Melancholic features and treatment outcome to selective serotonin reuptake inhibitors in major depressive disorder: A re-analysis of the STAR*D trial.

BACKGROUND: Melancholia has been positioned as a qualitatively different form of Major Depressive Disorder (MDD). Some studies have suggested that melancholic MDD patients may show lower remission when receiving treatment with Selective Serotonin Reuptake Inhibitors, but this has not yet been explored in large, representative samples of MDD. METHODS: We used data from the STAR*D, a multisite randomized controlled trial (n = 4041). We defined melancholia status through the BA Melancholia Empirical Index, constructed using items from the Inventory of Depressive Symptomatology (IDSC). The main outcome of interest was symptomatic remission defined as a Quick Inventory of Depressive Symptoms (Clinician version) (QIDS-C) below or equal to 5. Inverse probability weighting was used to control for confounding. RESULTS: 3827 patients were eligible for this study. Melancholic patients were more likely to be unemployed, never married, to self-report an African American race, and to have a higher depressive severity. The adjusted 4-month probability of remission was 26.9 % (22.0, 45.5) for melancholic and 53.8 % (53.2, 58.5), for nonmelancholic patients. Compared with nonmelancholic, the difference in 4-month probability of remission was -26.9 % (-37.0, -15.6). Results were consistent across sensitivity analyses. LIMITATIONS: Items from IDSC were used as a surrogate measure of the BA Melancholia Index, and extrapolation of the results to agents other than citalopram and to psychotic MDD patients requires caution. CONCLUSIONS: Melancholic MDD patients showed lower probabilities of remission at 4-months receiving treatment with citalopram. The results of this study show how validly subtyping episodes could contribute to the personalized treatment of depression.

Clinical Characteristics of Melancholic and Nonmelancholic Depressions.

ABSTRACT: This study aimed to compare clinical-demographic features of melancholic and nonmelancholic depressions. We included 141 depressed inpatients classified as melancholic and nonmelancholic by the Sydney Melancholia Prototype Index (SMPI) and Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM-5) criteria. Results were controlled for confounders, including severity measures. Melancholic patients by both diagnostic systems were more severely depressed and presented more psychotic symptoms, neurological soft signs, and psychomotor disturbances. Melancholic patients classified by the SMPI were also older at illness onset and had fewer suicide attempts. After controlling for confounders, although all differences remained significant for SMPI diagnosis, the DSM-5 diagnosis of melancholia was only associated with further impaired motor sequencing. The results obtained with the SMPI support the hypothesis that melancholia has clinical features qualitatively different from those of nonmelancholic depressions. Contrarily, the DSM-5 specifier seems to reflect the severity of depressive episodes rather than core clinical features of melancholia.

Correlates of neurological soft signs in bipolar disorder.

Recent research has focused on neurological soft signs (NSS) in bipolar disorder (BD), but there is still scarce evidence on their correlates with other relevant variables. The aim of this study was to explore the association between NSS and clinical demographic, neurocognitive, and functional variables. Eighty-eight euthymic BD patients were included in whom NSS were assessed using the Neurological Evaluation Scale. All subjects performed an extensive neurocognitive battery selected to assess premorbid IQ, attention, language, verbal memory, and executive functions. Psychosocial outcomes were assessed by means of the Functioning Assessment Short Test. Among the demographical variables, NSS tended to be negatively associated with years of education and increased with age. Likewise, NSS were higher in BD type I than in BD type II, but independent of age at illness onset and number of previous affective episodes. There was a negative correlation between NSS and premorbid IQ, as well as with performance in attention, language, and executive functions. Results tended to be unchanged when controlled for potential confounders. Although NSS were associated with poor psychosocial functioning in the bivariate analysis, when added to a multiple regression model including neurocognition, these neurological features did not significantly contribute to the variance of the functional outcome. Our findings contribute to a better characterization of NSS in BD; their potential clinical and theoretical implications are discussed in the manuscript.

Neurocognitive function of patients with melancholic and non-melancholic major depressive episodes: An exploratory study.

