Induced Abortion and Subsequent Preterm Birth:Evidence of Risk Association

Byron C. Calhoun, M.D., FACOG, FACS, MBAProfessor & Vice-Chair, Department of Obstetrics & Gynecology

West Virginia University-CharlestonMoira Gaul, M.P.H.

Director of Women’s and Reproductive HealthFamily Research Council, Washington, D.C.

U.S. Surgeon General’s Conference on the Prevention ofPreterm Birth

June 16 & 17, 2008

The abstracts, referenced studies, and two papers contained in this document provideevidence of a causal link between surgical induced abortion (IA) and subsequent pretermbirth. One of the two papers at the conclusion of the document gives further considerationto this causal link with respect to both cost consequences and impact on informedconsent.

ABSTRACTSThe following abstracts represent six studies which demonstrate the strongest and mostsignificant risk association between IA and later preterm birth.

(1.) Lumley J. The epidemiology of preterm birth. Bailliere's Clin Obstet Gynecology.1993;7(3):477-498ABSTRACTSecular trends in the prevalence of preterm birth and international comparisons of the rates ofpreterm birth are difficult to interpret because of differences, both formal and informal, in theregistration of extremely preterm births. Accurate estimation of gestational age is anotherproblem in the measurement of preterm birth. Preterm birth is heterogeneous in several ways. It isheterogeneous in terms of the extent to which the birth is preterm (20-27 weeks, 28-31 weeks or32-36 weeks of gestation); in whether the birth was elective or spontaneous; and amongspontaneous idiopathic preterm births, in whether there was preterm labour or premature ruptureof the membranes. Case-control study designs taking account of these subgroups have been arecent feature of epidemiologic approaches. The classic social associations of preterm birth--lowsocioeconomic status, extremes of maternal age, primiparity, being unmarried--apply toextremely preterm and moderately preterm births as well as to the mildly preterm group. Thestrength of these associations is small compared with factors in the prior reproductive history andwith medical and obstetric complications of the current pregnancy. Recent epidemiologicalresearch activities have focused on the ways in which risk factors such as physical workload,drugs and alcohol, lack of social support and infection might be mediating factors betweensociodemographic status and preterm birth. As Eastman (1947) pointed out almost 50 years ago,'only when the factors causing prematurity are clearly understood can any intelligent attempt atprevention be made'. [References: 74]

(2.) Lumley J. The association between prior spontaneous abortion, prior inducedabortion and preterm birth in first singleton births. Prenat Neonat Med 1998;3:21-24.DISCUSSION (in lieu of Abstract)

The evidence for and against a causal relationship between prior shortened pregnanciesand preterm birth is as follows. Selection bias into the study can be excluded since the data arepopulation-based with 99.6% of births captured in the data system. Measurement error of theexposures cannot be totally excluded. Though the data have been validated against hospitalrecords there is likely to be incomplete reporting of prior induced abortions in hospital records.Non-differential under-reporting would result in the relative risk being too low. Differentialunder-reporting is which prior abortions were more completely ascertained when there was apreterm birth could have occurred but the timing of data collection on prior pregnancies is at thefirst antenatal visit, before the outcome is known. Chance (random error) is unlikely since the95% confidence intervals for the association with prior abortions do not include 1.00 .

Confounding is likely. Maternal age, marital status and the presence of a birth defect inthe index birth can be excluded since the associations in Figures 1 to 3 are still present at a similarlevel in a restricted data stratum of married women, age 20-34 years, with no birth defect in theindex birth (unpublished data). Data on socioeconomic status, occupation, smoking, alcohol

consumption and other substance use are not collected routinely. The strength of association ofthese factors with preterm birth, however, is much smaller than the associations in Figures 1 to 3.The association of abortions with preterm birth cannot be explained by gravidity because therelative risks for prior births are different from prior pregnancies. Data on the gestation of theprior abortions, indications, complications, and inter-pregnancy intervals are not available in theroutine perinatal data. Termination of pregnancy has been legal under case interpretation of legislation since1969 and the procedure is covered by the federal government universal health insurance system[4].

The data meet four of the criteria for causality. The temporal sequence is clear: theabortions preceded the preterm birth. The association is a strong one. There is a dose-responserelationship: the greater the number of prior pregnancies the higher the relative risk. Theassociation is plausible: possible mechanisms exist which are outlined below. The other criteriacannot be fully met a present: the study design (a retrospective cohort) is rated as moderate.Reversibility of the exposure is not applicable to these exposures. No other papers have reportedan analysis stratified by gestation and by number of prior pregnancies so that the consistency ofthe findings cannot be assessed. One possible mechanism is that cervical instrumentation can facilitate the passage oforganisms into the upper part of the uterus, increasing the probability of inapparent infection andsubsequent preterm birth [5]. Another is the removal of the endometrium carries a small risk ofdamaging the decidual stroma in such a way as to impair trophoblast invasion, migration and thefull transformation of the maternal spiral arteries [6].

(3.) Moreau C, Kaminski M, Ancel PY, Bouyer J, Escande B, et al. Previous inducedabortions and the risk of very preterm delivery: results of the EPIPAGE study.British J Obstetrics Gynaecology 2005;112(4):430-437.ABSTRACTObjectives: To evaluate the risk of very preterm birth (22-32 weeks of gestation) associated withprevious induced abortion according to the complications leading to very preterm delivery insingletons. Design: Multicentre, case-control study (the French EPIPAGE study). Setting:Regionally defined population of births in France. Sample: The sample consisted of 1943 verypreterm live-born singletons (< 33 weeks of gestation), 276 moderate preterm live-bornsingletons (33-34 weeks) and 618 unmatched full-term controls (39-40 weeks). Methods: Datafrom the EPIPAGE study were analysed using polytomous logistic regression models to controlfor social and demographic characteristics, lifestyle habits during pregnancy and obstetric history.The main mechanisms of preterm delivery were classified as gestational hypertension, antepartumhaemorrhage, fetal growth restriction, premature rupture of membranes, idiopathic preterm laborand other causes. Main Outcome Measures: Odds ratios for very preterm birth by gestationalage and by pregnancy complications leading to preterm delivery associated with a history ofinduced abortion. Results: Women with a history of induced abortion were at higher risk of verypreterm delivery than those with no such history (OR + 1.5, 95% CI 1.1-2.0); the risk was evenhigher for extremely preterm deliveries (< 28 weeks). The association between previous inducedabortion and very preterm delivery varied according to the main complications leading to verypreterm delivery. A history of induced abortion was associated with an increased risk ofpremature rupture of the membranes, antepartum haemorrhage (not in association withhypertension) and idiopathic spontaneous preterm labour that occur at very small gestational ages(< 28 weeks). Conversely, no association was found between induced abortion and very pretermdelivery due to hypertension. Conclusion: Previous induced abortion was associated with anincreased risk of very preterm delivery. The strength of the association increased with decreasinggestational age.

