If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The proportion of US abortions performed in the second trimester has varied little since 1992. Although 30 years of cumulative data corroborate the safety of dilation and evacuation (D&E), the most commonly used method of second-trimester abortion in the United States, both D&E and alternative induction regimens continue to evolve such that the traditional safety gap between medical and surgical regimens has narrowed. Providers now have options that allow them to either expedite D&E by diminishing the cervical-ripening period or reduce induction abortion intervals during medical induction.
a contributing author in the text Second Trimester Abortion: Perspectives after a Decade of Experience wrote “…there is no reason, logically speaking, that second-trimester abortion should not join bubonic plague and poliomyelitis as practically historic medical conditions.” Dr Connell
predicted that within the coming decade the use of more effective contraceptives and increased access to first-trimester abortion would make second-trimester abortion obsolete–a “therapeutic memory” rather than a “medical reality.” Now, > 2 decades later, her vision seems no more probable than the end of poverty, hunger, or taxes. Second-trimester abortion comprises 10-15% of the 42 million abortions that occur worldwide each year.
The proportion of US abortions performed in the second trimester has varied little since 1992. According to surveillance data from the Centers for Disease Control and Prevention (CDC), 12% of abortions occur at or after 13 weeks' gestation. In all, 3.7% of all abortions occur at 16-20 weeks and 1.3% at ≥ 21 weeks.
Although second-trimester terminations represent a small percentage of total abortions, they still account for approximately 130,000 procedures annually in the United States.
has proved so elusive. In part, it is because her underlying premise that women would use effective contraception and access first-trimester abortion services has failed to materialize. Approximately 50% of all pregnancies in the United States are unintended and roughly 50% of unintended pregnancies are terminated. Although some second-trimester abortions occur because of maternal disease and fetal anomalies, the majority occur because of delay in obtaining first-trimester abortion in unintended pregnancies. At one large US public hospital, 58% of patients having second-trimester procedures were already beyond the first trimester by the time they obtained a pregnancy test. Second-trimester patients were less certain of their last menstrual period, had fewer pregnancy-related symptoms, and were more likely to report recent use of hormonal contraception than other patients.
reported similar findings in their evaluation of the reasons for delay in accessing abortion services. Second-trimester patients required more time to make arrangements for abortion (59%), more time to diagnose pregnancy (36%), and more time to decide whether to terminate (39%) than first-trimester patients. Difficulty securing financial resources–or finding a provider who accepted a particular insurance coverage–impeded many women's efforts to secure timely abortion. Ironically, legislation such as the Hyde Amendment, which has since 1977 forbidden the use of federal funds for abortion, continues to increase the need for second-trimester abortion services.
Although second-trimester abortion accounts for a relatively small proportion of all induced abortions, it is associated with disproportionate morbidity. Two-thirds of major abortion-related complications and half of abortion-related mortality occur in pregnancies terminated after 13 weeks of gestation, most commonly in countries that restrict access to safe abortion.
In countries with legal abortion, the risk of complications from second-trimester abortion–both medical and surgical–is low. How to perform abortion in the second trimester, particularly whether to induce labor or surgically evacuate the uterus, remains subject to regional variations that derive as much from custom and training as medical evidence. Although the old adage “If it isn't broke, don't fix it” might facilitate the continued delivery of safe abortion services, failure to implement new evidence-based practices denies patients and providers access to the full range of surgical and medical options now available throughout the second trimester.
Surgical abortion
Historical background
Dilation and evacuation (D&E), the most prevalent method of second-trimester pregnancy termination in the United States, accounts for > 98% of all second-trimester abortions.
Despite its general acceptance in the United States and 30 years of data confirming its safety, D&E remains a relatively recent surgical innovation that continues to evolve.
In 1973, limited data existed to compare the relative safety advantages of D&E vs medical abortion or hysterotomy in the second trimester. Data from the Joint Program on the Study of Abortion, a prospective chart review of thousands of abortions cosponsored by The Population Council and the CDC during the 1970s, suggested lower rates of hemorrhage and infection with D&E compared with other methods used at the time.
Indeed, patients undergoing abortion through instillation of urea or hypertonic saline experienced twice the rate of major complications than patients undergoing D&E.
As a result, the proportion of US abortions performed by D&E at ≥ 13 weeks' gestation increased from 31% in 1974 to 97% in 2004, whereas the percentage of abortions performed by intrauterine instillation at ≥ 13 weeks' gestation decreased from 57-0.5% during the same time period.
