For the pill to exhibit the characteristic of an abortifacient, one biological event is essential: ovulation. The crucial question is this--does break-through (or escape) ovulation occur during regular pill ingestion?
Grimes et al. (Obstet Gynecol, 1994) had previously reported that 'suppression of follicular development is incomplete with contemporary low-dose pill'.50
Grimes' study was characterized by a high rate of patient compliance, meaning that the women involved in the study adhered to the research protocol of daily ingestion of the pill.51 Yet, escape-ovulation was detected even within the context of a rigorously scrutinized scientific study.
These facts argue strongly in favour of escape-ovulation also occurring within the general populace of women on the pill. This latter group of women are not necessarily as highly motivated as those participating in a scientific study. To adhere to a tedious daily, monthly, yearly regime of pill ingestion, without supervision is, in the words of one feminist writer a 'bore and a chore'.52 Because daily pill ingestion is so onerous, patient compliance will be less than the necessary ideal. However, does the occasional failure to take the pill mean that 'escape ovulation' will increase in some proportional fashion?
In an attempt to determine the frequency of escape-ovulation under more realistic conditions, researchers have constructed experiments that required women in the study to deliberately miss one or more days of the pill. A variety of tests, including ultrasound of the ovaries, estradiol (E2), progesterone (P) levels and LH (leutinizing hormone) measurements were used to determine if ovulation had occurred.
Hedon and co-workers (1992) tested 47 young, healthy women who missed between 1 and 4 days tablets starting from day 1 of a new cycle. 'None of the patients experienced normal ovulation' though one, who missed 3 tablets at the beginning of the cycle, 'had a follicular rupture', but no LH surge or progesterone increases, factors usually associated with normal ovulation53 Note that this study was for only one cycle. Limiting the study to one cycle was a study weakness, because any follicles which may have ruptured during the normal 7 pill free days between cycles would not be detected.
Earlier, Hamilton (1989) had performed a similar study but extended the observations for two consecutive months. Of 30 women in the study, one had a probable ovulation, due to one deliberately omitted tablet on day one of the second cycle.54
More recently, Letterie (1998) published the results of a study employing a new, reduced dosage formulation of the pill. Ten women, divided into 2 groups, used two slightly different formulations comprising a delayed start, limited midcycle use of estrogen and progesterone. Each of the two treatment groups was monitored for 2 consecutive cycles. In total, 30% of cycles exhibited ovulation, all of which occurred in the second cycle.55
It is revealing to look more closely at the data for the two groups. In group one, ovulation occurred in 10% of cycles (1 in 10 cycles). This group took 50mcg ethinyl estradiol/lmg norethinodrone for days 6-10 and 0.7mg norethinodrone for days 11-19. Group two took 50mcg ethinyl estradiol/lmg norethinodrone for days 8-12, and 0.7mg norethinodrone only for days 13-21, 'five ovulation(s) occurred in 10 cycles'.56 This is an ovulation rate of 50%. This study did not investigate implantation; all participants used barrier contraceptives, or abstinence (Private correspondence).57
It should be noted that these research findings, conducted under ideal research conditions, represent the best possible outcome in terms of ovulation suppression by the pill. Yet these results do not faithfully replicate real-life because they do not take into account such common events as gastro-intestinal illness or drug interactions. Stomach upset decreases drug absorption, thus loosening the hold over ovulation otherwise exerted by the pill hormones. Likewise, drug interactions decrease the amount of active pill hormone available to act in a suppressant manner upon the ovaries.58 59 Other researchers and I are of the view that these two issues would contribute to an increase in the frequency of escape-ovulation.60
Based upon my 20 years experience as a community pharmacist, I believe the commonly held view is that the pill fully stops ovulation (anovulation). Yet this view is wrong. The recent work by Rabe et al. (1997) contradicts this common misunderstanding. Following are some salient points from this research.