(4.) Stang P, Hammond AO, Bauman P. Induced Abortion Increases the Risk of VeryPreterm Delivery; Results from a Large Perinatal Database. Fertility Sterility Sept2005;S159ABSTRACTObjective: 49% of pregnancies among American women are unintended; 1/2 of these areterminated by abortion. In 2000, 1.31 million abortions took place in the USA. There has beendebate over the impact of induced abortions (VTP) on future pregnancy outcome, particularly therisk of very preterm birth (<28 weeks). Sequelae of very preterm delivery remain a major publichealth burden. The objective of our study was to evaluate the association between prior inducedabortions and birth before 28 weeks gestation. Design: Retrospective analysis of a perinatal database. Materials and methods: Setting: 29 delivery units in the State of Schleswig-Holstein,Germany, between 1991 and 1997. Patients: 170,254 women with singleton pregnancies.Intervention: Univariate cross table analysis and logistic regression were conducted to determinethe association between previous induced first trimester abortions and delivery before 28 weeksgestation. Main outcome measures: Maternal age, smoking, cervical incompetence, pretermpremature rupture of membranes (PPROM), history of uterine hemorrhage before 28 weeks,maternal hypertension, delivery before 28 weeks. Results: 64586 nulliparous women wereincluded, 4572 (7.1 %) of whom had up to 9 VTPs . Maternal age was 27.4 ± 4.7 years in controlsand 29.1 ± 5.2 in the subjects. There was no difference in gestational age[(25.9 ± 1.9 weeks(controls), 25.4 ± 2.1 (subjects)]. Univariate relations between VTP and risk factors for pretermdelivery are shown in table 1. Logistic regression demonstrated the following associations withdelivery before 28 weeks: Cervical incompetence OR 4.46, CI 3.18-6.27; hemorrhage < 28 wks,OR 3.33, CI 2.37 - 2.69; hypertension, OR 3.18, CI 2.17-4.67; PPROM, OR 4.67, CI 3.45-6.33,first trimester miscarriage, OR 1.67, CI 1.25-2.21, and VTP, OR 1.55, CI 1.07-2.34. There was noassociation between smoking or AMA and very preterm delivery.

(5.) Smith GCS, Shah I, White IR, Pell JP, Crossley JA, Dobbie R. Maternal andbiochemical predictors of spontaneous preterm birth among nulliparous women: asystematic analysis in relation to degree of prematurity. Intl J Epidem2006;35(5):1169-1177.ABSTRACTBackground: Nulliparous women are at increased risk of spontaneous preterm birth. Othermaternal and biochemical risk factors have also been described. However, it is unclear whetherthese associations are strong enough to offer clinically useful prediction. It is also unclear whetherthe predictive power of these factors varies in relation to the degree of prematurity. Methods: Therisk of spontaneous preterm birth associated with maternal characteristics and second trimesterserum screening data was analysed in a dataset of 84 391 first births in Scotland between 1992and 2001 using Cox and logistic regression. Variation in the relative risk of preterm birth over theperiod 24–36 weeks was assessed using a test of the proportional hazards assumption. Results:The risk of spontaneous preterm birth was positively associated with maternal serum levels ofalpha-fetoprotein, socioeconomic deprivation, number of previous therapeutic abortions, smoking,and being unmarried and was negatively associated with height and body mass index. The risk ofpreterm birth at 24–28 weeks, but not later gestations, was increased in association with maternallevels of human chorionic gonadotrophin >95th percentile, maternal age <20, and two or moreprevious miscarriages. The area under the receiver operating characterise curve (95% CI) formodels based on these factors was 0.67 (0.63–0.71) for 24–28 weeks, 0.65 (0.62–0.68) for 29–32weeks, and 0.62 (0.61–0.63) for 33–36 weeks. Conclusions: Time to event analytic methods canidentify factors that are differentially associated with spontaneous preterm birth according to the

degree of prematurity. However, models based on maternal and biochemical data perform poorlyas a screening test for any degree of spontaneous preterm birth.

(6.) Levin AA, Schoenbaum SC, Monson RR, Stubblefield PG, Ryan JA. Association ofInduced Abortion with Subsequent Pregnancy Loss. JAMA 1980;243(24):2495-2499ABSTRACTWe compared prior pregnancy histories of two groups of multigravidas--240 women having apregnancy loss up to 28 weeks' gestation and 1,072 women having a term delivery. Women whohad had two or more prior induced abortions had a twofold to threefold increase in risk of first-trimester spontaneous abortion, loss between 14 to 19 and 20 to 27 weeks. The increased risk waspresent for women who had legal induced abortions since 1973. It was not explained by smokingstatus, history of prior spontaneous loss, prior abortion method, or degree of cervical dilatation.No increase in risk of pregnancy loss was detected among women with a single prior inducedabortion. We conclude that multiple induced abortions do increase the risk of subsequentpregnancy losses up to 28 weeks' gestation.

BIBLIOGRAPHYThe following list provides 51 peer-reviewed studies demonstrating a significant riskbetween IA and subsequent preterm birth since 1989. This list is not exhaustive.

* studies that included spontaneous and induced abortions but did not report preterm birth and low birth weight (LBW) risk separately for each.

