May 2013  << Back  

  Can't open the PDF? Click here for help.

 

 RESEARCH ARTICLE

Treatment resistance in severe unipolar depression: No association with psychotic or melancholic features

Leonardo Zaninotto, MD

Institute of Psychiatry, University of Bologna, Bologna, Italy, Department of Psychiatry, Catholic University of Sacred Heart, Rome, Italy

Daniel Souery, MD

Laboratoire de Psychologie Medicale, Université Libre de Bruxelles, Brussels, Belgium, Psy Pluriel, Centre Européen de Psychologie Medicale, Brussels, Belgium

Raffaella Calati, PsyD, PhD

I.R.C.C.S. Centro San Giovanni di Dio, Fatebenefratelli, Brescia, Italy

Othman Sentissi, MD, PhD

Département de Psychiatrie, Hôpitaux Universitaires de Genève, Faculté de Médecine de Genève, Geneva, Switzerland

Siegfried Kasper, MD

Department of Psychiatry and Psychotherapy, Medical University of Vienna, Vienna, Austria

Elena Akimova, MD

Department of Psychiatry and Psychotherapy, Medical University of Vienna, Vienna, Austria

Joseph Zohar, MD

Chaim Sheba Medical Center, Tel-Hashomer, Israel

Stuart Montgomery, MD

Imperial College School of Medicine, London, United Kingdom

Julien Mendlewicz, MD, PhD

Université Libre de Bruxelles, Brussels, Belgium

Alessandro Serretti, MD, PhD

Institute of Psychiatry, University of Bologna, Bologna, Italy

BACKGROUND: Depressive subtypes generally have been neglected in research on treatment efficacy. We studied a sample of 699 severe unipolar depressed patients to detect any association between depressive features and treatment resistance.

METHODS: Participants were divided into psychotic (PSY, n = 90), melancholic (MEL, n = 430) and non-melancholic (n = 179) subjects according to clinical features. Formal diagnostic criteria (Mini International Neuropsychiatric Interview items), and items from 17-item Hamilton Rating Scale for Depression (HRSD17) were compared across groups. Non-responders were defined by a HRSD17 cut-off score of ≥17 after the last adequate antidepressant treatment. Treatment-resistant depression (TRD) was defined as the failure to respond to ≥2 adequate antidepressant trials. Non-linear regression models were designed to detect associations between depressive subtypes and TRD.

RESULTS: PSY and MEL patients appeared to be more severely affected and to share some “core” melancholic symptoms. Both PSY and MEL patients reported a higher rate of seasonality. However, we found no clinical or illness course variable associated with TRD.

CONCLUSIONS: Our results indicate that psychotic and melancholic depression share some “core” melancholia symptoms, while no distinguishing psychopathological feature appears to be associated with TRD in severely depressed patients.

KEYWORDS: major depression, antidepressant, resistant, psychotic, melancholic

ANNALS OF CLINICAL PSYCHIATRY 2013;25(2):97-106

  INTRODUCTION

Since development of DSM-III,1 diagnostic manuals implicitly have supported a unitary nosographic model of depressive syndromes,2-5 according to which depressive features are considered more as a function of severity than as distinct subtypes.6

However, an alternative view7-9 sees melancholic and psychotic depression as distinct clinical entities, characterized by different psychopathological features,8,10-15 biological markers,16-24 and, above all, response to treatment.25-28

Significant empirical evidence suggests that psychotic depression is marked by poorer short-term outcome29,30 and by treatment resistance.28,31-33 Conversely, although a better response to treatment traditionally has been documented in melancholia,34-36 the impact of this pattern on treatment response has been debated.37-39 Previous research by our group reported a significant link between melancholic features and treatment resistance,40 but a subsequent study on the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) sample did not.39

Notwithstanding, researchers generally have neglected the role of depressive subtypes in treatment efficacy.16 After the unitary nosographic model was adopted, the heterogeneity of cases enrolled in most research trials increased,41 and the drug/placebo response rate decreased, leading to questions about the efficacy of today’s most widely used antidepressants.42,43

To address this issue, we examined a sample of severely depressed patients to determine the role of depressive subtypes in treatment resistance. The aims of the present study were to: 1) detect any difference across 3 groups of severely depressed patients, presenting psychotic (PSY), melancholic (MEL) and non-melancholic (NMEL) features, respectively; and 2) determine whether depressive subtypes might affect treatment resistance in severe unipolar depression.

  METHODS

Sample selection

Participants (N = 955) were recruited from January 2000 to February 2004 as part of the European multicenter project “Patterns of treatment resistance and switching strategies in unipolar affective disorder.” Inclusion criteria, recruiting centers, and study sample are described in detail elsewhere.40 The ethics committees of all participating centers approved the study protocol, and all participants gave their written informed consent.

Patients were included if they: 1) met diagnostic criteria for a primary (ie, not secondary to another Axis I disorder or to a medical condition) major depressive disorder according to DSM-IV,44 and 2) had been treated for their current major depressive episode (MDE) with ≥1 antidepressant trial of adequate dose and duration (≥4 weeks). Patients were treated with standard-dose antidepressants and the adequacy of doses for each treatment was defined according to licensed dose range (TABLE 1).