The estradiol level of 153 pg/ml, seen in pill users with enlarged follicles, is important, as it is close to the 'threshold level 150 to 200pg/ml', which, if persisting for approximately 36 hours, triggers ovulation.64
As a summary of their research, Rabe noted: 'Analysis of the ovarian activity in the current study demonstrated that the total number of developing follicles increased rather than diminished during OC use, without marked differences between OCs'.65
This research underscores the pill's precarious hold over ovulation suppression. It is an event endeavouring to occur. The intervention of a variety of 'lifestyle' factors such as missed doses, drug interactions or gastro-intestinal upset, can act to loose the hold exerted by the pill over natural ovarian function.
As a footnote to this discussion, the FDA approved, in late 1998, a low dose estrogen formulation of the pill (norethinodrone acetate, 1 mg; ethinyl estradiol 20 mcg). Similar low-dose estrogen formulations are also now avail-able in Australia.66 The frequency of escape ovulation can only be expected to increase in this situation of reduction hormonal ingestion.
Thus, the question arises: will a low dose pill, more inclined than not to permit escape-ovulation, increase the frequency of implantation failure due to a under-developed endometrium? The medical literature indicates that there is a critical thickness of the endometrium needed to sustain implantation of a human embryo.
Issacs (Fert Steril, 1996) reported that an endometrial thickness of at least 10mm or more, around the time of ovulation, 'defined 91% of conception cycles'.67 Spandorfer (Fertil Steril, 1996) noted that 97% of abnormal pregnancies, defined as Fallopian tube lodgment or spontaneous abortion, had endometrial thickness of 8mm or less.68 Shoham (Fert Steril, 1991) reported that a mid-luteal thickness of 11 mm or more 'was found to be a good prognostic factor for detecting early pregnancy' but no pregnancies were reported in an ovulation induction programme 'when the endometrial thickness was less than or equal to 7mm'.69
The mid-luteal phase of the menstrual cycle, around day 20, is referred to in the medical literature as the window of expected implantation.70 71
Gonen (Journ In Vitro Fert Embryo Transf, 1990) also reported that 'endometrial thickness was significantly greater in the group of patients who achieved pregnancy than in the group who did not'.72 Implantation failure was associated with endometrial thickness of approximately 7.5mm, success with endometrial thickness of approximately 8.5-9mm.
These study results, which indicate a normative endometrial thickness of around 8.5mm for successful implantation, are central to any claimed interceptive/abortifacient capacity of the pill. Research findings from Rabe and co-workers (1997) underscore this point.
Rabe reported that study subjects who took the triphasic levonorgestrel/ethinylestradiol formulation had the highest percentage of follicular cysts with a diameter greater than 20mm/73 but they failed to develop a median endometrial thickness in excess of 6mm.74 To recall, follicles of this size are 'thought to be associated with increased risk of escape ovulation'.75
The importance of these events is clear; follicles of a suit-able size can develop in women taking the pill daily, but endometrial thickness has been shown to be underdeveloped. In the event of follicle rupture and release of an 'ovum', implantation of a human embryo would be greatly hampered. Rabe confirms this very point: '. . . the occurrence of pregnancy would be unlikely because accessory contraceptive mechanisms such as cervical hostility and endometrial suppression are usually in effect'.76
It must be pointed out that in this quote Rabe has falsely defined pregnancy as beginning at implantation. Pregnancy begins with the fertilization of the female sex cell (ovum) by sperm, the restoration of the full complement of 23 pairs of chromosomes and thereby the creation of a new human person.
Based upon these findings, a number of issues present themselves:
It is important to note that these four observations exist independently of the impact of the pill on the various implantation factors involved in cell-signaling.
As the aforementioned research indicates, the last few years have seen a remarkable unveiling of the process of implantation of the human embryo into the uterine tissue. A large body of evidence now exists which demonstrates that the process of implantation, rather than being an accidental event dependent on chance, is in fact a multi-factorial, cascading bio-molecular,77 physiological and hormonal event of spectacular intricacy, complexity, refinement and interdependence.78 Implantation is not, as one might suppose, akin to two pieces of Velcro fortuitously touching and gripping together. Rather, implantation is, in every sense, as complex, and therefore susceptible to interference, as is the clotting mechanisms of the cardiovascular system.