+ studies that found a dose response effect (signifying an increased risk of preterm birth following increased number of induced abortions.)

1990s

+A01 Vasso L-K, Chryssa T-B, Golding J. Previous obstetric history and subsequentpreterm delivery in Greece. European J Obstetrics & Gynecology ReproductiveBiology 1990;37:99-109.

A02 Li YJ, Zhou YS. Study of factors associated with preterm delivery. Zhongjua Liu Xing Bing Xue Za Chi. Aug 1990;11(4):229-234.

A03 Harger JH, Hsing AW, Tuomala RE, Gibbs RS, Mead PG et al. Risk factors for preterm premature rupture of fetal membranes: A multicenter case-control study. Am J Obstet Gynec 1990;163:130-137.

A04 McGregor JA, French J, Richter R. Antenatal microbiologic and maternal risk factors associated with prematurity. Amer J Obstet Gynecol 1990;163:1465-1473

A05 Pickering RM, Deeks JJ. Risks of Delivery during 20th to the 36th Week of Gestation. Intl. J Epidemiology 1991;20:456-466.

*+A06 Zhang J, Savitz DA. Preterm Birth Subtypes among Blacks and Whites. Epidemiology 1992;3:428-433.

*A07 Michielutte R, Ernest JM, Moore ML, Meis PJ, Sharp PC, Wells HB, Buescher PA. A Comparison of Risk Assessment Models for Term and Preterm Low Birthweight. Preventive Medicine 1992;21:98-109.

A08 Gong JH. Preterm delivery and its risk factors. Zhounghua Fu Chan Ke Za Chi Jan. 1992;27(1):22-24

A09 Mandelson MT, Maden CP, Daling JR. Low Birth Weight in Relation Multiple Induced Abortions. Am J Public Health 1992;82;391-394

+A10 Lumley J. The epidemiology of preterm birth. Bailliere's Clin Obstet Gynecology. 1993;7(3):477-498

A11 Algert C, Roberts C, Adelson P, Frammer M. Low birth weight in New South Wales, 1987: a Population-Based Study. Aust New Zealand J Obstet Gynaecol 1993;33:243-248.

A12 Ekwo EE, Grusslink CA, Moawad A. Previous pregnancy outcomes and subsequent risk of premature rupture of amniotic sac membranes. Brit J Obstet Gynecol 1993;100(6):536-541.

A13 Lekea-Karanika V, Tzoumaka-Bangoula C. Past obstetric history of the mother and its association with low birth weight of a subseaquent child: a population-

based study. Paediatric Perinat Epidemiol 1994;8:173-187

A14 Guinn D, Goldenberg RL, Hauth JC, Andrews WA, Thom E, Romero R. Risk factors for the development of preterm premature rupture of membranes after arrest of preterm labor. AJOG 1995;173(4):1310-1315.

*A15 Hillier SL, Nugent RP, Eschenbach DA, Krohn MA,et al. Association Between Bacterial Vaginosis And Preterm Delivery Of A Low-Birth-Weight Infant. NEJM 1995;333:1737-1742.

A16 Khalil AK, El-Amrawy SM, Ibrahim AG, et al. Pattern of growth and development of premature children at the age of two and three years in Alexandria, Egypt. Eastern Mediterranean Health Journal 1995;1(2):186-193.

A17 Meis PJ, Michielutte R, Peters TJ, Wells HB. Factors associated with preterm birth in Cardiff, Wales. Amer J Obstet Gynecol 1995;173:590-596.

+A18 Lang JM, Lieberman E, Cohen A. A Comparison of Risk Factors for Preterm Labor and Term Small-for-Gestational-Age Birth. Epidemiology 1996;7:369-376.

*A19 Hagan R, Benninger H, Chiffings D. Evans S, French H. Very preterm birth - a regional study. Part 1: Maternal and obstetric factors. BJOG 1996;103:230-238.

A20 Chie-Pein Chen, Kuo-Gon Wang, Yuh-Cheng Yang, Lai-Chu See. Risk factors for preterm birth in an upper middle class Chinese population. Eur J Obstet Gynecol Reprod Bio 1996;70(1):53-59.

A21 Jacobsen G, Schei B, Bakketeig LS. Prepregnant reproductive risk and subsequent birth outcome among Scandinavian parous women. Norsk Epidemiol

1997;7(1):33-39.

+A22 Lumley J. The association between prior spontaneous abortion, prior induced abortion and preterm birth in first singleton births. Prenat Neonat Med 1998;3:21- 24.

+A23 Martius JA, Steck T, Oehler MK, Wulf K-H. Risk factors associated with preterm (<37+0 weeks) and early preterm (<32+0 weeks): univariate and multi-variate analysis of 106 345 singleton births from 1994 statewide perinatal survey of Bavaria. European J Obstetrics & Gynecology Reproductive Biology 1998;80:

183-189.

A24 Small Babies in Scotland A Ten Year Overview 1987-1996. Information and Statistics Division. The National Health Service in Scotland. Scottish Program

for Clinical Effectiveness. Edinburgh 1998 ISBN 1-902076-07-9.

A25 Lee KS, Lee WC, Meng KH, Lee Ch, Kim SP. Maternal Factors Associated with the Premature Rupture of Membrane in the Low Birth Weight Infant Deliveries. Korean J Prev Med 1998;21(2):207-216.

+A26 Ancel PY, Saurel-Cubizolles M-J, Renzo GCD, Papiernik E, Breart G. Very and moderate preterm births: are the risk factors different? British J Obstetrics and Gynaecology 1999;106:1162-1170.

+A27 Zhou W, Sorenson HT, Olsen J. Induced Abortion and Subsequent Pregnancy Duration. Obstetrics & Gynecology 1999;94:948-953.

A28 Ancel PY, Saurel-Cubizolles, Di Renzo GC, Papiernik E, Breart G Social Differences of very preterm birth in Europe: interaction with obstetric history. American J Epi 1999;149(10):908-915.

