Health professional's knowledge and awareness of perinatal depression: results of a national survey

INTRODUCTION

Postnatal depression affects 14% of women, occurring also antenatally, with potential long-term consequences, making it an important disorder to detect and manage early. In this study we sought to examine knowledge and awareness of perinatal depression in health professionals involved in perinatal care throughout Australia prior to the implementation of a comprehensive screening program, aimed at improving detection and access to appropriate management.

METHODS

A random sample of General Practitioners (GPs) and Maternal Child Health Nurses (MCHNs) and Midwives, in regions throughout Australia to be subsequently targeted by a screening and education program, were invited to participate. Responses to a hypothetical vignette and a knowledge questionnaire, as well as details of experience were completed.

FINDINGS

Questionnaires were completed by 246 GPs, 338 MCHNs and 569 midwives, with overall response rates; GP's 23%; MCHN's 55% and midwives 57%. Although knowledge level was similar among professional groups, MCHNs had higher levels of awareness of perinatal depression. Both GPs and MCHNs were more likely than midwives to recognize the need for providing help to women with emotional distress. Depression was more likely to be considered postnatally than antenatally in all groups, with GPs most likely to provide this diagnosis. GPs had a significant propensity to recommend antidepressants, and midwives to select non-specific medications.

CONCLUSIONS

Health professionals responding to this survey had a high awareness and similar knowledge base. Further education on antenatal depression and the safety risks and alternatives to medication is important for all groups, but particularly important for midwives and GPs. The latter is especially relevant given the preference for women with perinatal depression not to use pharmacological interventions to treat their emotional distress.

Gender differences in response to differing antidepressant drug classes: two negative studies

BACKGROUND

A recent US study presented data suggesting that depressed women are more likely to respond to selective serotonin reuptake inhibitor (SSRI) than tricyclic (TCA) antidepressant drug therapies. We have undertaken replication studies in two independent databases.

METHOD

We have examined for gender differences in SSRI and TCA antidepressant response in both retrospective and prospective naturalistic uncontrolled studies, and in subsets of melancholic and non-melancholic depressed subjects. As the US study had indicated that women under the age of 40 years were particularly likely to show a differential response to SSRIs, we examined for age, gender and interactional effects. In addition, we examined for differential SSRI and TCA responsiveness in a subset of patients who had received drugs from both classes.

RESULTS

We failed to find evidence of women having a preferential response to SSRI medication or, conversely, of men having a superior response to TCA medication. Older age, however, was associated with a superior TCA response and younger age with a superior SSRI response.

CONCLUSION

As few studies have examined for differential gender and age effects in response to narrow action and broad action antidepressant drugs across major depressive subtypes, gender differential effects remain to be established.

Indicators of suicide over 10 years in a specialist mood disorders unit sample

BACKGROUND

There are few firm data to guide the clinician in identifying individual depressed patients who may be at high risk for completing suicide. In particular, there have been few prospective studies of well-characterized depressed patients to determine indicators of such future events.

METHOD

Eight hundred thirteen patients with a major depressive episode (DSM-III, DSM-III-R, or DSM-IV criteria) were assessed in detail in a specialist Mood Disorders Unit (MDU) over a 10-year period. Follow-up at the end of that period (mean = 5.1 years) confirmed that 31 patients (3.8%) had completed suicide. The suicide completers were compared on a broad range of clinical and demographic variables obtained at baseline with (1) the total remaining depressed sample, (2) 31 age- and sex-matched subjects who were confirmed to be alive and had never attempted suicide, and (3) 24 age- and sex-matched living subjects who had made at least 1 suicide attempt.

RESULTS

The most consistent finding, across all 3 comparisons, was that the suicide completers were more likely to have been inpatients at the time of the index MDU assessment. Other characteristics of completers were a greater number of prior admissions for depression, being older and in a relationship, and being male and married or female and single. Somewhat paradoxically, suicide completers also evidenced fewer previous suicide attempts and less suicidal ideation compared with living subjects who had attempted suicide at the time of index assessment.

CONCLUSION

Overall, we were able to find few predictors of later suicide in this sample. Those who completed suicide demonstrated evidence of more severe illness over a lifetime (for example, having more admissions). but revealed less suicidal ideation at the time of the index MDU assessment. While these features were statistically significant, they are of limited usefulness in predicting suicide in an individual patient.

What interventions may reduce postpartum depression

Postnatal depression is a major public health problem affecting about one in seven women after childbirth. Depression is also common during pregnancy and throughout the perinatal period it is associated with symptoms of anxiety. Apart from the adverse consequences for women themselves becoming depressed when they are going through demanding physical and social changes, there are additional concerns. There is the possible negative impact of maternal depression on the relationship between mother and child and on the child's emotional, behavioural and cognitive development. Primary prevention and early intervention/secondary prevention strategies are potentially important in view of the frequent contact pregnant women, new mothers and infants have with health services, but the effectiveness of these strategies needs to be tested. In the past year there have been five new studies of antenatal screening for postnatal depression. These studies are consistent with nine earlier studies in showing that there is no evidence to support routine antenatal screening for postnatal depression. Seven new primary prevention/early intervention trials add evidence on a wide range of interventions ranging from practical support to individual interpersonal therapy, but without identifying significant differences in depression as an outcome. Two new trials of secondary prevention, one involving interpersonal therapy and the other including partners in a series of psychoeducational visits, show promise but neither is large enough to form a basis for practice change. Novel interventions, or promising findings, with a strong basis in theory need to be tested in trials which are appropriately sized and which comply with internationally accepted design and reporting guidelines.

Early and late onset depression in old age: different aetiologies, same phenomenology

BACKGROUND

Phenomenological differences between older patients with early onset (EO; onset of first major depressive episode before 60 years) and late onset (LO) depression have been inconsistent but, if real, may reflect differences in aetiology. We aimed to compare aetiological factors, phenomenology and cognitive function in older patients with depression by age of onset.

METHODS

Subjects were all patients > or =60 years old (n=73) from 407 consecutive attenders to a Mood Disorders Unit, diagnosed with DSM-III-R Major Depressive Episode, at or close to the nadir of their episode. Putative risk factors were assessed by structured interview. Psychological morbidity and depressive symptoms were assessed by the 21-item Hamilton Rating Scale for Depression, CORE rating of psychomotor disturbance, Newcastle Endogeneity Scale, Zung Depression Scale and General Health Questionnaire. Cognition was assessed by tests of memory, attention, executive function and motor speed.

RESULTS

Personality abnormalities, a family history of psychiatric illness and dysfunctional past maternal relationships were significantly more common in EO depression. The two age of onset groups were essentially similar in terms of depressive sub-type and severity, phenomenology, history of previous episode, and in neuropsychological performance.