TABLE 1

Classes of antidepressants used for the current major depressive episode

SSRIs Citalopram (≥20 mg)
Escitalopram (≥10 mg)
Fluoxetine (≥20 mg)
Fluvoxamine (≥50 mg)
Paroxetine (≥20 mg)
Sertraline (≥50 mg)
TCAs Amitriptyline (≥150 mg)
Clomipramine (≥150 mg)
Dosulepin (≥200 mg)
Dothiepin (≥200 mg)
Imipramine (≥150 mg)
Nortriptyline (≥100 mg)
Protriptyline (≥75 mg)
Trimipramine (≥50 mg)
MAOIs Phenelzine (≥60 mg)
Moclobemide (≥450 mg)
Tranylcypromine (≥30 mg)
SNRIs Venlafaxine (≥75 mg)
Milnacipran (≥100 mg)
Others Amoxapine (≥300 mg)
Desipramine (≥150 mg)
Maprotiline (≥150 mg)
Mianserin (≥30 mg)
Mirtazapine (≥15 mg)
Reboxetine (≥8 mg)
Tianeptine (≥37.5 mg)
Trazodone (≥150 mg)
Viloxazine (≥200 mg)
In parentheses: lowest dose defined as effective in the product datasheet.
MAOIs: monoamine oxidase inhibitors; SNRIs: serotonin-norepinephrine reuptake inhibitors; SSRIs: selective serotonin reuptake inhibitors; TCAs: tricyclic antidepressants.

Two researchers independently checked the case-by-case inclusion criteria. A detailed checklist of antidepressant treatments was completed for each participant, including information about type, dose, and duration of antidepressant, and the sequence of antidepressants received for the current MDE in case of multiple trials. We only recorded previous trials for which we confidently could confirm compliance. We also registered co-treatment with mood stabilizers, lithium, antipsychotics, sedatives, and electroconvulsive therapy during the episode and included this in our analyses. Although being retrospectively assessed, data regarding treatment for the current MDE were considered reliable.45

Current and lifetime diagnoses, suicidal risk level, psychopathological features (psychotic or melancholic), and severity of the episode all were assessed by a modified version of the Mini International Neuropsychiatric Interview (MINI) version 5.0.0.46 The MINI is a brief, structured interview designed to identify the major Axis I psychiatric disorders according to DSM-IV diagnostic criteria.44

For the present study, only patients meeting DSM-IV criteria (MINI interview, module “A II-Criteria for Severity for Current Major Depressive Episode”) for a severe MDE at intake were included (n = 699). We divided the sample into 3 groups according to psychopathological features of the episode: patients with psychotic features reported ≥1 psychotic symptom (ie, delusions or hallucinations), exclusively restricted to mood episodes. Nonpsychotic patients were defined as melancholic or non-melancholic by the number of melancholia symptoms reported (≥4), including anhedonia, lack of reactivity to pleasant stimuli, distinctive quality of mood, diurnal mood variations, late insomnia, psychomotor disturbances, anorexia or weight loss, and pathologic feelings of guilt.

We also evaluated each patient using a questionnaire investigating psychiatric and somatic comorbidities, as well as personal and family history of psychiatric disorders. The 17-item Hamilton Rating Scale for Depression (HRSD17)47 was administered to all participants in the post-treatment phase, ≥4 weeks after the last adequate antidepressant trial. Non-responders were defined by a HRSD17 cut-off score of 17 after ≥1 antidepressant treatment of adequate dose (TABLE 1) and duration (4 weeks). Treatment-resistant depression (TRD) was defined as the failure (HRSD17 score ≥17) to respond to ≥2 adequate antidepressant trials.40,48 Those who made diagnostic assessment were blind to treatment response status (ie, TRD vs responder)

Statistical analyses

Data were analyzed using STATISTICA software, version 8.0.49 Chi-square test (with Fisher exact correction, when necessary), analysis of variance (ANOVA), and analysis of covariance (ANCOVA) were applied to categorical and continuous variables, respectively. All P values were 2-tailed, and statistical significance was set at .01.

Univariate analyses were performed to provide a preliminary screening of appropriate candidate independent variables for some non-linear logistic regression models, designed to detect any association between depressive subtypes and treatment resistance. A Bonferroni corrected a was adopted to adjust for multiple comparisons.

  RESULTS

Clinical and psychopathological features across depressive subtypes

We found no difference regarding socio-demographic features across depressive subtypes (TABLE 2). PSY patients showed a higher lifetime prevalence of psychotic mood episodes, and both PSY and MEL groups reported a higher prevalence of seasonal recurrences and premenstrual dysphoric disorder (PMDD). Otherwise, we found no other differences in illness course, family history, or comorbidities, except for a lower prevalence of Axis II disorders in MELs (TABLE 2).


