Abortion Standards for Northern Ireland:
Deficiencies of the RCOG Guideline Pertaining to Medical Abortion

Martha Shuping
Christopher Gacek, J.D.,Ph.D.,
Donna Harrison, M.D.
April 17, 2007
Reproduced with Permission

The Draft Guidance on abortion references and includes the Royal College of Obstetrics and Gynaecology (RCOG) Guideline1 which encourages use of medical (non-surgical) abortion procedures using mifepristone. But the 2004 RCOG Guideline is out of date and gravely harmful to women's health regarding its recommendations for medical abortion. Even if the Guideline were amended according to information below, medical abortion remains much less safe and less effective than surgical abortion. Medical abortion endangers women's lives and health. Although there are many different protocols for medical abortion used worldwide and in the research literature, with various combinations of several different drugs, this report will focus specifically on the mifepristone and misoprostol combination used in the U.S. and recommended in the RCOG Guidelines, though as we will see, the specific doses and recommendations differ between the U.S. and U.K.

The 2004 RCOG Guideline is out of date, and was not informed by "real-world" experience. It was published prior to the 2006 report by Gary and Harrison2 which evaluated 607 unique mifepristone Adverse Event Reports submitted to the U.S. Food and Drug Administration (FDA). Results included 237 hemorrhages (1 fatal, 42 life threatening, 168 serious cases; 68 requiring transfusions); infections, including 7 cases of septic shock (3 fatal, 4 life threatening) and 43 cases requiring parenteral antibiotics; and ectopic pregnancies. Surgery was required in 513 cases, including 235 emergency surgeries. Since this study, additional reports have been received by the FDA, bringing the total reports to 1070 as of April 2006.3 U.S. physicians are not required to report adverse events, and the FDA estimates it receives reports for only 1 - 10% of drug complications.4 There is reason to believe that only about 3-4% of mifepristone adverse events have actually been reported.5

The FDA webpage displays copies of FDA required product labeling information,6 a Medication Guide,7 and a Patient Agreement,8 by which patients and providers are cautioned concerning risk of various complications associated with mifepristone abortions, including ectopic pregnancy, infections, septic shock, excessive hemorrhage, possible need for transfusions or D & C's. This information was updated on July 19, 20059 to reflect post-marketing experience including deaths related to mifepristone abortion. These warnings are not currently included or discussed within the RCOG Guideline.

A general problem in regard to safety and efficacy data is the scarcity of fully randomized controlled trials comparing medical to surgical abortions. In the few head to head comparisons, surgical abortion is shown to be safer and more effective.10,11,12,13,14,15

The U.S. House of Representatives Government Reform Committee in its report "The FDA and RU-486: Lowering the Standard for Women's Health" points out that the FDA typically "requires data from two clinical trials that are randomized, blinded and controlled against a 'comparator' - often a placebo but more typically an alternative therapy."16 This report discusses the necessity of matched controls so that there is a truly comparable control group. However, the report states that in regard to the French and U.S. studies on which FDA approval was based, "these trials were uncontrolled."17

The RCOG Guideline references a small review article in which Say et al. searched the literature for randomized studies in which surgical and medical abortion were compared. Say et al. found only five studies, comparing four different abortion methods, concluding, "The results are derived from relatively small trials," and that "there is inadequate evidence to comment on the acceptability and side effects of medical compared to surgical first trimester abortions."18

A few studies do exist which compare medical abortion with mifepristone to surgical abortion, though the following are not randomized trials as in each case women chose the type of abortion they preferred. Cabeza's study of 500 women found the medical abortion group had more side effects (including bleeding, cramping, nausea and vomiting).19 Elul et al. reported on data from 1,373 women in China, Cuba and India, finding that medical abortion patients at all locations experienced more side effects than surgical patients, particularly bleeding and pain.20 Jensen's study of 377 patients found more side effects (pain, nausea, vomiting and bleeding) in the medical abortion group.21 Each of these studies showed a higher failure rate for medical abortion.

