- Email: [email protected]
11 The effect of pregnancy termination on future reproduction
11
The effect of pregnancy termination on future reproduction H A N I K. A T R A S H C A R O L J. R O W L A N D H O G U E
Most women obtaining induced abortions are at the beginning of their reproductive life. More than a quarter of the abortions in the United States and other developed countries are performed on teenagers, and more than 60% are performed on women 24 years old or younger (Tietze and Henshaw, 1986; Centers for Disease Control, 1989). A survey by the Alan Guttmacher Institute reported that 70% of all women having abortions in the United States in 1987 said they intended to have children in the future (Henshaw and Silverman, 1988). Moreover, of all the women obtaining induced abortions in the United States in 1987, 54% have had no previous live births and 76% have had only one or no previous live birth (Centers for Disease Control, 1989). It is safe then to assume that most women seeking induced abortions intend to bear a child after their abortion. As a result, the effect of pregnancy termination on future reproduction is a crucial one for the very large number of women obtaining induced abortion. A variety of conditions have been ascribed anecdotally to induced abortion, including sterility, menstrual disorders, and psychiatric conditions as well as an increase in the incidence of premature births, tubal pregnancies, stillbirths, birth defects and spontaneous abortion in subsequent pregnancies. Because the number of induced abortions performed every year is very large, even a minor increase of risk associated with induced abortion would have significant public health implications. Although some reproductive complications may be associated with previous induced abortions, the recent scientific literature agrees that most induced abortions have little, if any, physical impact on the women. Many of the published studies have serious limitations; some were conducted in countries where abortion has not been legalized, such as Greece, Israel and Taiwan, whereas others failed to control adequately for factors known to contribute to the immediate and long-term complications of induced abortion. Such factors include the patient's demographic profile, the patient's medical risk factors, the operator's skill, the type of anaesthesia, the type and amount of dilatation, the type and size of instrumentation, the length of the operation procedure, the type of abortifacient, the type of Bailli~re's Clinical Obstetrics and Gynaecology--
Vol.4, No. 2, June 1990 ISBN 0-7020-1477-X
391 Copyright© 1990,byBaillirreTindall All rightsofreproductionin anyformreserved
392
l-I. K. A T R A S H A N D C. J. R O W L A N D H O G U E
facility or service, the use of prophylactic antibiotics and concurrent sterilization (Cates, 1979; Hogue et al, 1982). The effect of induced abortion on future reproduction has been extensively reviewed (Hogue et al, 1982,1983; Hogue, 1986). Researchers have reviewed the world literature, including studies from 21 countries, and have concluded that women who have their first pregnancy terminated by vacuum aspiration are at no increased risk of subsequent infertility or ectopic pregnancy when compared with women who carry their first pregnancy to term. They also concluded that a single induced abortion performed by vacuum aspiration does not increase the risk of complications during future pregnancies, the risk of having a low birthweight baby, or the risk of having a pregnancy result in a miscarriage, stillbirth, infant death or congenital malformations. Less research has been done to study the effects of multiple abortions. However, the reviews concluded that studies do not support a firm conclusion about whether the number of procedures in itself produces any increased risk of adverse outcomes in subsequent pregnancies (Hogue et al, 1982). In this chapter we review the literature to update findings from the previously published reviews and summarize major studies dealing with the subject. Special emphasis will be given to secondary infertility, ectopic pregnancy, mid-trimester spontaneous abortion, premature delivery and low birthweight. METHODOLOGICAL CONSIDERATIONS Several methodological problems arise in examining the risk of adverse pregnancy outcomes following an induced abortion. The most important considerations include the choice of an appropriate comparison group, the number of previous induced abortions, the type of procedure used to perform the abortion, and the gestational age at the time of abortion.
Comparison groups The choice of an appropriate comparison group has been a subject of dispute. A woman who becomes pregnant after an induced abortion (gravida 2 para 0) is unlike a woman who is pregnant for the first time (gravida 1 para 0) and also unlike a woman who is pregnant for the second time but whose first pregnancy was carried to term (gravida 2 para 1). Yet there is no perfect comparison group, and women who are pregnant after an induced abortion have been compared with women who are pregnant for the first time and with women who are pregnant for the second time and whose first pregnancy resulted in a live birth. In the following discussion, we summarize findings from studies using both comparison groups. Gravida 1 para 0 women are defined as comparison group 1, and gravida 2 para 1 women are defined as comparison group 2.
Number of previous abortions Very few studies have considered the number of previous abortions when
EFFECT OF ABORTION ON FUTURE REPRODUCTION
393
looking at the risk of adverse pregnancy outcomes. One study in Japan (Roht et al, 1976) found that, compared with women who had no history of induced abortion, women who had one previous induced abortion had a relative risk of spontaneous abortion of 0.99 (95% confidence intervals [CI] of 0.4 and 2.3), whereas women with a history of two induced abortions had a relative risk of 1.5 (95% CI of 1.2 and 1.9). Other studies examined the risk following multiple induced abortions and reported an increased risk of preterm delivery and low birthweight among women with two or more induced abortions compared with the risk among women undergoing one abortion before delivery; however, none of the differences were statistically significant (Chung et al, 1981; Lerner and Varma, 1981; Linnet al, 1983).
Type of procedure and gestational age The type of procedure used to terminate a pregnancy has been found to be an important contributing factor to the increased risk of adverse pregnancy outcomes in future pregnancies. Most studies have concluded that vacuum aspiration, commonly used to terminate pregnancies during the first trimester, carries no increased risk of adverse pregnancy outcomes, whereas dilatation and evacuation procedures, commonly used for second trimester abortions, are associated with significantly increased risks (Hogue, 1986). Because both gestational age and type of procedure have been determined to be independent risk factors for immediate short-term complications of induced abortion (Grimes, 1979), and because most studies failed to control for gestational age when reporting procedure-specific complications, it is not possible to determine whether the increased risk of adverse pregnancy outcomes among women whose pregnancies were terminated by using a dilatation and evacuation procedure results from the gestational age (second trimester) or the type of procedure. Previous reviews found that there were too few studies to warrant a conclusion about procedure-specific effects of second trimester abortions (Hogue, 1986). Recent studies do not add much to our knowledge about the subject. One study reported an increased (but not significant) risk of having a low birthweight baby for women undergoing abortion by dilatation and evacuation who were dilated to 16 mm compared with women who were dilated to 14 mm (Hogue and Peterson, 1984). Another study found an increased (again not significant) risk of having a low birthweight baby for women whose first pregnancy was terminated by prostaglandins compared with those whose first pregnancy was terminated by saline instillation (Meirik and Nygren, 1984). A study in England showed no increase in the incidence of spontaneous abortion and a slight (but not significant) increase in the incidence of low birthweight deliveries among women who had had second trimester abortion for neural tube defects or for fetal Down's syndrome (Seller and Hancock, 1985). Finally, a study looking at the effect of second trimester prostaglandin abortion on fertility found that 104 out of 105 women who wanted to conceive following second trimester abortion succeeded (MacKenzie and Fry, 1988). The study concluded that midtrimester abortion has no effect on future fertility. Thus, there is no evidence
394
H . K. A T R A S H A N D C. J. R O W L A N D H O G U E
from the published literature that second trimester abortions, other than dilatation and evacuation, are associated with significantly increased risk of adverse pregnancy outcomes. Our following discussion examines the effects of induced abortion on future reproduction by type of procedure. We divided procedures into two categories: vacuum aspiration and dilatation and evacuation procedures. One can safely assume that almost all vacuum aspiration procedures are performed during the first trimester and that dilatation and evacuation procedures are performed during the second trimester. For populationbased studies where the type of procedure was not stated, results were included among vacuum aspiration abortions because more than 90% of all induced abortions are performed by vacuum aspiration (Centers for Disease Control, 1989).
SECONDARY INFERTILITY Our knowledge of a potential association between infertility and induced abortion is based on a variety of reports including early case reports, reports from case control studies, cross-sectional surveys and prospective studies. Studies that followed a series of patients longitudinally reported an incidence of infertility between 0 and 7.6% (Hogue et al, 1982). However, case series do not permit comparison of the incidence of secondary infertility among women having induced abortion with that among women who have had other pregnancy outcomes. We summarized studies that assessed the relative risk of secondary infertility, and, except for one study from Greece where abortion is illegal (Trichopoulos et al, 1976), none of these studies found any increased risk of infertility following induced abortion (Table 1). On the basis of these reports, we concluded that induced abortion is not associated with an increased incidence of secondary infertility. Furthermore, in two studies (Hogue et al, 1978; Stubblefield et al, 1984), fertility was found to be greater among women following their abortions as indicated by significantly shorter interpregnancy intervals. In the Stubblefield study, this indication of greater fecundity was evident only for women who had had three or more previous induced abortions. Because women who experience an unwanted pregnancy are likely to be among the more fertile group of women (Hogue et al, 1978), finding shorter interpregnancy intervals among this group is not surprising. The most recent prospective studies on the association between induced abortion and secondary infertility were published in 1984 and 1985 (Stubblefield et al, 1984; WHO, 1984; Cramer et al, 1985; Daling et al, 1985a). All four studies concluded that there is no significant effect of induced abortion on secondary infertility. Other than the study by MacKenzie and Fry (1988) that reported no association between mid-trimester induced abortion and secondary infertility, an in-depth search of the literature for the years 1985 to 1989 did not yield any new reports on the possible association between induced abortion and subsequent secondary infertility.
