) 16.82 1 b0.0001Onset thyroid disorder (years) 34.4 (16.3) 43.7 (16.8) −2.74 120 0.0071Diabetes 85 (13.3%) 164 (10.8%) 2.58 1 0.1081Others 373 (62.4%) 758 (54.3%) 11.02 1 0.0009Use of lithium (lifetime) 101 (18.0%) 111 (8.3%) 37.45 1 b0.0001

Psychiatric comorbiditiesb

Panic disorder current 90 (14.3%) 165 (11.4%) 3.54 1 0.0599Panic disorder lifetime 135 (21.1%) 253 (16.8%) 5.55 1 0.0185Agoraphobia 103 (25.2%) 196 (18.0%) 9.71 1 0.0018Social phobia 75 (11.7%) 138 (9.0%) 3.56 1 0.0591OCD 58 (9.1%) 74 (4.9%) 14.23 1 0.0002GAD 116 (20.5%) 201 (15.6%) 6.68 1 0.0098PTSD 37 (5.8%) 66 (4.3%) 2.08 1 0.1492Alcohol dependence 43 (6.7%) 82 (5.4%) 1.43 1 0.2310Alcohol abuse 53 (8.3%) 80 (5.3%) 7.02 1 0.0081Drug dependence 26 (4.1%) 39 (2.6%) 3.51 1 0.0611Drug abuse 16 (2.5%) 37 (2.4%) 0.01 1 0.9113AN 5 (0.9%) 6 (0.5%) 1.15 1 0.2833BN 15 (2.7%) 20 (1.6%) 2.55 1 0.1105

Legend: PMD=Psychotic Mood Disorders; MD=non-psychotic Mood Disorders; MDD=Major Depressive Disorder; BP-I=Bipolar Disorder, type I; BP-II=Bipolar Disorder, type II; MDE=Major Depressive Episode; Hosp.=Hospitalization; OCD=Obsessive–Compulsive Disorder; GAD=Generalized AnxietyDisorder; PTSD=Post-Traumatic Stress Disorder; AN=Anorexia Nervosa; BN=Bulimia Nervosa.

a This information was available only for part of the sample (number in brackets).b When it is unspecified whether it is current or lifetime it is meant as current.

245D. Souery et al. / Journal of Affective Disorders 135 (2011) 241–250

current psychotic features (n=43; 26.7%) (χ2=82.81; d.f.=2;pb0.0001). No difference was found on “Delusion” factorbetween PMD with no psychotic symptoms and MD.

Considering HAM-D single items, patients with psychoticdepression exhibited significantly higher scores on: “DepressedMood” (item1), “Feelings ofGuilt” (item2), “Retardation” (item8),“Insight” (item 17), “Depersonalization/Derealization” (item 19),“Paranoid symptoms” (item 20) and “Obsessional–Compulsivesymptoms” (item 21) (Fig. 2).

PMD patients in general, both with and without currentpsychotic features, exhibited a higher YMRS total score(F=44.24; d.f.=2.621; pb0.0001). They also scored higheron “Motor Activity” (item 2), “Thought Disorder” (item 7) and,

above all, “Irritability” (item 5) items, where the differencewas also significant between psychotic and non-psychoticPMD patients (Fig. 3). Post-hoc analyses revealed also asignificantly higher severity of PMD in general on item 1(“Elevated Mood”), item 6 (“Speech Disorder”), item 9(“Aggressive behavior”), item 10 (“Appearance”) and item 11(“Insight”) (all pb0.0001).

Comparing symptom presentation of psychotic depression(n=256) between unipolar and bipolar patients, we found aworse insight in BP. Moreover, a trend for a higher severity ofpsychotic symptoms (items 15 and 20) and suicidality (item 3)and a higher severity of sub-thresholdmanic symptoms (items6–11) were also found in BP subjects (Figs. 4 and 5).

Table 3Family history of Psychotic Mood Disorders when compared to non-psychotic Mood Disorders. Significant variables (pb0.001) are indicated by bold font.