LIMITATIONS

Use of self-report data, moderate sample size, sample not age-matched, tertiary referral patients.

CONCLUSIONS

EO and LO depression are similar phenotypically, but differ aetiologically. The pursuit of mechanisms which predispose depressive episodes may be heuristically more valuable than further investigation of individual depressive features in distinguishing early from late onset depression.

Cognitive deficits in depression: possible implications for functional neuropathology

BACKGROUND

While depression is known to involve a disturbance of mood, movement and cognition, its associated cognitive deficits are frequently viewed as simple epiphenomena of the disorder.

AIMS

To review the status of cognitive deficits in depression and their putative neurobiological underpinnings.

METHOD

Selective computerised review of the literature examining cognitive deficits in depression and their brain correlates.

RESULTS

Recent studies report both mnemonic deficits and the presence of executive impairment--possibly selective for set-shifting tasks--in depression. Many studies suggest that these occur independent of age, depression severity and subtype, task 'difficulty', motivation and response bias: some persist upon clinical 'recovery'.

CONCLUSIONS

Mnemonic and executive deficits do no appear to be epiphenomena of depressive disorder. A focus on the interactions between motivation, affect and cognitive function may allow greater understanding of the interplay between key aspects of the dorsal and ventral aspects of the prefrontal cortex in depression.

The clinical features of bipolar depression: a comparison with matched major depressive disorder patients

BACKGROUND

Despite a resurgence of interest in the treatment of bipolar depression, there have been few controlled studies of the clinical characteristics of this condition. Identification of any distinctive clinical "signatures" of bipolar depression would be helpful in determining treatment options in the clinical setting.

METHOD

From a cohort of 270 inpatients and outpatients assessed in detail during a DSM-IV major depressive episode, 39 bipolar I disorder patients were identified and closely matched with 39 major depressive disorder patients for gender, age, and the presence or absence of DSM-IV melancholic subtype. Patients were compared on a broad range of parameters including the Hamilton Rating Scale for Depression (depression severity), 54 depressive symptoms, the Newcastle Endogenous Depression Diagnostic Index, 3 family history items, 2 physical health items, the CORE scale (psychomotor disturbance), and 5 history items.

RESULTS

Although the bipolar patients were no more severely depressed than the major depressive disorder controls, they were more likely to demonstrate psychomotor-retarded melancholic and atypical depressive features and to have had previous episodes of psychotic depression. These findings were largely duplicated even when the population was confined to those with DSM-IV melancholia.

CONCLUSION

The clinical admixture of psychomotor-retarded melancholic signs and symptoms, "atypical" features, and (less frequently) psychosis may provide a "bipolar signature" in clinical scenarios when there is uncertainty concerning the polarity of a depressive presentation.

Effect of apomorphine on motor and cognitive function in melancholic patients: a preliminary report

Given the many clinical parallels between melancholia and disorders associated with impaired dopaminergic function such as Parkinson's Disease (PD), it has been hypothesised that major depressive disorder, and in particular the psychomotor features of melancholic depression, may also be associated with a hypodopaminergic state. If this is the case, then the use of a dopamine agonist might lead to reversal of both the cognitive and motor impairments seen in these patients. A double-blind, placebo-controlled cross-over design was used to test the effect of apomorphine on motor and cognitive function in seven melancholic subjects (as defined by the CORE instrument) and five control subjects. The testing battery included the following items: finger tapping, rapid alternating movements, verbal fluency, Rey Auditory Verbal Learning Task, digit symbol substitution task and simple and complex reaction times. The independent t-test, after covarying for age, revealed significant impairment in melancholic subjects for the walking task and digit symbol substitution at baseline. Results of the ANCOVA revealed no impact of time or drug condition, either alone or in combination, upon task performance in either group whether assessed separately or jointly. Results of a MANCOVA revealed that apomorphine impaired performance on some cognitive tasks, and that this was seen to a lesser extent in melancholics than control subjects. There was no evidence that the dopamine agonist apomorphine improved cognitive or motor function in subjects with strictly defined melancholia, suggesting that psychomotor retardation is not associated with a hypodopaminergic state. Our conclusions, however, were limited by small sample size; minimal baseline task impairment in depressed compared to control subjects; mild sedation in many subjects during task performance; and lack of serum apomorphine levels.

Maternal stress and obstetric and infant outcomes: epidemiological findings and neuroendocrine mechanisms

This review examines the associations between antenatal maternal stress and obstetric and infant outcomes using preterm delivery as the key outcome indicator. This was done by means of a Medline search focusing predominantly on prospective, controlled studies which investigated both the associated epidemiological factors and putative neuroendocrine mechanisms. There is evidence from a number of United States studies in economically deprived African American women for an association between perceived maternal life event (LE) stress and preterm delivery. The findings from the European studies are conflicting, partly because they combine outcome measures ie. preterm delivery and low birth weight. However the three largest Scandinavian epidemiological studies examining preterm delivery and controlling for confounders such as smoking, age and obstetric history, have confirmed this association. These studies taken together suggest that this may be a robust finding not limited to socioeconomically deprived African American samples and independent of other significant risk factors. Two small prospective studies examining the relationship between the hypothalamic-pituitary-adrenal (HPA) axis, psychosocial status and premature delivery have reported a significant association between a set of adverse psychosocial factors on the one hand, and levels of adrenocorticotrophic hormone (ACTH), corticotrophin releasing hormone (CRH) and cortisol levels, and on the other hand, a significant correlation between CRH levels and premature delivery. Clearly, these findings remain preliminary and indicate a complex relationship between perceived stress in pregnancy, the HPA axis and premature delivery. The impact of antenatal maternal stress on infant temperament and psychopathology remains to be examined more fully in prospective controlled trials.

An exploration of links between early parenting experiences and personality disorder type and disordered personality functioning

Reports of early parenting were assessed using two measures, the Parental Bonding Index (PBI) and the Measure of Parenting Style (MOPS), in a sample of 265 patients with DSM-defined major depressive disorder. Psychiatrists then rated the extent to which sample members evidenced the personality "styles" underpinning 15 separate personality disorders, returning personality vignette scores. The extent of disordered functioning was also assessed across "parameters" and "domains" by psychiatrists, referrers, and family members, using a range of measures. Those with higher scores on vignettes measuring borderline, anxious, depressive, and self-defeating personality style rated parents as uncaring, overcontrolling, and abusive. When vignettes were consolidated into scores akin to the DSM clusters, the most consistent links between perceived dysfunctional parenting were with the Cluster C (anxious), and Cluster B (dramatic) styles and were nonsignificant for Cluster A (eccentric) style. Meeting criteria for an increasing number of personality disorder clusters was associated with increasing levels of adverse parenting. Multiple regression analyses indicated that disordered functioning (as assessed by the three independent rater groups) was most distinctly associated with paternal indifference and maternal overcontrol.

