Abstract
Establishing the proportion of fertilized oocytes and early human embryos that proceed to term may help policy makers in their evaluation of when the life of a new human individual begins and in determining the nature of protection to be accorded to it. The rate of spontaneous abortions, although increasing with age, overall does not exceed 15%. However, abortion rates refer only to ‘clinical pregnancy’, whereas early embryonic loss is more common than generally believed. Evidence of such wastage comes from many sources. Human fecundity rarely exceeds 35% and may be decreasing due to deterioration in semen quality. Embryological studies show that 50% of randomly recovered preimplantation embryos have severe anomalies. The study of sensitive markers of pregnancy, such as human chorionic gonadotrophin, indicates early embryo wastage in the order of 50%. Pregnancy wastage may be a function of the time lapse between ovulation and implantation as the implantation window extends between menstrual cycle days 20 and 24. Finally, data obtained with natural IVF cycles also indicate major losses, with an overall pregnancy rate of 7.2% per cycle and 15.8% per transfer. These data, however, are biased by a high cancellation rate and low oocyte retrieval in natural IVF cycles.
Establishing the proportion of fertilized oocytes and early human embryos that, under physiological conditions, proceed to term may help ethicists, philosophers and policy makers in their evaluation of when the life of a new human individual begins and in determining the nature of protection to be accorded to it at the various stages of development. In this respect, the rate of spontaneous abortions, although increasing with age, overall does not exceed 15%. The situation however, is much more complex, since abortion rates refer only to ‘clinical pregnancy’ and evidence has been accumulating that early embryonic loss among fertile women is much more common than generally believed. Evidence of early embryo wastage is of complex nature and comes from many sources: first, human fecundity rarely exceeding 35% and may be decreasing due to deterioration in semen quality; second, embryological studies showing that 50% of pre-implantation embryos had severe anomalies; third, sensitive markers of pregnancy indicate embryo wastage in the order of 50%, during the first 2 weeks of gestation. There is also evidence that pregnancy wastage may be a function of the time lapse between ovulation and implantation; the implantation window seems to extend between menstrual cycle day 20 and 24. Finally, there are data obtained with natural IVF cycles indicating major losses, with an overall pregnancy rate of 7.2% per cycle and 15.8% per transfer in a meta-analysis. These data however, are biased by a high cancellation rate and low oocyte retrieval in natural IVF cycles.
Keywords
Introduction
‘When exactly does a new human life begin?’ is probably the most fundamental question of reproductive biology. It is a question that permeates every aspect of bioethics of human reproduction and to which very different responses have been provided, depending on the philosophical or theological starting point selected. Assisted reproduction technologies, preimplantation genetic diagnosis (PGD), embryo freezing, emergency contraception and other recent advances in reproductive biology, such as use of cybrids for the prevention of diseases based on mitochondrial disorders and the use of chimeras have moved the debate on the definition of the beginning of human life from theoretical discussion to practical application (
Benagiano et al., 2010
). An answer to the question has now also become legally relevant in view of the need felt in most countries to regulate assisted reproduction treatment producing legislation that, even implicitly, utilizes some kind of definition of the status of a human embryo.One aspect of the problem that often seems to go unnoticed is what happens physiologically to a fertilized oocyte, because the public generally assumes that, once fertilization has taken place, most oocytes easily progress until they reach the stage of a viable fetus. This presumption is based on the clinical observation that spontaneous abortions after the 12th week of gestation, although increasing with age, in younger women do not exceed 15%. Miscarriage rates range from about 10% at age 20 to 50% at age 40–44 and more than 90% for women 45 years of age or older (
Nybo Andersen et al., 2000
).At close scrutiny however, the situation is much more complex, since abortion rates refer only to what is usually called clinical pregnancy, namely a pregnancy in which an ultrasound scan demonstrating an embryonic heart action, has confirmed definitively that a woman is pregnant. During the 2–3 weeks that precede the recognition of a live embryo, i.e. an ongoing gestation, much can happen and, indeed, evidence has been accumulating that early embryonic loss among fertile women is much more common than lay wisdom believes.
Over the last 50 years, a number of scientific publications have produced evidence of different nature showing a major pre-clinical human pregnancy wastage. As far as is known, this is the first systematic review of this evidence, which is of complex nature and from many sources. It is, therefore, hoped to provide a useful summary of present knowledge of the fate of a zygote/embryo in the first 2 weeks of its existence.
The importance of this information lies in the fact that, when debating ethical issues related to the beginning of a new human life, biological facts must be taken into consideration, because whereas biology cannot alone answer the question, no correct answer can be given if one ignores biological facts. It is proposed to hold such a debate in a subsequent paper, with the help of experts in bioethics.
Assessing embryo loss after fertilization
Currently, there are many observations, scientific experiments and objective data which demonstrate the existence of significant wastage of human fertilized ova and early embryos during the first 2 weeks after fertilization (i.e. before the first period is missed). This evidence is relatively new, because, until the middle of the 20th century, common wisdom believed that – since congenital anomalies at birth are fortunately rare () – conception was almost universally followed by the development of an embryo and then a fetus that, in the great majority of instances, would reach term and result in the birth of a healthy child. The reason for this major difference of opinion is methodological: 50 years ago, wastage could only be measured by establishing the total rate of miscarriages which, in many demographic surveys, did not exceed 15% and in some cases indicated even lower figures (
Bocciolone et al., 1989
). These percentages are influenced by maternal age and genetic and environmental factors, but if one considers only miscarriage of clinical pregnancies including late miscarriages, the inevitable conclusion appears to be that at least 80–85% of all conceptions become a viable fetus.Today it is known that the situation is much more complex and, before reaching any conclusion, a number of important potentially confounding variables must be analysed.