Total 2178 patients PMD MD χ2/T d.f. p

N=645 (29.6%) N=1533 (70.4%)

N (%)/Mean (SD) N (%)/Mean (SD)

MDD 1st degree relatives 311 (52.1%) 682 (46.9%) 4.56 1 0.0327Nr. of affected members 0.62 (0.72) 0.54 (0.66) 2.74 2049 0.0063

MDD 2nd degree relatives 163 (31.4%) 299 (23.1%) 13.46 1 0.0002Nr. of affected members 0.43 (0.77) 0.28 (0.60) 4.50 1816 b0.0001

BP 1st degree relatives 124 (21.8%) 133 (9.3%) 56.44 1 b0.0001Nr. of affected members 0.23 (0.46) 0.10 (0.33) 7.24 1998 b0.0001

BP 2nd degree relatives 69 (14.1%) 58 (4.5%) 49.91 1 b0.0001Nr. of affected members 0.16 (0.45) 0.05 (0.25) 6.84 1791 b0.0001

Suicide 119 (20.3%) 232 (16.3%) 4.63 1 0.0314Nr. of members 0.25 (0.61) 0.18 (0.43) 3.12 2011 0.0018

Attempted suicide (n=661) 65 (28.3%) 94 (21.8%) 3.42 1 0.0646Nr. of members 0.39 (0.78) 0.27 (0.58) 2.24 660 0.0253

SUD (n=789) 88 (28.5%) 119 (24.8%) 1.32 1 0.2505SCF/SCA (n=788) 7 (2.3%) 22 (4.6%) 2.85 1 0.0912Anxietya (n=788) 32 (10.4%) 44 (9.2%) 0.30 1 0.5869Others (n=788) 33 (10.7%) 33 (6.9%) 3.52 1 0.0608

Legend: PMD=Psychotic Mood Disorders; MD=non-psychotic Mood Disorders; MDD=Major Depressive Disorder; BP=Bipolar Disorder; SUD=Substance UseDisorders; SCF/SCA=Schizophrenia/Schizoaffective Disorder.

a Includes panic disorder, generalized anxiety disorder and obsessive–compulsive disorder.

246 D. Souery et al. / Journal of Affective Disorders 135 (2011) 241–250

Finally, to control for potential confounders, we intro-duced each significant factor and item as the dependentvariable into a General Linear Model. Predictors were chosenamong significant variables of preliminary analyses andclinically relevant factors for the course and severity ofdepression. They included: age, gender, age at illness onset,number ofmood episodes lifetime, overall symptom severity (asmeasured by HAM-D or YMRS total scores), BP diagnosis,presence of current psychotic features, positive lifetime historyof psychotic mood episodes, presence of psychiatric comorbid-ities and presence of somatic comorbidities.

Apart from overall depressive severity, “Core” (F=25.58;pb0.0001), “Somatic Anxiety” (F=13.63; p=0.0002) and

Fig. 2. Comparison of depressive symptoms among Psychotic Mood Disorders (PMpsychotic Mood Disorders (MD). HAM-D items were adjusted for HAM-D total scor

“Delusion” factors (F=91.46; pb0.0001) appeared to be allsignificantly associated with the presence of current psy-chotic features. “Delusion” factor was also influenced by age atillness onset (F=14.21; p=0.0002), while “Activity” factorscore resulted to be influenced by the presence of a comorbidpsychiatric disorder (F=12.42; p=0.0005). HAM-D totalscore was significantly affected by the presence of currentpsychotic features (F=10.80; p=0.0011) and of somaticcomorbidities (F=11.96; pb0.0001). Similarly, YMRS totalscore resulted to be influenced by the presence of currentpsychotic features (F=11.07; p=0.0009), but not by BPdiagnosis (F=8.17; p=0.0044). No association was foundbetween any predictor and single YMRS items.

D) with and without psychotic Major Depressive Episode (MDE) and none. *pb0.05; **pb0.005; ***pb0.0005.

Fig. 3. Comparison of sub-threshold manic symptoms in Psychotic Mood Disorders (PMD) with and without psychotic Major Depressive Episode (MDE) and nonpsychotic Mood Disorders (MD). YMRS items were adjusted for HAM-D and YMRS total score. *pb0.05; **pb0.005; ***pb0.0005.