Sub-grouping non-melancholic depression from manifest clinical features

AIM

To determine whether clinical symptoms manifested during an episode of major depression are sufficient to allow meaningful sub-groups of non-melancholic depression to be identified.

METHODS

A sample of 178 non-melancholic patients with a major depressive episode was studied. The initial set of clinical variables was refined to 38 (21 depression, 17 anxiety) items and a cluster analysis undertaken.

RESULTS

A four-cluster solution identified 'anxiety', 'irritability', 'depressed mood' and 'residual' clusters, with these labels clarified by reference to a large data bank of non-symptomatic variables. These analyses suggested that members of the first two clusters could be viewed as having spectrum conditions (whereby Axis I symptom states are able to be linked with precursor or prodromal states and personality).

CONCLUSIONS

We confirm the long-standing suggestion that the non-melancholic depressive class contains sub-groups of those with manifest states of anxious depression and of an irritable or 'hostile' depression, and that such manifest symptoms are likely to be rooted in and fed by temperament and personality characteristics. The delineation of such sub-groups should assist studies designed to identify underlying neurobiological underpinnings and clinical management of the non-melancholic depressive disorders.

Sub-grouping non-melancholic major depression using both clinical and aetiological features

OBJECTIVE

In previous papers we have considered the extent to which two contrasting analytic approaches, examining reported clinical symptom variables alone and aetiological variables alone, assist definition of subgroups of non-melancholic major depression. Here, we address the same objective but combine both sets of variables, and contrast the combined solution with each of the contributing ones.

METHOD

We study a sample of 185 subjects with a putative non-melancholic major depressive disorder, with analyses involving 13 aetiological and 38 symptom variables.

RESULTS

A four-class subgrouping was derived by use of a cluster analytic technique, with 'neurotic depression', non-anxious 'depressed', 'situational' and 'residual' groups. The largest group comprised 'neurotic depression' subjects, with characteristics compatible with a spectrum disorder encompassing both clinical features as well as an underlying temperament and personality style marked by anxiety.

CONCLUSIONS

Comparative advantages and properties of the three differing analytic approaches to defining 'meaningful' non-melancholic major depressive subgroupings are considered. As a 'neurotic depressive' class has been consistently identified across those three approaches, but with quite varying numbers of subjects circumscribed, it is clearly a 'fuzzy' entity which may benefit from a dimensional approach to its measurement. As many of the non-melancholic groupings appear secondary to a substantive predisposing factor such as anxiety or disordered personality functioning, the clinical importance and treatment utility in identifying and circumscribing such classes are clearly supported.

Characteristics of depressed patients who report childhood sexual abuse

OBJECTIVE

Depressed patients who had and had not been exposed to childhood sexual abuse were studied to determine differences in severity of depressed mood, lifetime histories of anxiety and depression, childhood environment, and disordered personality function.

METHOD

Data were obtained from 269 inpatients and outpatients with major depression (171 women and 98 men) by means of structured clinical interviews and self-report questionnaires.

RESULTS

Forty-six of the 269 patients reported childhood sexual abuse; 40 of these were women. These 40 women were compared with the 131 who did not report childhood sexual abuse. The patients who experienced abuse did not differ from those who had not on psychiatrist-rated mood severity estimates, but they did have higher self-report depression scores. They also evidenced more self-destructive behavior, more personality dysfunction, and more overall adversity in their childhood environment. Childhood sexual abuse status was associated with more borderline personality characteristics independently of other negative aspects of the patients' earlier parenting. Childhood sexual abuse status was linked strongly to adult self-destructiveness, as was early exposure to maternal indifference.

CONCLUSIONS

Multivariate analyses suggest that depression is unlikely to be a direct consequence of childhood sexual abuse. Childhood sexual abuse appears to be associated with a greater chance of having experienced a broadly dysfunctional childhood home environment, a greater chance of having a borderline personality style, and, in turn, a greater chance of experiencing depression in adulthood.

"Acting out" and "acting in" as behavioral responses to stress: a qualitative and quantitative study

In a sample of 270 depressed patients, we describe some behaviors in response to stress. One third acknowledged "acting out" behaviors--angry, destructive acts and "out of control" behaviors. Four fifths acknowledged "acting in" behaviors--most commonly withdrawal. As a percentage of subjects acknowledged both response styles (with alternate expression influenced by situation), we developed a dimensional self-report measure within a subsample of 177 who attended a follow-up reassessment. Assignment to four groups with contrasting expressions of "acting out" and "acting in" scores demonstrated differences in age, diagnostic status, age of onset of depression, and self-injurious behaviors.

Cognitive function in depression: a distinct pattern of frontal impairment in melancholia?

BACKGROUND

Although depressed patients demonstrate impaired performance on a range of neuropsychological tests, there is little research that examines either frontal cognitive deficits or possible differences in test performance between melancholic and non-melancholic subtypes.

METHODS

Depressed subjects were administered a broad neuropsychological battery. In an overall analysis, 77 depressed subjects were compared with 28 controls. In a second set of analyses, the depressed sample was divided into melancholic and non-melancholic subsets according to DSM-III-R, the CORE system and the Newcastle scale. These depressed subsets were contrasted to controls and with each other using ANCOVA controlling for age, IQ, simple reaction time and Hamilton Depression scores where appropriate.

RESULTS

The total depressed sample was impaired on most mnemonic tasks, simple reaction time and Trails B. Similar findings applied to DSM-III-R melancholic and non-melancholic subjects. When defined by the CORE and Newcastle (narrower definitions of melancholia), melancholic patients were additionally impaired on WCST (perseverative response) and (for Newcastle) digit symbol substitution. In contrast, the cognitive performance of the CORE and Newcastle-defined non-melancholic patients was largely unimpaired.

CONCLUSIONS

Using narrower definitions of melancholia, i.e. CORE and (in particular) Newcastle, melancholic patients were impaired on mnemonic tasks and tasks of selective attention, and set-shifting while non-melancholic subjects were largely unimpaired in their cognitive performance. These differences may be due to impairment of specific neuroanatomical regions in narrowly defined melancholic patients, in particular the anterior cingulate.

The influence of anxiety as a risk to early onset major depression

OBJECTIVE

we seek to identify and quantify any risk provided by several expressions of "anxiety" to major depression overall, and to separate melancholic and non-melancholic sub-types.