Human fecundity
When calculating early pregnancy it is necessary to take into account human fecundity, defined as ‘the probability to produce a vital term newborn per menstrual cycle during which there was normal sexual activity’ (
Olsen and Rachootin, 2003
). Data collected in the last few decades indicate that human fecundity rarely exceeds 35% and, even under ideal conditions, the greatest probability of achieving a clinical pregnancy per cycle is around 30–40% (Balakrishnan, 1979
, Charbonneaux, 1970
, Henripin, 1954
, Sheps, 1965
, Slama et al., 2002
, van Noord-Zaadstra et al., 1991
, Vessey et al., 1976
, Wang et al., 2003
, Zinamen et al., 1996
; Table 1).Table 1Human fecundity: the proportion of women who will deliver a term baby per menstrual cycle during which there were frequent acts of non protected intercourse.
| Population studied | Publication | Fecundity index |
|---|---|---|
| France, 17th and 18th centuries | Charbonneaux, 1970 | 0.21 |
| Peru, 20th century | Balakrishnan, 1979 | 0.17 |
| Mexico, 20th century | Balakrishnan, 1979 | 0.21 |
| USA, 20th century (Hutterite sect) | Sheps, 1965 | 0.28 |
| Canada (Québec), 18th century | Henripin, 1954 | 0.31 |
| Great Britain, 20th century (pluriparous women trying to conceive) | Vessey et al., 1976 | 0.21 |
| USA, 20th century | Zinamen et al., 1996 | 0.30 |
| The Netherlands, 20th century after 12 months and age 31 | van Noord-Zaadstra et al., 1991 | 0.54 |
| China, 20th century | Wang et al., 2003 | 0.40 |
Although successful human reproduction may be influenced by factors like the incidence of infertility in a given population, clearly there is a major discrepancy between the frequency of fertilization and that of implantation, miscarriage and live birth.
Some 35 years ago, two British researchers were among the first to point out this reality: in an article in The Lancet entitled ‘Where have all the conceptions gone?’ (
Roberts and Lowe, 1975
). Before them, in a speculative essay, Bishop, 1964
judged that in mammals there are several causes of prenatal loss and that much of what he called the ‘basal loss’ was the result of the selective elimination of genetically abnormal embryos. Any such loss should be regarded as unavoidable and indeed desirable as it prevents the birth of abnormal offspring and because the loss occurs early in gestation, thereby creates the opportunity for a further mating in animals with only one embryo (Wilmut et al., 1986
). In the human, this probably represents the principal mechanism through which nature prevents the incorporation of negative features created through mutations or meiotic errors, into the inheritable characteristics of that species. The merit of Roberts and Lowe, 1975
was undertaking for the first time a statistical estimation of early pregnancy wastage. Although their calculations can be criticized in a number of ways, the results are impressive: they estimate that at least 75% of all conceptions do not become a viable fetus or proceed to term (Table 2).Table 2Estimated fetal loss in England and Wales during 1971: married women aged 20–29 years.
| n (millions) | |
|---|---|
| Married women aged 20–29 | 2437 |
| Annual acts of coitus (assuming a mean of twice a week) | 253,448 |
| Annual acts of unprotected coitus (assuming one in four is unprotected) | 63,362 |
| Unprotected acts occurring within 48-h period around ovulation | 4526 |
| Assume one in two of these results in fertilization | 2263 |
| Actual number of infants (live and stillborn) born to these women | 505 |
| Estimated loss (2263–505) | 1758 |
| Percentage loss (1758/2263 × 100) | 78% |
Data reproduced from
Roberts and Lowe, 1975
with permission.a Derived from
Nishimura, 1970
.- Nishimura H.
Fate of human fertilized eggs during prenatal life: present status of knowledge.
Okajimas Folia Anat. Jpn. 1970; 46 (mentioned in: Nishimura, H., 1970. In: Fraser, F.C., McKusick, A. (Eds.), Proceedings of the IIIrd International Conference on Congenital Malformations. Excerpta Med, Amsterdam): 297-305
Another important factor in determining human fecundity is contained in reports of apparently decreased semen quality over the last 50 years. It is notable that despite an abundance of data, this apparent decline is still fiercely contested. Concern was raised for the first time some 30 years ago in Europe (
Bostofte et al., 1983
) and, since then, disturbing data have emerged from a number of European countries. The Danish group of Skakkaebek (Carlsen et al., 1992
) investigated whether semen quality had changed over the previous 50 years, reviewing a total of 61 papers published between 1938 and 1991, reporting on semen quality in 14,947 men without a history of infertility. They concluded that, taken together, these studies showed evidence of a significant decrease in mean sperm count from 113 × 106 ml in 1940 to 66 × 106 ml in 1990 (P < 0.0001) and in seminal volume from 3.40 ml to 2.75 ml (P = 0.027). Subsequently, observations in individual countries – Belgium (Van Waeleghem et al., 1996
), Italy (Bilotta et al., 1999
) and Slovenia (Zorn et al., 1999
) – confirmed these conclusions across Europe. In addition, the incidence of genitourinary abnormalities in human males had also increased during the past 50 years, including congenital abnormalities such as cryptorchidism and hypospadias, as well as testicular cancer (Giwercman et al., 1993
).By contrast, in the USA, no such a trend had been identified.
MacLeod and Wang, 1979
compared semen data from 1951 and 1966–77 and concluded that there had been no substantial change in the numerical aspect of semen quality over this period. Paulsen et al., 1996
found no decrease in sperm concentration, semen volume, total number of sperm per ejaculate and percentage normal sperm morphology in 510 healthy men studied between 1972 and 1993 in Seattle; finally, Fisch et al., 1996
also found no decline in sperm counts over a 25-year period in 1283 men from three different geographical areas.Furthermore, pronounced seasonal variations were found in sperm concentration and total sperm count when observed over five successive year-intervals (
Chen et al., 2003
, Gyllenborg et al., 1999
), as well as an age-related increase in the percentage of sperm tail defects.The group of Wilcox (
Sallmén et al., 2005
) attempted to determine whether semen quality and/or other factors, might have affected human fecundity. As they explained, ‘changes in the availability and use of effective contraception and induced abortion might bias the direct study of time trends in couple fertility’. They added: ‘In time-to-pregnancy studies, bias alone could produce more than a two-fold apparent increase in fecundity over recent decades. In studies of infertility rates, the bias works in the opposite direction: a 30% underestimation of infertility during earlier decades could produce an apparent decrease in fertility over time’. Their conclusion is disheartening: ‘We may never know’. Olsen and Rachootin, 2003
have also expressed doubt that any historical trends will be definitively quantified, adding that a good case could be made for more precise monitoring of this phenomenon in the future.A possible way out of this quandary is to correctly measure trends in human fertility is to evaluate time to pregnancy over a given period, although it must be born in mind that the occurrence of accidental pregnancies may confuse results.