247D. Souery et al. / Journal of Affective Disorders 135 (2011) 241–250

4. Discussion

Our results revealed a number of robust differences in theclinical picture of Psychotic Mood Disorders (PMD), whencompared to affective disorders with no history of psychoticsymptoms. In linewith previous findings, a number of negative

Fig. 4. Comparison of depressive symptoms in psychotic depression between Unipototal score. *pb0.05; **pb0.005; ***pb0.0005.

prognostic factors were associated with PMD: a higherprevalence of BP-I diagnosis (Goes et al., 2007), a younger ageat the first hospitalization (Jager et al., 2005), a higher numberof mood episodes lifetime (Leyton et al., 1995; Coryell et al.,1996; Coryell et al., 2001), a younger age at illness onset (Lenziet al., 1996) and, above all, a higher prevalence of early onset

lar (UP) and Bipolar (BP) subjects. HAM-D items were adjusted for HAM-D

Fig. 5. Comparison of sub-threshold manic symptoms in psychotic depression between Unipolar (UP) and Bipolar (BP) subjects. YMRS items were adjusted forHAM-D and YMRS total score. *pb0.05; **pb0.005; ***pb0.0005.

248 D. Souery et al. / Journal of Affective Disorders 135 (2011) 241–250

(before the age of 21) of mood symptoms. Other studies hadalso found a higher number of hospitalizations (Jager et al.,2005) in PMD patients, but in our sample lifetime duration ofhospitalization was not different between groups. However, asin previous studies, we did not observe any index of relational,educational or occupational impairment related to psychosis inmooddisorders (Coryell and Tsuang, 1982; Coryell et al., 1987).

PMD patients also had a higher comorbidity rate in terms ofboth, psychiatric and somatic diseases. Agoraphobia and OCDwere found to be associated with PMD in general (Cassano et al.,1998; Cassano et al., 1999). Though a relationship with eitherpanic disorder (Chen and Dilsaver, 1995b) or OCD (Chen andDilsaver, 1995a) seemstobe substantially stronger forBP than forMDD, it has also been suggested that the higher is the cumulativecomorbidity for anxiety disorders, the higher is the risk forpsychotic symptoms in both unipolar and bipolar subjects(Dilsaver et al., 2006; Dilsaver et al., 2008).

A higher medical burden (Vythilingam et al., 2003;Rothschild, 2003) was also significantly associated with PMD,while the higher incidence of thyroid dysfunctions resulted tobe affected by the use of lithium (Bocchetta and Loviselli, 2006;Grandjean and Aubry, 2009).

Unfortunately, because of the limited availability of thisinformation, no definite conclusions can be traced about thehigher incidence of post-partum onset and seasonal recur-rence of depressive episodes in PMD. However, both of thesefeatures have been repeatedly associated with bipolarity(Dallaspezia et al., 2011; Sharma, 2006; White et al., 1990;Simonsen et al., 2011).

Actually, most of the above cited variables, more than beingconnected with psychosis per se, seem to be an expression forthe higher prevalence of PMD among BP patients (Dunayevichand Keck, 2000). A further confirmation came from our logistic

regression model, where BP diagnosis resulted to be the onlysignificant predictor for psychosis in mood disorders.

As regards the features of the depressive syndrome, asignificantly more severe suicidal risk level was found amongPMD, independently from the presence of psychotic features.Most studies on PMD were performed during delusionalepisodes,where suicidalitymightbe influencedby thepresenceof current psychotic features (Thakur et al., 1999; Gaudiano etal., 2008). However, a 25-year retrospective analysis of suicidesamong patients previously hospitalized forMDDdemonstratedthat patientswhohad delusions during their index episode hada five-fold increased likelihood of suicide compared withpatients with non-psychotic depression (Roose et al., 1983).

While psychotically depressed patients weremore severe onmost psychopathological dimensions (Glassman and Roose,1981; Frances et al., 1981; Coryell et al., 1996; Coryell et al.,1994; Coryell, 1996), PMD subjects with no current psychoticfeatures exhibited similar – or even significantly lower – scoresthan MD individuals on most HAM-D factors. Covarying fordepressive severity, they apparently had less sleep disturbances,less somatic and psychic anxiety, and lower activity impairmentthan consistently non-psychotic individuals. No difference wasfound between PMDwith no current psychotic features andMDon “Delusion” factor, suggesting that no sub-threshold psychoticsymptoms may be found outside psychotic mood episodes.

In a study by Coryell et al. (1996) it was shown that non-psychotic episodes in previously psychotic individuals weresometimes significantly briefer than non-psychotic episodes ofindividuals with no history of psychosis. For the authors thisfinding supported the view that a particularly severe episodemay be necessary for psychotic features to emerge, even inindividuals with proven diathesis for them. An alternative andintriguing hypothesis is that PMD subjectsmay have a different

249D. Souery et al. / Journal of Affective Disorders 135 (2011) 241–250

clinical presentation, and, possibly, a different response totreatment when they have a non-psychotic mood episode. Inour sample the proportion of “non-responders” to antidepres-sant treatment was significantly lower among PMD with nocurrent psychotic features than in MD individuals.