METHOD

a sample of 269 patients with a current major depressive episode was assessed for rates of separate formalised anxiety disorders, both for lifetime and prior to the initial depressive episode. We also sought for evidence of familial anxiety and, early childhood expression of anxiety forerunners, measured both state and trait anxiety levels as well as anxiety at a "personality" level, and assessed use of anxiolytic medications. Depressive sub-typing was undertaken using DSM-IV criteria, while "early onset" (EO) depression was defined as an initial onset at 25 years or less, and subsequently re-examined with a cut-off age of 20 years or less.

RESULTS

overall. 42% of our sample were assigned as having EO depression, with there being a higher representation of non-melancholic than melancholic EO subjects (i.e., 51% vs. 29%), arguing for sub-type status being respected in the analyses. For both melancholic and non-melancholic subjects two trait anxiety items ("tense"; "keyed up/on edge") were over-represented, suggesting that such a tense anxiety style may provide an antecedent risk to depression (of either sub-type) or be a consequence of depression. Specificity was most evident in the non-melancholic sub-sample, where EO depression was associated with a family history of anxiety, early childhood expressions of anxiety and with two lifetime anxiety disorders (social phobia and obsessive-compulsive disorder). Broadly similar results were returned when "EO" definition was reduced to 20 years or less.

CONCLUSIONS

our study is consistent with previous research in identifying anxiety in the form of social inhibition or social avoidance as being particularly likely to precede and perhaps be a conduit to early onset non-melancholic major depression. This conclusion both sharpens risk factor research and indicates an important fulcrum that could be used to assist primary prevention of the depressive disorders.

Use of psychotropic medications in breast-feeding women: acute and prophylactic treatment

OBJECTIVES

The postnatal period is a time of increased onset and relapse of mental illness. It poses a clinical dilemma, as many mothers requiring medication acutely or prophylactically will also choose to breast feed. The present paper first reviews the safety of psychotropes in breast-fed infants and the usefulness of prophylaxis for women at risk of postpartum affective relapse and, second, provides guidelines in the use of psychotropic drugs in breast-feeding women.

METHODS

A Medline review was conducted reviewing all papers published during the period 1993-1998 (and their associated bibliographies) on the use of psychotropes in breast-feeding women and the prophylactic usefulness of medications in women at risk of affective postpartum relapse.

RESULTS

Findings are based on case reports and small, mostly uncontrolled studies. Both tricyclic antidepressants (TCA) and specific serotonin re-uptake inhibitors (SSRIs) appear to be relatively safe in breast feeding. Antidepressants commenced in the early postpartum period may reduce depressive relapse. While prophylactic lithium appears to significantly reduce relapse of affective psychosis in the puerperium, there have been no studies of the anticonvulsants in the puerperium. Finally, high dose antipsychotics should be avoided, as they may be associated with long-term adverse sequelae in the infant.

CONCLUSIONS

On the basis of current knowledge, the use of SSRIs, TCA, carbamazepine, sodium valproate and short-acting benzodiazepines in breast feeding is relatively safe. If lithium is to be used, close collaboration with a paediatrician is essential. The long-term risks of antipsychotics, especially at high doses, remain to be clarified. Before a decision can be made, the risk-benefit ratio must be clearly outlined and discussed with the mother and her partner.

On modelling personality disorders: are personality style and disordered functioning independent or interdependent constructs?

Current descriptors of personality disorder (PD) are an amalgam of two constructs, personality style and/or disorder. We seek to determine whether their intrinsic personality style descriptors are proxy measures of, or independent of, disordered personality functioning. In a sample of depressed patients, psychiatrists rated 16 differing PD personality style vignettes and assessed eight differing manifestations of disordered functioning. When "personality" vignettes and identified personality clusters were intercorrelated with "disorder" variables, interdependence was generally evident, suggesting that the personality descriptors underpinning current definition of the PDs actually act as proxy criteria for assessing disorder because they are, in and of themselves, descriptors of pathological functioning. The obsessional personality vignette provided an exception, seeming to be independent of disordered function. Such results assist consideration of how best to model, define and measure the personality disorders.

Qualitative and quantitative analyses of a 'lock and key' hypothesis of depression

BACKGROUND

We examine a 'lock and key' ('L-K') hypothesis to depression which posits that early adverse experiences establish locks that are activated by keys mirroring the earlier adverse experience to induce depression.

METHODS

Two-hundred and seventy clinically depressed patients were examined with open-ended and pre-coded interview questions to ascertain both early adverse experiences and precipitating life events. Qualitative and quantitative data analyses examined for any associations between developmental 'locks' and precipitating 'keys'.

RESULTS

Qualitative assessment suggested 'L-K' links in almost one-third of the sample, and examples are provided. While quantitative analyses indicated significant associations between several identical 'lock' and 'key' constructs, evidence of specificity was rare. When individual 'locks' and 'keys' were consolidated into three higher-order constructs, variable models were suggested, including a non-specific link, a specific link and absence of any link. 'L-K' links appeared more likely in those with 'non-melancholic' (versus 'melancholic') depression, with the seemingly greater relevance to 'reactive' (versus 'neurotic') depression in the quantitative analyses inviting speculation that that 'disorder' may be more a reaction to a salient rather than a severe stressor.

CONCLUSIONS

This exploratory study suggests that early adverse experiences may variably establish specific and non-specific patterns of vulnerability to having depression triggered by exposure to salient mirroring life event stressors.

Psychotropic medications in pregnant women: treatment dilemmas

OBJECTIVES

To review the evidence from all studies of adverse effects on infant outcome of psychotropic medications taken during pregnancy.

DATA SOURCES

MEDLINE January 1976-February 1998, EMBASE 1976-February 1998, and bibliographies of retrieved articles.

STUDY SELECTION AND DATA EXTRACTION

All studies focusing on adverse effects associated with psychotropic drug use during pregnancy, with a particular focus on prospective controlled studies. OUTCOME CRITERIA: Congenital anomalies, perinatal complications and neurobehavioural sequelae.

DATA SYNTHESIS

23 studies were identified, nine of which were prospective controlled studies: five involving antidepressants (tricyclic antidepressants [TCAs] and selective serotonin reuptake inhibitors [SSRIs]), one each involving lithium and carbamazepine, and two involving benzodiazepines. As statistical synthesis was not possible given the heterogeneity of outcome criteria, a qualitative review is provided. Neither the SSRIs nor the TCAs appear to cause major congenital anomalies, but both may be associated with a small increased risk of minor anomalies, prematurity and neonatal complications. Benzodiazepines, lithium, anticonvulsants and chlorpromazine do lead to an increased rate of congenital anomalies as well as neonatal problems. Studies of longer-term neurobehavioural sequelae of psychotropic medications are very limited, but at present do not indicate any adverse effects.

CONCLUSIONS

While some psychotropes are associated with congenital anomalies and perinatal complications, mental illness per se may also be associated with an adverse outcome in the infant. Clearly, the risks to both mother and infant need to be carefully weighed and discussed with the parents.