Key et al., 2009
have recently reviewed six European data sets to investigate fertility trends in Europe over the past 50 years, trying at the same time to determine whether there was evidence of the so-called ‘protection bias’, such as availability of effective contraception and induced abortion. This is important because, if they vary over time, then the results may bias the estimates of fertility trends over time. They confirmed that couples experiencing accidental pregnancies tended to be more fertile than the general population, but trends in accidental pregnancy rates were insufficient to produce substantial bias in fertility trends in simulated data. In their view, however, even the improved methodology will not provide specific data on changes in female fertility.In contradistinction to the situation in the male, no specific data have been published regarding possible overall trends in female fecundity over the last half century. In other words, no data exist on whether the rate of deterioration in oocyte quality produced by ageing has changed over the last 50 years.
Anatomical studies
The first data pointing to a high frequency of natural pregnancy wastage were gathered by the group led by the embryologist Hertig (
Adams et al., 1956
, Hertig et al., 1959
) who collected 34 human embryos aged between 1 and 17 days; eight were considered to be in the preimplantation stage and the remainder derived from the first 2 weeks post-implantation. Hertig and his group were surprised to observe that 50% of the eight preimplantation embryos had severe anomalies leading them to conclude that pregnancy could not have proceeded to term. Six (23%) of the 26 remaining embryos also had major anomalies incompatible with normal development. The study provided the first strong evidence that spontaneous loss of early human embryos by far exceeded clinically evident pregnancy losses. A subsequent publication (Hertig and Rock, 1973
) described a phenomenon called ‘disintegration of an ovum once fertilized’ and for the first time identified the period during which the greatest loss occurs as the first 2 weeks after fertilization (i.e. prior to the first missed period). It must be pointed out that studies such as the ones carried out by Hertig’s group are extremely complex and time consuming: their specimens were collected over a period of almost 17 years. This work relied upon randomly collected specimens in women of childbearing years, who did not suspect that they were pregnant at the time of their surgery. Regrettably, none of the present-day sophisticated technology, such as DNA analysis, existed. Twenty years later, Hustin et al., 1994
using techniques of histochemical staining, confirmed the observations of Hertig and colleagues, noting the presence of unsuspected pregnancy in women undergoing hysterectomy for various benign pathologies. It seems obvious from these studies that to further investigate early, pre-clinical pregnancy wastage it is necessary to utilize additional and different techniques.The use of biochemical markers
Today, new technologies have been developed to provide fresh insight into early loss of human embryos. The most important development in this respect has been the availability of highly specific and sensitive methods to detect the presence of even minute quantities of human chorionic gonadotrophin (HCG). This has made the diagnosis of pregnancy possible well before the first missed period.
Several investigators have applied these methods to examine early pregnancy wastage (Table 3), which is based on a publication of
Grudzinskas and Nysenbaum, 1985
and summarizes data obtained from the principal studies in which sensitive HCG assays have been applied. Estimates of early embryo wastage were considered to vary between 10% and 60% of all conceptions. Miller et al., 1980
followed 623 menstrual cycles in women who wished to become pregnant and recorded the detection of 152 conceptions. They documented that 43% of these pregnancies ended spontaneously, in 66.4% the pregnancies were clinically detectable and in the remaining 33.6% conception could be transiently detected only by measuring small amounts of HCG. A further study by Edmonds et al., 1982
followed 198 cycles in a population of normal women who wished to conceive; they calculated that, in these women the probability of conception was 59.6% per ovulatory cycle. However, 61.9% of all conceptions ended spontaneously within 12 weeks, with the vast majority (91.7%) being lost at the pre-clinical stage. Whittaker et al., 1983
examined 141 menstrual cycles of regular frequency and normal duration in women seeking pregnancy and – through HCG measurements – found that in 8% of them, gestation was ‘lost’ before the first missed period. This indicates failure of embryonic development between days 5 and 14 after fertilization. The data of Sharp et al., 1986
seem to indicate that early pregnancy wastage is – for the human species – a ‘physiological’ phenomenon, seen in normal women and not the consequence of a pathological condition. Indeed, in their study, the proportion of pregnancies which ceased to develop further at a pre-clinical stage, was higher among fertile than in sub-fertile women. Wilcox et al., 1985
, Wilcox et al., 1988
, in a well-documented study, utilized a method they specifically developed for this purpose. In a sample of 707 menstrual cycles from 221 healthy women desirous of pregnancy, they identified 198 pregnancies by detecting elevated HCG concentrations at around the time of implantation; 22% of these pregnancies ended spontaneously before they became clinically evident. When adding those ending with a clinical spontaneous pregnancy loss, the percentage rose to 31%. These were fertile women, since 95% of them eventually carried a baby to term. Additional prospective, observational studies (Wang et al., 2003
) examined the rates of conception and pregnancy loss and their relation with time to clinical pregnancy and reproductive outcomes in a cohort of 518 healthy newly married Chinese women who intended to conceive. Wang et al., 2003
found a conception rate per cycle of 40% over the first 12 months. Of the 618 detectable conceptions, 49 (7.9%) ended in clinical spontaneous abortion and 152 (24.6%) in early pregnancy loss, which was detected in 14% of all the cycles without a clinically recognized pregnancy. Frequencies were lower among women with delayed time to clinical pregnancy. Interestingly enough, early pregnancy loss in the preceding cycle was associated with increased odds of conception. In another study, 200 couples desiring pregnancy were evaluated and it was found that the maximal fertility rate was approximately 30% per cycle in the first two cycles (Zinamen et al., 1996
). This rate quickly declined over the remainder of the study. Pregnancy wastage accounted for 31% of the pregnancies detected. Of these losses, 41% were seen only by urine HCG testing and were categorized as ‘occult’ (pre-clinical). Furthermore, in a group of 217 women attempting to become pregnant, early loss rates were observed, from a low estimate of 11.0% to a high estimate of 26.9%, depending on the definition used and the subgroup analysed (Ellish et al., 1996
).Table 3Published studies on the incidence of pre-clinical pregnancy wastage.