2000-2007

A29 Foix-L'Helias L, Ancel PY, Blondel B. Changes in risk factors of preterm delivery in France between 1981 and 1995. Paediatric and Perinatal Epidemiology. Oct 2000;14(4):314-323.

A30 Foix-L'Helias L, Ancel, Blondel B. Risk factors for prematurity in France and

comparisons betweeen spontaneous prematurity and induced labor; results from the National Perinatal Survey 1995. J Gynecol Obstet Bio Reprod (Paris) Feb2000;29(1);55-65.

*A31 Gardosi J, Francis A. Early Pregnancy predictors of preterm birth: the role of a prolonged menstruation-conception interval. BJOG 2000;107(2):228-237.

A32 Bettiol H, Rona RJ, Chin S, Goldani M, Barberi M. Risk Factors Associated with preterm births in Southeast Brazil: a comparison of two birth cohorts born 15 years apart. Paediatric Perinatal Epidemiol 2000;14(1):30-38.

+A33 Henriet L, Kaminski M. Impact of induced abortions on subsequent pregnancy outcome: the 1995 French national perinatal survey. British J Obstetrics Gynaecology 2001;108:1036-1042.

A34 Letamo G, Majelantle RG. Factors Influencing Low Birth Weight and Prematurity in Botswana. J Biosoc Sci 2001;33(3):391-403.

A35 Grimmer I, Buhrer C, Dudenhausen JW. Preconceptional factors associated with very low birth weight delivery: a case-control study. BMC Public Health 2002; 2:10 [Germany].

A36 Balaka B, Boeta S, Aghere AD, Boko K, Kessie K, Assimadi K. Risk factors associated with prematurity at the University of Lme, Togo. Bull Soc Pathol Exot Nov 2002;95(4):280-283.

A37 Han WH, Chen LM, Li CY. Incidences of and Predictors for Preterm Births and Low Birth Weight Infants in Taiwan. Chinese Electronic Periodical Services 2003:131-141.

A38 Reime B, Schuecking BA, Wenzlaff P. Perinatal outcomes of teenage pregnancies according to gravidity and obstetric history. Annals of Epidemiology

2004;14(8):619-619 [German subjects].

+A39 Ancel PY, Lelong N, Papiernik E, Saurel-Cubizolles MJ, Kaminski M. History of induced abortion as a risk factor for preterm birth in European countries: results of EUROPOP survey. Human Repro 2004; 19(3): 734-740.

A40 Umeora OUJ, Ande ABA, Onuh SO, Okubor PO et al. Incidence and risk factors for preterm delivery in a tertiary health institution in Nigeria. J Obstet Gynaecol2004;24(8):895-896.

+A41 Moreau C, Kaminski M, Ancel PY, Bouyer J, Escande B, et al. Previous induced abortions and the risk of very preterm delivery: results of the EPIPAGE study. British J Obstetrics Gynaecology 2005;112(4):430-437.

A42 Conde-Agudelo A, Belizan JM, Breman R, Brockman SC, Rosas-Bermudez.

Effect of the interpregnancy interval after an abortion on maternal and perinatal health in Latin America. Int J Gynaecol & Obstet 2005;89 (Supp. 1):S34-S40.

A43 Stang P, Hammond AO, Bauman P. Induced Abortion Increases the Risk of Very Preterm Delivery; Results from a Large Perinatal Database. Fertility Sterility Sept 2005;S159.

A44 Etuk SJ, Etuk IS, Oyo-Ita AE. Factors Influencing the Incidence of Pre-term Birth in Calabar, Nigeria. Nigerian J Physiological Sciences 2005;20(1-2):63-68.

A45 Poikkens P. Unkila-Kallio L, Vilska S, Repokari L. et al. Impact of Infertility Characteristics and treatment modalities on singleton pregnancies after assisted reproduction. Reprodutive Biomed July 2006;13(1):135-144.

A46 Samin A, Al-Dabbagh, Wafa Y Al-Taee. Pregnancy and Childbirth. BMC 2006;6:13.

A47 Smith GCS, Shah I, White IR, Pell JP, Crossley JA, Dobbie R. Maternal and biochemical predictors of spontaneous preterm birth among nulliparous women: a systematic analysis in relation to degree of prematurity. Intl J Epidem 2006;35(5):1169-1177.

A48 Briunsma F, Lemley J, Tan J, Quinn M. Precancerious changes in the cervix and risk of subsequent preterm birth. BJOG Jan. 2007;114(1):70-80

A49 Jackson JE, Grobman WA, Haney E, Casele H. Mid-trimester dilation and evacuation with laminaria does not increase the risk for severe subsequent pregnancy complications. Intl J Gynecol Obstet 2007;96:12-15.

*+A50 Brown TS, Adera T, Masho SW. Previous abortion and the risk of low birth weight and preterm births. J Epidemiol Commun Health 2008;62:16-22

A51 Reime B, Schuecking BA, Wenzlaff P. Reproductive Outcomes in Adolescents Who Had a Previous Birth or an Induced Abortion Compared to Adolescents' First Pregnancies. BMC Pregnancy and Childbirth 2008;8:4.

Brent RooneyByron C. Calhoun, M.D.

ABSTRACT

At least 49 studies have demonstrated a statistically significant

increase in premature births (PB) or low birth weight (LBW) risk in

women with prior induced abortions (IAs). This paper will focus on

the risk of early premature births (EPBs) (< 32 weeks gestation)

and extremely early premature births (XPBs) (< 28 weeks gesta-

tion). Large studies have reported a doubling of EPB risk from two

prior IAs. Women who had four or more IAs experienced, on aver-

age, nine times the risk of XPB, an increase of 800 percent.

These results suggest that women contemplating IA should be

informed of this potential risk to subsequent pregnancies, and that

physicians should be aware of the potential liability and possible

need for intensified prenatal care.

Informed consent for an elective surgical procedure must gen-erally cover long-term consequences and not just immediate risk.Awoman considering an induced abortion (IA) should thus expect tobe informed of potential effects on her fertility and the health of fu-ture infants, as well as her own future health. An elevated risk ofbearing a child afflicted with a serious disability such as cerebralpalsy might influence her decision, as well as future liability deter-minations by courts.