TABLE 2

Sociodemographic and clinical features across depressive subtypes

N = 699 MDD subjects
(% column)
Total study sample PSY
90 (12.9%)
MEL
430 (61.5%)
NMEL
179 (25.6%)
χ2/F df P
Mean age (SD) 49.9 (14.5) 48.8 (14.6) 49.8 (14.6) 50.5 (14.1) 0.40 2.696 .6719
Males 171 (24.5%) 25 (27.8%) 102 (23.7%) 44 (24.6%) 0.66 2 .7173
White 679 (97.1%) 87 (96.7%) 422 (98.1%) 170 (95.0%) 4.65 2 .0980
Marital status (n = 688)
  Single 99 (14.4%) 10 (11.6%) 61 (14.4%) 28 (15.8%)      
  Married or cohabiting 423 (61.5%) 59 (68.6%) 264 (62.1%) 100 (56.5%) 4.49 6 .6105
  Separated or divorced 102 (14.8%) 11 (12.8%) 59 (13.9%) 32 (18.1%)      
  Widowed 64 (9.3%) 6 (7.0%) 41 (9.7%) 17 (9.6%)      
Education (n = 650)
  Below compulsory 67 (10.3%) 3 (3.8%) 43 (10.5%) 21 (13.0%)      
  Compulsory education 164 (25.2%) 17 (21.3%) 107 (26.2%) 40 (24.7%) 6.86 6 .3344
  Secondary school 280 (43.1%) 39 (48.8%) 173 (42.4%) 68 (42.0%)      
  University 139 (21.4%) 21 (26.3%) 85 (20.8%) 33 (20.4%)      
Occupationa (n = 680)
  Category 1 101 (14.9%) 14 (16.3%) 56 (13.4%) 31 (17.6%)      
  Category 2 223 (32.8%) 26 (30.2%) 133 (31.8%) 64 (36.4%) 4.44 4 .3503
  Category 3 356 (52.4%) 46 (53.5%) 229 (54.8%) 81 (46.0%)      
Path to care (n = 672)
  GP 155 (23.1%) 22 (24.7%) 80 (19.2%) 53 (31.9%)      
  Outpatient 148 (22.0%) 12 (13.5%) 87 (20.9%) 49 (29.5%) 27.30 6 .0001
  By himself 199 (29.6%) 32 (36.0%) 132 (31.7%) 35 (21.1%)      
  Otherb 170 (25.3%) 23 (25.8%) 118 (28.3%) 29 (17.5%)      
Inpatients 540 (77.3%) 79 (87.8%) 338 (78.6%) 123 (68.7%) 13.55 2 .0011
Illness course
  Recurrent MDD (n = 610) 220 (36.1%) 26 (32.5%) 131 (35.2%) 63 (39.9%) 1.55 2 .4604
  MDE LT (nr) (n = 648) 4.0 (5.4) 3.7 (3.0) 4.0 (5.3) 4.0 (6.5) 0.13 2.645 .8754
  Age 1st MDE (n = 676) 37.2 (16.0) 33.9 (14.7) 37.6 (16.1) 38.0 (16.4) 2.13 2.673 .1202
  Age 1st Hosp. (n = 489) 43.6 (15.4) 41.3 (14.4) 43.5 (15.6) 45.3 (15.5) 1.40 2.486 .2482
  Age 1st AD treatment (n = 666) 41.0 (22.1) 37.7 (14.4) 41.3 (25.3) 42.1 (15.8) 1.20 2.663 .3025
  Hosp. LT (nr) (n = 545) 2.8 (3.9) 2.7 (1.8) 3.1 (4.8) 2.1 (1.4) 2.94 2.542 .0535
  Hosp. LT (weeks) (n = 411) 16.3 (30.9) 20.7 (39.5) 16.0 (27.8) 14.7 (32.6) 0.74 2.408 .4780
  Psychotic features LT (n = 695) 62 (8.9%) 44 (49.4%) 13 (3.0%) 5 (2.8%) 206.24 2 <.0001
  Seasonal pattern (n = 450) 95 (21.1%) 16 (28.6%) 65 (23.7%) 14 (11.7%) 9.42 2 .0090
  PMDDc (n = 489) 131 (26.8%) 20 (32.3%) 91 (29.8%) 20 (16.4%) 9.11 2 .0105
  Postpartum episodec (n = 364) 71 (19.5%) 14 (31.8%) 41 (18.9%) 16 (15.5%) 5.33 2 .0694
  History of SA (n = 686) 221 (30.8%) 30 (33.3%) 128 (30.4%) 53 (30.3%) 0.32 2 .8507
Comorbidities
Anxiety disorders 266 (38.1%) 38 (42.2%) 162 (37.7%) 66 (36.9%) 0.80 2 .6717
  PD (n = 648) 146 (22.5%) 23 (26.1%) 92 (23.4%) 31 (18.7%) 2.22 2 .3294
  SP 71 (10.2%) 13 (14.4%) 46 (10.7%) 12 (6.7%) 4.29 2 .1171
  OCD (n = 697) 29 (4.2%) 2 (2.2%) 18 (4.2%) 9 (5.1%) 1.21 2 .5468
  PTSD 38 (5.4%) 6 (6.7%) 20 (4.7%) 11 (6.7%) 1.34 2 .5117
  GAD (n = 501) 58 (11.6%) 8 (14.3%) 36 (10.7%) 14 (12.8%) 0.82 2 .6648
Substance use disorders 75 (10.7%) 11 (12.2%) 43 (10.0%) 21 (11.7%) 0.64 2 .7276
  Alcohol abuse or dependence 52 (7.4%) 7 (7.8%) 26 (6.1%) 19 (10.6%) 3.85 2 .1461
  Drug abuse or dependence 31 (4.4%) 5 (5.6%) 19 (4.4%) 7 (3.9%) 0.38 2 .8257
Axis II disorders (n = 243) 84 (34.6%) 14 (41.2%) 48 (28.7%) 22 (52.4%) 9.05 2 .0108
Somatic disorders (n = 593) 319 (53.8%) 35 (49.3%) 194 (52.2%) 90 (60.0%) 3.31 2 .1914
  Diabetes (n = 694) 65 (9.4%) 6 (6.7%) 39 (9.1%) 20 (11.5%) 1.75 2 .4178
  Thyroid disorders (n = 688) 98 (14.2%) 13 (14.4%) 58 (13.6%) 27 (15.7%) 0.44 2 .8031
Family history
  MDD (n = 648) 329 (50.8%) 53 (62.4%) 196 (48.4%) 80 (50.6%) 5.48 2 .0647
  BP (n = 611) 42 (6.9%) 4 (5.0%) 27 (7.1%) 11 (7.2%) 0.51 2 .7767
  Suicide (n = 671) 105 (15.6%) 16 (18.4%) 64 (15.4%) 25 (14.8%) 0.61 2 .7388
aOccupation was categorized according to Hollingshead’s Two-Factor Index of Social Position: category 1: proprietors of large and medium-sized concerns, major professionals, business managers in large concerns; category 2: administrative personnel, owners of small independent businesses, minor professionals, clerical and sales workers; category 3: skilled/semi-skilled/unskilled manual employees, machine operators, farmers.
bPrivate psychiatrists; transfer from other clinics.
cOnly female patients.
BP: bipolar disorder; GAD: generalized anxiety disorder; GP: general practitioner; LT: lifetime; MDD: major depressive disorder; MDE: major depressive episode; MEL: melancholic; NMEL: non-melancholic; OCD: obsessive-compulsive disorder; PD: panic disorder; PMDD: premenstrual dysphoric disorder; PSY: psychotic; PTSD: posttraumatic stress disorder; SA: suicide attempts; SD: standard deviation; SP: social phobia.