Neural network subtyping of depression

OBJECTIVE

To examine the applicability of a neural network classification strategy to examine the independent contribution of psychomotor disturbance (PMD) and endogeneity symptoms to the DSM-III-R definition of melancholia.

METHOD

We studied 407 depressed patients with the clinical dataset comprising 17 endogeneity symptoms and the 18-item CORE measure of behaviourally rated PMD. A multilayer perception neural network was used to fit non-linear models of varying complexity. A linear discriminant function analysis was also used to generate a model for comparison with the non-linear models.

RESULTS

Models (linear and non-linear) using PMD items only and endogeneity symptoms only had similar rates of successful classification, while non-linear models combining both PMD and symptoms scores achieved the best classifications.

CONCLUSIONS

Our current non-linear model was superior to a linear analysis, a finding which may have wider application to psychiatric classification. Our non-linear analysis of depressive subtypes supports the binary view that melancholic and non-melancholic depression are separate clinical disorders rather than different forms of the same entity. This study illustrates how non-linear modelling with neural networks is a potentially fruitful approach to the study of the diagnostic taxonomy of psychiatric disorders and to clinical decision-making.

Non-melancholic depression: the contribution of personality, anxiety and life events to subclassification

BACKGROUND

We sought to develop a clinically useful subtyping system for the non-melancholic depressive disorders, and here we assess one weighted to central aetiological factors.

METHODS

We studied 185 patients meeting DSM-III-R and/or clinical criteria for non-melancholic depression. Data were obtained by self-report, interview of patients and from corroborative witnesses. We developed a set of variables for class definition, assessing: (i) 'P', disordered personality as a vulnerability factor; (ii) 'A', meeting criteria for a lifetime anxiety disorder or positive on probe questions about trait anxiety characteristics, so assessing anxiety as a vulnerability factor; and (iii) 'L', psychiatrist and consensually-rated life event stress prior to depression onset.

RESULTS

A latent class analysis generated a four-class solution for the P-A-L variables. Life event stressors had similar item probabilities across all four classes, and did not influence the four-class 'P-A' solution when deleted from the analysis, suggesting that life event stress may act more as a general provoking agent, rather than constituting any distinct 'reactive' or 'situational' depression class. Three classes generated clinically meaningful groupings, reflecting varying contributions of anxiety and disordered personality functioning, and with evidence of differential outcome over the following 12 months.

CONCLUSIONS

We suggest that a refined aetiologically-weighted model may assist definition of the non-melancholic depressive disorders, and provide the logic for exploring the comparative utility of differing treatments to identified vulnerability-based classes.

An aetiological model of non-melancholic depression: study design and validity of the measures

OBJECTIVE

The aim of this paper is to describe an approach to sub-typing non-melancholic depression and to determine which raters from a variety of backgrounds provided the most valid information on study variables.

METHOD

A sample of non-melancholic depressed patients is described. Multiple raters (i.e. patients, psychiatrists, referrers and corroborative witnesses) completed measures of the patient's trait anxiety levels, severity of recent life event stressors and personality functioning.

RESULTS

The study and representative data are reported. Congruence between several measures employed indicated that psychiatrist rating of disordered personality was superior to corroborative witness report. Assessment of anxiety traits indicated reasonable agreement between referrers and corroborative witnesses but poor agreement between those ratings and interview-elicited ratings. There were also discrepancies in quantifying 'severity' of life event stress, with patients and their corroborative witnesses rating such events as more severe than either the interviewing psychiatrist or psychiatrists involved in consensus rating sessions. Importantly, the psychiatrists' capacity to quantify the relative contribution of disordered personality, anxiety and life-event stress to the particular depressive episode was supported.

CONCLUSIONS

Results indicate some of the difficulties in operationalising determinants that may contribute to and sub-type the non-melancholic depressions, and demonstrate the advantages of using a range of rating strategies and raters. In this study, psychiatrist-generated judgements are clearly favoured, although the advantages of also assessing trait anxiety and life-event stress impact by self-report strategies are conceded. Some techniques for estimating the contribution of disordered personality function, anxiety and life-event stress are offered for both their research and their clinical utility.

Disordered personality style: higher rates in non-melancholic compared to melancholic depression

OBJECTIVE

We test whether there is differential representation of disordered personality function across melancholic and non-melancholic depressive sub-types, with levels of differentiation examined against differing sub-typing measures.

METHOD

In a sample of 245 subjects meeting criteria for a non-psychotic major depressive episode lasting less than 2 years, we examine for differential rates of disordered personality style across melancholic and non-melancholic depression, using four differing sub-typing measures (i.e., DSM-III-R and DSM-IV criteria, Newcastle Index, and the CORE measure). Disordered personality was assessed by psychiatrist ratings of 15 differing personality styles underpinning disorder classes, and several parameters and domains that reflect the arenas whereby disordered personality may be manifested.

RESULTS

However defined, those with non-melancholic depression were distinctly more likely to rate as showing disordered personality function, with over-representation to an avoidant personality disorder style being the most consistently and strongly identified personality disorder class. Because of limitations to several of the sub-typing measures, we focus on the DSM-IV system. Discriminant function analyses indicated that those so identified as having non-melancholic depression were significantly more likely to rate as having avoidant and schizoid personality styles. A principal components analysis of our 15 differing personality disorder classes identified three molar classes corresponding to the three-cluster DSM system. Again, non-melancholic (compared to melancholic) subjects were more likely to return higher scores: in order, an 'anxious and fearful' (Cluster C) personality style, then an 'eccentric' Cluster A, and somewhat less clearly, a 'dramatic' Cluster B personality disorder style. Specificity to the non-melancholic depressive class was again suggested in relation to a large number of the parameters and domains measuring disordered personality function.

CONCLUSIONS

Disordered personality function appears distinctly more likely in non-melancholic, compared to melancholic depression.