Low birth weight (LBW) and premature birth (PB) are the most

important risk factors for infant mortality or later disabilities aswell as for lower cognitive abilities and greater behavioral

problems and thus contribute importantly to the liability exposureof obstetricians.

A literature review retrieved 49 studies that demonstrated atleast 95 percent confidence in an increased risk of preterm birth(PB), or surrogates such as low birth weight or second-trimesterspontaneous abortion, in association with previous induced abor-tions. A list of these studies, which probably does not comprise allsuch studies, is appended to this article. If these 49 statistically sig-nificant associations were the result of chance alone, as may happenin 5 of 100 tests, IA should be associated with a reduction in PBs,with P<.05, in an equivalent number of tests. Not one such instancehas been found in the literature.

A MedLine search from 1966 to March, 2003, retrieved 8 stud-ies that purportedly failed to show a significant increase in prema-

ture births after IA. Most showed an increase that did not reachstatistical significance because of the small sample size: fewer than1,000 pregnancies following an IA. In one, an increased risk of PBin women who had had an IA was nonsignificant when controlled

for parity. These studies did not consider separately the risk ofEPBs or XPBs, or the effect of multiple IAs, except one that showeda statistically significant increase of EPB, despite the statement inthe abstract that “in the Netherlands there are no significant indica-tions that spontaneous midtrimester abortions or premature deliver-

ies are caused by a previous induced abortion.”

1

2

3-10

3

6

Induced Abortion andRisk of Later Premature Births

A 1986 review concluded that “more research is needed before

it is clear whether multiple induced abortions carry an increased

risk of adverse pregnancy outcomes.” The more recent, large stud-

ies discussed here help supply this lack.

A 1993 study in Victoria, Australia, involved 121,305 total

births and compared the risk of PB and XPB in women with various

numbers of IAs, compared with a control group of women who had

no prior pregnancies (see Table 1, derived from data in this report).

As Lumley explains:

The associations are different in the three gestation catego-

ries (20-27, 28-31, and 32-36 weeks), being particularly

striking for births before 28 weeks. In this category, there is

also evidence for a dose-response relationship between num-

ber of prior lost pregnancies and the prevalence of preterm

birth: relative risks of 1.66 and 1.55 for one spontaneous or

induced abortion, of 2.94 and 2.46 for two, and of 5.89 and

5.58 for three or more. These last four relative risks are sub-

stantially greater than any of those associated with maternal

age, marital status, parity or socioeconomic status: that is,

the association is most unlikely to be explained by con-

founding factors of a sociodemographic kind.

Lumley's argument that “small single possible confounders can-

not explain big risk factors such as 2.46 and 5.58” would also apply

to any attempt to pose smoking or drug abuse as an explanation for

the entire abortion-premature birth association.

The great majority of theAustralian IAs were via vacuum aspi-

ration; thus the PB risk cannot be attributed to dilation & curet-

tage IAs.

The author noted that cross-sectional studies show that the rel-

ative risk of preterm delivery increases with the number of the pre-

vious preterm births, but that the risk of subsequent preterm

births diminishes with each full-term delivery. Thus, IA re-

moves the protective potential of a full-term delivery, as others

have also observed.

11

12

12

12

13

Australian Study

Table 1: Premature birth risk by number of prior induced abortions (IAs)compared with outcome of first pregnancies, Victoria, 1986-1990

12

46 Journal of American Physicians and Surgeons Volume 8 Number 2 Summer 2003

In 1998, with twice the number of births (243,679) to analyze as

in 1993, Lumley validated her 1993 results and additionallyshowed that women with four or more prior IAs had an XPB risknine times that of primigravidas.

Another large study of 106,345 births in Bavaria, including 85percent of births in the state and 1,146 EPBs, showed a comparabledose-response curve (see Table 2, extracted from Table 2 in theBavarian study), confirming the Australian finding of the greatestincreased risk for the earliest premature infants.

In a multivariate analysis that included many of the possible con-founding variables, including previous stillborns, infertility treat-ment, age under 18 or over 35 years, malpresentation, prematurerupture of membranes, and preeclampsia, the effect of even a singleIAremained significant.

A 1999 study of Danish women is especially important be-cause it used an IAregistry, thus eliminating recall bias, the hypoth-esis that women with prior IAs who deliver prematurely are more ac-curate in reporting reproductive history than women who deliver atfull term, as a possible explanation for the results.

This study of 61,753 women found an odds ratio for pretermbirth at <34 weeks gestation of 1.99 (95% CI 1.64-2.43) for oneprior IA and 2.03 (95% CI 1.36-3.04) for two or more prior IAs.Vacuum aspiration (VA) was the method used in 92.3 percent of allabortions. For VA, PB (gestation <37 weeks) odds ratios for 1, 2, 3or more IAs were: 1.82, 2.45, and 2.00, respectively.

Dilation and evacuation increased the risk substantially. Oneevacuation was associated with a PB odds ratio of 2.27 whereas twoprior evacuations had a very large odds ratio of 12.55.

An accepted risk of surgical IA is incompetent cervix, which is

a PB risk factor. Nulliparous women who have multiple IAs boosttheir odds of being over age 35 at their first term delivery, a risk fac-

tor for PB. Additional risk factors for PB that may be increased

by abortion include uterine adhesions, infection, and mental

distress.The evidence meets four of the criteria for determining

causality: (1) the abortions preceded the premature births; (2) theassociation is strong; (3) there is a dose-response relationship; and(4) the association is plausible. A criterion for causality that couldnot be met in 1998 was confirmation by a prospective study.However, the Danish study identified all subsequent pregnanciesuntil 1994 of the women under study, whose first pregnancies oc-

14

15

16

17

14,15

14,23 17-19

17

14

German Study

Danish Study

Mechanisms forAbortions Causing a Premature Birth Risk

curred in 1980, 1981, and 1982. Reversibility of the exposure isnot applicable to this circumstance. Consistency of findings withearlier studies cannot be assessed because these were not stratifiedby length of gestation, number of prior pregnancies, and number of

IAs. However, the large studies in Germany, Denmark, andAustralia consistently support multiple prior IAs as boosting EPB

risk. Only theAustralian studies included an XPB category.