EFFECT OF ABORTIONON FUTURE REPRODUCTION
395
Table 1. Selected controlled studies of secondary infertility followinginduced abortion. From Hogue (1986). Source
Location
Abortion procedure
Roht et al (1973) Trichopoulos et al (1976)
Japan
Mostly D&E
Greece
Illegal D&E
Hogue et al (1978)
Yugoslavia
MostlyVA
Obel (1979)
Denmark
Mostly VA
Chung et at (1981) Daling et al (1981) Stubblefield et al (1984) WHO (1984)
Hawaii
Mostly VA
Washington
MostlyVA
Daling et al (1985a)
Massachusetts MostlyVA Koreaand Hungary Washington
Mostly D&E MostlyVA
Findings History of prior abortion not related to gravidity Increased risk for secondaryinfertility of > 18 months when controlling for spontaneous abortion history Interpregnancy intervals were significantly shorter followinginduced abortion Interpregnancy intervals were not significantly different No significantlyimpaired ability to conceive No increased risk for secondary infertility > 12 months No significant difference in conception rates up to 42 months after procedure No significant difference in conception rates up to 30 months after procedure History of prior abortion not related to infertility status
D&E = dilatation and evacuation; VA = vacuum aspiration.
ECTOPIC PREGNANCY From 1970 to 1986, the numbers and incidence rates of ectopic pregnancy in the United States increased more than threefold (Lawson et al, 1989). This increase coincided with legalization of abortion and with a seven times increase in the number of reported legal induced abortions, from 175 000 in 1970 to more than 1.3 million in 1986 (Centers for Disease Control, 1989). This temporal association raises a question of whether a portion of the ectopic pregnancy increase could be associated with prior induced abortion. Studies evaluating this issue have yielded inconsistent results, have often been carried out in settings with dissimilar practices, or have used methodology or samples that are inadequate to evaluate critical issues. Several reports have examined the relationship between ectopic pregnancy and induced abortion. Cervical pregnancy, a rare pregnancy complication, was first reported to be associated with previous induced abortions in a case control study (Shinagawa and Nagayama, 1969) and in a few other case reports (Dicker et al, 1985). However, many of the reported cases had other conditions known to be associated with increased risk of cervical pregnancy such as the use of intrauterine devices, pelvic pathology and Asherman's syndrome.
396
n.K.
ATRASH AND C. J. ROWLAND HOGUE
Available evidence does not support an overall aetiological relationship between legally induced abortion and subsequent ectopic pregnancy. However, only a few, small case control studies have attempted to evaluate this relationship. The strongest associations between induced abortion and ectopic pregnancy were observed in studies with small numbers of ectopic pregnancy cases, studies that failed to control for important risk factors, and studies conducted in countries where abortion was illegal (Sawazaki and Tanaka, 1966; Shinagawa and Nagayama, 1969; Panayotou et al, 1972; Dziewulska, 1973; Hren et al, 1974). Other studies reported a significant increase in the incidence of ectopic pregnancy after abortions that were complicated by infection or retained products of conception (Chung et al, 1982b). If there is an aetiological relationship, it is likely mediated through infection (either pre-existing and untreated or post-abortion infection). Because infection is a relatively rare complication of induced abortion, small case control studies could miss this subset of susceptible women. If real, the increased risk of ectopic pregnancy following induced abortion should be detected with a large case control study. However, recent relatively large case control studies confirm previous findings of no association between induced abortion and ectopic pregnancy. A report from the Women's Health Study (Burkman et al, 1988), which compared 462 cases of ectopic pregnancy with 2326 controls, found no increased risk of ectopic pregnancy among women with one previous induced abortion (odds ratio (OR) 1.0, 95% confidence limits (CL) 0.5, 1.8) or women with two or more previous induced abortions (OR 0.9, 95% CL 0.8, 1.1). In another report,
8.0-
(1)
4.0-
2.0-
1.0-
0.5-
, (9), I
,tl
I
II Ili
0.25
O. 125
Figure 1. Relative risks and 95% confidence intervals of ectopic pregnancy among women having induced abortion compared with women with no history of induced abortion (logarithmic scale). Numbers in parentheses are reference numbers for sources of data listed in the Appendix.
E F F E C T OF A B O R T I O N O N F U T U R E R E P R O D U C T I O N
397
Holt et al (1989) compared 396 women who had ectopic pregnancy with 893 controls. After adjusting for age, county, reference date, religion, gravidity, age at first pregnancy, lifetime number of sexual partners and miscarriage history, they concluded that legal abortion as performed in the United States since 1970 has little or no influence on a woman's risk of ectopic pregnancy in subsequent pregnancies (one previous abortion, OR 0.9, 95% CL 0.6, 1.3; two or more previous abortions, OR 1.2, 95% CL 0.6, 2.4). Most studies have not found a significant association between induced abortion and ectopic pregnancy (Figure 1); most of the reported odds ratios, however, were over 1.0, which suggest that a small association may exist. SPONTANEOUS ABORTION
Because it is very difficult to detect first trimester spontaneous abortions, most studies of first trimester spontaneous abortion following induced abortion have been seriously flawed (Hogue et al, 1982). As a result, most of our current knowledge pertains to the incidence of mid-trimester spontaneous abortion following induced abortion. Although Eastern European researchers observed no temporal rise in spontaneous abortion associated with the introduction of legalized abortion (Miltenyi, 1964; Bognar and Czeizel, 1976), some authors reported the incidence of spontaneous abortion following induced abortion to be as high as 30-40% (Kotasek, 1971). However, recent studies have reported spontaneous abortion rates ranging from 4.4 to 24% following instillation procedures and from 9 to 15% following evacuation procedures (Hogue et al, 1982). Most studies have concluded that vacuum aspiration, commonly used to terminate pregnancies during the first trimester, carries no increased risk of spontaneous abortion, whereas dilatation and evacuation, commonly used for second trimester abortion, is associated with a significantly increased risk of spontaneous abortion. Frank (1985) reported a relative risk for midtrimester spontaneous abortion of 0.97 (95% confidence intervals of 0.51, 1.8) among women with a history of induced abortion compared with other women independent of parity, and a relative risk of 1.5 (95% confidence intervals of 0.6, 3.5) compared to group 1 (gravida 1 para 0) women. In general, studies indicate that, compared with both group 1 and group 2 women, women who had their pregnancies terminated by vacuum aspiration (mostly first trimester) are at no increased risk of spontaneous abortion, whereas those whose pregnancies were terminated by dilatation and evacuation (mostly second trimester) had a significantly increased risk of spontaneous abortion in their subsequent pregnancy (Figure 2). PREMATURITY
Premature delivery and low birthweight are commonly used to measure prematurity. However, low birthweight is a more sensitive measure of
398
H.K.
A T R A S H A N D C. J . R O W L A N D
HOGUE
Dilatation a n d E v a c u a t i o n
Vacuum Aspiration
(9)
8.0
(11) i
£
(4)
(3) 4.0
i.
2.0 1.0 0.5O. 25 O. 125 -
(19) 8.0.
03)
4.0-
(20) (l._4)
(16) (17) (18) (15
2.0. 1.0.
i
i i
0.5.
i
0.25 I
O. 125
(21)
i
c.Q
S r"o Ix.) "O _.,x
Figure 2. Relative risks and 95% confidenceintervals of mid-trimester spontaneous abortion among women whose first pregnancy was terminated by induced abortion compared with womenin their first pregnancy(comparisongroup 1; gravida 1 para 0) and with womenin their second pregnancywhose first pregnancy was carried to term (comparison group 2; gravida 2 para 1) (logarithmic scale). Numbers in parentheses are reference numbers for sourcesof data listed in the Appendix.
prematurity, for it can be measured more accurately than gestational age. If induced abortion increases the risk of prematurity (as manifested by premature delivery, low birthweight, or both), this increased risk could be caused by cervical insufficiency or placental adhesions caused by uterine adhesions.
Premature delivery As in the case of spontaneous abortion, most studies indicate that there is no increased risk of premature delivery following an induced abortion, independent of the type of procedure and gestational age at the time of the
399
EFFECT OF ABORTION ON FUTURE REPRODUCTION
procedure (Figure 3). Independent of the type of procedure, the relative risks of premature delivery were higher when women with previous induced abortions were compared with group 2 women than when compared with group 1 women (Figure 3). In fact, all except one study reported that the relative risk of premature delivery among women with previous induced abortion compared with group 2 women was greater than 1.0. This observation suggests that women whose first pregnancy was terminated by induced abortion have a higher risk of low birthweight than women who carried their first pregnancy to term, and, therefore, induced abortion is not protective of the well-known risk of low birthweight for first-born offspring.
Vacuum aspiration
Dilatation a n d E v a c u a t i o n
8,0 4.0
O o
(1) (9) (2)
2,0
(3)
(4)
=,
(6)
nsnNn
1.0
(8)
,,,
I
I
0.5 0,25O, 125.
. . . . .
8.0-
. .
.
.
.
.
.
.
.
.
.
.
.
° °
. . . . . .
. l . . . . ° . .
.
.
.
.
.
.
.
.
.
.
. . . . . .
(11) (20)
4.02.01,00.5. 0.25 -
O o 3 -o
(12) (13) (14) (15) (21)
(la)
B
r
?o' o= 3 t-
-o I'o "12
O, 125.
Figure 3. Relative risks and 95% confidence intervals of premature delivery among women whose first pregnancy was terminated by induced abortion compared with women in their first pregnancy (comparison group 1; gravida I para 0) and with women in their second pregnancy whose first pregnancy was carried to term (comparison group 2; gravida 2 para 1) (logarithmic scale). Numbers in parentheses are reference numbers for sources of data listed in the Appendix.