According to the FDA's medical review, written prior to initial U.S. approval, medical abortion patients had "far more frequent" nausea, vomiting, and cramping, than surgical abortion patients.22 This report also stated failure rates were higher for medical abortion, and that blood loss was significantly higher with medical compared to surgical abortion.23

Bleeding that continues for a month or more after medical abortion has frequently been reported after medical abortion. Dr. James McGregor, speaking at a workshop presented by the U.S. Centers for Disease Control (CDC) and FDA, discussed a study from 1986 in which "many patients had prolonged, over a month of abnormal bleeding."24 The FDA product label information for mifepristone states, "According to data from the U.S. and French clinical trial studies, women should expect to experience vaginal bleeding or spotting for an average of nine to 16 days. Up to 8% of women may experience some type of bleeding for more than 30 days."25 The International Inquiry Commission on RU-486 expressed concern about uterine bleeding in over 90% of cases, lasting from 1 to 35 days, "in many cases an emergency" requiring surgery or transfusion.26

In some cases, the bleeding is of massive proportions well beyond the amount of bleeding typically experienced in usual gynecological cases. Dr. Donna Harrison, co-author of a published report on 607 of the Adverse Event Reports received by the FDA on mifepristone, testified before a U.S. Congressional committee regarding the severity of some of the cases: "In my experience as an ob-gyn, the volume of blood loss seen in the life-threatening cases is comparable to that observed in major surgical trauma cases like motor-vehicle accidents. This volume of blood loss is rarely seen in early surgical abortion without perforation of the uterus, and it is rarely seen in spontaneous abortions."27

In her testimony, discussing studies from the U.S. and a number of international locations, Harrison concluded that the risk of a hemorrhage serious enough to require transfusion was between 0.1% to 0.2% (one to two per thousand).28 Harrison concluded that the risk of hemorrhage from mifepristone abortions is much greater than the risk of hemorrhage from surgical abortion.29

While some cases of severe hemorrhage do require transfusion, more frequently cases of severe bleeding are managed surgically. The FDA Medication Guide for mifepristone states that "in about 1 out of 100 women, bleeding can be so heavy that it requires a surgical procedure (surgical abortion/D&C) to stop it."30 The Medication Guide gives specific patient information as to how much bleeding is too much and when the patient should seek emergency treatment for bleeding, since it is expected that some women will have serious bleeding requiring surgical intervention following medical abortion. In fact, the mifepristone product label states that in the clinical trials that led to FDA approval, 5% of U.S. women experienced "uterine hemorrhage" and 6% of the French experienced a "decrease in hemoglobin greater than 2 g/dL."31

RCOG Guideline does not address this serious risk of hemorrhage, stating, "The risk of hemorrhage at the time of abortion is low. It complicates around 1 in 1000 abortions overall."32 Here, the Guideline appears to be referring to the risk of hemorrhage from all types of abortions in the UK, as it states, "The above rates of haemorrhage at the time of abortion are taken from National Office for Statistics . . . via the notification process," adding that "criteria for reporting are ill-defined and it is not known whether or not all these cases required transfusion."33 It is not stated whether the National Office for Statistics is plagued by a similar degree of under-reporting as is recognized for the FDA, but in any case it seems clear that the report is considering hemorrhaging after abortion generally, and not the specific case of medical abortion. Additionally it seems clear that there is some uncertainty as to the reporting.

Further along in the Guideline, in the section specifically pertaining to methods of medical abortion, a large study by Ashok is cited.34,35 Conducted at the University of Aberdeen, this large study (4131 cases after one woman dropped out) reported 61 cases of serious bleeding, serious enough to require surgery, transfusion, or treatment with oxytocics. Although only 0.2% (two per thousand) required transfusion, a total of 1.5% actually had serious bleeding problems (calculations our own based on 61 women out of 4,131 reported as requiring surgery, transfusion or oxytocics for heavy bleeding). This is in the same range as that estimated by Dr. Harrison (above) and by the FDA's Medication Guide for mifepristone (as above). Within the RCOG Guideline, the Calman scheme is used to describe "the degree of risk of complications and sequelae associated with abortions."36 In this scheme, shown in Table 5.1 of the Guideline, a risk occurring between 1 in 1,000 and 1 in 100 is considered moderate, while a risk greater than 1 in 100 is considered high.37 By this definition, the risk of serious hemorrhage after medical abortion should not be described as "low."