Recent litigation by women who claim that they were not fullyinformed about all the risks of an elective abortion, especially a pos-sible increased lifetime risk of breast cancer, has drawn attention tothe process of obtaining informed consent for this procedure.Moreover, even a signed consent form does not suffice to relieve aphysician in the U.S. or Canada of the responsibility to withhold atreatment that he knows, or ought to know, is medically contraindi-

cated. What level of risk will courts determine to constitute a medi-cal contraindication?

Liability costs are especially high in cases involving damagedbabies. The median damage award in cases of medical negligence

in attending at childbirth was $2,050,000 between 1994 and 2000.Women are warned in a classic book covering 50 risk factors for

PB that “if you have had one or more induced abortions, your risk of

prematurity with this pregnancy increases by about 30 percent.”As shown here, the risk could be substantially higher than that, de-pending on the number of abortions and the method used.

It has been claimed that “induced abortion...is directly responsi-ble for many thousands of cases of cerebral palsy–in NorthAmerica

alone–that otherwise would not have occurred.” Supporting thisassertion is the fact that the cerebral palsy risk in XPB is about 38

times higher than in the overall population of newborns, in which

the risk of cerebral palsy is approximately 2-3 per 1,000 births. Asthe liability costs for cerebral palsy are exceptionally high, inducedabortion, particularly without very detailed informed consent, maycarry an unsupportable legal liability. Courts may not require defin-itive proof of causation; the existence of a number of positive stud-ies, in the absence of definitive refutation, may be sufficient reasonto include discussion of a potential serious adverse effect in obtain-ing informed consent.

Aconsent form that simply lists such items as “incompetent cer-vix” or “infection” as potential complications, but does not informwomen of the elevated future risk of a preterm delivery, and that thelatter constitutes a risk factor for devastating complications such ascerebral palsy, may not satisfy courts.

The authors of a recent CME review survey article, which eval-uated 24 studies of abortion and PB, strongly affirmed the need forinformed consent. They stated that prior IAs boost the risk of PBand that 7 of 12 significant studies that they reviewed identified a

dose-response effect, with risks increasing with the number of IAs.

16

14

14-16 12,14

20

21

22

23

1

24

25

Liability and Informed Consent

Brent Rooney

Byron C. Calhoun, M.D., F.A.C.O.G., F.A.C.S.

is a medical researcher. He may be contacted at the ReducePreterm Risk Coalition, 3456 Dunbar St. (146), Vancouver, Canada V6S2C2. E-mail address: stopcancer@yahoo.com or whatsup@vcn.bc.ca.

, is Director, PerinatalAssessment Unit, Rockford Memorial Medical Center in Rockford, IL;Visiting Clinical Professor in Obstetrics and Gynecology, University ofIllinois at Chicago; and Adjunct Professor of Obstetrics and Gynecology,Midwestern University, Chicago College of Osteopathic Medicine.

Editorial assistance by Jane M. Orient, M.D., is gratefully acknowledged.

Table 2: Odds ratio (OR) for premature births by numberof prior induced abortions (IAs)

15

47Journal of American Physicians and Surgeons Volume 8 Number 2 Summer 2003

REFERENCES:1

2

3

4

5

6

7

8

9

10

11

12

Escobar GJ, Littenberg B, Petitti DB. Outcome among surviving very lowbirthweight infants; a meta-analysis. 1991;66:204-211.

Bhutta AT, Cleves MA, Casey PH, Cradock MM, Anand KJS. Cognitive andbehavioral outcomes of school-aged children who were born preterm: ameta-analysis. 2002;288:728-737.

Park TK, Strauss LT, Hogue CJ, Kim IS. Previous experience of inducedabortion as a risk factor for fetal death and preterm delivery.

1984;22:195-202.

Meirik O, Bergstrom R. Outcome of delivery subsequent to vacuum-aspiration abortion in nulliparous women.1983;62:499-509.

Frank PI, Kay CR, Lewis TL, Parish S. Outcome of pregnancy following in-duced abortion. Report from the joint study of the Royal College ofGeneral Practitioners and the Royal College of Obstetricians andGynaecologists. 1985;92:308-316.

Van der Slikke JW, Treffers PE. Influence of induced abortion on gesta-tional duration in subsequent pregnancies. 1978;1(6108):270-272.

Daling JR, Emanuel I. Induced abortion and subsequent outcome of preg-nancy in a series of American women. 1977;297:1241-1245.

Schott G, Ehrig E, Wulff V. Prospective studies into pregnancies ofprimiparae with record of therapeutic termination of previous pregnanciesor of spontaneous abortion and assessment of fertility.1980;102:932-938, 939-944

Kalish RB, Chasen ST, Rosenzweig LB, Rashbaum WK, Chervenak FA.Impact of midtrimester dilation and evacuation on subsequent pregnancyoutcome. 2002;187:882-885.

Lao TT, Ho LF. Induced abortion is not a cause of subsequent preterm de-livery in teenage pregnancies 1998;13:758-761.

Hogue CJ. Impact of abortion on subsequent fecundity.1986;13:95-103.

Lumley J. The epidemiology of preterm birth.1993;7(3):477-498.

Arch Dis Child

JAMA

Int J Gynaecol

Obstet

Acta Obstet Gynecol Scand

Br J Obstet Gynaecol

BMJ

N Engl J Med

Zentralbl Gynaekol

Am J Obstet Gynecol

. Hum Reprod

Clin Obstet

Gynaecol

Bailliere's Clin Obstet

Gynecol

[German].

APPENDIX: Studies that showed a statistically

significant increase in preterm birth after induced

abortion

*Ancel P-V, Saurel-Cubizolles M-J, Renzo GCD,

Papiernik E, Breart G. Very and moderate

preterm births: are the risk factors different?