Observational data and several retrospective cohort trials in the 1980s consistently confirmed the safety advantages of D&E vs medical induction throughout much of the second trimester.
from the CDC noted a reduction from 10.4 deaths per 100,000 cases between 1972 and 1976 to 3.3 deaths per 100,000 cases between 1977 and 1982. Unfortunately, the CDC cannot calculate national abortion case-fatality rates for 1998-2002, the most recent study interval, because a substantial number of the abortions occurred in states not reporting data to the CDC. Thus, the total number of abortions, or denominator, is unknown. Nevertheless, only 10 US women died as a result of complications among the roughly 850,000 induced abortions reported to CDC in 2004, the vast majority of those procedures accomplished by D&E.
Recent trials further document the general safety of D&E, including its impact on subsequent pregnancy outcome. In a retrospective review by Kalish et al
of 600 patients undergoing D&E between 14 and 24 weeks, the overall rate of preterm birth in subsequent pregnancies was less than the overall rate of preterm birth for the general US population (6.5% vs 12.5%). Similarly, Jackson et al
compared subsequent pregnancy outcomes among 317 women undergoing second-trimester D&E with 170 matched control subjects who had experienced viable pregnancies without midtrimester D&E. Although patients with a history of prior D&E delivered slightly earlier in gestation than control subjects (38.9 vs 39.5 weeks' gestation; P = .001) there was no statistically significant difference in birth weight, spontaneous preterm delivery, abnormal placentation, or overall rates of perinatal complications.
In addition to safety, surgical abortion offers many perceived advantages compared with medical abortion. D&E affords both patients and clinicians more predictable procedure timing. The patient typically undergoes between 1 and 2 days of preoperative cervical preparation with osmotic dilators, chemical ripening agents, or a combination of the 2. Experienced clinicians can accomplish D&E in < 30 minutes as an outpatient procedure, and patients commonly return to work the day after the procedure. Many patients find that the predictability of surgical abortion and avoiding the memory of prolonged labor make D&E less emotionally burdensome than induction abortion.
attempted to perform a randomized clinical trial comparing D&E with medical induction, 62% of women did not consent to randomization because of the many apparent advantages of D&E. Unfortunately, many women in the US have little choice in method of second-trimester termination because of impaired access to second-trimester surgical abortion services. The most critical requirement for any safe D&E program is a surgeon skilled and experienced in D&E provision. Many of the most skilled providers will soon reach retirement age and it is unclear whether a new generation of trained providers will replace them. A national survey of obstetrics and gynecology residency program directors found that 51% of programs offered routine abortion training compared with only 12% in 1992. In programs offering routine training, however, most (64%) trained less than half of their residents in D&E techniques, and very few offered the volume of procedures necessary to attain competence.
The recent development of Ryan Training Programs, to assure resident training in comprehensive abortion services, and family planning fellowships, to create an academic subspecialty committed to training and providing comprehensive family planning services, both offer hope that US women will retain access to such a safe and convenient method of uterine evacuation.
Recent advances
Unfortunately, the performance of D&E at later gestational ages often requires multiple sets of osmotic dilators over multiple days increasing the financial burden related to travel, lodging, and time missed from work.
To address this issue, during the past decade, providers have increasingly used misoprostol, a synthetic PG E1 analogue, either as a sole ripening agent or as an adjunct to traditional mechanical and osmotic dilation.
Although many studies document the safety and efficacy of misoprostol preceding first-trimester aspiration procedures, few compare the safety and efficacy of misoprostol vs osmotic dilation preceding second-trimester abortion.
Most comparisons of these methods consist of retrospective, observational reports designed to establish the safety of misoprostol and adequacy of dilation before second-trimester abortion. For example, Todd et al
reported adequate cervical preparation in 32 women who received 600 μg of buccal misoprostol 2-4 hours before D&E at 14-16 weeks of gestation. Procedure time was comparable with women at 16-18 weeks of gestation who had cervical ripening with laminaria (n = 78). Patel et al
analyzed data from 2218 D&E procedures between 12 and 23 6/7 weeks' gestation in which clinicians at multiple clinic sites used cervical preparation consisting of various regimens of buccal misoprostol with or without osmotic dilators. Although the authors concluded that buccal misoprostol could safely reduce or eliminate the need for multiday laminaria treatments, they recognized that their retrospective series was not designed to truly compare methods. They also noted that adequacy of dilation was generally higher for laminaria vs no laminaria regardless of misoprostol use. When misoprostol was used alone, a strong association emerged between need for additional dilation and gestational age: patients at lower gestational ages were much more likely to require additional dilation. Although Patel et al
conclude that misoprostol is safe and holds promise as a primary cervical-ripening agent, the study's limitations suggest the need for caution before completely abandoning traditional osmotic dilation. The authors correctly underscore the need for randomized, prospective trials in this area.