RCOG Guideline does not discuss mifepristone's serious effects in disrupting the innate immune system. In an animal study, mifepristone increased death rate from sepsis from 13% to 100%.38 "Numerous studies have been done whichÉshow increased lethality in animal models in the presence of RU-486 or mifepristone."39 Immune system inhibition was a concern of the International Inquiry Commission on RU 486.40 "During a mifepristone-induced abortion, the blockade of glucocorticoid receptors by mifepristone results in the disruption of the innate immune system."41

There have now been six North American deaths from septic shock due to Clostridium sordellii in young women who have had medical abortion with mifepristone and misoprostol.42 Research presented at the Centers for Disease Control and Prevention (CDC)-FDA Workshop on "Emerging Clostridial Disease" supports a causal relationship between mifepristone immune suppression and death from Clostridial sepsis.43,44 This workshop also reported on additional cases of medical abortion patients who have died of Clostridium perfringens infection which were at the time still being investigated.45

Recommendations regarding infection control do not discuss possible infection with C. sordellii, nor atypical presentation of infection. At the CDC-FDA workshop on Clostridial disease, Dr. Marc Fischer said, "C. sordellii toxic shock syndrome. . . . is an acute and rapidly progressive disease that is characterized by a lack of or minimal fever."46 The FDA Public Health Advisory: Sepsis and Medical Abortion points out that deaths have occurred without fever and other usual signs and symptoms of an infection.47

Following the third and fourth mifepristone associated deaths, on July 19, 2005, the FDA updated the Mifeprex (mifepristone) Label,48 Medication Guide,49 and Patient Agreement50 for Mifeprex (mifepristone) to include information concerning infection with C. sordellii.51,52 The updated prescribing information alerts abortion providers and emergency room health care providers concerning how to evaluate for this diagnosis in patients who may not present with usual signs of infection. The information given to patients advises them as to what symptoms require emergency evaluation. Because the infections have had atypical presentations, and also have had a rapidly fatal course, it is of particular importance to caution patients and providers so that prompt diagnosis can be made.

The RCOG Guideline recommends prophylactic antibiotics to prevent infection from abortion, though several speakers at the CDC-FDA workshop recommended against routine use of prophylactic antibiotics due to issues relevant to Clostridial infection. Speaking of experience with C. difficile, (closely related to C. sordellii), Dr. Seligman pointed out that infections "occur mostly in association with antibiotic use,"53 and Dr. Gerding agreed that prior antibiotic exposure was "the largest risk factor," indicating that normal bacterial flora was "protective."54 Elimination of the normal flora could allow Clostridial infection to flourish. Dr. McGregor, speaking specifically about possible antibiotic prophylaxis at the time of mifepristone abortion, stated, "I do not suggest ... that antimicrobial prophylaxis, either short or long course which has been suggested, would be likely effective. Indeed, it might be just the opposite."55 He stated that routine antibiotic prophylaxis was a "bad idea."56 The FDA's Public Health Advisory: Sepsis and Abortion also recommended against this.57

Speaking at the CDC-FDA workshop, Dr. Soper raised the concern that these infections have "such a rapid downhill course that there really is little opportunity for any kind of therapeutic intervention at all,"58 commenting on the need for prevention, but also agreeing that antibiotic prophylaxis was not advisable. Dr. McGregor instead proposed, "I would suggest that we either reduce or eliminate mifepristone or at least consider that" as a means of prevention for abortion related Clostridial infection. And I note that there doesn't appear to be a safe or specific alternative that only gets the progesterone receptor"59 The best prevention of these fatal infections would be to eliminate use of mifepristone, and in addition, there is not going to be a safe alterative since there are alternatives which only act on the progesterone receptors without acting on the glucocorticoid receptors.

McGregor recommended aggressive management of C. sordellii infection should include "rapid" admission to hospital, that consultations should be obtained from infectious disease specialists and intensivists, and that multiple organ support is started "promptly."60 He also proposes that if Clostridia sordellii-associated toxic shock is definitely the diagnosis, that "immediate" hysterectomy or D & C be considered in order to remove the source of the infection.61

FDA's Dr. Sandra Kweder expressed concern that there could be additional deaths due to C. sordellii sepsis which may be undiagnosed and attributed to unknown causes. She pointed out that these cases were likely to be attended by an emergency physician rather than an infectious disease specialist or a gynecologist.62

The Guideline does not discuss issues related to pharmacokinetics of mifepristone including its long half-life, metabolic pathways, and drug interactions.

Dr. Miech, speaking at the CDC-FDA workshop stated that mifepristone normally has a half-life of 20-30 hours, so that it normally would take about four or five days before 95% of a single dose of mifepristone would be broken down. However, he said in some individuals who process the drug more slowly, it can have a 90 hour half-life and would then be active for 18 days.63

Meich stated that mifepristone is broken down in the liver by cytochrome P450-3A4. "There is evidence that the enzyme itself can be inactivated during the metabolism of mifepristone. This probably accounts for the relatively long biologically half-life seen in humans."64 So mifepristone is normally broken down very slowly, but in addition, if mifepristone causes the suppression of the enzyme responsible for its breakdown, then that would explain mifepristone still being active in the body days or weeks later.