1999;106:1162-1170.

Barsy G, Sarkany J. Impact of induced abortion

on the birth rate and infant mortality.

1963;6:427-467.

Berkowitz GS. An epidemiologic study of

preterm del ivery.

1981;113:81-92.

Bognar Z, Czeizel A. Mortality and morbidity asso-

ciated with legal abortions in Hungary, 1960-

1973. 1976;66:568-575.

Czeizel A, Bognar Z, Tusnady G, et al. Changes in

mean birth weight and proportion of low-

weight births in Hungary

1970;24:146-153.

Drac P, Nekvasilova Z. Premature termination of

pregnancy after previous interruption of preg-

nancy. 1970;35:332-333.

Furusawa Y, Koya Y. The influence of artificial

abortion on delivery. In: Koya Y, ed.

Tokyo: Family

Planning Federation of Japan; 1966:74-83.

Br

J Obstet Gynaecol

Demografia

Am J Epidemiol

Am J Public Health

. Br J Prev Soc Med

Cesk Gynekol

Harmful

Effects of Induced Abortion.

48 Journal of American Physicians and Surgeons Volume 8 Number 2 Summer 2003

13

14

15

16

17

18

19

20

21

22

23

24

25

Hogue CJ, Cates W Jr., Tietze C. Impact of vacuum aspiration on futurechildbearing: a review. 1983;15:119-126.

Lumley J. The association between prior spontaneous abortion, prior in-duced abortion and preterm birth in first singleton births.

1998;3:21-24.

Martius JA, Steck T, Oehler MK, Wulf K-H. Risk factors associated withpreterm (<37+0 weeks) and early preterm (<32+0 weeks): univariateand multi-variate analysis of 106,345 singleton births from 1994 statewideperinatal survey of Bavaria.1998;80:183-189.

Zhou W, Sorenson HT, Olsen J. Induced Abortion and subsequent preg-nancy duration. 1999;94:948-953.

Berkowitz GS, Papiernik E. Epidemiology of preterm birth.1993;15:414-443.

Muhlemann K, Germain M, Krohn M. Does an abortion increase the risk ofintrapartum infection in the following pregnancy? 1996;7:194-198.

Daling JR, Krohn MA, Miscarriage or termination in the immediately pre-ceding pregnancy increases the risk of intraamniotic infection in the fol-lowing pregnancy 1992;136:1013 SER Abstracts.

Picard E, Robertson G.. Scarsborough, Ontario: Carswell; 1996:264-265.

Zimmerman R, Oster C. Assigning liability: insurers' missteps helped pro-voke malpractice “crisis.” 2002;June 24:A1.

Luke B. .New York, N.Y.: Times Books; 1995.

Ring-Cassidy E, Gentles I. Toronto,Ontario: de Veber Institute; 2002.

Smith WS, Camfield C, Camfield P. Living with cerebral palsy and tubefeeding: a population-based follow-up study. 1999;135:307-310.

Thorp JM, Hartmann KE, Shadigian E. Physical and psychological healthconsequences of induced abortion: review of the evidence.

2003;58(1):66-79.

Fam Planning Perspect

Prenat Neonat

Med

Eur J Obstet Gynecol Reprod Biol

Obstet Gynecol

Epidemiol Rev

Epidemiol

. Am J Epidemiol

Legal Liabilities of Doctors and Hospitals in

Canada

Wall St J

Every Pregnant Woman's Guide to Preventing Premature Birth

Woman's Health after Abortion.

J Pediatr

Obstet

Gynecol Survey

Grindel B, Lubinski H, Voigt M. Induced abortion

in primigravidae and subsequent pregnancy,

with particular attention of underweight.

1979;101:1009-1114.

Harlap S, Davies AM. Late sequelae of induced

abortion: complications and outcome of preg-

nancy and labor.

1975;102:219-224.

Henriet L, Kaminski M. Impact of induced abor-

tions on subsequent pregnancy outcome: the

1995 French national perinatal survey.

2001;108:1036-1042.

*Hillier SL, Nugent RP, Eschenbach DA, Krohn

MA, et al. Association between bacterial

vaginosis and preterm delivery of a low-

birth-weight infant. 1995;333:

1737-1742.

Hungar ian Cent ra l Sta t is t ica l Of f ice .

Perinatalishalazons. Budapest: Hungarian

Central Statistical Office; 1972.

Koller O, Eikhom SN. Late sequelae of induced

abortion in primigravidae.

1977;56:311-317.

Kreibich H, Ludwig A. Early and late complica-

tions of abortion in juvenile primigravidae (in-

cluding recommended measures).

(Jena) 1980;74:311-316.

Zentralbl Gynaekol

Am J Ep idemio l

Br J

Obstet Gynaecol

N Engl J Med

Acta Obstet Gynecol

Scand

Z Aerztl

Fortbild

Lang JM, Lieberman E, Cohen A. A Comparison

of Risk Factors for Preterm Labor and Term

Small-for-Gestational-Age Birth.

1996;7:369-376.

Lean TH, Hogue CJR, Wood J. Low birth weight

after induced abortion in Singapore. Presented

at the 105th Annual Meeting of the American

Public Health Association, Washington D.C.,

Oct. 31, 1977.

Legrillo V. Quickenton P, Therriault GD, et al.

Effect of induced abortion on subsequent re-

productive function. Final report to NICHD.

Albany, N.Y.: New York State Health

Department; 1980.

Lerner RC, Varma AO. Prospective study of the

outcome of pregnancy subsequent to previous

induced abortion. Final report, Contract no.

(N01-HD-62803). New York: Downstate

Medical Center, SUNY, January 1981.

Levin A, Schoenbaum S, Monson R, Stubblefield

P, Ryan K. Association of abortion with subse-

quent pregnancy loss. 1980;243(24):

2495-2499.

Lieberman E, Ryan KJ, Monson RR,

Schoenbaum SC. Risk factors accounting for

racial differences in the rate of premature birth.

1987;317: 743-748.