400
H . K . A T R A S H A N D C. J. R O W L A N D H O G U E
Recent studies confirm findings reported earlier. Frank (1985) compared the outcome of pregnancy among 470 women with previous induced abortion to 1075 women with no history of induced abortion (regardless of parity) and found no significant difference in the incidence of premature delivery between the two groups (relative risk [RR] I. 13, 95% CI of 0.79 and 1.62). When women with a history of induced abortion were compared with women in group 1 (gravida 1 para 0), the relative risk of having a premature delivery increased to 1.22, but was still not significant (95% CI of 0.74 and 2.01). Linnet al (1983) studied 12 718 postpartum women to assess the risk of premature delivery following induced abortion. They reported that women who gave a history of one induced abortion had a 1.07 times increased risk of premature delivery (95% CI of 0.86 and 1.34), whereas women with a history of two or more induced abortions had a relative risk of 1.31 (95% CI of 0.91 and 1.89). Pickering and Forbes (1985) compared 3000 women who had an induced abortion with group 1 (gravida 1 para 0) and group 2 (gravida 2 para 1) women delivering in Scotland in 1980-1981. After controlling for maternal height, age, sex of infant, marital status and social class, they found that the risk of premature delivery among women with a history of induced abortion was the same as the risk among comparison group 1 women (gravida 1 para 0), but significantly lower than the risk among comparison group 2 women (gravida 2 para 1) (RR 0.79, 95% CI of 0.74 and 0.85). The type of procedure in the reports by Frank (1985) and by Pickering and Forbes (1985) were not specified, and their findings were included among vacuum aspiration procedures in Figure 3.
Low birthweight Overall, studies have shown no significantly increased risk of low-birthweight following abortions performed using vacuum aspiration. However, the dilatation and evacuation method has been found in most reports to be associated with an increased risk of low birthweight (Figure 4). The effect is more remarkable when women with previous induced abortion are compared with women in group 2 (gravida 2 para 1). Similar to the risk of premature delivery (Figure 3), the reported relative risk of low birthweight was higher when women with abortions induced by vacuum aspiration were compared with women in group 2 than when they were compared with women in group 1 (Figure 4). In addition, the relative risk of low birthweight following dilatation and evacuation procedures was higher in comparison to women in group 2 than in comparison to women in group 1. Recent studies reporting the effect of induced abortion on premature delivery also reported its effect on low birthweight. Frank (1985) reported that the relative risk of low birthweight among women with previous induced abortion was 1.39 (95% CI of 0.82 and 2.36) compared with other pregnant women independent of parity and 1.14 (95% CI of 0.56 and 2.34) compared with group 1 women (gravida 1 para 0). Linnet al (1983) reported that the relative risk of low birthweight among women with one previous induced abortion compared with women with no history of induced abortion was 0.93 (95% CI of 0.51 and 1.13), whereas the relative risk among women
401
EFFECT OF A B O R T I O N ON F U T U R E R E P R O D U C T I O N
Vacuum aspiration
8.0
Dilatation and Evacuation o 0 3
4.0
(2) (3) (4)
(11)
B"°'Brim"" "UU.l:l rlll o Q
2.0
""
1.0
1
0.5-
--~.
(14,
0
i
"o
O. 25 O, 125 -
8.0(26) 4,0-
2.0-
1.0-
0.5
(15)
(17) (18) (19) (20)
HHHi--tLIFI FtB
'""'UUH
O
(23)
~D C9 0
I-1
0 c-
o
O. 25
O, 125
Figure 4. Relative risks and 95% confidenceintervals of havinglow birthweight infants among women whose first pregnancywas terminated by induced abortion compared with women in their first pregnancy (comparisongroup 1; gravida 1 para 0) and with women in their second pregnancy whose first pregnancy was carried to term (comparisongroup 2; gravida2 para 1) (logarithmic scale). Numbersin parentheses are reference numbersfor sourcesof datalisted in the Appendix.
with history of two or more previous induced abortions compared with w o m e n with no previous induced abortion was 1.25 (95% CI of 0.88 and 1.79). Finally, Pickering and Forbes (1985) reported that the adjusted relative risk of low birthweight among w o m e n with previous induced abortion compared with w o m e n in group 1 was 0.68 (95% CI of 0.64 and 0.71). The type of procedure in the reports by Frank (1985) and by Pickering and Forbes (1985) were not specified, and their findings were included among the vacuum aspiration procedures in Figure 4.
402
H . K. ATRASH A N D C. J. R O W L A N D H O G U E
OTHER COMPLICATIONS
Placenta praevia has also been studied as a possible late sequela of induced abortion. One case control study reported a tenfold higher risk (Barrett et al, 1981). However, a later study which accounted for potential confounding risk factors of gravidity and age in a predominantly black population found the relative risk to be 1.1 and not statistically significant (Grimes and Techman, 1984). In a prospective cohort study, Frank et al (1987) conducted a ten-year follow-up of 6418 women who had an induced abortion and 8059 women who had an unplanned pregnancy that was not terminated with an induced abortion. Further pregnancies had occurred among 729 women in the case group and 1754 women in the control group. The researchers reported that prior induced abortion had no material effect on the rate of pregnancy-related morbidity, including haemorrhage before labour ( R R 0.94, 95% CI of 0.53 and 1.65), mental illness ( R R 0.54, 95% CI of 0.23 and 1.27), pre-eclampsia ( R R 0.80, 95% CI of 0.49 and 1.30), haemorrhage in labour (RR 0.73, 95% CI of 0.36 and 1.47), disproportion (RR 1.12, 95% CI of 0.80 and 1.56), trauma at delivery (RR 0.94, 95% CI of 0.61 and 1.45), forceps delivery ( R R 1.01, 95 % CI of 0.64 and 1.60), caesarean section (RR 0.72, 95% CI of 0.48 and 1.09) and fetal distress (RR 0.91, 95% CI of 0.43 and 1.93). However, the incidence of anaemia of pregnancy was found to be significantly reduced ( R R 0.25, 95% CI of 0.07 and 0.84), whereas the incidence of urinary tract infection was significantly increased (RR 2.26, 95% CI of 1.29 and 3.95). SUMMARY
A variety of conditions have been anecdotally ascribed to induced abortion, including subsequent reproductive complications. Since most women obtaining induced abortions are at the beginning of their reproductive life, the effect of induced abortion on subsequent reproduction becomes a very significant one. Our review of the literature confirms findings reported previously. First, except in the case where an infection complicates induced abortion, there is no evidence of an association between induced abortion and secondary infertility or ectopic pregnancy. Second, the risk of midtrimester abortion, premature delivery and low birthweight in women whose first pregnancy is terminated by vacuum aspiration is not higher than that in women in their first pregnancy or women in their second pregnancy whose first pregnancy was carried to term. However, the risk of having a premature delivery or a low birthweight baby tends to be higher (but not significantly) among women whose first pregnancy is terminated by induced abortion when compared with women in their second pregnancy than when compared with women in their first pregnancy. This suggests that an induced abortion does not protect a woman against the known risk of low birthweight for first-born offspring. Finally, women whose pregnancy is terminated by dilatation and evacuation may have an increased risk of subsequent premature delivery and a low birthweight baby. Very little has been
EFFECT OF ABORTION ON FUTURE REPRODUC'HON
403
p u b l i s h e d a n d n o c o n c l u s i o n s c a n b e m a d e r e g a r d i n g t h e effects o f instill a t i o n p r o c e d u r e s a n d r e p e a t a b o r t i o n s o n f u t u r e r e p r o d u c t i o n . In conclusion, e x c e p t for t h e a s s o c i a t i o n b e t w e e n p r e g n a n c i e s f o l l o w i n g d i l a t a t i o n a n d e v a c u a t i o n p r o c e d u r e s a n d p r e m a t u r e d e l i v e r y a n d low b i r t h w e i g h t , n o significantly i n c r e a s e d risk o f a d v e r s e r e p r o d u c t i v e h e a l t h has b e e n o b s e r v e d following i n d u c e d a b o r t i o n .
APPENDIX: SOURCES OF DATA FOR FIGURES 1 TO 4 F/gure 1 1, 3: Daling et al, (1985b); 2: Chung et al (1982b); 4: Levin et al (1982); 5, 10: Holt et al (1989); 6, 7: Hren et al (1974); 8, 9: Burkman et al (1988). F/gure 2 1, 15: Frank (1985); 2, 9: Harlap et al (1979); 3, 6, 10, 11, 14, 17, 19, 21: WHO (1979); 4, 13: Lerner and Varma (1981); 5, 16: Chung et al (1982a); 7, 18: Bracken et al (1986); 8: Drac and Nekvasilova (1970); 12, 20: Logrillo et al (1980). F/gure 3 1, 4, 8, 9,11, 15, 19, 20: WHO, (1979); 3, 13: Lerner and Varma (1981); 5: Daling (1977); 6, 10, 17, 21: Logrillo et al (1980); 7, 14: Chung et al (1981); 2, 16: Frank (1985); 18: Pickering and Forbes (1985); 12: Meirik and Bergstrom (1983). Figure 4 1, 13, 16, 26: Logrillo et al (1980); 2, 3, 12, 14, 17, 20, 25, 27: WHO (1979); 4, 19: Frank (1985); 5, 21: Chung et al (1981); 6, 15: Lerner and Varma (1981); 7, 18: Meirik and Bergstrom (1983); 8, 22: Obel (1979); 9: Pickering and Forbes (1985); 10, 24: Hungarian Central Statistical Office (1972); 11, 23: Roht et al (1976).