| Publication | Technique utilized | Pregnancy loss (%) |
|---|---|---|
Hertig et al., 1959 | Histological evaluation | 43 |
Braunstein et al., 1977 | Presence of HCG | 15 |
Morton et al., 1977 | – | 61 |
Block, 1978 | – | 37.5 |
Chartier et al., 1979 | – | 30 |
Seppälä et al., 1979 | Presence of β1 glycoprotein | 1 case |
Miller et al., 1980 | Presence of HCG | 43 |
Edmonds et al., 1982 | – | 62 |
Edwards and Steptoe, 1983 | – | 25–35 |
Jones et al., 1983 | – | 33 |
Whittaker et al., 1983 | – | 20 |
Ahmed and Klopper, 1984 | Presence of β1 glycoprotein | Not mentioned |
– = not reported; HCG = human chorionic gonadotrophin.
Modified from
Grudzinskas and Nysenbaum, 1985
.Considering these data together, one must conclude that early embryonic wastage is in the order of 50%. In addition, it must be remembered that early embryos do not begin secreting HCG in quantities sufficient to be detected in vivo before the end of the first week (
Fishel and Edwards, 1984
), which limits the ability to identify preimplantation losses from HCG measurements of serum and urine. Therefore, not even the most recent data can provide unequivocal figures when trying to measure early pregnancy wastage. However the bulk of evidence obtained with different techniques, indicates that, in the human species, the majority of fertilized ova do not produce viable offspring.Pregnancy wastage as a function of the time of ovulation and of implantation
A method of evaluating the effects of the timing of conception on the risk of spontaneous abortion was tested by
Gray et al., 1995
for 868 pregnancies for which there were natural family planning charts recording day of intercourse and indices of ovulation. A total of 88 spontaneous abortions were recorded (10.1%). The spontaneous abortion rate was similar for 361 optimally timed conceptions (9.1%) and 507 non-optimally timed conceptions (10.9%). However, among 171 women who had experienced a spontaneous abortion in a prior pregnancy, the rate of spontaneous abortion in the index pregnancy was significantly higher with non-optimally timed conceptions (22.6%) as compared with optimally timed conceptions (7.3%). These data indicate that, once a pregnancy is clinically evident, the likelihood of subsequent failure does not exceed 10%. The relationship between delayed implantation and pre-clinical pregnancy loss has been addressed by several investigators; Lenton et al., 1991
, Liu and Rozenwaks, 1991
have proposed that doubling peri-implantation time decreases the chances of reaching a proper fetal–maternal equilibrium, thereby decreasing the probability of a successful implantation. Finally, Stewart et al., 1993
tried to correlate relaxin concentrations to implantation rates and concluded that in failing pregnancies there is an ample variability in the relationship between HCG and relaxin, with relaxin tending to be higher.Wilcox et al., 1999
studied the first appearance of HCG (utilized as the index of the time of implantation) in 221 women attempting to conceive; the appearance of HCG was found to vary between 6 and 12 days after ovulation; when implantation occurred on days 8, 9 or 10; the rate at which pregnancies proceeded was 84%, with the highest probability occurring on day 9; in contrast to this, if implantation occurred >12 days post-ovulation, no pregnancy proceeded.Some researchers speculate that delayed implantation may be either the cause or the result of abnormal events occurring soon after fertilization. Thus, pathological embryos may develop slowly or implant abnormally, leading to delayed and reduced production of HCG. As a consequence, the corpus luteum would begin its involution process, preventing pregnancy from proceeding further.