TABLE 3 describes psychopathological features of the current MDE. A higher prevalence of suicide risk was detected in psychotic patients. Both PSY and MEL groups reported a higher—and almost comparable—prevalence of the classical symptoms of melancholia. HRSD17 items’ scores—adjusted for depressive severity—also were compared across groups (FIGURE). Differences were found on items 2 (feelings of guilt; F = 64.48; P < .0001), 5 (middle insomnia; F = 10.89; P < .0001), 6 (late insomnia; F = 4.96; P = .0073), 11 (somatic anxiety; F = 10.64; P < .0001), 13 (somatic symptoms in general; F = 10.56; P < .0001), and 16 (weight loss; F = 4.59; P = .0104).


TABLE 3

Psychopathological features and treatment adopted across depressive subtypes

n = 699 MDD subjects
(% column)
PSY
90 (12.9%)
MEL
430 (61.5%)
NMEL
179 (25.6%)
χ2/F df P
  Abrupt onset (n = 682) 33 (37.5%) 141 (33.5%) 44 (25.4%) 5.08 2 .0787
  Duration (days) (n = 640) 194.1 (195.1) 210.8 (203.7) 253.0 (252.8) 2.87 2.637 .0576
  Dysthymia (n = 606) 7 (10.0%) 37 (9.7%) 19 (12.4%) 0.91 2 .6358
Symptoms of melancholia
  Anhedonia (n = 603) 81 (94.2%) 373 (98.9%) 120 (85.7%) 39.22 2 <.0001
  Lack of reactivity (n = 595) 79 (90.8%) 339 (91.4%) 83 (60.6%) 74.65 2 <.0001
  Distinctive quality of mood (n = 638) 58 (69.1%) 348 (85.1%) 51 (35.2%) 131.56 2 <.0001
  Diurnal mood variations (n = 639) 47 (56.6%) 268 (65.2%) 31 (21.4%) 83.16 2 <.0001
  Late insomnia (n = 636) 44 (52.4%) 261 (64.0%) 18 (12.5%) 112.91 2 <.0001
  Psychomotor disturbances (n = 638) 70 (83.3%) 369 (90.0%) 69 (47.9%) 117.15 2 <.0001
  Anorexia or weight loss (n = 595) 56 (68.3%) 302 (79.3%) 47 (35.6%) 85.97 2 <.0001
  Pathologic guilt feelings (n = 635) 57 (67.9%) 222 (54.4%) 14 (9.8%) 103.21 2 <.0001
HRSD17 total score 24.4 (6.4) 20.1 (6.7) 17.7 (6.7) 29.91 2.696 <.0001
Suicidal risk (MINI items) (n = 691) 73 (81.1%) 285 (66.9%) 106 (60.6%) 11.40 2 .0034
Level of risk (n = 682)
  Low 27 (37.0%) 114 (40.7%) 37 (36.3%)      
  Moderate 13 (17.8%) 68 (24.3%) 27 (26.5%) 3.54 4 .4719
  High 33 (45.2%) 98 (35.0%) 38 (37.3%)      
Treatment
  Delayed treatment (days) (n = 403) 64.1 (137.3) 66.4 (111.4) 189.5 (532.2) 6.93 2.400 .0011
  Nr. treatments 2.4 (1.6) 2.1 (1.3) 2.0 (1.1) 2.92 2.696 .0547
  Antidepressants 89 (98.9%) 421 (97.9%) 177 (98.9%) 0.94 2 .6257
  Mood stabilizers 8 (8.9%) 26 (6.1%) 14 (7.8%) 1.28 2 .5266
  Lithium 1 (1.1%) 7 (1.6%) 3 (1.7%) 0.14 2 .9303
  CMZ (n = 546) 3 (4.1%) 6 (1.8%) 5 (3.8%) 2.42 2 .2985
  VP (n = 545) 2 (2.7%) 5 (1.5%) 2 (1.5%) 0.62 2 .7341
  Antipsychotics 37 (41.1%) 90 (20.9%) 32 (17.9%) 20.50 2 <.0001
  ECT 3 (3.3%) 19 (4.4%) 7 (3.9%) 0.26 2 .8804
  AD of a different class (n = 411) 44 (84.6%) 199 (76.8%) 75 (75.0%) 1.92 2 .3824
  Psychotherapy 19 (21.8%) 87 (21.2%) 24 (14.6%) 3.51 2 .1725
Treatment response
  Non-responders 74 (82.2%) 308 (71.6%) 105 (58.7%) 17.76 2 .0001
  TRD (n = 389) 34 (82.9%) 134 (57.3%) 44 (38.6%) 25.71 2 <.0001
Response to 1st AD treatmenta (n = 655)
  Good 49 (56.3%) 200 (49.9%) 73 (43.7%)      
  Partial 25 (28.7%) 140 (34.9%) 65 (38.9%) 4.05 4 .3998
  No response 13 (14.9%) 61 (15.2%) 29 (17.4%)      
aAt the first MDE.
AD: antidepressants; CMZ: carbamazepine; ECT: electroconvulsive therapy; HRSD17: 17-item Hamilton Rating Scale for Depression; MDD: major depressive disorder; MDE: major depressive episode; MEL: melancholic; MINI: Mini International Neuropsychiatric Interview; NMEL: non-melancholic; PSY: psychotic; TRD: treatment-resistant depression; VP: valproate.