REFERENCES Barrett JM, Boehm FH & Killam AP (1981) Induced abortion: a risk factor for placenta previa. American Journal of Obstetrics and Gynecology 141: 769-772. Bognar Z & Czeizel A (1976) Mortality and morbidity associated with legal abortions in Hungary, 1960-1973. American Journal of Public Health 66: 568-575. Bracken MB, Bryce-Buchanan C, Srisuphan W et al (1986) Risk of late first and second trimester miscarriage after induced abortion. American Journal of Perinatology 3: 84-91. Burkman RT, Mason KJ & Gold EB (1988) Ectopic pregnancy and prior induced abortion. Contraception 37: 21-27. Cates W Jr (1979) Late effects of induced abortion: hypothesis or knowledge? Journal of Reproductive Medicine 22: 207-212. Centers for Disease Control (1989) Abortion surveillance: preliminary analysis---United States, 1986 and 1987. Morbidity and Mortality Weekly Report 38: 662-663. Chung CS, Steinhoff PG & Smith RG (1981) Effects of Induced Abortion on Subsequent Reproductive Function and Pregnancy Outcome. Final report, Contract number (N01-HD62801). Honolulu: University of Hawai. Chung CS, Steinhoff PG, Ming-Pi Met al (1982a) Induced abortion and fetalloss in subsequent pregnancies. American Journal of Public Health 72: 548-554. Chung CS, Smith RG, Steinhoff PG et al (1982b) Induced abortion and ectopic pregnancy in subsequent pregnancies~ American Journal of Epidemiology 115- 879-887.
404
H. K. ATRASH AND C. J. ROWLAND HOGUE
Cramer DW, Schiff I, Schoenbaum SC et al (1985) Tubalinfertility and the intrauterine device. New England Journal of Medicine 312" 941-947. Daling JR (1977) SuJsequent pregnancy outcome following induced abortion. PhD dissertation, Department of Epidemiology, University of Washington. Daling JR, Spadoni LR & Emanuel I (1981) Role of induced abortion in secondary infertility. Obstetrics and Gynecology 57: 59-61. Daling JR, Weiss NS, Voigt L e t al (1985a) Tubal infertility in relation to prior induced abortion. Fertility and Sterility 43: 389-394. Daling JR, Chow WH, Weiss NS et al (1985b) Ectopic pregnancy in relation to previous induced abortion. Journal of the American Medical Association 253: 1005-1008. Dicker D, Feldberg D, Samuel N & Goldman JA (1985) Etiology of cervical pregnancy: association with abortion, pelvic pathology, IUDs and Asherman's syndrome. Journal of Reproductive Medicine 30: 25-27. Drac P & Nekvasilova Z (1970) Premature termination of pregnancy after previous interruption of pregnancy. Ceskoslovenska Gynekologie 35: 332-333. Dziewulska W (1973) Abortion in the past versus the fate of the subsequent pregnancy. Ginekologia Polska 44: 1003-1011. Frank PI (1985) Sequelae of induced abortion. Ciba Foundation Symposium 115: 67-82. Frank PI, Kay CR, Scott LM, Hannaford PC & Haran D (1987) Pregnancy following induced abortion: maternal morbidity, congenital abnormalities and neonatal death. Royal College of General Practitioners/Royal College of Obstetricians and Gynaecologists Joint Study. British Journal of Obstetrics and Gynaecology 94(9): 836-842. Grimes DA (1979) Complications from legally induced abortion: a review. Obstetrical and Gynecological Survey 34: 177-191. Grimes DA & Techman T (1984) Legal abortion and placenta previa. American Journal of Obstetrics and Gynecology 149: 501-504. Harlap S, Shinono P, Ramcharan Set al (1979) A prospective study of spontaneous fetal losses after induced abortions. New England Journal of Medicine 301: 677-681. Henshaw SK & Silverman J (1988) The characteristics and prior contraceptive history of US abortion patients. Family Planning Perspectives 20: 158--168. Hogue CJR (1986) Impact of abortion on subsequent fecundity. Clinics in Obstetrics and Gynecology 13: 95-103. Hogue C JR & Peterson W (1984) Outcome of pregnancy following mid-trimester abortion by dilation and evacuation. Paper presented at the Risk Management Seminar, National Abortion Federation, September 17, Vancouver. Hogue CJ, Schoenfelder JR, Gesler WM et al (1978) The interactive effects of induced abortion, inter-pregnancy interval, and contraceptive use on subsequent pregnancy outcome. American Journal of Epidemiology 107: 15-26. Hogue CJR, Cares W Jr & Tietze C (1982) The effects of induced abortion on subsequent reproduction. Epidemiologic Reviews 4: 66-94. Hogue CJR, Cates W Jr & Tietze C (1983) Impact of vacuum aspiration abortion on future childbearing: a review. Family Planning Perspectives 15: 119-126. Holt VL, Daling JR, Voigt LF et al (1989) Induced abortion and the risk of subsequent ectopic pregnancy. American Journal of Public Health 79: 1234-1288. Hren M, Tomazevic T & Seigel D (1974) Ectopic pregnancy. In Andolsck L (ed.) The Ljubljana Abortion Study, 1971-1973, pp 34-38. Bethesda, Maryland: Center for Population Research, National Institutes of Health. Hungarian Central Statistical Office (1972) Perinatalis Halalozas. Budapest: Hungarian Central Statistical Office. Kotasek A (1971) Artificial termination of pregnancy in Czechoslovakia. International Journal of Gynaecology and Obstetrics 9: 118-119. Lawson HW, Atrash HK, Saftlas AF & Finch EL (1989) Ectopic pregnancy in the United States, 1970-1986. Morbidity and Mortality Weekly Report Surveillance Summaries 38(SS-2): 1-15. Lerner RC & Varma AO (1981) Prospective Study of the Outcome of Pregnancy Subsequent to Previous lnducedAbortion. Final report, Contract number (NO1-HD-62803). New York: Downstate Medical Center, State University of New York, January. Levin AA, Schoenbaum SC, Stubblefield PG et al (1982) Ectopic pregnancy and prior induced abortion. American Journal of Public Health 72: 253-256.
EFFECT OF ABORTION ON FUTURE REPRODUCTION
405
Linn S, Schoenbaum SC, Monson RR et al (1983) The relationship between induced abortion and outcome of subsequent pregnancies. American Journal of Obstetrics and Gynecology 146: 136-140. Logrillo V, Quickenton P, Therriault GD et al (1980) Effect of Induced Abortion on Subsequent Reproductive Function. Final report to National Institute of Child Health and Development. Albany, New York: New York State Health Department. MacKenzie IZ & Fry A (1988) A prospective self-controlled study of fertility after secondtrimester prost aglandin-induced abortion. American Journal of Obstetrics and Gynecology 158(5): 113%1140. Meirik O & Bergstrom R (1983) Outcome of delivery subsequent to vacuum aspiration abortion in nulliparous women. Acta Obstetrica et Gynecologica Scandinavica 62: 499-509. Meirik O & Nygren KG (1984) Outcome of first delivery after 2nd trimester two-step induced abortion: controlled historical cohort study. Acta Obstetrica et Gynecologica Scandinavica 63: 45-50. Miltenyi K (1964) On the effects of induced abortion. Demographia 7: 73-87. Obel E (1979) Pregnancy complications following legally induced abortion. Acta Obstetrica et Gynecologica Scandinavica 58: 485-490. Panayotou PP, Kaskarelis DB, Miettinen OS et al (1972) Induced abortion and ectopic pregnancy. American Journal of Obstetrics and Gynecology 114: 507-510. Pickering RM & Forbes JF (1985) Risks of preterm delivery and small-for-gestational age infants following abortion: a population study. British Journal of Obstetrics and Gynaecology 92:1106-1112. Roht LH, Aoyama H, Leinen GE et al (1976) The association of multiple induced abortions with subsequent prematurity and spontaneous abortion. Acta Obstetrica et Gynecologica Japonica 23: 140-145. Sawazaki C & Tanaka S (1966) The relationship between artificial abortion and extrauterine pregnancy. In Koya Y (ed.) Harmful Effects of Induced Abortion, pp 49-63. Tokyo: Family Planning Association of Japan. Seller MJ & Hancock PC (1985) Effects of mid-trimester induced abortion on the subsequent pregnancy. Prenatal Diagnosis 5: 375-380. Shinagawa S & Nagayama M (1969) Cervical pregnancy as a possible sequela of induced abortion. Report of 19 cases. American Journal of Obstetrics and Gynecology 105: 282284. Stubblefield PG, Monson RR, Schoenbaum SC et al (1984) Fertility after induced abortion: a prospective follow-up study. Obstetrics and Gynecology 63: 186-193. Tietze C & Henshaw HK (1986) Induced Abortion: a World Review, 1986 6th edn. New York: The Allan Guttmacher Institute. Trichopoulos D, Handanos N, Danezis J et al (1976) Induced abortion and secondary infertility. British Journal of Obstetrics and Gynaecology 83: 645-650. WHO (1979) Gestation, birthweight, and spontaneous abortion in pregnancy after induced abortion. World Health Organization Task Force on the Sequelae of Abortion. Lancet i: 142-145. WHO (1984) Secondary infertility following induced abortion. World Health Organization Task Force on Sequelae of Abortion, Special :Programme of Research, Development and Research Training in Human Reproduction. Studies in Family Planning 15: 291-295.