Other differences emerged in terms of manic symptoms,which scored significantly higher in both psychotic and non-psychotic PMD subjects. Indeed, PMD in general had anincreased motor activity, a more impaired thought process,and, above all, a higher irritability thanMD patients. Althoughin our sample YMRS total score resulted to be influencedmore by the psychotic nature of the episode than by BPdiagnosis, sub-threshold mixed features are probably to beascribed to the higher prevalence of BP diagnosis among PMD(Akiskal and Benazzi, 2003, Souery et al., 2011b).

One of the major limitations of this study is the lack ofinformation about the number and polarity of previouspsychotic mood episodes. It has been shown that psychoticfeatures may bemuch less predictive of future psychosis whenthey occurwithin amanic episode thanwhen they occurwithindepression (Coryell et al., 2001). Nevertheless, our firstobjective was to test the effect of psychotic features in general,with no consideration of mood polarity. Another limitation hasto deal with the use of positive or negative family history as adichotomous indicator of familial loading, as such a proceduredoes not take into account the different risk of non-affectedrelatives (Verdoux et al., 1996). Moreover, no information wasavailable about the presence of PMD among relatives. Anothertheoretic limitation to our evaluation of the depressivesyndrome could be the influence of previous pharmacotherapyon psychopathological features. Finally, the use of a rating scaleformania in a sample of depressedpatientmaybequestionable,but it should be stated that to date there are no availablevalidated instruments for evaluating the hypomanic dimensionof depression.

In conclusion, our results confirm the view that the bestpredictor of psychosis in the course of mood disorders isbipolarity. Most of the clinical features that we found todifferentiate PMD from MD patients in literature wereconsistently related to a diagnosis of BP. Secondly, psychoticdepressed patients resulted to be more severe than non-psychotic depressed, not only in the depressive but also in themanic dimension. The research and clinical implications (e.g.for treatment) of sub-threshold manic features in the courseof psychotic depression have yet to be explored. Finally, ourstudy suggests the importance for research and clinicalpractice of examining the nature and severity pattern ofnon-psychotic mood episodes in PMD patients, since theymay present with a distinctive symptom profile and, possibly,a different response to treatment.

Role of funding sourceThis workwas supported by a research grant from the National Found for

Scientific Research, Belgium—Fonds National de la Recherche Scientifique,Convention no. 3.4553.01. It was also supported by an unrestricted grantfrom Janssen-Cilag.

Conflict of interestAll the authors have no conflict of interest to declare.

AcknowledgementsNone to declare.

References

A.P.A. (Ed.), 2000. Diagnostic and Statistical Manual of Mental Disorders, 4thed. American Psychiatric Association, Washington DC. Text Revision(DSM-IV-TR).

Akiskal, H.S., Benazzi, F., 2003. Family history validation of the bipolar natureof depressive mixed states. J. Affect. Disord. 73, 113–122.

Benazzi, F., 2000. Depressive mixed states: unipolar and bipolar II. Eur. Arch.Psychiatry Clin. Neurosci. 250, 249–253.

Benazzi, F., 2003. Depressive mixed state: dimensional versus categoricaldefinitions. Prog. Neuropsychopharmacol. Biol. Psychiatry 27, 129–134.

Benazzi, F., 2009. Classifying mood disorders by age-at-onset instead ofpolarity. Prog. Neuropsychopharmacol. Biol. Psychiatry 33, 86–93.

Bocchetta, A., Loviselli, A., 2006. Lithium treatment and thyroid abnormal-ities. Clin. Pract. Epidemiol. Ment. Health 2, 23.

Brockington, I.F., Altman, E., Hillier, V., Meltzer, H.Y., Nand, S., 1982. Theclinical picture of bipolar affective disorder in its depressed phase. Areport from London and Chicago. Br. J. Psychiatry 141, 558–562.

Buchanan, R.W., Carpenter, W.T., 1994. Domains of psychopathology: anapproach to the reduction of heterogeneity in schizophrenia. J. Nerv.Ment. Dis. 182, 193–204.

Canuso, C.M., Bossie, C.A., Zhu, Y., Youssef, E., Dunner, D.L., 2008. Psychoticsymptoms in patients with bipolar mania. J. Affect. Disord. 111, 164–169.