Macklon et al., 2002
have also reviewed existing evidence on the respective roles in determining the time of implantation, of the embryo and the endometrium. From published data, Bergh and Navot, 1992
concluded that the implantation window extends between menstrual cycle day 20 and 24. Within this window, the timing of implantation seems dependent on the age of the embryo and not that of the endometrium. Finally, they proposed that further progress in understanding of where the missing pregnancies ‘have gone’ will come from a substance secreted into the maternal serum 12–16 h after fertilization (Morton et al., 1977
), ‘early pregnancy factor’ (EPF). Unfortunately, following initial reports, no further work in humans seems to have taken place (Somodevilla-Torres et al., 2003
).Nature and frequency of anomalies leading to loss of embryos
In contradistinction to the variety of information available on the high incidence of first trimester losses, spontaneous abortion material is often poorly described, although it has been known for years that more than one-half of early spontaneous abortion specimens contain no embryonic/fetal parts. If an embryo is present at all, it is often either severely damaged or fragmented (
Kalousek, 1987
).Burgoyne et al., 1991
analysed 271 specimens from pregnancies presumed normal and terminated by induced abortion at 7–8 weeks gestation and, in this population, they estimated that 4.6% had numerical anomalies at 43–56 days of gestation. There were 12.9% anomalies among women over 35, compared with 3.5% in younger women. Since the frequency of numerical chromosome anomalies is 0.53% for live and stillbirths combined, the incidence of such anomalies in spontaneous abortions after 7 weeks was estimated at 51.9%. This figure was arrived at by adjusting data by maternal age and multiplying the anomaly frequency by the total number of spontaneous abortions for each week using data from published reports. Both of these variables differ in various reports, because of sampling difficulties. At the same time they pointed out that cytogenetic data from surveys of spontaneous abortion suggest a lower incidence of 39.7%, although – in their view – this figure is likely to be an underestimate because chromosome anomalies are almost certainly over-represented among the many early abortuses which lack embryonic tissue and hence are not karyotyped. Two years later, Kalousek et al., 1993
evaluated the morphology of 3912 cases of spontaneous abortions. Overall, they collected 1716 ruptured or fragmented sacs and 599 specimens consisting of only decidua. Of the remaining 1597 specimens, 251 (16%) consisted of normally developing, 279 (17%) malformed and 318 (20%) degenerating embryos. An intact empty sac was found in 350 (22%) specimens; in an additional 200 cases (12.5%) there was nodular embryonic tissue with no recognizable external features and no retinal pigment, whereas cylindrical embryos with absent limb buds were observed in 175 (11%) specimens and developmentally delayed embryos for crown-rump-length (CRL) were recorded in 24 (1.5%) instances.Some information is also available on the kind of abnormalities leading to embryo wastage and their frequency. Twenty-five years ago,
Byrne et al., 1985
evaluated the morphological features of 3472 fetuses from spontaneous abortions and reported that 21% consisted of well-formed fetuses, 27.9% had no identifiable fetal tissues, 34.2% consisted of fetal membranes only and the remainder, 16.8%, consisted of a variety of embryonic types, 16.4% of which were focal embryo and fetus. The overall rate of chromosome anomalies in the 1356 karyotyped specimens was 39.8%. Interestingly, the vast majority of chromosome anomalies (94%) were found in embryos less than 30 mm. The rate of chromosome anomalies among non-malformed fetuses (greater than 30 mm) was only 1.7%. Slightly over half (56%) of all specimens less than 30 mm long had chromosome anomalies; for individual classes of such specimens the rate ranged from 45% to 81%. The morphological category with the highest rate of karyotypic anomalies had an excess of monosomy X abortuses. Focal malformations were multiple and severe in abortuses with triploidy, trisomies 13 and 18 and monosomy X and mild in trisomy 21. With the exception of monosomy X, the malformations were similar to, and not more severe than those reported from, term births with the same anomaly. They concluded that the presence of a large fraction of chromosomally normal abortions with the same degree of rudimentary development suggests the existence of early and profound developmental problems that are not associated with anomalies of the chromosome complement.Schmidt-Sarosi et al., 1998
conducted another study in which serial chromosomal analysis of spontaneously aborted fetal tissues was carried out in 336 pregnancies in which βHCG concentrations were measured at 4–5 weeks of gestation (with concentrations >40 mIU/ml); gestations were then followed by an initial transvaginal sonography at 5–8 weeks. Of these pregnancies, 52 ended in a first trimester loss; chromosomal aberrations were found in 75.0% of abortuses after failure to detect fetal heart motion (FHM) by 7 weeks of gestation and 65.6% in women where the loss of FHM occurred at a mean of 2.6 weeks after initial screening. Different types of chromosomal abnormalities were present in each of these groups and their frequency varied significantly with maternal age, with normal fetal karyotypes in seven of 11 (63.6%) women <35 years, but only in nine of 41 (22.0%) women aged 35 years or older. Detection of FHM was associated with pregnancies continuing beyond the first trimester in 284/316 (90.0%) overall, but differed significantly with age (166/174, 95.4% women younger than 35 versus 118/142, 83.1% women 35 or older). Philipp et al., 2003
, using a technique called transcervical embryoscopy, evaluated 272 patients with missed abortion using morphology and cytogenetic analysis of chorionic villi. They found that 75% of the cases with missed abortion had an abnormal karyotype and 18% had a morphological defect with a normal karyotype, while no embryonic or chromosomal abnormality could be diagnosed in 7% of the cases. Among 233 examined cases with an observable fetus, 33 had normal external features, 71 were classified as growth-disorganized and 129 had either isolated or multiple defects. Finally, Ljunger et al., 2005
performed chromosomal analysis in chorionic villus sampling of 259 first-trimester miscarriages, An abnormal karyotype was found in 61% of the cases, with autosomal trisomies being the most frequently detected (37%), followed by polyploidies (9%) and monosomy X (6%). Cases with an extra sex chromosome constituted approximately 5% of the karyotyped abortions, with a remarkably high frequency of 47,XXY (3.4%), that is approximately 40 times greater the prevalence of Klinefelter syndrome among live births. The single most common aberration was trisomy 16, which was found in 14% of the chromosomally abnormal abortions.The analysis of oocytes failing to fertilize in vitro has provided evidence that natural selection against chromosome abnormalities may occur even prior to fertilization.
Ma et al., 1994
studied the chromosomal complement of 227 human oocytes in oocytes failing IVF and found a normal haploid chromosome complement in 54.6% of them; chromosomal abnormalities consisting of diploid sets were identified in 16.7% and aneuploidy was observed in 26%. Finally, premature condensation of sperm chromosomes of the GI-phase was observed in 22.9% of the oocytes.A different approach to the problem is that taken by
Stabile et al., 1987
who reported on 624 women who had been referred to an emergency gynaecological ultrasound clinic with a provisional diagnosis of threatened miscarriage. In these women they attempted to identify features which might indicate imminent fetal death. Interestingly, in 158 women there was no evidence of pregnancy, whereas 60 women had an ectopic pregnancy. In the remaining 406 women, ultrasound examination correctly identified the underlying cause of vaginal bleeding at first presentation in all but the six who subsequently aborted: 3.9% of the patients had a second empty sac and 5.4% had an intrauterine haematoma; none of these women subsequently aborted. Two patients had early-onset oligohydramnios and spontaneous abortion occurred in both.Assisted reproduction technology and embryo wastage
The need to develop universally accepted criteria for the diagnosis of pregnancy was critical in the early days of IVF/embryo transfer and assisted reproduction centres used the detection of urine/blood HCG for the diagnosis of pregnancy, as well as an index of implantation. They reported their success rates accordingly, rates which in fact were often unintentionally erroneously elevated.