Dysfunctional parenting: over-representation in non-melancholic depression and capacity of such specificity to refine sub-typing depression measures

We examine the proposition that dysfunctional parenting is more likely to be experienced by those with non-melancholic (compared to melancholic) depression and that, as a consequence, such specificity allows the validity of varying definitions of melancholia to be examined and their utility sharpened. We study a sample of 245 non-psychotic patients meeting DSM-III-R criteria for a major depressive episode and assign them to melancholic and non-melancholic sub-sets according to five separate sub-typing measures (DSM-III-R; DSM-IV; Newcastle; 'Clinical' and CORE criteria). We assess dysfunctional parenting by use of the Parental Bonding Instrument (PBI), and by structured psychiatrist assessment and self-report ratings of a range of dysfunctional parental experiences, with independent assessment and self-report ratings of a range of dysfunctional parental experiences, with independent assessment of the last by reports from corroborative witnesses and from the patients' referring therapists. The five sub-typing measures assigned varying percentages of the sample (24-42%) to a 'melancholic' sub-type. When Newcastle Index assignments were examined, there was no evidence that dysfunctional parenting had any specificity to non-melancholic depression. Neither the DSM-III-R nor DSM-IV systems demonstrated specificity in relation to PBI scores, but several interview-assessed dysfunctional parenting characteristics were over-represented in their non-melancholic sub-sets. 'Clinical' definition showed the greatest over-representation of dysfunctional parenting to those assigned as having non-melancholic depression. The CORE measure, a behaviourally weighted measure of psychomotor disturbance, was the next most differentiating. Importantly, those assigned as having non-melancholic depression by all five measures were more likely to be rated by corroborative witnesses as being exposed to anomalous parenting, validating the subjects' self-reports, arguing against results being an artefact of clinician-based assessment, and supporting the specificity of dysfunctional parenting to a depressive sub-type.

The development of a refined measure of dysfunctional parenting and assessment of its relevance in patients with affective disorders

BACKGROUND

The Parental Bonding Instrument (PBI) measures fundamental parenting dimensions (care and over-protection), but does not directly assess abusive parenting.

METHODS

We describe the development of the Measure of Parenting Style (the MOPS), comprising refined PBI scales assessing parental indifference and over-control, as well as a scale assessing parental abuse.

RESULTS

We examine psychometric properties of the MOPS, while several analyses build to the concurrent validity of the abuse scale as an experimental measure. We examine the extent to which both the PBI and the MOPS scales showed specificity of dysfunctional parenting to the non-melancholic depressive subtype, and across a range of anxiety disorders. Non-melancholic depressed patients returned anomalous parenting scale scores (compared to melancholic subjects), but only when such subtyping decisions were clinician-generated. Those receiving DSM-III-R lifetime anxiety diagnoses of panic disorder and of social phobia returned higher PBI protection and MOPS over-control scores than non-anxious subjects, while differences were not established for those with generalized anxiety disorder or obsessive compulsive disorder.

CONCLUSIONS

We consider the likely utility of the MOPS scale and note the module capacity of separate MOPS and PBI scales, which allow a set of options for assessing perceived parenting characteristics.

Increased rate of psychosis and psychomotor change in depression with age

BACKGROUND

We examined the phenomenology of depression in younger (< 60 years old) versus older (> or = 60 years) subjects and, more specifically, the interaction between age and psychomotor disturbance associated with depression.

METHOD

Two hundred and eighty-five patients with a DSM-III-R diagnosis of unipolar major depression referred to a mood disorders unit were assessed using the CORE rating scale, a sign-based system for defining melancholia. Subjects were also assessed using the Hamilton Rating Scale for Depression, Zung Depression Scale, Newcastle Endogenous Depression Inventory and the General Health Questionnaire.

RESULTS

The total CORE score (and each of its subscales) was found to interact with age. Rates of psychotic and melancholic depression increased with age. Elderly depressives suffered more severe depression (higher HRSD scores), appetite loss and weight loss. Level of psychomotor disturbance and rates of psychosis did not differ between those elderly subjects with an early onset (before the age of 60 years) and those with a late onset (at or after 60 years) of depression.

CONCLUSIONS

There appear to be robust phenomenological differences in depression between older and younger subjects. The association between age and psychomotor change may assist our understanding of the neurobiology of depression.

Distinguishing psychotic depression from melancholia

We contrast 25 patients with "psychotic depression" (PD) against two age- and sex-matched groups of melancholic depressed patients. In terms of clinical features, specificity of PD was suggested for several features, including delusions, morbid cognitions (involving guilt and a sense of deserving punishment), hallucinations and constipation. In addition, the PDs had significantly higher levels of behaviourally rated psychomotor disturbance. A comprehensive list of risk factors to depression (e.g., socio-demographic, family history, parental influences, medical disorders, anxiety, stressors and personality style) were examined, without clear differentiation between the comparison groups apart from the suggestion that being a "worrier" and having tenuous stability under stress was over-represented in the PDs. Findings favour the view that psychotic depression is a sub-type of melancholic depression (accounting for the similar expression of the majority of clinical and possible aetiological variables across our contrasted "types'). Findings also suggest possible benefits from future phenomenological studies of psychotic depression relying more on observer-based rather than self-report or symptom data sets. Aetiological studies would benefit from focussing on those features identified as distinguishing the condition from melancholic depression.

Plumbing the depths: some problems in quantifying depression severity

We first report further studies of the psychometric properties of the AUSSI), a brief (11-item) self-report depression measure assessing mood state severity and disability in a sample of 270 subjects. Total scores (combining mood state and disability) correlated well with another self-report measure of depression severity, the Beck, but minimally with both Hamilton and DSM-III-R Global Assessment of Function (GAF) scores. For a sub-sample where we had complete data sets (including a corroborative witness [CW] form of the AUSSI), we established moderate levels of agreement between self-report and CW AUSSI scores, important in supporting validity of its disability sub-scale. Against expectation, neither mood nor disability AUSSI scale scores differed across clinically meaningful depressive sub-types, while differentiation was demonstrated for the Hamilton and GAF measures. Differentiation for the Hamilton measure was clearly driven by several type-specific component clinical features, but also presumably contributed to by the clinician-rater format. Our study raises questions about the varying reference bases and semantics used by patients, corroborative witnesses and clinicians in rating depression severity, and which may generate similar or contrasting severity estimates across varying depressive sub-types.

Defining the personality disorders: description of an Australian database

OBJECTIVE

We seek to improve the definition and classification of the personality disorders (PDs) and derive a large database for addressing this objective.

METHOD

The paper describes the rationale for the development of a large set of descriptors of the PDs (including all DSM-IV and ICD-10 descriptors, but enriched by an additional 109 items), the design of parallel self-report (SR) and corroborative witness (CW) measures, sample recruitment (of 863 patients with a priori evidence of personality disorder or disturbance) and preliminary descriptive data.

RESULTS

Analyses (particularly those comparing ratings on molar PD descriptions with putative PD dimensions) argue for acceptable reliability of the data set, while both the size of the sample and the representation of all PD dimensions of interest argue for the adequacy of the database.

CONCLUSIONS

We consider in some detail current limitations to the definition and classification of the PDs, and foreshadow the analytic techniques that will be used to address the key objectives of allowing the PDs to be modelled more clearly and, ideally, measured with greater precision and validity.

Sub-typing depression, I. Is psychomotor disturbance necessary and sufficient to the definition of melancholia?