performed the only randomized, double-blinded, controlled trial comparing misoprostol with the traditional practice of overnight laminaria before second-trimester surgical abortion. Patients at 13.0-16.0 weeks' gestation (n = 84) received either 400 μg of vaginal misoprostol 3-4 hours preoperatively or overnight laminaria. The primary outcome was procedure time; secondary outcomes included completion of the procedure on the first attempt, procedural difficulty, and patients' pain scores and preferences. Second-trimester abortions after same-day misoprostol took longer and were technically more challenging than those after overnight laminaria. Patients, however, preferred a same-day regimen, and the vast majority of procedures in both groups were accomplished safely and with adequate dilation. No study has directly compared same-day second-trimester abortion procedures using preoperative osmotic dilators vs misoprostol.
Regardless of misoprostol's safety and efficacy as a sole dilating agent, it often serves as a useful adjunct to traditional osmotic dilation, particularly when the cervix resists placement of adequate numbers of osmotic dilators. However, it remains unclear exactly when to use misoprostol as an adjunct, in what dose, and via what route of delivery. To ascertain whether adjuvant buccal misoprostol truly improves cervical preparation with laminaria, Edelman et al
Cervical preparation using laminaria with adjunctive buccal misoprostol before second-trimester dilation and evacuation procedures: a randomized clinical trial.
performed a randomized, double-blind, placebo-controlled trial comparing preoperative cervical preparation with overnight laminaria and either buccal placebo or 400 μg of buccal misoprostol 90 minutes before second-trimester D&E at 16-21 weeks' gestation. Although some surgeons subjectively reported easier dilation after pretreatment with misoprostol, the study found no objective differences in cervical dilation measured by passage of rigid dilators, need for additional dilation, or duration of procedures at < 19 weeks' gestation. In addition, women receiving buccal misoprostol reported more side effects. The authors concluded that use of misoprostol as an adjunct to laminaria augments cervical dilation in gestations beyond 19 weeks; earlier use might increase discomfort without conferring an objective benefit.
Investigators have evaluated several other pharmacologic agents as primary or adjunctive ripening agents including gemeprost, meteneprost, PG F2α, and PG E2 suppositories.
Although expense or lack of availability currently make mifepristone an impracticable cervical-ripening agent for physicians in the United States or Canada, it has clinical value either alone or in combination with misoprostol or laminaria. Carbonell et al
Vaginal vs sublingual misoprostol with mifepristone for cervical priming in second-trimester abortion by dilation and evacuation: a randomized clinical trial.2007; 75: 230-237
evaluated the efficacy of mifepristone among 900 women undergoing elective termination of pregnancy by D&E at 12-20 weeks' gestation. They randomized patients to 1 of 4 groups: mifepristone 200 mg plus sublingual misoprostol 600 μg before D&E; mifepristone 200 mg plus vaginal misoprostol 600 μg before D&E; sublingual misoprostol 600 μg before D&E; and vaginal misoprostol 600 μg before D&E. Administering mifepristone before D&E with misoprostol in the second trimester decreased operating time and decreased the risk of cervical injury. However, mifepristone use also increased procedure cost by approximately 25 euros per procedure, total patient visits, and number of pre-D&E fetal expulsions. Noted advantages of adjuvant mifepristone included decreased waiting time after administering misoprostol (1.7 ± 0.6 vs 2.1 ± 0.7 hours; P < .001), a significant decrease in the number of osmotic dilators used, and greater cervical dilatation obtained after preoperative treatment with mifepristone. The difference in degree of mean cervical dilation obtained after mifepristone was noted in both the sublingual group (12.6 ± 2.1 vs 8.9 ± 3.0 mm) and the vaginal group (12.4 ± 3.3 vs 8.1 ± 3.3 mm). Given the inverse relationship between cervical dilatation and operative morbidity, it is possible that adjuvant mifepristone could decrease overall surgical morbidity during D&E.