FIGURE: Least squares mean scores from individual baseline HRSD17 items, adjusted for total HRSD17 score
aP < .0001.
bP < .01.
Scheffé’s post-hoc tests reported the following results: Depressed Mood: PSY > MEL (P < .0001) and MEL > NMEL (P = .0061); Guilt: PSY > MEL (P < .0001) and MEL > NMEL (P < .0001); Suicide: PSY > MEL (P = .0019) and MEL > NMEL (P = .0002); Insomnia Late: PSY > NMEL (P = .0039) and MEL > NMEL (P < .0001); Activities: PSY > NMEL (P < .0001) and MEL > NMEL (P = .0003); Retardation: PSY > MEL (P < .0001) and PSY > NMEL (P < .0001); Psychic Anxiety: PSY > MEL (P < .0001) and PSY > NMEL (P < .0001); Weight Loss: MEL > NMEL (P = .0077); Insight: PSY > MEL (P = .0004) and PSY > NMEL (P = .0003).
GI: gastrointestinal; HRSD17: 17-item Hamilton Rating Scale for Depression; MEL: melancholic; NMEL: non-melancholic; PSY: psychotic.

We also detected no difference in overall treatment for the episode, except for a higher prevalence in antipsychotic use in PSY patients. NMEL patients were characterized by a longer delay in initiating treatment. When exploring response to treatments, we found a significant trend in the prevalence of non-responders and TRD across depressive subtypes (TABLE 3).

Variables associated with treatment resistance in severely depressed patients

Comparing TRD patients with responders, our team found no difference in socio-demographic or clinical variables. TRD patients exhibited longer hospitalizations over their lifetime (24.1 weeks ± 44.6 vs 13.2 weeks ± 19.0; t = 2.38; P = .0182) and a longer duration of the current episode (280.7 days ± 235.3 vs 211.7 days ± 216.2; t = 2.83; P = .0049). A higher prevalence of comorbid panic disorder also was found in the TRD group (28.1% vs 16.4%; χ2 = 6.96; P = .0084), while no other difference was detected regarding illness course, family history, or somatic comorbidities.

A higher rate of melancholic (74.1% vs 59.9%; χ2 = 8.84; P = .0029) and psychotic (16.0% vs 3.9%; χ2 = 15.07; P = .0001) features was evidenced in TRD patients. Late insomnia was the only melancholia symptom that was more prevalent in this group (54.4% vs 34.2%; χ2 = 14.16; P = .0002).

We also detected a higher rate of patients at risk for suicide (72.1% vs 59.3%; χ2 = 7.00; P = .0082) and higher scores on HRSD17 item 3 (suicide ideation; F = 6.72; P = .0099) in the TRD group. Otherwise, we saw no difference regarding treatments administered for the episode, except for a higher number of antidepressant trials in the TRD group (3.2 ± 1.2 vs 2.1 ± 1.2; t = 8.93; P < .0001).

Significant variables in linear analyses were incorporated as predictors into a non-linear logistic regression model having TRD as the outcome variable (Bonferroni corrected α = .0014). Independent variables were: lifetime duration of hospitalizations, duration of the current episode, panic disorder comorbidity, psychotic features, melancholic features, suicide risk, and late insomnia. No clinical or illness course variable, including melancholic and psychotic features, was found to be associated with TRD in severe unipolar depression. Only a small trend was detected for an association between psychotic features and treatment resistance (Est. = 0.89; Wald. = 7.31; P = .0483).

  DISCUSSION

Our results support the view that psychotic and melancholic depression share some “core” melancholia features, such as late insomnia and psychomotor disturbances. Nonetheless, in our sample there was no association between these features and treatment resistance in severe unipolar depression.

