Background: Maternal perinatal mental health has been shown to be associated with adverse consequences for the mother and the child. However, studies considering the effect of DSM-IV anxiety disorders beyond maternal self-perceived distress during pregnancy and its timing are lacking. Aims: To examine the role of maternal anxiety disorders with an onset before birth and self-perceived distress during pregnancy for unfavourable maternal, obstetric, neonatal and childhood outcomes. Study design: DSM-IV mental disorders and self-perceived distress of 992 mothers as well as obstetric, neonatal and childhood outcomes of their offspring were assessed in a cohort sampled from the community using the Munich-Composite International Diagnostic Interview. Logistic regression analyses revealed associations (odds ratios) between maternal anxiety disorders and self-perceived distress during pregnancy with maternal depression after birth and a range of obstetric, neonatal and childhood psychopathological outcomes. Results: Lifetime maternal anxiety disorders were related to offspring anxiety disorders, but not to offspring externalizing disorders. Analyses focussing on maternal DSM-IV anxiety disorders before birth yielded associations with incident depression after birth. In addition, self-perceived distress during pregnancy was associated with maternal depression after birth, preterm delivery, caesarean section, separation anxiety disorder, ADHD, and conduct disorder in offspring. Conclusion: Findings confirm the transmission of anxiety disorders from mother to offspring. Apart from maternal anxiety, self-perceived distress during pregnancy also emerged as a putative risk factor for adverse outcomes. The finding that maternal anxiety disorders before birth yielded less consistent associations, suggests that self-perceived distress during pregnancy might be seen as a putative moderator/mediator in the familial transmission of anxiety.

Die Psychoneuroimmunologie beschäftigt sich unter anderem mit der Frage, ob und unter welchen Bedingungen psychosozialer Stress krank macht. Humanstudien, die dieser Frage nachgehen, können dabei meist nur korrelative Zusammenhänge aufdecken. Um trotzdem Aussagen zu Mechanismen, Ursache-Wirkungsbeziehungen und klinische Relevanz treffen zu können, muss z.B. auf Befunde aus der Tierforschung oder aus in vitro-Studien zurückgegriffen werden. Ziel der vorliegenden Arbeit war es, eine Methode zu finden, welche eine breitere Interpretationsgrundlage für korrelative Befunde aus Humanstudien liefert. Als besonders viel versprechend kann die Untersuchung von Patienten mit Morbus Addison angesehen werden. Patienten mit dieser Erkrankung können aufgrund zerstörter Nebennierenrinden kein Cortisol produzieren. Dieses Fehlen von Cortisol wird medikamentös ausgeglichen. Da bislang keine experimentellen Daten zu den aus der Substitutionstherapie resultierenden freien Cortisolkonzentrationen sowie den endokrinologischen Reaktionen auf Stress vorliegen, wurden in einem ersten Schritt diese beiden Fragestellungen bearbeitet. Anschließend wurde die Methode auf immunologische Fragestellungen angewandt. Die Ergebnisse der durchgeführten Studien zeigen zum einen, dass die medikamentöse Morbus Addison-Therapie in der momentanen Form nicht geeignet ist, den bei Gesunden zu beobachtenden zirkadianen Cortisolrhythmus optimal nachzustellen. Des weiteren bestätigte sich, dass psychosozialer Stress bei Patienten mit Morbus Addison lediglich eine Noradrenalinantwort auslöst, stress-induzierte Anstiege in Cortisol und Adrenalin jedoch ausbleiben. Eine Injektion von 0.03mg/kg Hydrocortison nach einem akuten Stresstest zeigte sich als geeignet, normale stressinduzierte Cortisolanstiege und Maximalwerte nachzustellen. In den untersuchten Immunparametern unterschieden sich gesunde Probanden und Patienten mit Morbus Addison vor Stressinduktion nicht. Stress-bedingte Veränderungen in peripheren Lymphozytenzahlen lassen sich dahingehend interpretieren, dass bei Patienten zwar akut eine Einwanderung von Lymphozyten aus dem Gewebe in das Blut stattfindet, aber Cortisol mitverantwortlich ist für die anschließende Redistribution. Bleibt eine Cortisolstressantwort aus, deutet dieser Befund auf ein erhöhtes Infektionsrisiko hin. Die Verläufe zur stimulierten Produktion des Entzündungsmediators Interleukin-6 stehen im Einklang mit der Hypothese, dass stress-induzierte Cortisolkonzentrationen vor einer überschießenden Entzündungsreaktion schützen. Des weiteren lässt sich bei gesunden Probanden das Absinken der NF-?B-Aktivität nach Stress durch Cortisolwerte direkt nach Stress, bei Morbus Addison-Patienten hingegen durch Veränderungen im Noradrenalin vorhersagen. Das vorliegende Dissertationsprojekt konnte zeigen, dass die Untersuchung von Patienten mit