Melancholia is most commonly distinguished from non-melancholic depression by the presence of psychomotor disturbance (PMD) and a set of 'endogeneity' symptoms. We examine the capacity of an operationalized clinician-rated measure of PMD (the CORE system) to predict diagnostic assignment to 'melancholic/endogenous' classes by the DSM-III-R and Newcastle systems. Examining a pre-established CORE cut-off score (> or = 8) against independent diagnostic assignment, PMD was present in 51% of those assigned as melancholic by DSM-III-R, and 85% of those assigned as endogenous by the Newcastle system, quantifying the extent to which it is 'necessary' to the two definitions of 'melancholia'. Additionally, multivariate analyses established that the addition of a refined set of historically suggested endogeneity symptoms added only slightly to overall discrimination of melancholic and non-melancholic depressives. While only few endogeneity symptoms independent of psychomotor disturbance were suggested, their specific relevance varied against system definition of melancholia (appetite/weight loss and terminal insomnia being identified for DSM-III-R; anhedonia for Newcastle; and diurnal variation in mood and energy for both systems). Results allow consideration of the relative importance of two domains (psychomotor disturbance and 'endogeneity' symptoms) to clinical definition of melancholia, and have the potential to assist both classification and pursuit of neurobiological determinants. We interpret findings as suggesting a 'core and mantle' model for conceptualizing the clinical features of melancholia, with psychomotor disturbance as the core and with independent endogeneity symptoms as only a thin mantle.

Sub-typing depression, II. Clinical distinction of psychotic depression and non-psychotic melancholia

We have attempted to clarify clinical differentiating features of psychotic depression. Forty-six depressed subjects meeting DSM-III-R criteria for major depression with psychotic features were compared with (i) DSM-defined melancholic, (ii) Newcastle-defined endogenous, and (iii) a residual DSM-defined major depressive episode group. Additionally, a 'bottom up' latent class analysis (LCA) suggested a larger sample of 82 'psychotic depressive' subjects, and multivariate analyses contrasted these subjects with both LCA-identified melancholic and all residual depressed subjects. Analyses suggested that, in addition to two features with absolute specificity (delusions and hallucinations), both the DSM-defined and LCA-defined 'psychotic depressive' subjects were significantly more likely to demonstrate marked psychomotor disturbance, to report two morbid cognitions (feeling sinful and guilty; feeling deserving of punishment), as well as be more likely to report constipation, terminal insomnia, appetite/weight loss and (variable across the defined 'psychotic depressive' groups) loss of interest and pleasure. The study identifies a wider set of potentially discriminating clinical variables than previous studies, as well as both indicating the existence and assisting identification of 'true' psychotic depression in the absence of formal psychotic features being acknowledged or elicited.

A clinical perspective on SPECT

The potential clinical utility of SPECT (Single Photon Emission Computed Tomography) brain imaging to clarifying certain diagnostic dilemmas faced by clinical psychiatrists is considered generally and is illustrated by several case vignettes. Three case histories consider dementia vs depressive pseudodementia, two the possibility of a cerebral vasculitis in patients with auto-immune conditions, and two whether the patient had a "type" of depression likely to benefit from a course of ECT. Published studies reviewing the utility of SPECT in dementia, depression, depressive "pseudodementia" and cerebral lupus are considered. It is suggested that SPECT is an important investigatory technique providing additional information that may assist some diagnostic decisions, while its utility in assisting other clinical decisions awaits clarification.

Subcortical hyperintensities on magnetic resonance imaging: clinical correlates and prognostic significance in patients with severe depression

In 39 hospital inpatients with severe primary depressive disorders, we evaluated the relationships between subcortical hyperintensities on magnetic resonance imaging (MRI) and clinical features, neuropsychological impairment and response to standard therapies. Both white matter and gray nuclei lesions were associated with older age and the absence of a family history of affective disorder. White matter hyperintensities were also associated with onset of first affective episode after the age of 50 years and impaired psychomotor speed. Most importantly, the severity of white matter hyperintensities predicted a poorer response to treatment (r = -0.44, p < .01). Negative correlations of the same order were detected in those (n = 20) who received electroconvulsive therapy (r = -0.42, p = .06) and those (n = 19) who received pharmacotherapy alone (r = -.49, p < .05). This study provides preliminary evidence supporting the clinical and prognostic significance of extensive white matter hyperintensities in patients with severe depression.

Uptake of 99mTc-exametazime shown by single photon emission computerized tomography in obsessive-compulsive disorder compared with major depression and normal controls

Twelve patients with obsessive-compulsive disorder (OCD) were investigated at rest using single photon emission computerized tomography with 99mTc-exametazime. The uptake of 99mTc-exametazime was expressed relative to calcarine/occipital cortex. Patients were matched for drug treatment with 12 patients with a major depressive episode and the patient groups were compared with a control group. Significant bilateral decreases in tracer uptake were confined to basal ganglia in the OCD group. There was a paradoxical positive correlation between anxiety ratings and tracer uptake to basal ganglia in the OCD group. The findings confirm that the functional topography of OCD implicates altered function in the basal ganglia.

Treatment resistant depression in an Australian context. I: The utility of the term and approaches to management

The concept of "treatment resistant depression" (TRD) has generally been defined in terms of failure to respond to a standard course of somatic therapy with little reference to diagnostic sub-types or relevant psychosocial factors. In this paper we examine problems with the use of the term "treatment resistant depression" and then outline an approach to TRD employed in an Australian mood disorders unit. After discussing the need for a biopsychosocial assessment, multimodal management strategies for melancholic and non-melancholic TRD patients are described.

State changes in brain activity shown by the uptake of 99mTc-exametazime with single photon emission tomography in major depression before and after treatment

Twenty-eight patients with a major depressive episode previously investigated at rest using Single Photon Emission Tomography (SPET or SPECT) with 99mTc-exametazime, were followed up at an interval of 9-28 months with the same investigation after full recovery. All patients were unipolar and were rated on the Newcastle scale and with the 17-item Hamilton scale. The uptake of 99mTc-Exametazime was expressed relative to calcarine/occipital cortex. Sixteen patients were scanned when optimally matched for drug treatment (4) or on both occasions drug free (12). The other 12 patients were fully recovered but could not be matched for drug status; these patients showed significantly more retardation, diurnal mood variation and guilt at presentation. Significant bilateral increases in tracer uptake were confined to basal ganglia and inferior anterior cingulate cortex in the matched group, where there were additional increases in thalamus and posterior cingulate cortex on the right side. There were no statistically discernible changes in the neocortex in the matched sample. The unmatched sample yielded inconclusive evidence of increased tracer uptake in left temporal cortex. The findings give a potential focus to the neuropharmacological analysis of depressive illness because the topography of the state change in brain function implicates dopamine function.