Many terms crept into common usage to describe these phenomena, scientists fiercely debating whether the transient detection of HCG within 4 weeks of embryo transfer constituted evidence of pregnancy, which was useful either to the infertile couple or to their attending physician and the treatment team. It soon became clear that clinical action/intervention for miscarriage of these early pregnancies was not necessary and that surgical procedures to remove the small amounts of embryonic tissues were meddlesome and not without the risk of complications of such interventions. Regulators in each country held differing views on what constituted evidence of pregnancy, something required for the purposes of reporting. Eventually, in the UK, the Human Fertilisation and Embryology Authority ruled that clinical pregnancy as identified by a fetal heart action ultrasonically and live birth were the only statistical parameters of value. Whereas, such clear definitions avoided ambiguities introduced by evidence of pregnancy which was presumptive (HCG detection) and possibly based on artefacts, ultrasound observer errors or false positive results of the tests used, these strict criteria ignored pregnancies for which there was ample definitive evidence, e.g. ectopic pregnancy and anembryonic pregnancy, to name two examples.
Nevertheless, the terms biochemical/chemical pregnancy, occult pregnancy, sub/pre-clinical pregnancy came into and remain in usage today.
Grudzinskas and Nysenbaum, 1985
argued for adoption of the terms ‘sub/pre-clinical pregnancy’ to indicate that there was no action required for pregnancies which miscarried in these early gestations.The hypotheses and speculative arguments, which the work of
Hertig and Rock, 1973
and later Roberts and Lowe, 1975
provoked, curiously were supported by the epidemiological data on fecundity and live birth rates following embryo transfer. Human fecundity is lower than other mammals, lower rates of conception being cited as the most likely explanation. But assisted reproduction treatment turned this view on its head. True, the fertilization rates of human gametes in vitro were slightly lower than in other mammals, but there was a very real and substantial lower rate of pregnancy and live birth following transfer of the embryos, in great contrast to the very high rates in other mammals. Faced with the need to explain why there were such differences, it was and is still argued that a majority of early human embryos are abnormal, not destined to implant and develop into clinical pregnancies and live births. Subsequent availability of fluorescent in-situ hybridization cytogenetic technology provided evidence to support these hypotheses. As it is now possible to sustain embryonic development in vitro for more than 6 days after fertilization, the rate of growth and development as well as the morphology of these embryos are used to identify the candidate embryo(s) for transfer to the uterus. The later that a cleaving embryo is transferred to the uterus, the more likely it is to implant and result in a live birth, providing more evidence that fertilization of the human oocyte does not always result in a normal live birth.Recently,
Boomsma et al., 2009
carried out a prospective cohort study to explore whether a cytokine profile could be identified in endometrial secretions aspirated immediately prior to embryo transfer following IVF, that might predict implantation and clinical pregnancy. They found significant associations, negative and positive respectively, between monocyte chemo-attractant protein MCP-1 and 10 kDa protein IP-10 concentrations and implantation (P = 0.005 and P = 0.037, respectively) and between cytokine IL-1β and tumour necrosis factor TNFα concentrations (P = 0.047 and P = 0.023, respectively) and clinical pregnancy. The predictive value for pregnancy of IL-1β and TNFα was observed to be equivalent and additive to that of embryo quality. It is not known whether this profile may also be predictive of implantation in spontaneous pregnancies.Although the study of success rates utilizing IVF models may yield useful information, the natural process takes place in a hormonal and physical milieu that is totally different from that existing in assisted reproduction treatment and therefore any comparison is open to criticism. In this light, the model that comes closer (or, perhaps, is less distant from the natural condition) is the situation created by the use of natural cycles.
Pelinck et al., 2002
carried out a systematic review of the literature to evaluate the effectiveness of natural cycle IVF. They analysed a total of 1800 cycles resulting in 819 embryo transfers (45.5% per cycle) and 129 ongoing pregnancies (7.2% per cycle and 15.8% per transfer, with a range between 0 and 33.0). The age of the women ranged between 32 and 40 years, a period during which natural fertility is known to decline. Although these results seem fairly low, the authors pointed out that in natural cycle IVF there is a high cancellation rate (with an average of 28.9%) due to premature LH surge and ovulation. In addition, the percentage of oocyte retrieval is low (57.1–100%).Taking these facts into account, it seems that – for the purpose of calculating human fecundity – two main parameters must be considered: implantation and cumulative pregnancy rates. Implantation rates in the studies evaluated were more than acceptable, up to 50%. With regard to cumulative pregnancy rates, in three studies where these could be calculated: they were 43 and 41.7% after three and five oocyte aspirations (
Aboulghar et al., 1995
, Paulson et al., 1992
) and 46% after four started cycles (Nargund et al., 2001
).More recently,
Phillips et al., 2007
compared retrospectively a series of 134 controlled natural IVF cycles with 370 stimulated IVF cycles. The clinical pregnancy rate per embryo transfer, for women aged less than 35 years, was 27% following natural cycles, whereas it rose to 47% in stimulated cycles. The percentage for natural cycles decreased to 8% per embryo transfer over 35 years of age.In conclusion, data obtained utilizing IVF confirm the low fecundity of the human species, as well as the rapidly declining fertility after age 35–37. The likely cause for this phenomenon seems to be oocyte aneuploidy leading to maldevelopment of the embryo (
Staessen et al., 2004
). In women aged 35–41 undergoing preimplantation genetic screening, more than 60% of embryos are aneuploid and about two-thirds of the products of conception after miscarriage have abnormal karyotype (Heffner, 2004
).Conclusions
Early, pre-clinical pregnancy wastage is today well established through a variety of methods and approaches. Although no exact percentage can be provided, its extent seems to be in the order of at least 50%. This information may have applications outside biology and clinical obstetrics and serve to guide decision makers, as well bioethicists trying to determine the true beginning of a human life. A discussion of these aspects will be the topic of a subsequent paper.