Beside PAF, the interleukin system and other factors mentioned briefly in the introduction, the class of cell adhesion molecules known as integrins also play a critical role in successful implantation of the human embryo into the endometrium.
As the description of the molecule suggests, the role of integrins is to bind cells together. Etzioni has suggested that integrin facilitated cell adhesion is 'a process that is essential for anchorage' of cells to each other (Lancet, 1999).79
There are a variety of different types of integrins found within the body--one that plays an essential role in implantation is known as avb3. The medical literature now contains many research papers demonstrating the vital role of this integrin in the process of binding the 5-7 day old human embryo to the endometrium (lining of the womb).
Somkuti and co-workers (Fert Steril, 1996) for example reported that integrins 'might prove useful as markers of normal endometrial receptivity'80 because they have been shown to be absent in women with unexplained infertility and endometriosis.81
Similarly, Lessey (Am J Reprod Immunol, 1996) reported 'aberrant expression of this integrin is associated with infertility in women'.82 Widra (Mol Hum Reprod, 1997) noted 'the absence of endometrial avb3 during the critical period of implantation ... in women with unexplained infertility and endometriosis'.83 Others had also commented on the absence or diminution of avb3 in women with recurrent pregnancy loss84 or unexplained infertility.85
Assessing the role of the pill, Somkuti (1996) compared endometrial sampling from women on the pill with samples from non-users and reported integrin expression 'to be altered grossly in OC users'.86
Complementing this work were the observations of Yoshimura (1997):' a loss of normal avb3 expression is associated with primary infertility and milder forms of the disease. These observations suggest that this integrin plays a significant role in the implantation process'.87
Eric Widra and colleagues (1997), at Georgetown University investigated the role of physiological levels of estrogen and progesterone on the endometrial levels of avb3. They reported that estrogen caused a down-regulation in the expression of avb3,88 an important finding in the light of the fact that 'expression of the avb3 integrin may, in fact, be necessary for normal implantation to occur.89 Castelbaum and co-workers (J Clin Endo Metab, 1997) reported the endometrial expression (presence) of avb3 was 'reduced by E2 treatment and further suppressed by E2 plus P...'90
These results indicate a link between the impact of hormones on the expression of integrins, and the role of integrins in implantation. Whilst the inter-relationship between hormones, integrins and implantation is not yet fully understood,91 sufficient evidence exists to conclude that the inter-relationship is significant from the perspective of implantation. This is because implantation occurs only 'on or about day 20 of an idealized 28-day menstrual cycle'92 and the avb3 integrin 'is expressed on endometrial epithelial cells only at the opening of the implantation window, on postovulatory day 6'.93
The IGF system is an important growth factor, playing a key role in the monthly development of the endometrium and in the process of implantation.94 There are two subsets, IGF-1 and IGF-2. The first is believed to facilitate the mitotic action of estradiol [E2] in the endometrium, whilst IGF-2 'expressed abundantly in mid-late secretory endometrium, may be a mediator of progesterone action'.95 Aside from this hormonal aspect, the most abundant expression if IGF-11 is in the columns of the invading trophoblast in the anchoring ville.