The effect of pregnancy termination on future reproduction H A N I K. A T R A S H C A R O L J. R O W L A N D H O G U E
Most women obtaining induced abortions are at the beginning of their reproductive life. More than a quarter of the abortions in the United States and other developed countries are performed on teenagers, and more than 60% are performed on women 24 years old or younger (Tietze and Henshaw, 1986; Centers for Disease Control, 1989). A survey by the Alan Guttmacher Institute reported that 70% of all women having abortions in the United States in 1987 said they intended to have children in the future (Henshaw and Silverman, 1988). Moreover, of all the women obtaining induced abortions in the United States in 1987, 54% have had no previous live births and 76% have had only one or no previous live birth (Centers for Disease Control, 1989). It is safe then to assume that most women seeking induced abortions intend to bear a child after their abortion. As a result, the effect of pregnancy termination on future reproduction is a crucial one for the very large number of women obtaining induced abortion. A variety of conditions have been ascribed anecdotally to induced abortion, including sterility, menstrual disorders, and psychiatric conditions as well as an increase in the incidence of premature births, tubal pregnancies, stillbirths, birth defects and spontaneous abortion in subsequent pregnancies. Because the number of induced abortions performed every year is very large, even a minor increase of risk associated with induced abortion would have significant public health implications. Although some reproductive complications may be associated with previous induced abortions, the recent scientific literature agrees that most induced abortions have little, if any, physical impact on the women. Many of the published studies have serious limitations; some were conducted in countries where abortion has not been legalized, such as Greece, Israel and Taiwan, whereas others failed to control adequately for factors known to contribute to the immediate and long-term complications of induced abortion. Such factors include the patient's demographic profile, the patient's medical risk factors, the operator's skill, the type of anaesthesia, the type and amount of dilatation, the type and size of instrumentation, the length of the operation procedure, the type of abortifacient, the type of Bailli~re's Clinical Obstetrics and Gynaecology--
Vol.4, No. 2, June 1990 ISBN 0-7020-1477-X
391 Copyright© 1990,byBaillirreTindall All rightsofreproductionin anyformreserved
392
l-I. K. A T R A S H A N D C. J. R O W L A N D H O G U E
facility or service, the use of prophylactic antibiotics and concurrent sterilization (Cates, 1979; Hogue et al, 1982). The effect of induced abortion on future reproduction has been extensively reviewed (Hogue et al, 1982,1983; Hogue, 1986). Researchers have reviewed the world literature, including studies from 21 countries, and have concluded that women who have their first pregnancy terminated by vacuum aspiration are at no increased risk of subsequent infertility or ectopic pregnancy when compared with women who carry their first pregnancy to term. They also concluded that a single induced abortion performed by vacuum aspiration does not increase the risk of complications during future pregnancies, the risk of having a low birthweight baby, or the risk of having a pregnancy result in a miscarriage, stillbirth, infant death or congenital malformations. Less research has been done to study the effects of multiple abortions. However, the reviews concluded that studies do not support a firm conclusion about whether the number of procedures in itself produces any increased risk of adverse outcomes in subsequent pregnancies (Hogue et al, 1982). In this chapter we review the literature to update findings from the previously published reviews and summarize major studies dealing with the subject. Special emphasis will be given to secondary infertility, ectopic pregnancy, mid-trimester spontaneous abortion, premature delivery and low birthweight. METHODOLOGICAL CONSIDERATIONS Several methodological problems arise in examining the risk of adverse pregnancy outcomes following an induced abortion. The most important considerations include the choice of an appropriate comparison group, the number of previous induced abortions, the type of procedure used to perform the abortion, and the gestational age at the time of abortion.
Comparison groups The choice of an appropriate comparison group has been a subject of dispute. A woman who becomes pregnant after an induced abortion (gravida 2 para 0) is unlike a woman who is pregnant for the first time (gravida 1 para 0) and also unlike a woman who is pregnant for the second time but whose first pregnancy was carried to term (gravida 2 para 1). Yet there is no perfect comparison group, and women who are pregnant after an induced abortion have been compared with women who are pregnant for the first time and with women who are pregnant for the second time and whose first pregnancy resulted in a live birth. In the following discussion, we summarize findings from studies using both comparison groups. Gravida 1 para 0 women are defined as comparison group 1, and gravida 2 para 1 women are defined as comparison group 2.
Number of previous abortions Very few studies have considered the number of previous abortions when
EFFECT OF ABORTION ON FUTURE REPRODUCTION
393
looking at the risk of adverse pregnancy outcomes. One study in Japan (Roht et al, 1976) found that, compared with women who had no history of induced abortion, women who had one previous induced abortion had a relative risk of spontaneous abortion of 0.99 (95% confidence intervals [CI] of 0.4 and 2.3), whereas women with a history of two induced abortions had a relative risk of 1.5 (95% CI of 1.2 and 1.9). Other studies examined the risk following multiple induced abortions and reported an increased risk of preterm delivery and low birthweight among women with two or more induced abortions compared with the risk among women undergoing one abortion before delivery; however, none of the differences were statistically significant (Chung et al, 1981; Lerner and Varma, 1981; Linnet al, 1983).
Type of procedure and gestational age The type of procedure used to terminate a pregnancy has been found to be an important contributing factor to the increased risk of adverse pregnancy outcomes in future pregnancies. Most studies have concluded that vacuum aspiration, commonly used to terminate pregnancies during the first trimester, carries no increased risk of adverse pregnancy outcomes, whereas dilatation and evacuation procedures, commonly used for second trimester abortions, are associated with significantly increased risks (Hogue, 1986). Because both gestational age and type of procedure have been determined to be independent risk factors for immediate short-term complications of induced abortion (Grimes, 1979), and because most studies failed to control for gestational age when reporting procedure-specific complications, it is not possible to determine whether the increased risk of adverse pregnancy outcomes among women whose pregnancies were terminated by using a dilatation and evacuation procedure results from the gestational age (second trimester) or the type of procedure. Previous reviews found that there were too few studies to warrant a conclusion about procedure-specific effects of second trimester abortions (Hogue, 1986). Recent studies do not add much to our knowledge about the subject. One study reported an increased (but not significant) risk of having a low birthweight baby for women undergoing abortion by dilatation and evacuation who were dilated to 16 mm compared with women who were dilated to 14 mm (Hogue and Peterson, 1984). Another study found an increased (again not significant) risk of having a low birthweight baby for women whose first pregnancy was terminated by prostaglandins compared with those whose first pregnancy was terminated by saline instillation (Meirik and Nygren, 1984). A study in England showed no increase in the incidence of spontaneous abortion and a slight (but not significant) increase in the incidence of low birthweight deliveries among women who had had second trimester abortion for neural tube defects or for fetal Down's syndrome (Seller and Hancock, 1985). Finally, a study looking at the effect of second trimester prostaglandin abortion on fertility found that 104 out of 105 women who wanted to conceive following second trimester abortion succeeded (MacKenzie and Fry, 1988). The study concluded that midtrimester abortion has no effect on future fertility. Thus, there is no evidence
394
H . K. A T R A S H A N D C. J. R O W L A N D H O G U E
from the published literature that second trimester abortions, other than dilatation and evacuation, are associated with significantly increased risk of adverse pregnancy outcomes. Our following discussion examines the effects of induced abortion on future reproduction by type of procedure. We divided procedures into two categories: vacuum aspiration and dilatation and evacuation procedures. One can safely assume that almost all vacuum aspiration procedures are performed during the first trimester and that dilatation and evacuation procedures are performed during the second trimester. For populationbased studies where the type of procedure was not stated, results were included among vacuum aspiration abortions because more than 90% of all induced abortions are performed by vacuum aspiration (Centers for Disease Control, 1989).
SECONDARY INFERTILITY Our knowledge of a potential association between infertility and induced abortion is based on a variety of reports including early case reports, reports from case control studies, cross-sectional surveys and prospective studies. Studies that followed a series of patients longitudinally reported an incidence of infertility between 0 and 7.6% (Hogue et al, 1982). However, case series do not permit comparison of the incidence of secondary infertility among women having induced abortion with that among women who have had other pregnancy outcomes. We summarized studies that assessed the relative risk of secondary infertility, and, except for one study from Greece where abortion is illegal (Trichopoulos et al, 1976), none of these studies found any increased risk of infertility following induced abortion (Table 1). On the basis of these reports, we concluded that induced abortion is not associated with an increased incidence of secondary infertility. Furthermore, in two studies (Hogue et al, 1978; Stubblefield et al, 1984), fertility was found to be greater among women following their abortions as indicated by significantly shorter interpregnancy intervals. In the Stubblefield study, this indication of greater fecundity was evident only for women who had had three or more previous induced abortions. Because women who experience an unwanted pregnancy are likely to be among the more fertile group of women (Hogue et al, 1978), finding shorter interpregnancy intervals among this group is not surprising. The most recent prospective studies on the association between induced abortion and secondary infertility were published in 1984 and 1985 (Stubblefield et al, 1984; WHO, 1984; Cramer et al, 1985; Daling et al, 1985a). All four studies concluded that there is no significant effect of induced abortion on secondary infertility. Other than the study by MacKenzie and Fry (1988) that reported no association between mid-trimester induced abortion and secondary infertility, an in-depth search of the literature for the years 1985 to 1989 did not yield any new reports on the possible association between induced abortion and subsequent secondary infertility.