References
- In vitro fertilization in a spontaneous cycle: a successful simple protocol.J. Obstet. Gynaecol. 1995; 21: 337-340
- A description of 34 human ova within the first 17 days of development.Am. J. Anat. 1956; 98: 435-493
- Detection of subclinical abortion by assay of pregnancy specific beta 1 glycoprotein.Br. Med. J. (Clin. Res. Ed.). 1984; 288: 113
- Probability of conception, conception delay and estimates of fecundability in rural and semi-urban areas of certain Latin-American countries.Soc. Biol. 1979; 26: 226-231
- Sex and reproduction: an evolving relationship.Hum. Reprod. Update. 2010; 16: 96-107
- The impact of embryonic development and endometrial maturity on the timing of implantation.Fertil. Steril. 1992; 58: 537-542
- Analisi del declino nella qualità del liquido seminale nella popolazione Italiana negli ultimi 15 anni [Analysis of decline in seminal fluid in the Italian population during the past 15 years].Minerva Ginecol. 1999; 51: 223-231
- Paternal contribution to embryonic death.J. Reprod. Fertil. 1964; 1: 383-396
- Occult pregnancy as a factor in unexplained infertility.Reprod. Med. 1978; 21: 251-253
- L’epidemiologia dell’aborto spontaneo: una revisione della letteratura. [Epidemiology of spontaneous abortion: a literature revision].Ann. Ost. Ginec. Med. Perin. 1989; 110: 323-334
- Endometrial secretion analysis identifies a cytokine profile predictive of pregnancy in IVF.Hum. Reprod. 2009; 24: 1427-1435
- Has the fertility of Danish men declined through the years in terms of semen quality? A comparison of semen qualities between 1952 and 1972.Int. J. Fertil. 1983; 28: 91-99
- Subclinical spontaneous abortion.Obstet. Gynecol. 1977; 50: 41s-44s
- Incidence of numerical chromosome anomalies in human pregnancy estimation from induced and spontaneous abortion data.Hum. Reprod. 1991; 6: 555-565
- Morphology of early fetal deaths and their chromosomal characteristics.Teratology. 1985; 32: 297-315
- Evidence for decreasing quality of semen during past 50 years.Br. Med. J. 1992; 305: 609-613
- Tourouvre-au-Perche au XVIIème et XVIIIème Siècle [Tourouvre-au-Perche during the XVIIth and XVIIIth centuries].Presses Universitaires de France, Paris1970
- Measurement of plasma human chorionic gonadotropin (hCG) and beta-hCG activities in the late luteal phase: evidence of the occurrence of spontaneous menstrual abortions in infertile women.Fertil. Steril. 1979; 31: 134-137
- Seasonal variation and age-related changes in human semen parameters.J. Androl. 2003; 24: 226-231
- Early embryonic mortality in women.Fertil. Steril. 1982; 38: 447-453
- Current status of in-vitro fertilisation and implantation of human embryos.Lancet. 1983; 2: 1265-1269
- A prospective study of early pregnancy loss.Hum. Reprod. 1996; 11: 406-412
- Semen analyses in 1,283 men from the United States over a 25-year period: no decline in quality.Fertil. Steril. 1996; 65: 1009-1014
- HCG secreted by pre-implantation embryos cultured in vitro.Science. 1984; 223: 816-818
- Evidence for increasing incidence of abnormalities of the human testis: a review.Environ. Health Perspect. 1993; 101: 65-71
- Timing of conception and the risk of spontaneous abortion among pregnancies occurring during the use of natural family planning.Am. J. Obstet. Gynecol. 1995; 172: 1567-1572
- Failure of pregnancy after implantation.Ann. NY Acad. Sci. 1985; 442: 38-44
- Secular and seasonal changes in semen quality among young Danish men: a statistical analysis of semen samples from 1927 donor candidates during 1977–1995.Int. J. Androl. 1999; 22: 28-36
- Advanced maternal age – how old is too old?.N. Engl. J. Med. 2004; 351: 1927-1929
- La population Canadienne au début du XVIIIéme Siècle [Canadian population at the beginning of the XVIIIth century].Presses Universitaires de France, Paris1954
- Searching for early fertilized human ova.Gynecol. Invest. 1973; 4: 121-139
- Thirty four fertilized human ova, good bad and indifferent, recovered from 210 women of known fertility. A study of biologic wastage in early human pregnancy.Pediatrics. 1959; 23: 202-211
- Immuno-histochemical localization of two endometrial proteins in the early days of human pregnancy.Placenta. 1994; 15: 701-708
- Anatomical and chromosomal abnormalities in specimens of early spontaneous abortions: seven years experience.Birth Defects. 1987; 23: 153-168
- Early spontaneous abortion: morphologic and karyotypic findings in 3,912 cases.Birth Defects Orig. Artic Ser. 1993; 29: 53-61
- Methodological issues in analyzing time trends in biologic fertility: protection bias.Am. J. Epidemiol. 2009; 169: 285-293
- Epidemiology of birth defects.Epidem. Rev. 1989; 11: 244-248
- What is a pregnancy? A question for programs of in vitro fertilization.Fertil. Steril. 1983; 40: 728-733
- Normal and abnormal implantation in spontaneous in-vivo and in-vitro human pregnancies.J. Reprod. Fertil. 1991; 92: 555-565
- Early pregnancy wastage in IVF (in vitro fertilization) patients.J. In vitro Fertil. Embryo Transf. 1991; 8: 65-72
- Chromosomal anomalies in first-trimester miscarriages.Acta Obstet. Gynecol. Scand. 2005; 84: 1103-1107
- Chromosome investigation in in vitro fertilization failure.J. Assist. Reprod. Genet. 1994; 11: 445-451
- Conception to ongoing pregnancy: the ‘black box’ of early pregnancy loss.Hum. Reprod. Update. 2002; 8: 333-343