Although we had no standardized assessment of severity, the higher proportion of inpatients, the higher prevalence of suicide risk, and, above all, the higher rate of “core” depressive features all appear to be markers of greater severity in both PSY and MEL patients. Psychotic features are associated with severe mood episodes by definition.30,50-52 However, non-psychotic symptoms also are generally more severe in these patients.13,53-56 In our sample, the PSY group shared with melancholia some “core” symptoms, such as “distinctive quality of mood,” “late insomnia,” and “psychomotor disturbances.”8,10,18,57,58 Previous research has pointed to psychomotor disturbances and psychotic features as the defining features in a hierarchical model of depressive syndromes.8,11,14,15 However, the degree to which some symptoms are specific to depressive subtypes or dependent on overall severity still remains a matter of debate.5,8,10,59-61

Interestingly, both PSY and MEL subjects reported a higher prevalence of seasonal mood episodes and PMDD. A recent study found evidence of a trend for melancholic features to be associated with seasonal depressive episodes,62 supporting the traditional view of the “endogenous,” non-reactive nature of these forms.

The higher number of non-responders and TRD patients in the PSY and MEL groups also suggests more severe illness. A previous study has shown that severity and melancholic features both independently predicted treatment resistance.40 However, our regression analyses did not confirm this association. We did not find any effect of psychotic features on TRD. These results seem to be partly in line with more recent findings on the STAR*D sample, which did not support any specific effect of melancholic features over treatment response.39 Compared with non-psychotic depressed patients, those with psychotic depression seem to have a poorer response to antidepressants.63 However, these findings have been viewed as based only on limited data,64 and an association of psychotic features with TRD has not yet been explicitly reported.

Our study has several limitations. Patients were recruited from specialized units for mood disorders and may not represent the general population. The DSM-IV definition for severe MDEs is neither simple in its structure nor uniform in its validity.65 Further, HRSD17 scores and items were only available in the post-treatment phase, ≥4 weeks after the last antidepressant trial, so that response to treatment only could be assessed retrospectively. Prospective studies capturing clinical characteristics at each stage of the depressive episode and the sequence of treatments clearly are needed to confirm our results.

  CONCLUSIONS

Our study suggests limited differences across depressive subtypes in terms of comorbidities, family history, or illness course variables. These findings indirectly might support a unitary nosographic model of depressive syndromes,2,3 where specific symptoms such as melancholic and psychotic features become more prominent as severity increases.5 Further, no specific effect of melancholic or psychotic features over treatment resistance was detected in our sample of severe unipolar depressed patients.

DISCLOSURES: Dr. Souery has received grant or research support from GlaxoSmithKline and Lundbeck and has served as a consultant or on advisory boards for AstraZeneca, Bristol-Myers Squibb, Eli Lilly and Company, Janssen, and Lundbeck. Prof. Kasper has received grant or research support from Bristol-Myers Squibb, Eli Lilly and Company, GlaxoSmithKline, Lundbeck, Organon, Sepracor, and Servier; has served as a consultant or on advisory boards for AstraZeneca, Bristol-Myers Squibb, Eli Lilly and Company, GlaxoSmithKline, Janssen, Lundbeck, Novartis, Organon, Pfizer, Schwabe, Sepracor, and Servier; and has served on speakers’ bureaus for AstraZeneca, Eli Lily and Company, Janssen, Lundbeck, Neuraxpharm, Pierre Fabre, Schwabe, Sepracor, and Servier. Prof. Zohar has received grant or research support from Lundbeck, Pfizer, and Servier; has served as a consultant or on advisory boards for Actelion, Pfizer, Servier, and Solvay; and has served on speakers’ bureaus for GlaxoSmithKline, Jazz, Lundbeck, and Solvay. Prof. Montgomery has been a consultant or served on advisory boards for AstraZeneca, Bionevia, Bristol-Myers Squibb, Eli Lilly and Company, Forest, GlaxoSmithKline, Grunenthal, Intellect Pharma, Johnson & Johnson, Lundbeck, Merck, Merz, M’s Science, Neurim, Otsuka, Pierre Fabre, Pfizer, PharmaNeuroBoost, Richter, Roche, sanofiaventis, Sepracor, Servier, Shire, Synosis, Takeda, Theracos, Targacept, Transcept, UBC, Xytis, and Wyeth. Prof. Mendlewicz is a member of the Board of the Lundbeck International Neuroscience Foundation and is on the advisory board of Servier. Prof. Serretti is or has been a consultant or speaker for Abbott, AstraZeneca, Clinical Data, Boehringer, Bristol-Myers Squibb, Eli Lilly and Company, GlaxoSmithKline, Janssen, Lundbeck, Pfizer, sanofiaventis, and Servier. Drs. Zaninotto, Calati, Sentissi, and Akimova report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

ACKNOWLEDGEMENTS: This study has been supported by an unrestricted grant to the Group for the Study of Resistant Depression, by Lundbeck A/S and the Belgian National Fund for Scientific Research (FNRS) (FNRS; 3.4.530.07 F). The sponsors had no role in the study design, analysis, and writing of the paper. All authors were involved actively in the analytical method of the study and selection and review of all scientific content and had full editorial control during the writing of the manuscript.