Medical abortion
Historical background
Women have attempted to induce abortion throughout history using whatever devices and folk remedies nature and tradition allowed. Plato, Aristotle, Hippocrates, and Soranus all discussed methods of medically inducing abortion.
Even today, an Internet search of “herbal” and “induced abortion” identifies hundreds of sites touting pennyroyal with blue cohosh, mugwort, cotton root bark, and other herbal products either as teas or decoctions to medically evacuate the uterus. Fortunately, a far safer array of second-trimester abortifacients has evolved during the past century, some so effective as to have narrowed the traditional safety gap between medical and surgical abortion.
In the 1940s, physicians began injecting hypertonic agents into the amniotic cavity to “salt out” the fetus. Hypertonic saline, the first agent commonly used for intraamniotic induction termination, became the mainstay of second-trimester medical abortion through the 1970s, although most physicians have abandoned its use because of risks such as hypernatremia, coagulopathy, and massive hemorrhage.
In 1 recent randomized trial comparing hypertonic saline to misoprostol induction, hypertonic saline resulted in longer induction times and higher rates of both blood transfusion and retained placenta.
A randomized comparison of misoprostol to intrauterine instillation of hypertonic saline plus a prostaglandin F2 analogue for second trimester induction termination in Uzbekistan.
Instillation regimens using hyperosmolar urea are less commonly associated with hypernatremia or coagulopathy than hypertonic saline but have not been directly compared with more recent induction regimens.
Hyperosmolar regimens usually require concomitant use of medical induction agents to stimulate contractions and delivery. Oxytocin, the most commonly used induction agent at later gestational ages, fails to induce labor effectively as single-agent therapy at midtrimester as the number of myometrial oxytocin receptors increase with advancing gestational age. Compared with misoprostol, oxytocin requires longer induction times and has higher rates of serious side effects, particularly water intoxication.
The most important recent advance in medical abortion procedures has been the use of PG such as carboprost, sulprostone, gemeprost, and misoprostol. Of these, carboprost, an F2α PG analogue, is less well tolerated than E analogues and infrequently used. Sulprostone, an E2 analogue, is associated with myocardial infarction and is also no longer used.
Gemeprost and misoprostol, both E1 analogues, remain the most common PG analogues currently used for induction abortion. Both induce labor more effectively and with fewer side effects than intraamniotic PG F2α and extraamniotic PG E2.
The use of 16,16-dimethyl-trans-delta-2 prostaglandin-E1 methyl-ester (gemeprost) vaginal pessaries for the termination of pregnancy in the early 2nd trimester: a comparison with extra-amniotic prostaglandin-E2.
PG E1 is also more effective than second-trimester induction regimens using single-agent oxytocin or oxytocin in combination with other types of classes of PG.
A metaanalysis of randomized trials comparing misoprostol with gemeprost in midtrimester abortion, using various regimens of each, demonstrated that vaginal misoprostol compared with gemeprost vaginal suppositories was associated with reduced need for narcotic analgesia and surgical evacuation of the uterus.
No other statistically significant differences were observed. Gemeprost is still used in some settings for cervical preparation prior to a surgical uterine evacuation, midtrimester abortion, and treatment after intrauterine fetal death.
Misoprostol has the advantage of being effective at initiating uterine contractions and cervical ripening at any gestational age (unlike agents such as oxytocin) and is used in decreasing doses as the pregnancy advances.
reported second-trimester pregnancy termination using 200 μg of misoprostol every 12 hours. Numerous investigations since that time report superior results using higher doses of misoprostol.
The evidence-based regimen of the Royal College of Obstetricians and Gynecologists (RCOG) is 400 μg of vaginal misoprostol every 3 hours, up to 5 doses, in pregnancies between 13-22 weeks.
After 22 weeks' gestation, the misoprostol dose and frequency of administration should be reduced. In regimens using misoprostol or gemeprost alone, median induction abortion intervals often approach 12-16 hours.
The abortion time interval is decreased with more frequent dosing: 2 randomized clinical trials demonstrate a significantly shorter induction time with 3-hourly vaginal administration of 400 μg of misoprostol than 6-hourly administration, without a significant increase in side effects.
