Although it generally is accepted that second-trimester abortion procedures are safe in legal settings, few studies describe complication rates in contemporary practice. Of particular interest is the impact of the rising cesarean delivery rate on second-trimester abortion morbidity. This retrospective cohort study describes complications from second-trimester surgical abortions in a large, urban, academic institution, with a focus on the impact of previous cesarean delivery on the risk of procedural complications.
Immediate complications of second-trimester surgical abortions include hemorrhage, cervical laceration, uterine atony, perforation, and retained tissue. However, most studies addressing complication rates are limited to a narrow gestational age range,1–5 do not include procedures beyond 20 weeks of gestation,1,2,4 or have a small sample size.1–3,5,6 In addition, most studies were published in the 1970s and 1980s and do not reflect current practices involving prophylactic antibiotics,1,2 mechanical or medical cervical preparation,2,7,8 or intraoperative ultrasonography.2,7,8
Although increasing gestational duration is a widely recognized independent risk factor for complications in second-trimester surgical abortions,9 other risk factors have not been clearly established. The rising cesarean delivery rate and its impact on abortion-related morbidity is of particular concern to many abortion providers. The cesarean delivery rate increased from 21% in 1996 to a record high of 32% in 2007.10 It is well documented that previous cesarean delivery predisposes women to abnormal placentation,11 uterine rupture,12 and hemorrhage13 in subsequent pregnancies. It also is associated with hemorrhage and uterine rupture in second-trimester medical abortions.14 However, the previously published literature addressing this issue in second-trimester surgical terminations found no increased risk associated with a history of one or multiple cesarean deliveries.15,16 This study reexamines this issue among a larger cohort from an academic center in which more than half of abortions are performed beyond 14 weeks of gestation and more than 20% of women have a history of one or more cesarean deliveries. Our objective was to estimate second-trimester surgical abortion complication rates and to estimate the effect of past cesarean delivery on the risk of complications.
MATERIALS AND METHODS
We included all second-trimester surgical abortions performed between October 2004 and March 2007 at the San Francisco General Hospital Women's Options Center. The Women's Options Center is a referral site for medically and surgically complicated cases for northern and central California. The clinic provides an average of 2,200 abortions annually, of which more than half are second-trimester procedures. Second-trimester abortions were defined as dilation and evacuation (D&E) performed after 14 0/7 weeks of gestation dated by ultrasonography.
Second-trimester abortion providers include attending physicians, family planning fellows, and third-year gynecology residents. Abortions at or beyond 14 weeks of gestation are performed as a 2-day outpatient procedure. Laminaria are placed after ultrasonographic dating and a thorough counseling and consent process. If the initial dilation is thought to be insufficient, the patient undergoes additional laminaria placement, receives misoprostol, or undergoes mechanical dilation at the time of the procedure. Patients take doxycycline after the D&E and repeat the dose 12 hours later. All second-trimester procedures are performed with ultrasonographic guidance by D&E using moderate or deep sedation and local anesthesia. At the conclusion of the abortion, the patient is transferred to the recovery room, where she is monitored for a minimum of 90 minutes before discharge.
We began prospectively collecting each patient's basic demographic information, obstetric history, procedure details, and complications in September 2004 as part of internal review. Complications recorded in the database included cervical laceration, atony, hemorrhage, disseminated intravascular coagulation, anesthetic complications, perforation, retained products of conception, and “other.” A cervical laceration was defined as a tear requiring suture repair. Atony was defined as uterine relaxation and bleeding requiring more than one dose of uterotonic medication or additional intervention or both. Hemorrhage was defined as blood loss requiring transfusion. Recorded interventions included reaspiration; placement of an intrauterine balloon; suture repair of a cervical laceration; administration of uterotonic medications or immediate antibiotic therapy; transfusion; hospitalization; and reoperation by means of uterine artery embolization, laparoscopy, or laparotomy, including hysterectomy. A complication was labeled “major” if it resulted in transfusion; disseminated intravascular coagulation; or reoperation by means of uterine artery embolization, laparoscopy, or laparotomy.