Cassano, G.B., Pini, S., Saettoni, M., Dell'osso, L., 1999. Multiple anxietydisorder comorbidity in patients with mood spectrum disorders withpsychotic features. Am. J. Psychiatry 156, 474–476.

Cassano, G.B., Pini, S., Saettoni, M., Rucci, P., Dell'osso, L., 1998. Occurrenceand clinical correlates of psychiatric comorbidity in patients withpsychotic disorders. J. Clin. Psychiatry 59, 60–68.

Chen, Y.W., Dilsaver, S.C., 1995a. Comorbidity for obsessive–compulsivedisorder in bipolar and unipolar disorders. Psychiatry Res. 59, 57–64.

Chen, Y.W., Dilsaver, S.C., 1995b. Comorbidity of panic disorder in bipolar illness:evidence from the Epidemiologic Catchment Area Survey. Am. J. Psychiatry152, 280–282.

Cohen, J. (Ed.), 1988. Statistical Power Analysis for the Behavioral Sciences.Lawrence Erlbaum Associates, Hillsdale, New Jersey.

Committee for Medicinal Products for Human Use Chmp, 2002. Note forGuidance on Clinical Investigation of Medicinal Products in theTreatment of Depression. http://www.emea.europa.eu/pdfs/human/ewp/051897en.pdf.

Coryell, W., 1996. Psychotic depression. J. Clin. Psychiatry 57 (Suppl 3),27–31 discussion 49.

Coryell, W., Endicott, J., Keller, M., 1987. The importance of psychotic featuresto major depression: course and outcome during a 2-year follow-up.Acta Psychiatr. Scand. 75, 78–85.

Coryell, W., Lavori, P., Endicott, J., Keller, M., Vaneerdewegh, M., 1984.Outcome in schizoaffective, psychotic, and nonpsychotic depression.Course during a six- to 24-month follow-up. Arch. Gen. Psychiatry 41,787–791.

Coryell, W., Leon, A., Winokur, G., Endicott, J., Keller, M., Akiskal, H., Solomon,D., 1996. Importance of psychotic features to long-term course in majordepressive disorder. Am. J. Psychiatry 153, 483–489.

Coryell, W., Leon, A.C., Turvey, C., Akiskal, H.S., Mueller, T., Endicott, J., 2001.The significance of psychotic features in manic episodes: a report fromthe NIMH collaborative study. J. Affect. Disord. 67, 79–88.

Coryell, W., Tsuang, M.T., 1982. Primary unipolar depression and theprognostic importance of delusions. Arch. Gen. Psychiatry 39,1181–1184.

Coryell, W., Tsuang, M.T., Mcdaniel, J., 1982. Psychotic features in majordepression. Is mood congruence important? J. Affect. Disord. 4, 227–236.

Coryell, W., Winokur, G., Shea, T., Maser, J.D., Endicott, J., Akiskal, H.S., 1994.The long-term stability of depressive subtypes. Am. J. Psychiatry 151,199–204.

Craddock, N., O'donovan, M.C., Owen, M.J., 2006. Genes for schizophrenia andbipolar disorder? Implications for psychiatric nosology. Schizophr. Bull.32, 9–16.

Dallaspezia, S., Lorenzi, C., Pirovano, A., Colombo, C., Smeraldi, E., Benedetti,F., 2011. Circadian clock gene Per3 variants influence the postpartumonset of bipolar disorder. Eur. Psychiatry 26, 138–140.

Dilsaver, S.C., Akiskal, H.S., Akiskal, K.K., Benazzi, F., 2006. Dose–responserelationship between number of comorbid anxiety disorders in adoles-cent bipolar/unipolar disorders, and psychosis, suicidality, substanceabuse and familiality. J. Affect. Disord. 96, 249–258.

Dilsaver, S.C., Benazzi, F., Akiskal, K.K., Akiskal, H.S., 2008. Differentialpatterns of lifetime multiple anxiety disorder comorbidity betweenLatino adults with bipolar I and major depressive disorders. Bull.Menninger Clin. 72, 130–148.

Double, D.B., 1990. The factor structure of manic rating scales. J. Affect.Disord. 18, 113–119.

250 D. Souery et al. / Journal of Affective Disorders 135 (2011) 241–250

Dunayevich, E., Keck Jr., P.E., 2000. Prevalence and description of psychoticfeatures in bipolar mania. Curr. Psychiatry Rep. 2, 286–290.