Misoprostol inductions less commonly require surgical intervention for delivery of the placenta than induction with other agents. Providers have previously performed curettage if the placenta remains undelivered 30-120 minutes after delivery of the fetus. This is based on reports that retained placenta after saline and PG E2-induced abortion was associated with higher rates of complications with increasing time after delivery.
of misoprostol for second-trimester induction, 59% of placentas delivered within 1 hour of fetal expulsion and the overall rate of surgical intervention for retained placenta was 6.14% (95% confidence interval: 3.00-9.26). The authors recommended that physicians observe nonbleeding patients for at least 4 hours for spontaneous placental expulsion after fetal delivery.
Because of its efficacy and favorable side-effect profile, single-agent misoprostol has already become the most common method of second-trimester abortion in many parts of the world.
United Nations Development Program/United Nations Fund for Population Activities/World Health Organization/World Bank Special Program for Research and Development and Research Training in Human Reprod Annual technical report 1997.
World Health Organization,
Geneva, Switzerland1998
Use of PG E1 analogues in combination with mifepristone offers the safest and most expeditious method to induce abortion in the second trimester. Both the World Health Organization and the RCOG recommend regimens in which mifepristone precedes use of either misoprostol or gemeprost.
Although such regimens represent standard of care throughout much of the world, obstetricians in the United States and many other countries confront numerous obstacles accessing mifepristone for patients undergoing labor induction.
Mifepristone is an antiprogestin that competitively blocks both progesterone and glucocorticoid receptors. By opposing the activity of progesterone, mifepristone elicits a variety of effects that make the uterus more susceptible to abortion. These effects include cervical dilation, decidual necrosis, increased endogenous PG production, and increasing uterine sensitivity to exogenously administered PG. Indeed, mifepristone administration gradually elicits a 5-fold increase in sensitivity to PG 24-48 hours after its administration.
The synergy between mifepristone and PG permits greater efficacy of PG at lower doses, minimizing side effects.
Inductions using mifepristone followed by PG require roughly half the time as inductions using PG alone, and roughly 95% of patients deliver within 24 hours.
Studies of second-trimester medical terminations using mifepristone and misoprostol confirm its safety, reporting no maternal deaths and complication rates at or below those reported with other induction regimens.
The decrease in induction-to-abortion time and side effects removes the most serious psychological and clinical hurdles obstetricians have previously confronted when considering induction.
The Figure summarizes the RCOG recommended regimen for second-trimester induction using mifepristone and misoprostol. Physicians unfamiliar with mifepristone ask several questions when first contemplating its use:
•
Can providers substitute vaginal for oral misoprostol if patients prefer that route of administration?
FIGURERoyal College of Obstetricians and Gynecologists recommended regimen
Induction of abortion in the second trimester by a combination of misoprostol and mifepristone: a randomized comparison between two misoprostol regimens.
reported no difference in efficacy of vaginal vs oral misoprostol for second-trimester abortion after an 800-μg vaginal loading dose. Other studies have confirmed similar safety and efficacy for vaginal and oral misoprostol.
studied sublingual misoprostol for second-trimester abortion. Vaginal administration was more successful at 24 hours, but both regimens were similarly effective at 48 hours. Many women prefer oral to vaginal administration, but providers may administer according to patient preference.
•
Is 200 mg of mifepristone equally effective as 600 mg?
As with first-trimester abortion, randomized studies consistently indicated 200 mg of mifepristone is equally effective as 600 mg for termination of pregnancy in the second trimester.
Can providers shorten the interval between administration of mifepristone and PG analogue without sacrificing efficacy or increasing side effects?
When the interval between mifepristone and misoprostol administration was reduced to 24 hours, the induction-to-abortion interval was somewhat longer than after a 48-hour interval. In a retrospective study, time to fetal expulsion was 9.8 hours in the 24-hour interval group compared with 7.5 hours when the interval was 48 hours (P < .01), and in a randomized study it was 7.3 vs 6.2 hours (P < .05), respectively.
This latter study also reported a higher rate of uterine curettage with the 24-hour interval (P < .001).
•
Does patient parity impact efficacy of recommended regimens?
Studies using mifepristone with a PG E1 analogue have demonstrated statistically lower induction-to-abortion intervals among parous women when compared with nulliparous women.
This is not a typical finding with PG E1analogue alone.
•
Does use of osmotic dilators, such as laminaria tents, or mifepristone impact efficacy or increase side effects?
