Male reproductive disorders in humans and prenatal indicators of estrogen exposure: A review of published epidemiological studies, 2006
Estrogens may exert an adverse effect via FSH secreted from the fetal pituitary gland. FSH stimulates multiplication of fetal Sertoli cells, which are crucial for the sperm production after puberty. Exogenous estrogens, i.e. xenohormones, may reduce the number of Sertoli cells produced during fetal life, by enhancing the negative feedback of estrogenic control of the fetal pituitary FSH secretion.
Eight studies focussed on prenatal estrogen exposure and semen quality or time to pregnancy. Five of these were based upon diethylstilbestrol (DES) prescribed during pregnancy. DES is a potent estrogen (10 times more potent than 17β-estradiol) used widely from 1940s to 1971.
Gill et al. and Stenchever et al. found a significantly reduced sperm count in sons of mothers who had received DES from their first pregnancy trimester, while Andonian et al. and Leary et al. did not. Gill’s study encompassed 70% of both the placebo and DES exposed male offspring from the original Dieckmann’s randomised controlled DES trial from 1950 to 1952. The study is also published by Bibbo et al. who presented the preliminary results and by Schumacher et al., who included secondary semen samples from participants who initially were not invited to deliver a semen sample. This latter part of the study carries a risk of selection bias, as pointed out by the author, because of differential participation of DES exposed males with genital anomalies. These three studies have been considered as three independent studies, which they are not.
The study of Stenchever et al. only included 17 DES exposed and 12 non-exposed men, recruited through newspapers. The non-exposed group were students or physicians. To account for selection bias only men without knowledge of their fertility were included. Ascertainment of dose and time of administration is uncertain. We believe that this small study adds little additional information, if any.
The study of Leary et al. found no differences in sperm counts between DES exposed and non-exposed males. The study has a high number of participants and the assessment of the prenatal exposure is based upon medical records from the Mayo clinic, which are almost complete. The differences between Gill and Leary can be explained by the differences in the mean total DES dose (the Gill study 12.046 mg; the Leary study 1.429 mg).
The fifth DES study by Andonian et al. found no differences in semen quality measured by using Eliasson scores (a score based on sperm count, motility, motility grade and percent morphological normal sperm). The study only included 22 exposed and 22 non-exposed recruited through news media and exposure information was only obtained for 55% of the participants through copies of prescription, letters from attending obstetricians or copies of medical records.
It is possible that DES in high dosage given early in pregnancy (in first trimester) lower sperm counts, but it seems not to prolong TTP, as showed by Wilcox et al.. TTP is, however only strongly correlated to sperm counts for lower levels of counts.
In a case control study of 79 men with unexplained infertility and 104 men with normal fertility attending a fertility clinic, Ozturk et al. examined birth weight and semen quality by using mothers pregnancy records. No relation to semen quality was found.
One study examined the semen quality among monozygotic twins and dizygotic twins using single born brothers as a reference group. Although the estrogen level is substantial higher in twin pregnancies no differences in sperm count were found among the groups.
We identified only one semen study based upon prenatal xenoendocrine exposure only, namely a study from Taiwan where contamination of cooking oil with polychlorinated biphenyls (PCB’s) and polychlorinated dibenzofurans (PCDF) took place in 1979. However, lack of exposure assessment, uncertainty with respect to timing of exposure and the limited study size jeopardise causal inference.
In conclusion, few studies that are not flawed by methodological problems consistently indicate that DES administrated in high dose in first trimester (total 12 g) is associated with a 20% reduction of sperm count while DES at lower doses is not.
- Male reproductive disorders in humans and prenatal indicators of estrogen exposure. A review of published epidemiological studies, Reproductive toxicology (Elmsford, N.Y.)., NCBI PubMed PMID: 16005180, 2006 Jan.
- Male factor infertility featured image credit nycivf.