OBJECTIVE: The aim of this exploratory study was to compare the neurocognitive performance of patients undergoing melancholic and non-melancholic major depressive episodes. Considering potential limitations of the Diagnostic and Statistical Manual of Mental Disorders (5th ed.) specifier, we employed an additional tool that has proven useful in identifying melancholia (the Sydney Melancholia Prototype Index). METHODS: One hundred forty-one depressed inpatients were classified as melancholic or non-melancholic according to the Sydney Melancholia Prototype Index and the Diagnostic and Statistical Manual of Mental Disorders (5th ed.) criteria and compared on a neurocognitive battery selected to assess attention and processing speed, verbal memory, working memory and executive functions. Results were controlled for several potential confounders. RESULTS: Patients diagnosed as melancholic by the two diagnostic systems displayed lower scores in executive measures, semantic verbal fluency and phonological verbal fluency. On attention and processing speed, patients with melancholia underperformed those with non-melancholic depression only when diagnosed by the Sydney Melancholia Prototype Index. After controlling for confounders, associations between melancholic status and executive dysfunction remained significant for the Sydney Melancholia Prototype Index but not for the Diagnostic and Statistical Manual of Mental Disorders (5th ed.) diagnosis. CONCLUSION: In this study, melancholia diagnosed by the Sydney Melancholia Prototype Index (but not by the Diagnostic and Statistical Manual of Mental Disorders [5th ed.] criteria) was characterized by a greater compromise of tests assessing executive functions than non-melancholic depressions, even after controlling for depressive severity. These preliminary results might contribute to generating hypotheses about differences in the cognitive profile and pathophysiological substrate between melancholic and non-melancholic depressions. Likewise, the pattern of findings supports the hypothesis that the Diagnostic and Statistical Manual of Mental Disorders (5th ed.) melancholia specifier might identify more severe forms of depressive episodes rather than a qualitatively different subtype.

Clinical Course in Patients With Melancholic and Nonmelancholic Bipolar Depression.

ABSTRACT: The aim of the current study was to compare the longitudinal clinical course of patients with a history of melancholic and nonmelancholic bipolar depression. Seventy-seven euthymic outpatients with bipolar disorder were categorized as melancholic or nonmelancholic through the clinician-rated Sydney Melancholia Prototype Index. Clinical course was assessed for a period longer than 48 months by time spent ill, density of affective episodes, severity and duration of depressive episodes, and time to depressive recurrence. The mean follow-up time was 69.05 months. Patients with melancholic depressions had more severe and longer depressive episodes during follow-up, whereas patients with nonmelancholic depressions had a shorter time to depressive recurrence and more subsyndromal depressive symptoms and affective instability. These findings highlight the heterogeneity inherent to the current construct of bipolar depression and position melancholia as an interesting target for comparison with nonmelancholic depressions in other external validators in the field of bipolar disorder.

Neurocognitive predictors of long-term clinical course in bipolar disorder.

OBJECTIVE: The aim of this study was to estimate whether neurocognitive deficits are predictors of the long-term clinical course of patients with bipolar disorder. METHODS: A total of 76 outpatients with bipolar disorder performed a neurocognitive assessment at baseline and were followed for a period of at least 48 months. The clinical course during the follow-up period was documented by two measures: (1) number of affective episodes and (2) time spent ill. RESULTS: Patients had lower performance than controls in the domains of verbal memory and executive functions, and they were followed-up for a mean period of 70.73 months. Global cognitive deficits (performance 1.5 standard deviations below the mean in two or more cognitive domain) were independent predictors of both hypo/manic episode density and time spent with hypo/manic symptoms during follow-up. On the contrary, no neurocognitive measure showed a relationship with depressive morbidity during follow-up. CONCLUSION: Our findings suggest that cognitive deficits could be useful for predicting hypo/manic morbidity in the long-term clinical course. Theoretical and clinical implications of these findings are discussed.

Neurocognitive functioning in first-episode Bipolar Disorder: Relationship with functional status.

BACKGROUND: The aim of this study was to assess if an association between neurocognitive deficits and psychosocial functioning exists in first-episode BD patients. METHODS: Twenty-five euthymic first-episode BD patients and thirty-seven healthy controls were recruited. History of suicide attempts, psychiatric comorbidities, pharmacological exposure, and previous depressive episodes were investigated. Performances on neurocognitive domains (verbal memory, attention, processing speed, and executive functions) as well as a measure of psychosocial functioning were used as outcomes. RESULTS: First-episode BD patients showed medium-to-large size deficits on measures of attention, processing speed, and executive functions. A significant association between verbal memory and psychosocial functioning at the moment of BD diagnosis was detected (beta coefficient -3.9, IC 95% -6.7 to -1.2, p < 0.01). CONCLUSIONS: A relationship between cognitive performance at the moment of BD diagnosis and psychosocial functioning was detected. Possible therapeutic implications of this finding are discussed.