Date of Award


Degree Type


Degree Name

Master of Science (Human Biology)


Medical and Health Sciences

First Advisor

Associate Professor Peter Roberts

Second Advisor

Dr Phillip Matson

Third Advisor

Dr Narelle Hadlow


The first trimester of pregnancy is a dynamic time of change involving the implantation and development of the fetus, together with a wide range of related endocrine changes arising from the ovary and feto-placental unit, as well as other changes in maternal and fetal physiology. The present longitudinal study has investigated two main areas of endocrinology in circulating maternal serum from positive pregnancy test to the detection of a fetal heartbeat, namely (i) AMH and (ii) thyroid function. Women attending Fertility North and conceiving during fertility investigation or treatment underwent phlebotomy twice weekly and AMH, TSH, fT3 and fT4 levels were measured, in addition to the reproductive hormones oestradiol (E2), progesterone (P4) and human chorionic gonadotropin (hCG).

AMH is a well-established biomarker for assessing the age-related decline of the oocyte pool in healthy women, but has many other clinical uses including the assessment of polycystic ovary syndrome (PCOS), estimating response to ovarian stimulation with exogenous gonadotrophins, detection of granulosa cell tumours and identification of premature ovarian insufficiency. AMH also influences follicle growth, but the precise mechanism by which it does this is unknown and its role during pregnancy has not been adequately characterised.

In this study the key findings regarding AMH were the differences between preovulatory and luteal phase AMH levels in pre pregnant women and the three distinct patterns of changing AMH noted in women once pregnancy was confirmed. Prior to pregnancy, mean AMH levels peaked at time of ovulation (gestation week 2) and then dropped at mid-luteal phase (gestation week 3). 86.1% of women had their maximum AMH level at or before ovulation, with 13.9% of women having their maximum AMH at mid-luteal phase. However, once pregnancy was achieved, AMH levels for each woman did not remain constant, but moved significantly away from their first non-pregnant measurement, with a trend in viable pregnancies of an AMH level that either consistently increased or decreased from gestation week 4 (time of first positive hCG measurement) through to week 7. The difference between women with rising AMH compared to those with falling AMH significantly increased (p=0.000) during early pregnancy. In general terms women with rising AMH levels, had continually rising AMH iii levels and these levels became increasingly divergent from those women who had falling AMH levels and whose AMH levels continued to decrease. In contrast, AMH levels in women with non-viable pregnancies were very inconsistent, with no definitive rising or falling trend observed. Instead, AMH levels in women with non-viable pregnancies showed erratic and sporadic changes, both rising and falling in the same individual from gestation weeks 4 to 7. There was a significant negative correlation between baseline (day 2 of the menstrual cycle) AMH and patient age (r=-0.507, p=0.000). No associations were observed between AMH and patient body mass index (BMI), stimulation medications (follicle stimulating hormone dose), treatment type or fetal sex. There were weak negative correlations between AMH and P4 (r=-0.220, p=0.000,) and TSH (r=-0.155, p=0.001), but not E2, hCG, fT3 or fT4. This study was not designed to determine causative factors for the AMH changes observed and further investigations would be required to address possible causes of these reported changes.

The importance of normal maternal thyroid function in pregnancy and fetal development is well characterised and thyroid dysfunction can result in adverse effects on the unborn child, including a higher risk of miscarriage. Throughout the first trimester, the fetus is dependent on maternal thyroxine as the fetal thyroid is not fully functional until approximately gestation week 16. Despite this important role, the changes, if any, occurring during this vibrant period are poorly documented and often are limited to cross-sectional sampling.

In this study the key findings regarding thyroid function were of stable fT3, fT4 and TSH levels between gestation weeks 0 to 4 (pre pregnancy) with subsequent changes in thyroid function once pregnancy was established at gestation week 4 (hCG >25mU/ml). From gestation weeks 4 to 6.5 some significant changes in thyroid hormone levels were observed with a gradual decrease in fT3 (r=-0.104, p=0.005) and TSH levels (r=-0.123 p=0.013). In contrast, fT4 levels remained constant during early pregnancy (gestation weeks 4 to 6.5). No markers of thyroid function appeared to affect pregnancy outcome. Despite TSH levels ranging from 0.27 - 4.93mU/L, only 6 patients (7.0%) had TSH levels >4.0mU/L, of whom only 3 (3.5%) miscarried. Reference ranges were calculated for fT3, fT4 and TSH for gestation weeks 4 to 6.5 and were iv found to be comparable to later first trimester ranges reported in other studies. There were two patients who tested positive for thyroid peroxidase antibody (TPOAb) who were excluded from the study (both were viable pregnancies with normal thyroid hormone levels).

In summary, this study was able to directly follow the changes in AMH and thyroid hormone levels within the same individual over time, resulting in the identification of unique changes in very early pregnancy.