Placement of osmotic dilators at the beginning of induction does not shorten the induction-to-abortion interval or decrease complication rates with either gemeprost or misoprostol.
Two randomized trials evaluating laminaria before misoprostol induction demonstrated longer median abortion intervals, and women who received laminaria experienced greater discomfort during induction.
compared Dilapan (Dilapan-S; Gel-Med International, Prague, Czech Republic) dilators placed 6 hours prior to gemeprost induction with patients who received pretreatment with mifepristone or no pretreatment. Dilapan demonstrated no benefit when compared with gemeprost alone, but mifepristone significantly shortened induction-abortion intervals.
•
How common is retained placenta with regimens using mifepristone-PG E1 analogue?
Studies demonstrate rates of surgical intervention between 2.5% and 10% when using mifepristone-misoprostol regimens.
A randomized trial of mifepristone in combination with misoprostol administered sublingually or vaginally for medical abortion at 13-20 weeks gestation.
These rates are similar to those reported with single-agent misoprostol therapy and lower than those reported for other induction regimens.
•
Can providers substitute gemeprost for misoprostol without sacrificing efficacy or increasing side effects?
Several trials compare gemeprost and misoprostol as single induction regimens. In most trials, misoprostol is as effective or more effective than gemeprost and better tolerated.
randomized 100 women to receive either misoprostol or gemeprost administered at 6-hour intervals after preadministration of mifepristone. Complete abortion rates, induction-to-abortion intervals, surgical evacuation rates, and side effects were similar in the 2 groups. Gemeprost, however, requires refrigeration, has more limited shelf life than misoprostol, and is more expensive than misoprostol. It is currently unavailable in the United States.
Making choices
Although D&E is a very safe and effective procedure, its safety profile derives from the surgeon's knowledge, experience, and skills.
In the past, institutions that have lacked skilled D&E providers have had to refer patients to other facilities or choose medical induction regimens characterized by long induction-abortion intervals and relatively high rates of complications. This is no longer the case. Although few studies directly compare surgical abortion with recent medical regimens, newer medical regimens using PG E1 analogue with or without mifepristone offer sufficiently effective, well-tolerated abortion-induction intervals so that clinicians have far greater latitude choosing a medical vs surgical approach. Furthermore, studies of first-trimester medical abortion suggest that when a nonsurgical option for abortion exists, many women will choose it in hopes of avoiding instrumentation or assuring greater privacy.
This suggests that some women might also prefer medical induction to surgical abortion in the second trimester. Several common clinical situations warrant further consideration.
Fetal demise
Low cost, ease of administration, and presumed efficacy have made misoprostol a drug of choice for inducing labor in the setting of second-trimester fetal demise. During the past decade, many trials have evaluated misoprostol in this setting, but the trials differ remarkably in study population, dose, route of administration, and definition of therapeutic success. Many studies, for example, have included patients with live fetuses. Some studies have limited success to delivery within 24 hours whereas others have permitted patients up to 72 hours. Finally, some studies have restricted successful induction to those patients delivering only through the use of misoprostol while other studies have included patients requiring a rescue induction agent, such as oxytocin. To date, no formal dose-finding studies have been performed in the setting of second-trimester fetal demise.
Although at least 3 studies have evaluated pretreatment with mifepristone before misoprostol induction for second-trimester demise, there remains no proven advantage to routinely justify the additional cost.
Given the widespread use of misoprostol in this setting, more studies are urgently needed.
Although many physicians believe that induced fetal demise facilitates both second-trimester medical and surgical abortion, few studies substantiate this claim.
The 2 most common methods of inducing demise are the administration of intraamniotic or intrafetal digoxin, usually 24 hours before induction or D&E, or the administration of intracardiac potassium chloride (KCl), immediately before induction or D&E.
demonstrated no difference in blood loss or complications among women receiving intraamniotic digoxin before D&E vs those who received placebo. Although women receiving digoxin were more likely to experience vomiting (16% vs 3%) than those receiving placebo, 91% of patients preferred preoperative administration of a feticidal agent for emotional reasons.
Although several investigators have established the safety and favorable side-effect profile of digoxin preceding induction abortion, no randomized trials have evaluated whether it significantly influences induction abortion interval.
reported that women receiving feticidal KCl the day prior to induction with vaginal PG E2 experienced similar side effects but shorter induction times than women not receiving not receiving feticidal injection.