Endicott, J., Nee, J., Andreasen, N., Clayton, P., Keller, M., Coryell, W., 1985.Bipolar II. Combine or keep separate? J. Affect. Disord. 8, 17–28.

Faul, F., Erdfelder, E., Lang, A.G., Buchner, A., 2007. G*Power 3: a flexiblestatistical power analysis program for the social, behavioral, andbiomedical sciences. Behav. Res. Methods 39, 175–191.

Fava, M., 2003. Diagnosis and definition of treatment-resistant depression.Biol. Psychiatry 53, 649–659.

Feinstein, A.R. (Ed.), 1977. A Critical Overview of Diagnosis in Psychiatry.Brunner, New York.

Fisfalen, M.E., Schulze, T.G., Depaulo Jr., J.R., Degroot, L.J., Badner, J.A., Mcmahon,F.J., 2005. Familial variation in episode frequency in bipolar affectivedisorder. Am. J. Psychiatry 162, 1266–1272.

Flaum, M., O'leary, D.S., Swayze 2nd, V.W., Miller, D.D., Arndt, S., Andreasen,N.C., 1995. Symptom dimensions and brainmorphology in schizophreniaand related psychotic disorders. J. Psychiatr. Res. 29, 261–276.

Frances, A., Brown, R.P., Kocsis, J.H., Mann, J.J., 1981. Psychotic depression: aseparate entity? Am. J. Psychiatry 138, 831–833.

Gaudiano, B.A., Young, D., Chelminski, I., Zimmerman, M., 2008. Depressivesymptom profiles and severity patterns in outpatients with psychotic vsnonpsychotic major depression. Compr. Psychiatry 49, 421–429.

Glahn,D.C., Bearden, C.E., Cakir, S., Barrett, J.A., Najt, P., SerapMonkul, E.,Maples,N., Velligan, D.I., Soares, J.C., 2006. Differential working memory impair-ment in bipolar disorder and schizophrenia: effects of lifetime history ofpsychosis. Bipolar Disord. 8, 117–123.

Glassman, A.H., Roose, S.P., 1981. Delusional depression. A distinct clinicalentity? Arch. Gen. Psychiatry 38, 424–427.

Goes, F.S., Sadler, B., Toolan, J., Zamoiski, R.D., Mondimore, F.M., Mackinnon,D.F., Schweizer, B., Raymond Depaulo Jr., J., Potash, J.B., 2007. Psychoticfeatures in bipolar and unipolar depression. Bipolar Disord. 9, 901–906.

Goes, F.S., Sanders, L.L., Potash, J.B., 2008. The genetics of psychotic bipolardisorder. Curr. Psychiatry Rep. 10, 178–189.

Grandjean, E.M.,Aubry, J.M., 2009. Lithium:updatedhumanknowledgeusing anevidence-based approach: part III: clinical safety. CNS Drugs 23, 397–418.

Grant, M.M., Thase, M.E., Sweeney, J.A., 2001. Cognitive disturbance inoutpatient depressed younger adults: evidence of modest impairment.Biol. Psychiatry 50, 35–43.

Guze, S.B., Woodruff Jr., R.A., Clayton, P.J., 1975. The significance of psychoticaffective disorders. Arch Gen. Psychiatry 32, 1147–1150.

Hamilton, M., 1960. A rating scale for depression. J. Neurol. Neurosurg.Psychiatry 23, 56–62.

Haro, J.M., Van Os, J., Vieta, E., Reed, C., Lorenzo, M., Goetz, I., 2006. Evidencefor three distinct classes of ‘typical’, ‘psychotic’ and ‘dual’ mania: resultsfrom the EMBLEM study. Acta Psychiatr. Scand. 113, 112–120.

Harrow, M., Grossman, L.S., Herbener, E.S., Davies, E.W., 2000. Ten-yearoutcome: patients with schizoaffective disorders, schizophrenia, affectivedisorders and mood-incongruent psychotic symptoms. Br. J. Psychiatry177, 421–426.

Henry, C., Etain, B., 2010. New ways to classify bipolar disorders: going fromcategorical groups to symptom clusters or dimensions. Curr. PsychiatryRep. 12, 505–511.

Hill, S.K., Keshavan, M.S., Thase, M.E., Sweeney, J.A., 2004. Neuropsychologicaldysfunction in antipsychotic-naive first-episode unipolar psychotic de-pression. Am. J. Psychiatry 161, 996–1003.