EFFECT OF ABORTIONON FUTURE REPRODUCTION
395
Table 1. Selected controlled studies of secondary infertility followinginduced abortion. From Hogue (1986). Source
Location
Abortion procedure
Roht et al (1973) Trichopoulos et al (1976)
Japan
Mostly D&E
Greece
Illegal D&E
Hogue et al (1978)
Yugoslavia
MostlyVA
Obel (1979)
Denmark
Mostly VA
Chung et at (1981) Daling et al (1981) Stubblefield et al (1984) WHO (1984)
Hawaii
Mostly VA
Washington
MostlyVA
Daling et al (1985a)
Massachusetts MostlyVA Koreaand Hungary Washington
Mostly D&E MostlyVA
Findings History of prior abortion not related to gravidity Increased risk for secondaryinfertility of > 18 months when controlling for spontaneous abortion history Interpregnancy intervals were significantly shorter followinginduced abortion Interpregnancy intervals were not significantly different No significantlyimpaired ability to conceive No increased risk for secondary infertility > 12 months No significant difference in conception rates up to 42 months after procedure No significant difference in conception rates up to 30 months after procedure History of prior abortion not related to infertility status
D&E = dilatation and evacuation; VA = vacuum aspiration.
ECTOPIC PREGNANCY From 1970 to 1986, the numbers and incidence rates of ectopic pregnancy in the United States increased more than threefold (Lawson et al, 1989). This increase coincided with legalization of abortion and with a seven times increase in the number of reported legal induced abortions, from 175 000 in 1970 to more than 1.3 million in 1986 (Centers for Disease Control, 1989). This temporal association raises a question of whether a portion of the ectopic pregnancy increase could be associated with prior induced abortion. Studies evaluating this issue have yielded inconsistent results, have often been carried out in settings with dissimilar practices, or have used methodology or samples that are inadequate to evaluate critical issues. Several reports have examined the relationship between ectopic pregnancy and induced abortion. Cervical pregnancy, a rare pregnancy complication, was first reported to be associated with previous induced abortions in a case control study (Shinagawa and Nagayama, 1969) and in a few other case reports (Dicker et al, 1985). However, many of the reported cases had other conditions known to be associated with increased risk of cervical pregnancy such as the use of intrauterine devices, pelvic pathology and Asherman's syndrome.
396
n.K.
ATRASH AND C. J. ROWLAND HOGUE
Available evidence does not support an overall aetiological relationship between legally induced abortion and subsequent ectopic pregnancy. However, only a few, small case control studies have attempted to evaluate this relationship. The strongest associations between induced abortion and ectopic pregnancy were observed in studies with small numbers of ectopic pregnancy cases, studies that failed to control for important risk factors, and studies conducted in countries where abortion was illegal (Sawazaki and Tanaka, 1966; Shinagawa and Nagayama, 1969; Panayotou et al, 1972; Dziewulska, 1973; Hren et al, 1974). Other studies reported a significant increase in the incidence of ectopic pregnancy after abortions that were complicated by infection or retained products of conception (Chung et al, 1982b). If there is an aetiological relationship, it is likely mediated through infection (either pre-existing and untreated or post-abortion infection). Because infection is a relatively rare complication of induced abortion, small case control studies could miss this subset of susceptible women. If real, the increased risk of ectopic pregnancy following induced abortion should be detected with a large case control study. However, recent relatively large case control studies confirm previous findings of no association between induced abortion and ectopic pregnancy. A report from the Women's Health Study (Burkman et al, 1988), which compared 462 cases of ectopic pregnancy with 2326 controls, found no increased risk of ectopic pregnancy among women with one previous induced abortion (odds ratio (OR) 1.0, 95% confidence limits (CL) 0.5, 1.8) or women with two or more previous induced abortions (OR 0.9, 95% CL 0.8, 1.1). In another report,
8.0-
(1)
4.0-
2.0-
1.0-
0.5-
, (9), I
,tl
I
II Ili
0.25
O. 125
Figure 1. Relative risks and 95% confidence intervals of ectopic pregnancy among women having induced abortion compared with women with no history of induced abortion (logarithmic scale). Numbers in parentheses are reference numbers for sources of data listed in the Appendix.
E F F E C T OF A B O R T I O N O N F U T U R E R E P R O D U C T I O N
397
Holt et al (1989) compared 396 women who had ectopic pregnancy with 893 controls. After adjusting for age, county, reference date, religion, gravidity, age at first pregnancy, lifetime number of sexual partners and miscarriage history, they concluded that legal abortion as performed in the United States since 1970 has little or no influence on a woman's risk of ectopic pregnancy in subsequent pregnancies (one previous abortion, OR 0.9, 95% CL 0.6, 1.3; two or more previous abortions, OR 1.2, 95% CL 0.6, 2.4). Most studies have not found a significant association between induced abortion and ectopic pregnancy (Figure 1); most of the reported odds ratios, however, were over 1.0, which suggest that a small association may exist. SPONTANEOUS ABORTION
Because it is very difficult to detect first trimester spontaneous abortions, most studies of first trimester spontaneous abortion following induced abortion have been seriously flawed (Hogue et al, 1982). As a result, most of our current knowledge pertains to the incidence of mid-trimester spontaneous abortion following induced abortion. Although Eastern European researchers observed no temporal rise in spontaneous abortion associated with the introduction of legalized abortion (Miltenyi, 1964; Bognar and Czeizel, 1976), some authors reported the incidence of spontaneous abortion following induced abortion to be as high as 30-40% (Kotasek, 1971). However, recent studies have reported spontaneous abortion rates ranging from 4.4 to 24% following instillation procedures and from 9 to 15% following evacuation procedures (Hogue et al, 1982). Most studies have concluded that vacuum aspiration, commonly used to terminate pregnancies during the first trimester, carries no increased risk of spontaneous abortion, whereas dilatation and evacuation, commonly used for second trimester abortion, is associated with a significantly increased risk of spontaneous abortion. Frank (1985) reported a relative risk for midtrimester spontaneous abortion of 0.97 (95% confidence intervals of 0.51, 1.8) among women with a history of induced abortion compared with other women independent of parity, and a relative risk of 1.5 (95% confidence intervals of 0.6, 3.5) compared to group 1 (gravida 1 para 0) women. In general, studies indicate that, compared with both group 1 and group 2 women, women who had their pregnancies terminated by vacuum aspiration (mostly first trimester) are at no increased risk of spontaneous abortion, whereas those whose pregnancies were terminated by dilatation and evacuation (mostly second trimester) had a significantly increased risk of spontaneous abortion in their subsequent pregnancy (Figure 2). PREMATURITY
Premature delivery and low birthweight are commonly used to measure prematurity. However, low birthweight is a more sensitive measure of
398
H.K.
A T R A S H A N D C. J . R O W L A N D
HOGUE
Dilatation a n d E v a c u a t i o n
Vacuum Aspiration
(9)
8.0
(11) i
£
(4)
(3) 4.0
i.
2.0 1.0 0.5O. 25 O. 125 -
(19) 8.0.
03)
4.0-
(20) (l._4)
(16) (17) (18) (15
2.0. 1.0.
i
i i
0.5.
i
0.25 I
O. 125
(21)
i
c.Q
S r"o Ix.) "O _.,x
Figure 2. Relative risks and 95% confidenceintervals of mid-trimester spontaneous abortion among women whose first pregnancy was terminated by induced abortion compared with womenin their first pregnancy(comparisongroup 1; gravida 1 para 0) and with womenin their second pregnancywhose first pregnancy was carried to term (comparison group 2; gravida 2 para 1) (logarithmic scale). Numbers in parentheses are reference numbers for sourcesof data listed in the Appendix.
prematurity, for it can be measured more accurately than gestational age. If induced abortion increases the risk of prematurity (as manifested by premature delivery, low birthweight, or both), this increased risk could be caused by cervical insufficiency or placental adhesions caused by uterine adhesions.
Premature delivery As in the case of spontaneous abortion, most studies indicate that there is no increased risk of premature delivery following an induced abortion, independent of the type of procedure and gestational age at the time of the
399
EFFECT OF ABORTION ON FUTURE REPRODUCTION
procedure (Figure 3). Independent of the type of procedure, the relative risks of premature delivery were higher when women with previous induced abortions were compared with group 2 women than when compared with group 1 women (Figure 3). In fact, all except one study reported that the relative risk of premature delivery among women with previous induced abortion compared with group 2 women was greater than 1.0. This observation suggests that women whose first pregnancy was terminated by induced abortion have a higher risk of low birthweight than women who carried their first pregnancy to term, and, therefore, induced abortion is not protective of the well-known risk of low birthweight for first-born offspring.
Vacuum aspiration
Dilatation a n d E v a c u a t i o n
8,0 4.0
O o
(1) (9) (2)
2,0
(3)
(4)
=,
(6)
nsnNn
1.0
(8)
,,,
I
I
0.5 0,25O, 125.
. . . . .
8.0-
. .
.
.
.
.
.
.
.
.
.
.
.
° °
. . . . . .
. l . . . . ° . .
.
.
.
.
.
.
.
.
.
.
. . . . . .
(11) (20)
4.02.01,00.5. 0.25 -
O o 3 -o
(12) (13) (14) (15) (21)
(la)
B
r
?o' o= 3 t-
-o I'o "12
O, 125.
Figure 3. Relative risks and 95% confidence intervals of premature delivery among women whose first pregnancy was terminated by induced abortion compared with women in their first pregnancy (comparison group 1; gravida I para 0) and with women in their second pregnancy whose first pregnancy was carried to term (comparison group 2; gravida 2 para 1) (logarithmic scale). Numbers in parentheses are reference numbers for sources of data listed in the Appendix.