Additionally, Miech stated that cytochrome P450-3A4 which breaks down mifepristone is also responsible for the metabolism of codeine. He stated, "In mifepristone abortions, women are frequently prescribed codeine for pain. Thus, codeine would compete for the enzyme that metabolizes mifepristone and prolong its biological half-life."65

In addition, Miech stated that mifepristone is broken down into six metabolites, and that some of these metabolites have biological activity.66 McGregor also stated there are active metabolites, which he stated are active at both progesterone receptors and at glucocorticoid receptors.67 It is through its action on the glucocorticoid receptors that mifepristone disrupts the innate immune system68 but here we see that not only is it mifepristone but also its breakdown products which exert this effect.

Further, even "small concentrations of the molecule are quite efficient at saturating both progesterone receptors and glucocorticoid receptors,"69 meaning that even when time has permitted the breakdown of much of the original dose, small amounts that remain may continue to have an effect. McGregor gave an example of one study in which the effects on the innate immune system persisted for one week.70

Recommendation 16 of the RCOG Guideline states that "abortion is safer than continuing a pregnancy to term and that complications are uncommon."71 However, more recent reports have shown that the death rate from infection following medical abortion (mifepristone with misoprostol) has been ten times higher than the death rate from infections following surgical abortion 72,73,74,75 and 50 times more compared to childbirth.76 Dr. McGregor, speaking at the CDC-FDA workshop, stated that "surgical termination, which appears to be better studied, appears to have a lower risk in terms of patient mortality.77 He also stated that informed consent forms should be changed to acknowledge the risk of serious infections and complications.78 Recommendation 16 would inform women that complications of abortion are "uncommon,"79 yet the FDA mandated labeling information for mifepristone states that "nearly all of the women who receive Mifeprex and misoprostol will report adverse reactions, and many can be expected to report more than one such reaction."80 The mifepristone label provides a list of 27 adverse effects which occurred in at least 1% of participants in the French and U.S. clinical trials.81 Of these listed symptoms, the most frequent were abdominal pain (reported by 96% of U.S. participants), cramping (reported by 83% of the French), nausea (61 % U.S.), headache (31% U.S.), vomiting (26% U.S.) and diarrhea (20% U.S.).82

Especially worrisome, though lower in number, were the 5% who experienced uterine hemorrhage, the 6% (French) who experienced decrease in hemoglobin greater than 2 g/dl. and the 1% (U.S.) who experienced endometritis, salpingitis, and/or pelvic inflammatory disease.83 The mifepristone label indicates that by day 14 reports of adverse effects were rare except for bleeding which is expected for an average of 9 days but which will continue for more than 30 days in 8% of the cases.84

In regard to the severity of the various side effects the label states: "The percentage of women who considered any particular adverse event as severe ranged from 2 to 35% in the U.S. and French trials."85 The FDA's Medical Officer's review indicates that most patients reported more than one adverse event, and that "approximately 23% of the adverse events in each gestational age group were judged to be severe."86

The Guideline does not provide guidance on which patients must be excluded from use of mifepristone. For example, the FDA Medication Guide cautions the patients not to take Mifeprex if the patient is taking certain steroid medications, is taking anticoagulant medication, has a bleeding problem, if she cannot return for the next two visits, or if she could not easily get emergency medical help during the two weeks after taking Mifiprex.87 The FDA mandated labeling in the U.S., under the heading "Contraindications," cautions physicians that mifepristone must not be used in cases of chronic adrenal failure, in patients with hemorrhagic disorders or concurrent anticoagulant therapy, or in patients with receiving long-term corticosteroid therapy.88 For the complete list of exclusions, refer to the Mifeprex Guide89 and Mifeprex Label,90 as these are examples and not a comprehensive list.

In a separate section titled "Precautions," the FDA product label points out that "there are no data on the safety and efficacy of mifepristone in women with chronic medical conditions such as cardiovascular, hypertensive, hepatic, respiratory or renal disease; insulin dependent diabetes mellitus; severe anemia or heavy smoking."91

The Spitz study specifically excluded "women with liver, respiratory, renal, adrenal, or cardiovascular disease, thromboembolism, hypertension, anemia, insulin-dependent diabetes mellitus, coagulopathy, or known allergy to prostaglandins . . . women less than 18 years. . . or those more than 35 years who smoked more than 10 cigarettes per day and had another cardiovascular risk factor."92 Spitz had seven other exclusion criteria.

Next Page: continuation...
1, 2, 3