THE HYPOTHALAMIC-PITUITARY-GONADAL AXIS IN MALE PSYCHIATRIC INPATIENTS

THE HYPOTHALAMIC-PITUITARY-GONADAL AXIS IN MALE PSYCHIATRIC INPATIENTS

Bilyana Brdaroska

A thesis submitted in accordance with the requirements for the admission to the degree of Doctor of Medicine of The University of Sydney

Department of Psychological Medicine, August, 2006

Abstract

A large number of neuroendocrine studies indicate a possible relationship between the Hypothalamo-Pituitary-Gonadal (HPG) axis and major depressive illness in men. This observation is not surprising, considering the similarities between the symptom profiles of depression and hypogonadism. However, owing to the strong likelihood that a number of other demographic, clinical and treatment covariates may potentially obscure a

possible relationship between HPG and depression, studies in this area have produced somewhat inconsistent results. The main objective of the present study was to investigate the relationship between depression and HPG hormone levels in a population of

hospitalised men. Another objective was to examine the relationship of a number of demographic, behavioural, clinical and treatment variables with HPG hormone levels and depression.

METHOD: Serum hormones of the HPG axis (Luteinizing Hormone (LH), Follicle Stimulating Hormone (FSH), Free Testosterone (free T), Total Testosterone (total T) and Sex Hormone Binding Globulin (SHBG)) were compared between fifty-two male

patients with Major Depressive Disorder (mean age = 42.04; SD = 14.1) and twenty-five male patients with other psychiatric conditions (mean age = 40.72; SD = 13.8) on

admission into hospital. In addition, to elucidate the possible relationship between clinical outcome of depression and gonadal function, HPG parameters were measured in patients with depression 3 to 6 months following discharge. Based on their HDRS (Hamilton Depression Rating Scale) score, patients were categorised as remitters and non-remitters. Demographic, behavioral, clinical and treatment variables were also

examined as possible correlates of hormone levels. i

RESULTS: Comparison between patients with depression and patients with other diagnoses indicated a significantly lower free T and total T in patients with depression.

There were no differences in other hormone parameters between the two diagnostic groups. Correlational analyses indicated significant negative relationships between free T and total T and severity as well as duration of depression. Age was inversely correlated to both free T and total T, whereas BMI was negatively correlated with Total T and SHBG.

There was a positive relationship between Total T as well as Free T and measures of sexual dysfunction. While no difference in hormone parameters was observed as a function of psychotic features, patients with melancholic features exhibited significantly lower levels of free T and total T compared to patients with no melancholic features. In the multiple regression analyses, age, duration and severity of depression were the strongest predictors of both free and total T. In separate regression analyses somatic features, over and above other features of depression were found to account most in the variability in free T and total T. Longitudinal analysis revealed significantly higher free T and total T levels on follow-up compared to baseline in the patients who remitted. There was no significant change in any of the hormones studies in the non-remitting group.

CONCLUSION: The main findings of the present study support previous results that both total and free testosterone levels are lower during depression and that concentrations of free T and total T parallel changes in severity of depressive symptomatology. Further investigations into the mechanism for this observation, and perhaps examinations of testosterone supplementation for treatment of depression are in order.

ii

Acknowledgements:

I am deeply grateful to my supervisor, Professor Gordon Johnson for his advice and continual support as well as the opportunity to conduct this study at The Northside Clinic.

Professor Johnson provided me with invaluable knowledge, advice and practical

instruction in the methodology of this project as well as clinical background in psychiatry.

It was an honor working with such a well-respected and knowledgeable person. I would also like to express my gratitude to my co-supervisor, Professor David Handelsman, ANZAC Research Institute, who provided assistance in theoretical and practical aspects in the area of andrology.

Thank you to all staff members at The Northside Clinic, particularly the nurses, who were always available for practical information, despite their busy schedules. Also a big thank you to Pauline, Sandra, Lisa and other staff members from Douglas Hanly Moir, who provided much appreciated assistance in phlebotomy, logistis and all the associated blood collection issues. I enjoyed the gossip too!!

Thank you, Dr Glenn Hunt, for providing assistance in research design and statistical issues, and for organizing and encouraging attendance to the monthly PhD meetings.

Looking back now, I realize they were my Number One source of motivation and support as they brought me closer to students going through the same process.

I would also like to thank my family and friends who have listened to my complaints and provided valuable support, friendship and motivation as well as the opportunity for time- out from the thesis. I am grateful to the many people who provided discussions, ideas and

xii

suggestions essential to the planning and execution of this research, notably: Roman Montanary, Beth Kotze, Sophie Kavanagh, Adrianne Whitall and Warren Smith.

And finally, thank you to my husband, and friend, Paul, for the support he provided over the past two years.

xiii

APPENDIX I

Diagnostic Information

Diagnosis Group 1 (n = 52) Group 2 (n = 25)

Major Depressive Disorder 52 0

Schizophrenia 0 13

Generalised Anxiety Disorder 0 6

Panic Disorder 0 6

APPENDIX II

Author’s contribution to the thesis’ research design and methodology

Study design: The design of the study was performed in collaboration and under the guidance of the supervisor and co-supervisor. The author was solely responsible for researching and reviewing the literature, creating the study files and information collection papers.

Recruitment and data collection: The author was responsible for recruitment of all of the patients, which included monitoring when prospective subjects were admitted and approaching patients, obtaining consent form as well as conducting interviews and gathering all necessary information.

Blood collection: The author attended a one-week certified course on venipuncture which allowed her to gain experience and knowledge in venipuncture. The author had major involvement in blood collection: she performed almost all blood collections – there were few patients who it was difficult to collect blood from, in these cases blood collection was performed by a trained nurse employed by the clinic. The author performed centrifugation and pipetting. The author communicated directly with the pathology company regarding logistics and transport of blood samples and results.

Statistical design and analysis: The design, planning and execution of the statistical

analysis were mainly performed by the author, under the supervision and guidance by the

supervisor. The author was responsible for collecting data, data entry and management,

application and interpretation of statistical analyses.

APPENDIX III

Medication Usage

Group 1

(n = 52)

Group 2 (n = 25) SSRI

20 (38.5 %) 6 (24.0 %) SNRI

15 (28.9 %) 0 (0.0 %) Tricyclic

5 (9.6 %)

2 (8.0 %) Tetracyclic

12 (23.1 %)

2 (8.0 %)

RIMA

0 (0.0 %)

1 (4.0 %) Mood stabilizers

10 (19.2 %)

6 (24.0 %) Antipsychotics

19 (36.5 %) 12 (48.0 %) Antianxiety

9 (17.3 %)

7 (28.0 %)