Morbus Addison ein Erfolg versprechender Ansatz ist, um die Gültigkeit von psychoneuroimmunologischen in vitro- und Tierbefunden im intakten menschlichen Organismus zu überprüfen. Die vorliegenden Daten deuten zudem auf eine erstaunliche Anpassungsfähigkeit des Immunsystems hin, was die Notwendigkeit deutlich macht, komplexe psychoneuroimmunologische Prozesse auch im Humanbereich unter Anwendung der Kriterien Hormonsubtraktion und -substitution zu untersuchen. Die Untersuchung von Morbus Addison-Patienten wird in zukünftigen Studien sicherlich entscheidende Hinweise zur klinischen Relevanz einer normalen endokrinen Stressantwort liefern können. / The field of psychoneuroimmunology is among other things dedicated to the question, whether and on what terms psychosocial stress results in disease. Human studies investigating this question often only reveal associations. Consequences regarding mechanisms, causes, and clinical relevance, are thus usually deduced from animal or in vitro studies. The aim of the present thesis was to find an approach, which provides a broader basis for interpretations of correlative findings from human studies. The investigation of patients with Addison´s disease was regarded as the most promising approach. Due to destroyed adrenal glands, these patients are not able to produce cortisol, which therefore has to be substituted. Since no experimental data regarding free cortisol levels resulting from substitution therapy as well as endocrine stress responses are available, in a first step these two question formulations were investigated. Subsequently, this approach was used to investigate two immunological question formulations. The present studies found patients with Addison´s disease to be over-treated. Psychosocial stress resulted in noradrenaline but not in cortisol or adrenaline responses. An injection of 0.03mg/kg hydrocortisone after stress was suitable to induce increases in cortisol levels and cortisol maximums comparable to healthy subjects. Healthy subjects and patients with Addison´s disease did not differ in any baseline immune measures. Stress-induced changes in lymphocyte numbers suggest cortisol being necessary for lymphocyte redistribution subsequent to stress-induced migration into peripheral blood. Without stress-induced cortisol increases, patients are at higher risk for infectious diseases. The trajectories of stimulated interleukin-6 production support the hypothesis of stress-induced cortisol levels protecting the organism against an over-reacting inflammatory immune reaction. Furthermore, in healthy subjects post-stress cortisol levels predicted stress-induced decreases in NF-´B activity, whereas in patients with Addison´s disease noradrenaline predicted such changes. The results of the present thesis supports the assumption of Addison´s disease being a promising approach to test the validity of psychoneuroimmunological in vitro and animal data in human whole organism. Additionally, the present data emphasize the astonishing adaptability of the immune system. This further emphasizes the necessity to investigate psychoneuroimmunological processes utilizing the criteria of hormone subtraction and hormone substitution also in human research. In future studies investigation of patients with Addison´s diseases will certainly provide crucial evidence regarding the clinical relevance of a normal endocrine stress response.

Background: A recent epidemiological analysis on premenstrual dysphoric disorder (PMDD) in the community revealed increased rates of DSM-IV posttraumatic stress disorder (PTSD) among women suffering from PMDD. Aims: To explore whether this association is artifactual or might have important pathogenic implications. Methods: Data come from a prospective, longitudinal community survey of an original sample of N¼1488 women aged 14–24, who were followed-up over a period of 40 to 52 months. Diagnostic assessments are based on the Composite International Diagnostic Interview (CIDI) using the 12-month PMDD diagnostic module. Data were analyzed using logistic regressions (odds ratios) and a case-by-case review. Results: The age adjusted odds ratio between PTSD and threshold PMDD was 11.7 (3.0–46.2) at baseline. 10 women with full PTSD and at least subthreshold PMDD were identified at follow-up. Most reported an experience of abuse in childhood before the onset of PMDD. Some had experienced a life-threatening experience caused by physical attacks, or had witnessed traumatic events experienced by others. 3 women reported more than one traumatic event. Conclusions: A case-by-case review and logistic regression analyses suggest that women with traumatic events and PTSD have an increased risk for secondary PMDD. These observations call for more in-depth analyses in future research.