    REFERENCES

  1.  Diagnostic and statistical manual of mental disorders, 3rd ed. Washington DC: American Psychiatric Association; 1980.
  2. Akiskal HS, McKinney WT Jr. Depressive disorders: toward a unified hypothesis. Science. 1973;182:20–29.
  3. Akiskal HS, McKinney WT Jr. Overview of recent research in depression. Integration of ten conceptual models into a comprehensive clinical frame. Arch Gen Psychiatry. 1975;32:285–305.
  4. Shankman SA, Klein DN. Dimensional diagnosis of depression: adding the dimension of course to severity and comparison to the DSM. Compr Psychiatry. 2002;43:420–426.
  5. Judd LL, Schettler PJ, Akiskal HS. The prevalence clinical relevance, and public health significance of subthreshold depressions. Psychiatr Clin North Am. 2002;25:685–698.
  6. Quinn C, Harris A, Kemp A. The Interdependence of subtype and severity: contributions of clinical and neuropsychological features to melancholia and non-melancholia in an outpatient sample. J Int Neuropsychol Soc. 2012;18:361–369.
  7. Schatzberg AF, Rothschild AJ. Psychotic (delusional) major depression: should it be included as a distinct syndrome in DSM-IV? Am J Psychiatry. 1992;149:733–745.
  8. Parker G. Classifying depression: should paradigms lost be regained? Am J Psychiatry. 2000;157:1195–1203.
  9. Parker G, Fink M, Shorter E, et al. Issues for DSM-5: whither melancholia? The case for its classification as a distinct mood disorder. Am J Psychiatry. 2010;167:745–747.
  10. Parker G, Hadzi-Pavlovic D. Melancholia. A disorder of movement and mood. Cambridge United Kingdom: Cambridge University Press; 1996.
  11. Parker G. Defining melancholia: the primacy of psychomotor disturbance. Acta Psychiatr Scand Suppl. 2007;21–30.
  12. Parker G, Hadzi-Pavlovic D, Wilhelm K, et al. Defining melancholia: properties of a refined sign-based measure. Br J Psychiatry. 1994;164:316–326.
  13. Parker G, Hadzi-Pavlovic D, Brodaty H, et al. Sub-typing depression, II. Clinical distinction of psychotic depression and non-psychotic melancholia. Psychol Med. 1995;25:825–832.
  14. Sobin C, Sackeim HA. Psychomotor symptoms of depression. Am J Psychiatry. 1997;154:4–17.
  15. Parker G, Roussos J, Mitchell P, et al. Distinguishing psychotic depression from melancholia. J Affect Disord. 1997;42:155–167.
  16. Carroll BJ. Neurobiologic dimensions of depression and mania. In: Angst J ed. The origins of depression: current concepts and approaches. Berlin, Germany: Springer Verlag; 1983:163-186.
  17. Fountoulakis KN, Iacovides A, Gerasimou G, et al. The relationship of regional cerebral blood flow with subtypes of major depression. Prog Neuropsychopharmacol Biol Psychiatry. 2004;28:537–546.
  18. Taylor MA, Fink M. Melancholia. The diagnosis pathophysiology, and treatment of depressive illness. Cambridge, United Kingdom: Cambridge University Press; 2006.
  19. Thase ME, Kupfer DJ, Ulrich RF. Electroencephalographic sleep in psychotic depression. A valid subtype? Arch Gen Psychiatry. 1986;43:886–893.
  20. Brown RP, Stoll PM, Stokes PE, et al. Adrenocortical hyperactivity in depression: effects of agitation, delusions, melancholia, and other illness variables. Psychiatry Res. 1988;23:167–178.
  21. Nelson JC, Davis JM. DST studies in psychotic depression: a meta-analysis. Am J Psychiatry. 1997;154:1497–1503.
  22. Keller J, Flores B, Gomez RG, et al. Cortisol circadian rhythm alterations in psychotic major depression. Biol Psychiatry. 2006;60:275–281.
  23. Contreras F, Menchon JM, Urretavizcaya M, et al. Hormonal differences between psychotic and non-psychotic melancholic depression. J Affect Disord. 2007;100:65–73.
  24. Potash JB. Carving chaos: genetics and the classification of mood and psychotic syndromes. Harv Rev Psychiatry. 2006;14:47–63.
  25. Frank E, Kupfer DJ, Hamer T, et al. Maintenance treatment and psychobiologic correlates of endogenous subtypes. J Affect Disord. 1992;25:181–189.
  26. Parker G, Hadzi-Pavlovic D, Brodaty H, et al. Predicting the course of melancholic and nonmelancholic depression. A naturalistic comparison study. J Nerv Ment Dis. 1992;180:693–702.
  27. Thase ME, Friedman ES. Is psychotherapy an effective treatment for melancholia and other severe depressive states? J Affect Disord. 1999;54:1–19.
  28. Manschreck TC, Khan NL. Recent advances in the treatment of delusional disorder. Can J Psychiatry. 2006;51:114–119.
  29. Coryell W, Tsuang MT. Primary unipolar depression and the prognostic importance of delusions. Arch Gen Psychiatry. 1982;39:1181–1184.
  30. Coryell W. Psychotic depression. J Clin Psychiatry. 1996;57(suppl 3):27–31; discussion 49.
  31. Spiker DG, Weiss JC, Dealy RS, et al. The pharmacological treatment of delusional depression. Am J Psychiatry. 1985;142:430–436.
  32. Parker G, Roy K, Hadzi-Pavlovic D, et al. Psychotic (delusional) depression: a meta-analysis of physical treatments. J Affect Disord. 1992;24:17–24.
  33. Chan CH, Janicak PG, Davis JM, et al. Response of psychotic and nonpsychotic depressed patients to tricyclic antidepressants. J Clin Psychiatry. 1987;48:197–200.
  34. Coryell W, Turner R. Outcome with desipramine therapy in subtypes of nonpsychotic major depression. J Affect Disord. 1985;9:149–154.