Many providers have begun to induce fetal demise prior to abortion to avoid violating recent US legislation such as the Partial Birth Abortion Act of 2003. Given the general safety of these procedures, this is a reasonable consideration, but proper informed consent requires that the patient understand the paucity of data describing a clear medical benefit on her behalf. Furthermore, although intracardiac KCl elicits immediate fetal asystole, intrafetal digoxin may fail to effect demise in up to 5% of cases whereas intraamniotic digoxin may fail to cause demise in up to one third of cases.
Increasing rates of cesarean delivery mean that more women with a history of uterine scarring will confront second-trimester induction. Although the absolute risk of uterine rupture during second-trimester induction is unknown, most studies suggest that misoprostol can safely be used in postcesarean pregnancies.
Case reports describe uterine rupture among women with and without prior uterine scars undergoing PG induction throughout a wide range of gestational ages.
for example, conducted a retrospective study of 252 women between 15 and 35 weeks, 50 of whom had uterine scars. Although prior cesarean section did not increase the risk of hemorrhage (2% vs 0.9%; P = .56) or the median induction-abortion interval (8.5 vs 9.0 hours; P = .26), 1 case of uterine rupture and 1 case of uterine dehiscence were noted, both in women with prior cesarean section. Conversely, a series reported by Daskalakis et al
of 108 women with prior cesarean undergoing second-trimester misoprostol induction reported only 1 case of uterine rupture but it occurred in 1 of the 216 nonscarred control subjects. In most series, rates of uterine rupture in midtrimester among patients with scarred uteri are < 1%, although the impact of PG dose and dosing interval remain unclear. Because rates of rupture may increase with advancing gestational age some authors have advocated avoiding its use in the third trimester.
Furthermore, because previous studies excluded women with prior scarring from myomectomy or surgical correction of uterine anomalies, there is currently little evidence to guide physicians caring for these patients.
reported no increased risk of hemorrhage during second-trimester D&E among patients with placenta previa. No patient experienced significant blood loss after placement of laminaria or as a result of the surgical procedure. Although the risk of hemorrhage associated with third-trimester placenta previa causes most physicians to avoid second-trimester induction in this setting, several authors have suggested reconsidering this practice. Nakayama
evaluated the impact of placenta previa on blood loss among 158 patients undergoing second-trimester gemeprost pregnancy termination. Among the 11 patients with placenta previa, 4 underwent D&E and 7 underwent medical termination. Although 1 patient undergoing induction required transfusion, differences in mean blood loss were not significant. The authors noted a larger published series by Thong et al
and an unpublished series by Deguchi that also suggest safety of gemeprost induction in cases of second-trimester placenta previa. Until further data become available, D&E remains the safest and most prudent option in most of these cases.
Fetal anomalies
The widespread use of prenatal diagnosis has increased the number of patients undergoing second-trimester termination for fetal anomalies. Patients desiring the opportunity to hold the fetus sometimes prefer medical induction, although prolonged and unpredictable induction abortion intervals seen with older induction regimens can augment the emotional burden of an already troubled pregnancy. Newer regimens involving the use of mifepristone before PG offer more reasonable induction-abortion intervals and might decrease the emotional trauma of some inductions. Because fetal disarticulation does not always preclude anatomic diagnosis of suspected fetal anomalies, many patients with suspected fetal anomalies choose to undergo D&E.
Haskell M. Dilation and extraction for late second trimester abortion. Presented at: National Abortion Federation Risk Management Seminar. Dallas, TX, September 13, 1992. Oral presentation.
removes the intact or nearly intact fetus through the cervix. The technique minimizes the number of instrumental passes and permits the operator to perform a greater portion of the extraction under direct visualization within the vagina, decreasing the likelihood of uterine perforation and cervical abrasion. Because the fetus is removed intact, it permits patients the opportunity to hold their intact fetus and might allow superior morphologic evaluation to further clarify prenatal diagnosis.
retrospectively compared outcomes in 383 women undergoing surgical abortion at ≥ 20 weeks' gestation. A total of 263 women underwent D&E with disarticulation, and 120 women underwent D&X. D&X was associated with higher parity, later gestational age (median, 23 vs 21 weeks), and more preoperative cervical dilation (median, 5 vs 3 cm) than disarticulation. There was no difference in estimated blood loss or operative time between the intact extraction and disarticulation groups. The overall rate of complications was 5% in both groups. Only 4 major complications, requiring blood transfusion, laparotomy, or hospitalization, occurred in the group undergoing D&E. No major complications occurred in those undergoing D&X.