Ivleva, E.I., Morris, D.W., Moates, A.F., Suppes, T., Thaker, G.K., Tamminga,C.A., 2010. Genetics and intermediate phenotypes of the schizophrenia–bipolar disorder boundary. Neurosci. Biobehav. Rev. 34, 897–921.

Jager, M., Bottlender, R., Strauss, A., Moller, H.J., 2005. Fifteen-year follow-upof Diagnostic and Statistical Manual of Mental Disorders, Fourth Editiondepressive disorders: the prognostic significance of psychotic features.Compr. Psychiatry 46, 322–327.

Jeste,D.V.,Heaton,S.C., Paulsen, J.S., Ercoli, L.,Harris, J., Heaton,R.K., 1996.Clinicaland neuropsychological comparison of psychotic depression with non-psychotic depression and schizophrenia. Am. J. Psychiatry 153, 490–496.

Koukopoulos, A., Koukopoulos, A., 1999. Agitated depression as a mixed stateand the problem of melancholia. Psychiatr. Clin. North Am. 22, 547–564.

Kraepelin, E. (Ed.), 1921. Manic-Depressive Insanity and Paranoia. E and SLivingstone, Edinburgh.

Lattuada, E., Serretti, A., Cusin, C., Gasperini, M., Smeraldi, E., 1999.Symptomatologic analysis of psychotic and non-psychotic depression.J. Affect. Disord. 54, 183–187.

Lenzi, A., Rinaldi, A., Bianco, I., Balestri, C., Marazziti, D., 1996. Psychoticsymptoms inmood disorders: evaluation of 159 inpatients. Eur. Psychiatry11, 396–399.

Leyton, M., Corin, E., Martial, J., Meaney, M., 1995. Psychotic symptoms andvulnerability to recurrent major depression. J. Affect. Disord. 33, 107–115.

Maj, M., Pirozzi, R., Di Caprio, E.L., 1990. Major depression with mood-congruent psychotic features: a distinct diagnostic entity or a moresevere subtype of depression? Acta Psychiatr. Scand. 82, 439–444.

Mitchell, P.B., Wilhelm, K., Parker, G., Austin, M.P., Rutgers, P., Malhi, G.S., 2001.The clinical features of bipolar depression: a comparison with matchedmajor depressivedisorder patients. J. Clin. Psychiatry 62, 212–216 quiz 217.

Murray, R.M., Sham, P., Van Os, J., Zanelli, J., Cannon, M., Mcdonald, C., 2004. Adevelopmental model for similarities and dissimilarities betweenschizophrenia and bipolar disorder. Schizophr. Res. 71, 405–416.

Parker, G., Hadzi-Pavlovic, D., Hickie, I., Boyce, P.,Mitchell, P.,Wilhelm, K., Brodaty,H., 1991. Distinguishing psychotic and non-psychotic melancholia. J. Affect.Disord. 22, 135–148.

Parker, G., Roy, K., Wilhelm, K., Mitchell, P., Hadzi-Pavlovic, D., 2000. Thenature of bipolar depression: implications for the definition ofmelancholia. J. Affect. Disord. 59, 217–224.

Pope Jr., H.G., Lipinski Jr., J.F., 1978. Diagnosis in schizophrenia and manic-depressive illness: a reassessment of the specificity of ‘schizophrenic’symptoms in the lightof current research.Arch.Gen.Psychiatry 35, 811–828.

Potash, J.B., Toolan, J., Steele, J.,Miller, E.B., Pearl, J., Zandi, P.P., Schulze, T.G., Kassem,L., Simpson, S.G., Lopez, V., Mackinnon, D.F., Mcmahon, F.J., 2007. The bipolardisorder phenome database: a resource for genetic studies. Am. J. Psychiatry164, 1229–1237.

Robins, E., Guze, S.B., 1970. Establishment of diagnostic validity in psychiatricillness: its application to schizophrenia. Am. J. Psychiatry 126, 983–987.

Roose, S.P., Glassman, A.H., Walsh, B.T., Woodring, S., Vital-Herne, J., 1983.Depression, delusions, and suicide. Am. J. Psychiatry 140, 1159–1162.

Rothschild, A.J., 2003. Challenges in the treatment of depression withpsychotic features. Biol. Psychiatry 53, 680–690.