Epidemiology

JAMA

N Engl J Med

49Journal of American Physicians and Surgeons Volume 8 Number 2 Summer 2003

Lumley J. Very low birth-weight (less than 1500g)

and previous induced abortion: Victoria 1982-

1983 1986;26:268-272.

Lumley J. The epidemiology of preterm birth.

1993;7(3): 477-498.

Lumley J. The association between prior sponta-

neous abortion, prior induced abortion and

preterm birth in first singleton births.

1998;3:21-24.

Martius JA, Steck T, Oehler MK, Wulf K-H. Risk fac-

tors associated with preterm (<37+0 weeks)

and early preterm (<32+0 weeks): univariate

and multivariate analysis of 106 345 singleton

births from 1994 statewide perinatal survey of

Bavaria.

1998;80:183-189.

Meirik O, Bergstrom R. Outcome of delivery sub-

sequent to vacuum aspiration abortion in

nulliparous women.

1982;61:415-429.

*Michielutte R, Ernest JM, Moore ML, Meis PJ,

Sharp PC, Wells HB, Buescher PA. A

Comparison of Risk Assessment Models for

Term and Preterm Low Birthweight.

1992;21:98-109.

Miltenyi K. On the effects of induced abortion

1964;7:73-87.

Mocsary P, Csapo AI. Effect of menstrual induc-

tion on prematurity rate. 1978;1:1159-

1160.

Mueller-Heubach E, Guzick DS. Evaluation of

risk scoring in a preterm birth prevention study

of indigent patients.

1989;160:829-837.

Obel E, et al. Pregnancy complications following

legally induced abortion with special reference

to abortion technique.

1979;58:147-152.

Pantelakis SN, Papadimitriou GC, Doxiadis SA.

Influence of induced and spontaneous abor-

tions on the outcome of subsequent pregnan-

cies. 1973;116:799-805.

. Aust NZ J Obstet Gynecol

Bailliere's Clin Obstet Gynecol

Prenat

Neonat Med

Eur J Obstet Gynecol Reprod Biol

Acta Obstet Gynecol

Scand

Prev Med

.

Demografia

Lancet

Am J Obstet Gynecol

Acta Obstet Gynecol

Scand

Am J Obstet Gynecol

Papaevangelou G, Vrettos AS, Papadatos D,

Alexiou C. The effect of spontaneous and in-

duced abort ion on prematuri ty and

birthweight.

. 1973;80(May):418-422.

Pickering RM, Forbes J. Risk of preterm delivery

and small-for-gestational age infants following

abortion: a population study.

1985;92:1106-1112.

Pickering RM, Deeks JJ. Risks of Delivery during

20th to the 36th Week of Gestation.

1991;20:456-466.

Puyenbroek J, Stolte L. The relationship between

spontaneous and induced abortions and the

occurrence of second-trimester abortion in sub-

sequent pregnancies

1983;14:299-309.[This is the only

study in this complete list that uses second-

trimester abortion as a surrogate for PTB.]

Ratten G et al. Effect of abortion on maturity of

subsequent pregnancy.

1979( :479-480.

Richardson JA, Dixon G. Effect of legal termina-

tion on subsequent pregnancy.

1976;1:1303-1304.

Roht LH, Aoyama H, Leinen GE, et al. The associ-

ation of multiple induced abortions with subse-

quent prematurity and spontaneous abortion.

1976;23:140-145

Schoenbaum LS, Monson RR. No association be-

tween coffee consumption and adverse out-

comes of pregnancy.

1982;306:141-145.

Seidman DS, Ever-Hadani P, Slater PE, Harlap S,

et al. Child-bearing after induced abortion: reas-

sessment of risk.

1988;42:294-298.

Shiono PH, Lebanoff MA. Ethnic differences and

very preterm delivery.

1986;76:1317-1321.

J Obstet Gynaecol Br

Commonwealth

Br J Obstet

Gynecol

Int J

Epidemiol

. Eur J Obstet Gynecol

Reprod Biol

Med J Aust

BMJ

Acta Obstet Gynaecol Jpn

N Engl J Med

J Epidemiology Community

Health

Am J Public Health

June)

Slater PE, Davies AM, Harlap S. The effect of abor-

tion method on the outcome of subsequent

pregnancy. 1981;28:123-128.

Van der Slikke JW, Treffers PE. Influence of in-

duced abortion on gestational duration in sub-

sequent pregnancies. 1978;1:270-272.

Vasso L-K, Chryssa T-B, Golding J. Previous ob-

stetric history and subsequent preterm deliv-

ery in Greece.

1990;37:99-109.

World Health Organization. Special Programme

of Research, Development and Research

Training in Human Reproduction: Seventh

Annual Report, Geneva, Nov. 1978.

World Health Organization Task Force on the

Sequelae of Abortion. Gestation, birthweight

and spontaneous abor t ion .

1979;1:142-145.

*Zhang J, Savitz DA. Preterm birth subtypes

among blacks and whites.

1992;3:428-433.

Zwahr C, Voigt M, Kunz L, et al. Relationships be-

tween interruption abortion, and premature

birth and low birth weight.

1980;102: 738-747.

Zhou W, Sorenson HT, Olsen J. Induced abortion

and subsequent pregnancy duration.

col 1999;94:948-953.

*Studies that included spontaneous and in-

duced abortions but did not report PTB/LBW

risk separately for each

J Reprod Med

BMJ

Eur J Obstet Gynecol Reprod

Biol

Lance t

Epidemiology

Zentralbl Gynaekol

Obstet

Gyne

Medical Controversy

“Since among practitioners there will prove to be so much difference of opinion about acute diseases thatthe remedies which one physician gives in the belief that they are the best are considered by a second to bebad, laymen are likely to object to such that their art resembles divination; for diviners too think that thesame bird, which they hold to be a happy omen on the left, is an unlucky one when on the right, while otherdiviners maintain the opposite.”

[Fortunately AAPS can discern right from left, and open debate is welcome.]

Hippocrates Regimen in Acute Diseases

– Lawrence R. Huntoon, M.D., Ph.D.