From this it can be seen that IGF has a promotional effect upon the process of implantation. But IGF is in turn regulated. 'The biological actions of IGFs are modulated by a family of binding proteins (IGFBPs). The demonstration of IGF and IGFBP transcripts [copying facilities] in pre-implantation embryos indicates that the influence of IGFs and IGFBPs in fetal development begins even prior to implantation'.96
Thus far, it can be seen that these factors have a key role to play in both the preparation for and process of implantation. As Han et al. have noted: 'Presumably, IGF-II and IGFBPs are used for cell to cell communications between fetal trophoblasts and maternal decidual cells at the feto-maternal interface for placental development and/or function'.97
Against this background, the role of the hormones in the pill, particularly their influence over implantation, is important. A number of researchers have shown that the pill causes an increase in IGFBP-1 levels and a decrease in plasma concentrations of IGF-1.98 99 More specifically, during the pill free-week 'IGFBP-1 was significantly lower on the medication-free day than on day 14 of the cycle . . . The short absence of exogenous estrogen and progestin during the medication-free week also affected IGF-1 levels, which were significantly increased'.100
The superabundance of IGFBP induced by the pill has, from an implantation perspective, significance. Giudice has reported that: 'IGFBP's bind IGF's with high affinity and, for the most part, inhibit IGF bioavailability to their receptors for action in their target organs'.101 Thus, the supraphysiological levels of IGFBP, induced by the pill, may be detrimental to the process of implantation via an inhibitor action on the levels of IGF. Giudice highlights this point: 'IGFBP-1 has been shown to inhibit trophoblast invasion into decidualised endometrial stromal cultures, suggesting that this IGFBP-1 is a maternal "restraint" on trophoblast invasion'.102
Aside from the indirect anti-implantation effect of excessive levels of IGFBP upon IGF, IGFBP also has a direct, anti-attachment effect upon the human embryo. 'IGFBP-1 specifically binds to first trimester trophoblast and that it binds to the 'a5b1 integrin in trophoblast. Furthermore, it inhibits trophoblast attachment to fibronectin; another RGB ligand found in the placental bed.'103
In summary, the pill causes an increase in IGFBP levels, leading to a decrease in IGF levels. This may have a negative impact upon implantation. IGFBP also may have a direct effect at the level of trophoblast/endometrial integrin binding. More research is required to understand fully the roles of IGF and IGFBP. This represents a new, emerging field of research into the multitudinous factors involved in the process of implantation. Whilst the above research indicates that the pill facilitates anti-implantation endometrial environment, confirming evidence is yet to be found. Hence there exists a reasonable suspicion only, a point made by key researchers in the field.104
This discussion has had as its focus the multifactorial nature of embryo implantation. On occasion, this discussion has required detailed analysis of the relevant factors influencing the success of this event. Sometimes it is not possible to speak of these events, centred as they are on the maintenance of human life, without a certain measure of complexity and detail. To those readers who have struggled with this material I apologize.
This paper does not presume to be the final word on this complex and evolving branch of medical knowledge. New research appears almost monthly to illuminate further and sometimes confuse this emerging medical discipline. Nevertheless, I hope I have briefed the reader on issues related to the first right of all humans the right to stay alive. Some may seek to discount the interceptive/ abortifacient capacity of the pill. For three reasons, this would be a scientifically precarious position to adopt.
First, I am of the view that the preceding evidence strongly argues the case in favour of the pill possessing an interceptive/abortifacient capacity. At the very least, the evidence is repetitive and circumstantial. Indeed, how more clear and straightforward could the issue be than the following statement from Eric Widra and colleagues? 'Demonstration of complimentary integrin expression on preimplantation embryos has further buttressed the argument that these molecules are important for the initiation of pregnancy'.105
Second, even researchers view as the new arena of 'contraceptive' research the interrelated system of implantation factors. Carlos Simon and colleagues (Fertil Sterility, 1998), after discussing the interdependent relationship between the interleukin-1 system, the avb3 integrin adhesion system and implantation, conclude by stating that the interleukin-1 system would be a promising new area of research apropos the development of new 'contraceptives'.106 Given this sentiment, I am of the view that anti-interleukin chemicals will be the RU-486 of the next decade.
Third, and most tellingly, the abortifacient capacity of the pill is recognized by those who support abortion. Consider the following, taking from the Guttmacher Report. 'The best scientific evidence suggests that ECP's [emergency contraceptive pill] most often work by suppressing ovulation. But depending on the timing of intercourse in relation to a woman's hormonal cycle, they--as is the case with all hormonal contraceptive methods also may prevent pregnancy either by preventing fertilization or by preventing implantation of a fertilized egg in the uterus' (my emphasis).107
Need any more be said?