  35. Angst J, Stabl M. Efficacy of moclobemide in different patient groups: a meta-analysis of studies. Psychopharmacology (Berl). 1992;106(suppl):S109–S113.
  36. Peselow ED, Sanfilipo MP, Difiglia C, et al. Melancholic/endogenous depression and response to somatic treatment and placebo. Am J Psychiatry. 1992;149:1324–1334.
  37. Perry PJ. Pharmacotherapy for major depression with melancholic features: relative efficacy of tricyclic versus selective serotonin reuptake inhibitor antidepressants. J Affect Disord. 1996;39:1–6.
  38. Mallinckrodt CH, Watkin JG, Liu C, et al. Duloxetine in the treatment of major depressive disorder: a comparison of efficacy in patients with and without melancholic features. BMC Psychiatry. 2005;5:1.
  39. McGrath PJ, Khan AY, Trivedi MH, et al. Response to a selective serotonin reuptake inhibitor (citalopram) in major depressive disorder with melancholic features: a STAR*D report. J Clin Psychiatry. 2008;69:1847–1855.
  40. Souery D, Oswald P, Massat I, et al. Clinical factors associated with treatment resistance in major depressive disorder: results from a European multicenter study. J Clin Psychiatry. 2007;68:1062–1070.
  41. Carroll BJ. Bringing back melancholia. Bipolar Disord. 2012;14:1–5.
  42. Ioannidis JP. Effectiveness of antidepressants: an evidence myth constructed from a thousand randomized trials? Philos Ethics Humanit Med. 2008;3:14.
  43. Pigott HE, Leventhal AM, Alter GS, et al. Efficacy and effectiveness of antidepressants: current status of research. Psychother Psychosom. 2010;79:267–279.
  44.  Diagnostic and statistical manual of mental disorders, 4th ed. Washington DC: American Psychiatric Association; 1994.
  45. Posternak MA, Zimmerman M. How accurate are patients in reporting their antidepressant treatment history? J Affect Disord. 2003;75:115–124.
  46. Sheehan DV, Lecrubier Y, Sheehan KH, et al. The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry. 1998;59(suppl 20):22–33;quiz 4–57.
  47. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry. 1960;23:56–62.
  48.  European Agency for the Evaluation of Medicinal Products. Note for guidance on clinical investigation of medicinal products in the treatment of depression. http://www.emea.europa.eu/pdfs/human/ewp/051897en.pdf. Published April 25 2002. Accessed January 29, 2013.
  49.  StatSoft I, STATISTICA for Windows. 1995.
  50.  Diagnostic and statistical manual of mental disorders, 4th ed, text rev. Washington DC: American Psychiatric Association; 2000.
  51. Coryell W, Winokur G, Shea T, et al. The long-term stability of depressive subtypes. Am J Psychiatry. 1994;151:199–204.
  52. Coryell W, Leon A, Winokur G, et al. Importance of psychotic features to long-term course in major depressive disorder. Am J Psychiatry. 1996;153:483–489.
  53. Glassman AH, Roose SP. Delusional depression. A distinct clinical entity? Arch Gen Psychiatry. 1981;38:424–427.
  54. Frances A, Brown RP, Kocsis JH, et al. Psychotic depression: a separate entity? Am J Psychiatry. 1981;138:831–833.
  55. Parker G, Hadzi-Pavlovic D, Hickie I, et al. Psychotic depression: a review and clinical experience. Aust N Z J Psychiatry. 1991;25:169–180.
  56. Souery D, Zaninotto L, Calati R, et al. Phenomenology of psychotic mood disorders: lifetime and major depressive episode features. J Affect Disord. 2011;135:241–250.
  57. Hildebrandt MG, Stage KB, Kragh-Soerensen P. Danish University Antidepressant Group. Gender differences in severity symptomatology and distribution of melancholia in major depression. Psychopathology. 2003;36:204–212.
  58. Khan AY, Carrithers J, Preskorn SH, et al. Clinical and demographic factors associated with DSM-IV melancholic depression. Ann Clin Psychiatry. 2006;18:91–98.
  59. Gold PW, Chrousos GP. Organization of the stress system and its dysregulation in melancholic and atypical depression: high vs low CRH/NE states. Mol Psychiatry. 2002;7:254–275.
  60. Angst J, Gamma A, Benazzi F, et al. Melancholia and atypical depression in the Zurich study: epidemiology, clinical characteristics, course, comorbidity and personality. Acta Psychiatr Scand Suppl. 2007;72–84.
  61. Coryell W. The facets of melancholia. Acta Psychiatr Scand Suppl. 2007;31–36.
  62. Winthorst WH, Post WJ, Meesters Y, et al. Seasonality in depressive and anxiety symptoms among primary care patients and in patients with depressive and anxiety disorders; results from the Netherlands Study of Depression and Anxiety. BMC Psychiatry. 2011;11:198.
  63. Gaudiano BA, Beevers CG, Miller IW. Differential response to combined treatment in patients with psychotic versus nonpsychotic major depression. J Nerv Ment Dis. 2005;193:625–628.
  64. Harald B, Gordon P. Meta-review of depression subtyping models. J Affect Disord. 2012;139:126–140.
  65. Lux V, Aggen SH, Kendler KS. The DSM-IV definition of severity of major depression: inter-relationship and validity. Psychol Med. 2010;40:1691–1701.

CORRESPONDENCE: Alessandro Serretti, MD, PhD Institute of Psychiatry University of Bologna Viale Carlo Pepoli 5 40123 Bologna, Italy E-MAIL: alessandro.serretti@unibo.it