D&X is a safe procedure associated with a low rate of major complications. Because the Partial Birth Abortion Act of 2003 criminalizes variants of D&E and provides no legal exception for cases performed to protect maternal health, physicians may want to consider preoperative feticidal injection whenever they intend to perform an intact extraction.
Multifetal gestation
Operators may terminate multifetal gestation surgically or medically, depending on patient preference and any other mitigating factors. There is no evidence that patients with multifetal pregnancy undergoing medical induction require modification of medical regimen or surgical technique.
Conclusion
What is the future of second-trimester abortion? Given the high rate of unintended pregnancy in both the developed and developing world as well as the increasing use of prenatal diagnosis to diagnose anomalies in midtrimester it is unlikely that second-trimester abortion will soon join medical history's dust heap. The question is not really whether second-trimester abortion will exist, but how physicians will adapt to current and developing technologies. This review suggests several short term challenges for all current practitioners.
•
Practitioners should become increasingly comfortable offering patients either modern methods of labor induction or D&E. Institutional protocols often derive as much from local custom as evidence. Institutions lacking skilled D&E providers should recruit them. Institutions inclined toward D&E should reconsider induction as a comparable option in light of more rapid induction-abortion intervals offered by regimens combining mifepristone and misoprostol.
•
Large scale abortion providers should exercise prudence in using misoprostol as sole cervical-ripening agent in the late second trimester until better evidence documents its safety relative to traditional osmotic dilators.
•
Given the demonstrated safety and efficacy of mifepristone in the second trimester, individuals and agencies caring for women should advocate for improved access. Physicians should become registered prescribers, hospitals should add it to formularies, and public agencies that restrict its use should become aware of its vast potential to improve women's health.
References
Connell E.B.
Preventing the need for second-trimester abortion.
in: Berger G. Brenner W. Keith L.G. Second trimester abortion: perspectives after a decade of experience. John Wright PSG Inc,
Boston1971: 325
Cervical preparation using laminaria with adjunctive buccal misoprostol before second-trimester dilation and evacuation procedures: a randomized clinical trial.
Vaginal vs sublingual misoprostol with mifepristone for cervical priming in second-trimester abortion by dilation and evacuation: a randomized clinical trial.2007; 75: 230-237
A randomized comparison of misoprostol to intrauterine instillation of hypertonic saline plus a prostaglandin F2 analogue for second trimester induction termination in Uzbekistan.
The use of 16,16-dimethyl-trans-delta-2 prostaglandin-E1 methyl-ester (gemeprost) vaginal pessaries for the termination of pregnancy in the early 2nd trimester: a comparison with extra-amniotic prostaglandin-E2.
United Nations Development Program/United Nations Fund for Population Activities/World Health Organization/World Bank Special Program for Research and Development and Research Training in Human Reprod
Annual technical report 1997.
World Health Organization,
Geneva, Switzerland1998
Induction of abortion in the second trimester by a combination of misoprostol and mifepristone: a randomized comparison between two misoprostol regimens.
A randomized trial of mifepristone in combination with misoprostol administered sublingually or vaginally for medical abortion at 13-20 weeks gestation.
Haskell M. Dilation and extraction for late second trimester abortion. Presented at: National Abortion Federation Risk Management Seminar. Dallas, TX, September 13, 1992. Oral presentation.
In his review of recent advances in second-trimester abortion, Cassing Hammond1 did not include several specifically relevant clinical reports. In his discussion, he mentioned inadequate information in the medical literature concerning the use of misoprostol as an adjunct in second-trimester surgical abortion that includes dose and route of delivery, the effect on induction-to-abortion interval of the combination of osmotic dilators with misoprostol, the paucity of comparative studies that involve dilapan as an osmotic dilator, the lack of dose-finding studies of digoxin in the induction of fetal death in second-trimester abortion, the relative safety of second-trimester dilation and evacuation methods of abortion, the safety of various procedures for patients with multiple gestation, and the risks of uterine rupture in second-trimester abortion in the presence of a history of cesarean delivery with different treatment regimens.
02214-X&id=gr1.jpg){kind=link}