- Male fertility potential in terms of semen quality: a review of the past, a study of the present.Fertil. Steril. 1979; 31: 103-116
- Fetal loss after implantation. A prospective study.Lancet. 1980; 2: 554-556
- An early pregnancy factor detected in human serum by the rosette inhibition test.Lancet. 1977; 1: 394-397
- Cumulative conceptions and live birth rates in natural (unstimulated) IVF cycles.Hum. Reprod. 2001; 16: 259-262
- Fate of human fertilized eggs during prenatal life: present status of knowledge.Okajimas Folia Anat. Jpn. 1970; 46 (mentioned in: Nishimura, H., 1970. In: Fraser, F.C., McKusick, A. (Eds.), Proceedings of the IIIrd International Conference on Congenital Malformations. Excerpta Med, Amsterdam): 297-305
- Maternal age and fetal loss: population based register linkage study.Br. Med. J. 2000; 320: 1708-1712
- Invited commentary: Monitoring Fecundity over Time – If We Do It, Then Let’s Do Right.Am. J. Epidemiol. 2003; 157: 94-97
- Data from men in greater Seattle area reveals no downward trend in semen quality: further evidence that deterioration of semen quality is not geographically uniform.Fertil. Steril. 1996; 65: 909-911
- In vitro fertilization in unstimulated cycles: the University of Southern California experience.Fertil. Steril. 1992; 57: 290-293
- Efficacy of natural cycle IVF: a review of the literature.Hum. Reprod. Update. 2002; 8: 129-139
- Transcervical fetoscopic diagnosis of structural defects in four first-trimester monochorionic twin intrauterine deaths.Prenat. Diagn. 2003; 23: 964-969
- Controlled natural cycle IVF: experience in a world of stimulation.Reprod. Biomed. Online. 2007; 14: 356-359
- Where have all the conceptions gone?.Lancet. 1975; 305: 498-499
- Has human fertility declined over time? Why we may never know.Epidemiology. 2005; 16: 494-4999
- Chromosomal analysis of early fetal losses in relation to transvaginal ultrasonographic detection of fetal heart motion after infertility.Fertil. Steril. 1998; 69: 274-277
- Early detection of implantation by pregnancy-specific beta 1-glycoprotein secretion in an infertile woman treated by artificial insemination and human chorionic gonadotropin.Fertil. Steril. 1979; 32: 608-609
- Early conceptual loss in subfertile patients.Br. J. Obstet. Gynaecol. 1986; 93: 1072-1077
- An analysis of reproductive patterns in an American isolate.Popul. Stud. 1965; 19: 65-77
- Time to pregnancy and semen parameters: a cross-sectional study among fertile couples from four European cities.Hum. Reprod. 2002; 17: 503-515
- Purification and characterisation of functional early pregnancy factor expressed in Sf9 insect cells and in Escherichia coli.Protein Expr. Purif. 2003; 32: 276-285
- Ultrasonic assessment of complications during first trimester of pregnancy.Lancet. 1987; 2: 1237-1240
- Comparison of blastocyst transfer with or without preimplantation genetic diagnosis for aneuploidy screening in couples with advanced maternal age: a prospective randomized controlled trial.Hum. Reprod. 2004; 19: 2849-2858
- The relationship between hCG and relaxin secretion in normal pregnancies vs. peri-implantation spontaneous abortions.Clin. Endocrinol. (Oxf.). 1993; 38: 379-385
- Delaying childbearing: effect of age on fecundity and outcome of pregnancy.Br. Med. J. 1991; 302: 1361-1365
- Deterioration of sperm quality in young, healthy Belgian men.Hum. Reprod. 1996; 11: 325-329
- A long-term follow-up study of women using different methods of contraception. An interim Report.J. Biosoc. Sci. 1976; 8: 373-427
- Conception, early pregnancy loss, and time to clinical pregnancy: a population-based prospective study.Fertil. Steril. 2003; 79: 1517-1521
- Unsuspected pregnancy loss.Lancet. 1983; 1: 1126-1127
- Time of implantation of the conceptus and loss of pregnancy.N. Engl. J. Med. 1999; 319: 1796-1799
- Measuring early pregnancy loss: laboratory and field methods.Fertil. Steril. 1985; 44: 366-374
- Incidence of early loss of pregnancy.N. Engl. J. Med. 1988; 319: 189-194
- Maternal and embryonic factors associated with prenatal loss in mammals.J. Reprod. Fertil. 1986; 76: 851-864
- Estimates of human fertility and pregnancy loss.Fertil. Steril. 1996; 65: 503-509
- Semen quality changes among 2343 health Slovenian men included in an IVF-ET programme from 1983 to 1996.Int. J. Androl. 1999; 22: 178-183
Article info
Publication history
Published online: September 07, 2010
Accepted:
August 25,
2010
Received in revised form:
August 25,
2010
Received:
January 21,
2010
Declaration: The authors report no financial or commercial conflicts of interest.Footnotes
Professor Giuseppe Benagiano is the Dean of the Post-graduate School of Gynaecology and Obstetrics, University ‘la Sapienza’ in Rome. Here, he became professor in 1980 and acted as Director of the First Institute of Obstetrics and Gynaecology during 1982–1993. He was nominated Director of the Special Programme of Research in Human Reproduction of WHO in 1993 and became Director General of the Istituto Superiore di Sanità in 1994. He has studied, Karolinska Institute, Stockholm and Population Council, Rockefeller University. His long-standing interest is hormonal contraception and he pioneered the use of GnRH analogues for gynaecological disorders of endocrine origin, such as endometriosis, adenomyosis and leiomyomata.
Identification
Copyright
© 2010 Reproductive Healthcare Ltd. Published by Elsevier Inc. All rights reserved.