The elevation of plasma beta-endorphin levels in major depression

Day-time plasma beta-endorphin/beta-lipotrophic hormone (beta-ENDO/beta-LPH), ACTH and cortisol have been determined in 26 patients with major depression and 25 controls. beta-ENDO/beta-LPH and cortisol were significantly elevated in patients, while ACTH was not. Cortisol levels were significantly negatively correlated with age in controls as were beta-ENDO/beta-LPH and ACTH. In patients, by contrast, cortisol levels were positively, albeit not significantly, related to age. Peptide levels were not related to age in the patient group. Instead, beta-ENDO/beta-LPH was negatively correlated with clinical ratings of symptom severity in patients and positively associated with an acute psychosocial precipitant. The findings cast further light on beta-ENDO/beta-LPH as a measure of hypothalamic-pituitary over activity in depressive illness. A negative association with symptom severity suggests that beta-ENDO/beta-LPH responses are, like those of ACTH, down-regulated in the course of depressive illness.

Proactive interference and the neuropsychology of schizophrenia

Gray, Felden, Rawlins, Hemsley & Smith (1991) have proposed a theoretical model of the neuropsychology of schizophrenia. A major feature of this model is that it is a weakening of the influences of memories of previous input on current perception/learning which is basic to the phenomenon of acute schizophrenia. In the present study, proactive interference (PI) was used as a paradigm to test this hypothesis. PI occurs when new learning is diminished as a consequence of previously learned material. According to our reading of the Gray et al. (1991) model, acutely ill unmedicated patients with schizophrenia should demonstrate reduced PI relative to controls. Ten acutely ill unmedicated patients with schizophrenia, 20 patients suffering from major depressive disorder, and 20 healthy controls were assessed using a PI paradigm. No significant differences in PI emerged between the groups. The results do not support this specific feature of the neuropsychological model of acute schizophrenia proposed by Gray et al. (1991).

A single photon emission computerised tomography study of regional brain function in elderly patients with major depression and with Alzheimer-type dementia

The uptake, at rest, of 99mTc-exametazime into different brain regions was compared using SPECT for 20 elderly subjects with major depressive disorder, 20 with Alzheimer-type dementia, and 30 age-matched normal volunteers. Uptake was referred to calcarine-occipital cortex as a reference sensory area. Cross-sectional differences between the three groups were highly statistically significant, but reflected primarily the reductions in cortical uptake in the Alzheimer group. A detailed comparison of depressed patients and controls identified decrements in anterior cingulate, temporal and frontal cortex and in caudate and thalamus in men only. These decrements were correlated with impairment of performance on a trail-making task, but were also associated with continuing treatment with antidepressants or benzodiazepines. However, most depressed patients had quantitatively normal scans for posterior parietal association cortex, and this suggests that SPECT may find a limited role in the differential diagnosis of depression and dementia. The reduced brain function in some depressed patients may parallel the findings from studies of brain structure in elderly depressives; there was between good outcome at 6-18 months and increased tracer uptake in subcortical areas.

Puerperal affective psychosis: is there a case for lithium prophylaxis?

It is well known that women with a history of manic-depressive or puerperal affective psychosis are at particularly high risk of relapse in the puerperium. This paper describes the use of lithium given during or after pregnancy to women with a history of bipolar illness or puerperal affective psychosis. The rate of puerperal relapse in these subjects was compared with that in a similar group of women not on lithium. The significantly better outcome of the treatment group highlights the need for a prospective controlled trial looking at the effectiveness of lithium in minimising puerperal bipolar relapse.

Single photon emission tomography with 99mTc-exametazime in major depression and the pattern of brain activity underlying the psychotic/neurotic continuum

Forty patients with a major depressive episode were investigated at rest using Single Photon Emission Tomography (SPET or SPECT) with 99mTc-exametazime, an intravenous ligand taken into brain in proportion to regional cerebral blood flow, thereby providing an estimate of regional metabolism. All patients were unipolar and were rated on the Newcastle scale and with the 17-item Hamilton scale. They also completed a range of neuropsychological tests. They were compared with 20 control subjects matched for age, gender, premorbid intelligence and education. The uptake of 99mTc-exametazime was expressed for a range of anatomically defined regions of interest relative to calcarine/occipital cortex. The depressed group showed reduced uptake in the majority of cortical and sub-cortical regions examined, most significantly in temporal, inferior frontal and parietal areas. Unexpectedly, there was a strong positive association between uptake and scores on the Newcastle scale, especially in cingulate areas and frontal cortex. After removing the variance attributable to the Newcastle ratings, however, there emerged the expected negative association between Hamilton scores and anterior tracer uptake. The associations between neuropsychological impairment and regional brain uptake of tracer in part reflected the pattern seen with the Newcastle scale: for example, impairment of memory function correlated with higher uptake into posterior cingulate areas. We propose that depressive illness may be characterised by two processes. One leads to an overall reduction in anterior neocortical function, perhaps related to symptom severity. The other mechanism is manifest as relatively increased function, most notably within cingulate and frontal areas of the cerebral cortex in association with psychotic symptoms. The findings offer new understanding of the brain states underlying depressive illness and a potential focus to subsequent neuropharmacological analysis.

Cognitive function in major depression

Forty patients with a major depressive episode were divided into equal endogenous and neurotic sub-groups using the Newcastle scale. They were all rated on the 17-item Hamilton scale and with a variety of neuropsychological tests. They were compared with 20 age- and education-matched control subjects. Both endogenous and neurotic groups had impaired memory function on the auditory verbal learning test; recall and recognition were equally impaired suggesting that effort was not a major determinant of performance. The endogenous group was more impaired on digit symbol substitution and the Trail making test (A and B). Impairment was correlated with symptom scores on the Hamilton and Newcastle scales, even after allowing for the effect of age. It is concluded that the conventional distinction between organic and functional impairment breaks down in severe depressive illness. The implications of this for clinical neuropsychological testing and the anatomy of the brain dysfunction in depressive illness are discussed.

Lithium augmentation in antidepressant-resistant patients. A quantitative analysis

A quantitative analysis was used to examine the efficacy of lithium augmentation in the acute treatment of depressed patients resistant to a standard trial of an antidepressant. Effect sizes were measured by the odds ratio using the Mantel-Haenszel method. Only controlled trials were included in order to minimise bias in method. A highly statistically significant effect for lithium augmentation was found, the pooled odds ratio being 0.146 and its 95% confidence interval 0.05-0.44 (i.e. the odds of remaining ill are reduced by between 56% and 95% with the use of lithium treatment). While these results support the case for lithium augmentation in treatment-resistant depression, there remains considerable uncertainty over the duration of treatment necessary to see and sustain the treatment response.