400
H . K . A T R A S H A N D C. J. R O W L A N D H O G U E
Recent studies confirm findings reported earlier. Frank (1985) compared the outcome of pregnancy among 470 women with previous induced abortion to 1075 women with no history of induced abortion (regardless of parity) and found no significant difference in the incidence of premature delivery between the two groups (relative risk [RR] I. 13, 95% CI of 0.79 and 1.62). When women with a history of induced abortion were compared with women in group 1 (gravida 1 para 0), the relative risk of having a premature delivery increased to 1.22, but was still not significant (95% CI of 0.74 and 2.01). Linnet al (1983) studied 12 718 postpartum women to assess the risk of premature delivery following induced abortion. They reported that women who gave a history of one induced abortion had a 1.07 times increased risk of premature delivery (95% CI of 0.86 and 1.34), whereas women with a history of two or more induced abortions had a relative risk of 1.31 (95% CI of 0.91 and 1.89). Pickering and Forbes (1985) compared 3000 women who had an induced abortion with group 1 (gravida 1 para 0) and group 2 (gravida 2 para 1) women delivering in Scotland in 1980-1981. After controlling for maternal height, age, sex of infant, marital status and social class, they found that the risk of premature delivery among women with a history of induced abortion was the same as the risk among comparison group 1 women (gravida 1 para 0), but significantly lower than the risk among comparison group 2 women (gravida 2 para 1) (RR 0.79, 95% CI of 0.74 and 0.85). The type of procedure in the reports by Frank (1985) and by Pickering and Forbes (1985) were not specified, and their findings were included among vacuum aspiration procedures in Figure 3.
Low birthweight Overall, studies have shown no significantly increased risk of low-birthweight following abortions performed using vacuum aspiration. However, the dilatation and evacuation method has been found in most reports to be associated with an increased risk of low birthweight (Figure 4). The effect is more remarkable when women with previous induced abortion are compared with women in group 2 (gravida 2 para 1). Similar to the risk of premature delivery (Figure 3), the reported relative risk of low birthweight was higher when women with abortions induced by vacuum aspiration were compared with women in group 2 than when they were compared with women in group 1 (Figure 4). In addition, the relative risk of low birthweight following dilatation and evacuation procedures was higher in comparison to women in group 2 than in comparison to women in group 1. Recent studies reporting the effect of induced abortion on premature delivery also reported its effect on low birthweight. Frank (1985) reported that the relative risk of low birthweight among women with previous induced abortion was 1.39 (95% CI of 0.82 and 2.36) compared with other pregnant women independent of parity and 1.14 (95% CI of 0.56 and 2.34) compared with group 1 women (gravida 1 para 0). Linnet al (1983) reported that the relative risk of low birthweight among women with one previous induced abortion compared with women with no history of induced abortion was 0.93 (95% CI of 0.51 and 1.13), whereas the relative risk among women
401
EFFECT OF A B O R T I O N ON F U T U R E R E P R O D U C T I O N
Vacuum aspiration
8.0
Dilatation and Evacuation o 0 3
4.0
(2) (3) (4)
(11)
B"°'Brim"" "UU.l:l rlll o Q
2.0
""
1.0
1
0.5-
--~.
(14,
0
i
"o
O. 25 O, 125 -
8.0(26) 4,0-
2.0-
1.0-
0.5
(15)
(17) (18) (19) (20)
HHHi--tLIFI FtB
'""'UUH
O
(23)
~D C9 0
I-1
0 c-
o
O. 25
O, 125
Figure 4. Relative risks and 95% confidenceintervals of havinglow birthweight infants among women whose first pregnancywas terminated by induced abortion compared with women in their first pregnancy (comparisongroup 1; gravida 1 para 0) and with women in their second pregnancy whose first pregnancy was carried to term (comparisongroup 2; gravida2 para 1) (logarithmic scale). Numbersin parentheses are reference numbersfor sourcesof datalisted in the Appendix.
with history of two or more previous induced abortions compared with w o m e n with no previous induced abortion was 1.25 (95% CI of 0.88 and 1.79). Finally, Pickering and Forbes (1985) reported that the adjusted relative risk of low birthweight among w o m e n with previous induced abortion compared with w o m e n in group 1 was 0.68 (95% CI of 0.64 and 0.71). The type of procedure in the reports by Frank (1985) and by Pickering and Forbes (1985) were not specified, and their findings were included among the vacuum aspiration procedures in Figure 4.
402
H . K. ATRASH A N D C. J. R O W L A N D H O G U E
OTHER COMPLICATIONS
Placenta praevia has also been studied as a possible late sequela of induced abortion. One case control study reported a tenfold higher risk (Barrett et al, 1981). However, a later study which accounted for potential confounding risk factors of gravidity and age in a predominantly black population found the relative risk to be 1.1 and not statistically significant (Grimes and Techman, 1984). In a prospective cohort study, Frank et al (1987) conducted a ten-year follow-up of 6418 women who had an induced abortion and 8059 women who had an unplanned pregnancy that was not terminated with an induced abortion. Further pregnancies had occurred among 729 women in the case group and 1754 women in the control group. The researchers reported that prior induced abortion had no material effect on the rate of pregnancy-related morbidity, including haemorrhage before labour ( R R 0.94, 95% CI of 0.53 and 1.65), mental illness ( R R 0.54, 95% CI of 0.23 and 1.27), pre-eclampsia ( R R 0.80, 95% CI of 0.49 and 1.30), haemorrhage in labour (RR 0.73, 95% CI of 0.36 and 1.47), disproportion (RR 1.12, 95% CI of 0.80 and 1.56), trauma at delivery (RR 0.94, 95% CI of 0.61 and 1.45), forceps delivery ( R R 1.01, 95 % CI of 0.64 and 1.60), caesarean section (RR 0.72, 95% CI of 0.48 and 1.09) and fetal distress (RR 0.91, 95% CI of 0.43 and 1.93). However, the incidence of anaemia of pregnancy was found to be significantly reduced ( R R 0.25, 95% CI of 0.07 and 0.84), whereas the incidence of urinary tract infection was significantly increased (RR 2.26, 95% CI of 1.29 and 3.95). SUMMARY
A variety of conditions have been anecdotally ascribed to induced abortion, including subsequent reproductive complications. Since most women obtaining induced abortions are at the beginning of their reproductive life, the effect of induced abortion on subsequent reproduction becomes a very significant one. Our review of the literature confirms findings reported previously. First, except in the case where an infection complicates induced abortion, there is no evidence of an association between induced abortion and secondary infertility or ectopic pregnancy. Second, the risk of midtrimester abortion, premature delivery and low birthweight in women whose first pregnancy is terminated by vacuum aspiration is not higher than that in women in their first pregnancy or women in their second pregnancy whose first pregnancy was carried to term. However, the risk of having a premature delivery or a low birthweight baby tends to be higher (but not significantly) among women whose first pregnancy is terminated by induced abortion when compared with women in their second pregnancy than when compared with women in their first pregnancy. This suggests that an induced abortion does not protect a woman against the known risk of low birthweight for first-born offspring. Finally, women whose pregnancy is terminated by dilatation and evacuation may have an increased risk of subsequent premature delivery and a low birthweight baby. Very little has been
EFFECT OF ABORTION ON FUTURE REPRODUC'HON
403
p u b l i s h e d a n d n o c o n c l u s i o n s c a n b e m a d e r e g a r d i n g t h e effects o f instill a t i o n p r o c e d u r e s a n d r e p e a t a b o r t i o n s o n f u t u r e r e p r o d u c t i o n . In conclusion, e x c e p t for t h e a s s o c i a t i o n b e t w e e n p r e g n a n c i e s f o l l o w i n g d i l a t a t i o n a n d e v a c u a t i o n p r o c e d u r e s a n d p r e m a t u r e d e l i v e r y a n d low b i r t h w e i g h t , n o significantly i n c r e a s e d risk o f a d v e r s e r e p r o d u c t i v e h e a l t h has b e e n o b s e r v e d following i n d u c e d a b o r t i o n .
APPENDIX: SOURCES OF DATA FOR FIGURES 1 TO 4 F/gure 1 1, 3: Daling et al, (1985b); 2: Chung et al (1982b); 4: Levin et al (1982); 5, 10: Holt et al (1989); 6, 7: Hren et al (1974); 8, 9: Burkman et al (1988). F/gure 2 1, 15: Frank (1985); 2, 9: Harlap et al (1979); 3, 6, 10, 11, 14, 17, 19, 21: WHO (1979); 4, 13: Lerner and Varma (1981); 5, 16: Chung et al (1982a); 7, 18: Bracken et al (1986); 8: Drac and Nekvasilova (1970); 12, 20: Logrillo et al (1980). F/gure 3 1, 4, 8, 9,11, 15, 19, 20: WHO, (1979); 3, 13: Lerner and Varma (1981); 5: Daling (1977); 6, 10, 17, 21: Logrillo et al (1980); 7, 14: Chung et al (1981); 2, 16: Frank (1985); 18: Pickering and Forbes (1985); 12: Meirik and Bergstrom (1983). Figure 4 1, 13, 16, 26: Logrillo et al (1980); 2, 3, 12, 14, 17, 20, 25, 27: WHO (1979); 4, 19: Frank (1985); 5, 21: Chung et al (1981); 6, 15: Lerner and Varma (1981); 7, 18: Meirik and Bergstrom (1983); 8, 22: Obel (1979); 9: Pickering and Forbes (1985); 10, 24: Hungarian Central Statistical Office (1972); 11, 23: Roht et al (1976).