Serretti, A., Jori, M.C., Casadei, G., Ravizza, L., Smeraldi, E., Akiskal, H., 1999.Delineating psychopathologic clusters within dysthymia: a study of 512out-patients without major depression. J. Affect. Disord. 56, 17–25.

Serretti, A., Lattuada, E., Cusin, C., Macciardi, F., Smeraldi, E., 1998. Analysis ofdepressive symptomatology in mood disorders. Depress Anxiety 8, 80–85.

Serretti, A., Lattuada, E., Lorenzi, C., Lilli, R., Smeraldi, E., 2000. Dopaminereceptor D2 Ser/Cys 311 variant is associated with delusion anddisorganization symptomatology in major psychoses. Mol. Psychiatry5, 270–274.

Sharma, V., 2006. A cautionary note on the use of antidepressants inpostpartum depression. Bipolar Disord. 8, 411–414.

Sheehan, D.V., Lecrubier, Y., Sheehan, K.H., Amorim, P., Janavs, J., Weiller,E., Hergueta, T., Baker, R., Dunbar, G.C., 1998. The Mini-InternationalNeuropsychiatric Interview (M.I.N.I.): the development and validationof a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J. Clin. Psychiatry 59 (Suppl 20), 22–33 quiz 34–57.

Simonsen, H., Shand, A.J., Scott, N.W., Eagles, J.M., 2011. Seasonal symptomsin bipolar and primary care patients. J Affect Disord. 132 (1–2), 200–208.

Sobin, C., Sackeim, H.A., 1997. Psychomotor symptoms of depression. Am.J. Psychiatry 154, 4–17.

Souery, D., Amsterdam, J., De Montigny, C., Lecrubier, Y., Montgomery, S., Lipp,O., Racagni, G., Zohar, J., Mendlewicz, J., 1999. Treatment resistantdepression: methodological overview and operational criteria. Eur.Neuropsychopharmacol. 9, 83–91.

Souery, D., Oswald, P.,Massat, I., Bailer, U., Bollen, J., Demyttenaere,K., Kasper, S.,Lecrubier, Y., Montgomery, S., Serretti, A., Zohar, J., Mendlewicz, J., 2007.Clinical factors associated with treatment resistance in major depressivedisorder: results from a European multicenter study. J Clin Psychiatry 68,1062–1070.

Souery, D., Serretti, A., Calati, R., Oswald, P.,Massat, I., Konstantinidis, A., Linotte,S., Bollen, J., Demyttenaere, K., Kasper, S., Lecrubier, Y., Montgomery, S.,Zohar, J., Mendlewicz, J., 2011a. Switching antidepressant class does notimprove response or remission in treatment resistant depression. J. Clin.Psychopharmacol. 31, 512–516.

Souery, D., Zaninotto, L., Calati, R., Linotte, S.,Mendlewicz, J., Sentissi, O., Serretti,A., 2011b. Depression across mood disorders: review and analysis in aclinical sample. Compr Psychiatry.

Statsoft, I. (Ed.), 1995. STATISTICA per Windows.Tafalla, M., Sanchez-Moreno, J., Diez, T., Vieta, E., 2009. Screening for bipolar

disorder in a Spanish sample of outpatients with current majordepressive episode. J. Affect. Disord. 114, 299–304.

Thakur, M., Hays, J., Krishnan, K.R., 1999. Clinical, demographic and socialcharacteristics of psychotic depression. Psychiatry Res. 86, 99–106.

Thase, M.E., 2001. The need for clinically relevant research on treatment-resistant depression. J. Clin. Psychiatry 62, 221–224.

Verdoux, H., Van Os, J., Sham, P., Jones, P., Gilvarry, K., Murray, R., 1996. Doesfamiliality predispose to both emergence and persistence of psychosis? Afollow-up study. Br. J. Psychiatry 168, 620–626.

Vythilingam, M., Chen, J., Bremner, J.D., Mazure, C.M., Maciejewski, P.K.,Nelson, J.C., 2003. Psychotic depression and mortality. Am. J. Psychiatry160, 574–576.

White, D.M., Lewy, A.J., Sack, R.L., Blood, M.L., Wesche, D.L., 1990. Is winterdepression a bipolar disorder? Compr. Psychiatry 31, 196–204.

Young, R.C., Biggs, J.T., Ziegler, V.E., Meyer, D.A., 1978. A rating scale formania: reliability, validity and sensitivity. Br. J. Psychiatry 133, 429–435.