REFERENCES Barrett JM, Boehm FH & Killam AP (1981) Induced abortion: a risk factor for placenta previa. American Journal of Obstetrics and Gynecology 141: 769-772. Bognar Z & Czeizel A (1976) Mortality and morbidity associated with legal abortions in Hungary, 1960-1973. American Journal of Public Health 66: 568-575. Bracken MB, Bryce-Buchanan C, Srisuphan W et al (1986) Risk of late first and second trimester miscarriage after induced abortion. American Journal of Perinatology 3: 84-91. Burkman RT, Mason KJ & Gold EB (1988) Ectopic pregnancy and prior induced abortion. Contraception 37: 21-27. Cates W Jr (1979) Late effects of induced abortion: hypothesis or knowledge? Journal of Reproductive Medicine 22: 207-212. Centers for Disease Control (1989) Abortion surveillance: preliminary analysis---United States, 1986 and 1987. Morbidity and Mortality Weekly Report 38: 662-663. Chung CS, Steinhoff PG & Smith RG (1981) Effects of Induced Abortion on Subsequent Reproductive Function and Pregnancy Outcome. Final report, Contract number (N01-HD62801). Honolulu: University of Hawai. Chung CS, Steinhoff PG, Ming-Pi Met al (1982a) Induced abortion and fetalloss in subsequent pregnancies. American Journal of Public Health 72: 548-554. Chung CS, Smith RG, Steinhoff PG et al (1982b) Induced abortion and ectopic pregnancy in subsequent pregnancies~ American Journal of Epidemiology 115- 879-887.
404
H. K. ATRASH AND C. J. ROWLAND HOGUE
Cramer DW, Schiff I, Schoenbaum SC et al (1985) Tubalinfertility and the intrauterine device. New England Journal of Medicine 312" 941-947. Daling JR (1977) SuJsequent pregnancy outcome following induced abortion. PhD dissertation, Department of Epidemiology, University of Washington. Daling JR, Spadoni LR & Emanuel I (1981) Role of induced abortion in secondary infertility. Obstetrics and Gynecology 57: 59-61. Daling JR, Weiss NS, Voigt L e t al (1985a) Tubal infertility in relation to prior induced abortion. Fertility and Sterility 43: 389-394. Daling JR, Chow WH, Weiss NS et al (1985b) Ectopic pregnancy in relation to previous induced abortion. Journal of the American Medical Association 253: 1005-1008. Dicker D, Feldberg D, Samuel N & Goldman JA (1985) Etiology of cervical pregnancy: association with abortion, pelvic pathology, IUDs and Asherman's syndrome. Journal of Reproductive Medicine 30: 25-27. Drac P & Nekvasilova Z (1970) Premature termination of pregnancy after previous interruption of pregnancy. Ceskoslovenska Gynekologie 35: 332-333. Dziewulska W (1973) Abortion in the past versus the fate of the subsequent pregnancy. Ginekologia Polska 44: 1003-1011. Frank PI (1985) Sequelae of induced abortion. Ciba Foundation Symposium 115: 67-82. Frank PI, Kay CR, Scott LM, Hannaford PC & Haran D (1987) Pregnancy following induced abortion: maternal morbidity, congenital abnormalities and neonatal death. Royal College of General Practitioners/Royal College of Obstetricians and Gynaecologists Joint Study. British Journal of Obstetrics and Gynaecology 94(9): 836-842. Grimes DA (1979) Complications from legally induced abortion: a review. Obstetrical and Gynecological Survey 34: 177-191. Grimes DA & Techman T (1984) Legal abortion and placenta previa. American Journal of Obstetrics and Gynecology 149: 501-504. Harlap S, Shinono P, Ramcharan Set al (1979) A prospective study of spontaneous fetal losses after induced abortions. New England Journal of Medicine 301: 677-681. Henshaw SK & Silverman J (1988) The characteristics and prior contraceptive history of US abortion patients. Family Planning Perspectives 20: 158--168. Hogue CJR (1986) Impact of abortion on subsequent fecundity. Clinics in Obstetrics and Gynecology 13: 95-103. Hogue C JR & Peterson W (1984) Outcome of pregnancy following mid-trimester abortion by dilation and evacuation. Paper presented at the Risk Management Seminar, National Abortion Federation, September 17, Vancouver. Hogue CJ, Schoenfelder JR, Gesler WM et al (1978) The interactive effects of induced abortion, inter-pregnancy interval, and contraceptive use on subsequent pregnancy outcome. American Journal of Epidemiology 107: 15-26. Hogue CJR, Cares W Jr & Tietze C (1982) The effects of induced abortion on subsequent reproduction. Epidemiologic Reviews 4: 66-94. Hogue CJR, Cates W Jr & Tietze C (1983) Impact of vacuum aspiration abortion on future childbearing: a review. Family Planning Perspectives 15: 119-126. Holt VL, Daling JR, Voigt LF et al (1989) Induced abortion and the risk of subsequent ectopic pregnancy. American Journal of Public Health 79: 1234-1288. Hren M, Tomazevic T & Seigel D (1974) Ectopic pregnancy. In Andolsck L (ed.) The Ljubljana Abortion Study, 1971-1973, pp 34-38. Bethesda, Maryland: Center for Population Research, National Institutes of Health. Hungarian Central Statistical Office (1972) Perinatalis Halalozas. Budapest: Hungarian Central Statistical Office. Kotasek A (1971) Artificial termination of pregnancy in Czechoslovakia. International Journal of Gynaecology and Obstetrics 9: 118-119. Lawson HW, Atrash HK, Saftlas AF & Finch EL (1989) Ectopic pregnancy in the United States, 1970-1986. Morbidity and Mortality Weekly Report Surveillance Summaries 38(SS-2): 1-15. Lerner RC & Varma AO (1981) Prospective Study of the Outcome of Pregnancy Subsequent to Previous lnducedAbortion. Final report, Contract number (NO1-HD-62803). New York: Downstate Medical Center, State University of New York, January. Levin AA, Schoenbaum SC, Stubblefield PG et al (1982) Ectopic pregnancy and prior induced abortion. American Journal of Public Health 72: 253-256.
EFFECT OF ABORTION ON FUTURE REPRODUCTION
405
Linn S, Schoenbaum SC, Monson RR et al (1983) The relationship between induced abortion and outcome of subsequent pregnancies. American Journal of Obstetrics and Gynecology 146: 136-140. Logrillo V, Quickenton P, Therriault GD et al (1980) Effect of Induced Abortion on Subsequent Reproductive Function. Final report to National Institute of Child Health and Development. Albany, New York: New York State Health Department. MacKenzie IZ & Fry A (1988) A prospective self-controlled study of fertility after secondtrimester prost aglandin-induced abortion. American Journal of Obstetrics and Gynecology 158(5): 113%1140. Meirik O & Bergstrom R (1983) Outcome of delivery subsequent to vacuum aspiration abortion in nulliparous women. Acta Obstetrica et Gynecologica Scandinavica 62: 499-509. Meirik O & Nygren KG (1984) Outcome of first delivery after 2nd trimester two-step induced abortion: controlled historical cohort study. Acta Obstetrica et Gynecologica Scandinavica 63: 45-50. Miltenyi K (1964) On the effects of induced abortion. Demographia 7: 73-87. Obel E (1979) Pregnancy complications following legally induced abortion. Acta Obstetrica et Gynecologica Scandinavica 58: 485-490. Panayotou PP, Kaskarelis DB, Miettinen OS et al (1972) Induced abortion and ectopic pregnancy. American Journal of Obstetrics and Gynecology 114: 507-510. Pickering RM & Forbes JF (1985) Risks of preterm delivery and small-for-gestational age infants following abortion: a population study. British Journal of Obstetrics and Gynaecology 92:1106-1112. Roht LH, Aoyama H, Leinen GE et al (1976) The association of multiple induced abortions with subsequent prematurity and spontaneous abortion. Acta Obstetrica et Gynecologica Japonica 23: 140-145. Sawazaki C & Tanaka S (1966) The relationship between artificial abortion and extrauterine pregnancy. In Koya Y (ed.) Harmful Effects of Induced Abortion, pp 49-63. Tokyo: Family Planning Association of Japan. Seller MJ & Hancock PC (1985) Effects of mid-trimester induced abortion on the subsequent pregnancy. Prenatal Diagnosis 5: 375-380. Shinagawa S & Nagayama M (1969) Cervical pregnancy as a possible sequela of induced abortion. Report of 19 cases. American Journal of Obstetrics and Gynecology 105: 282284. Stubblefield PG, Monson RR, Schoenbaum SC et al (1984) Fertility after induced abortion: a prospective follow-up study. Obstetrics and Gynecology 63: 186-193. Tietze C & Henshaw HK (1986) Induced Abortion: a World Review, 1986 6th edn. New York: The Allan Guttmacher Institute. Trichopoulos D, Handanos N, Danezis J et al (1976) Induced abortion and secondary infertility. British Journal of Obstetrics and Gynaecology 83: 645-650. WHO (1979) Gestation, birthweight, and spontaneous abortion in pregnancy after induced abortion. World Health Organization Task Force on the Sequelae of Abortion. Lancet i: 142-145. WHO (1984) Secondary infertility following induced abortion. World Health Organization Task Force on Sequelae of Abortion, Special :Programme of Research, Development and Research Training in Human Reproduction. Studies in Family Planning 15: 291-295.