All cases in which a complication or intervention was recorded were reviewed by the first author (A.C.F.) to ensure the case met standard criteria for a complication as defined above. The review included nursing intake notes, history and physical examination forms, procedure and recovery room documentation, and any relevant hospital records. Additional information regarding the specific management and resolution of the complication was abstracted.
Demographic, medical, and procedural variables were compared between women with and without major complications. Categorical data were calculated as percentage frequencies, and differences between proportions were compared using the χ2 test; t tests were performed to evaluate the relationship between major complication and continuous variables. Models for major complication, hemorrhage, cervical laceration, and atony were then created. Chi-square and t tests were used to select appropriate independent variables for multivariable logistic regression models. Variables assessed for each model include age, ethnicity, insurance status, history of vaginal delivery, history of cesarean delivery, history of induced abortion, estimated gestational age of 20 weeks or more, and need for additional dilation. Age was treated as a continuous variable. Ethnicity was dichotomized into white and nonwhite, whereas insurance status, history of vaginal delivery, history of induced abortion, estimated gestational age 20 weeks or more, and need for additional dilation were dichotomized. History of cesarean delivery was included as a three-part variable of none, one, or two or more cesarean deliveries. A value of P<0.1 was required for inclusion in the multivariable logistic regression model.
Approval for the study was obtained from the Committee on Human Research through the University of California, San Francisco. Statistical analysis was performed using STATA 10.0 (STATA Corp., College Station, TX).
The cohort includes 2,973 D&Es completed between September 2004 and March 2007 at the Women's Options Center of San Francisco General Hospital. Women were ethnically diverse: 32% were Latinas, 29% were whites, 25% were African Americans, 13% were Asian/Pacific Islanders, and 1% represented other ethnicities (Table 1). The mean age was 26 years (range 14–47 years). More than 92% of abortions were funded by public insurance, whereas 6% were funded by private insurance and 1% were self-pay.
The data describe D&Es performed from 14 to 27 weeks of gestation, with a mean gestational age of 20.2 (±2.5) weeks. Abortions beyond 23.1 weeks (n=38) were performed only in cases of fetal anomalies or maternal health compromise. Thirty-eight percent (n=1,098) of abortions were completed before 20 weeks of gestation, whereas 32% (n=987) were performed at 20–21 weeks and 30% (n=888) were performed at or beyond 22 weeks of gestation. We observed that 7.2% (95% confidence interval [CI] 6.3–8.1%) of all procedures were notable for at least one complication, whereas 1.4% (95% CI 1.0–1.8%) involved more than one (Table 2).
A major complication resulting in transfusion; disseminated intravascular coagulation; or reoperation with uterine artery embolization, laparoscopy, or laparotomy was encountered in 38 (1.3%, 95% CI 0.9–1.7%) cases. In multivariable analysis, a history of two or more cesarean deliveries (odds ratio [OR] 7.4, 95% CI 3.4–15.8), gestational age of 20 or more weeks (OR 4.4, 95% CI 2.0–11.4), and insufficient initial cervical dilation (OR 2.6, 95% CI 1.2–5.4) remained independent risk factors for a major complication (Table 3). Thirty patients (1.0%, 95% CI 0.6–1.4%) experienced hemorrhage requiring transfusion. The same variables were independently correlated with hemorrhage: a history of two or more cesarean deliveries (OR 7.4, 95% CI 3.0–17.6), gestational age of 20 or more weeks (OR 6.5, 95% CI 2.4–22.9), and insufficient initial cervical dilation (2.6, 95% CI 1.1–5.7).
In 99 cases (3.3%, 95% CI 2.7–3.9), a cervical laceration required suturing. Eighty-seven (88%) of 99 cervical lacerations were repaired adequately with suture alone. Seven patients had lacerations that required that required application of Monsels or a Foley bulb placement, two of whom were admitted for prolonged observation. Two women had cervical lacerations in association with perforation necessitating subsequent hysterectomy. An additional two patients had cervical lacerations in association with uterine atony and hemorrhage and required admission, one of whom underwent a successful uterine artery embolization. The second patient was treated conservatively. In addition, one patient had a high cervical laceration requiring a Foley bulb placement, admission, and a blood transfusion. A logistic regression model revealed that, although gestational age of 20 or more weeks and a history of vaginal delivery increased the risk of cervical laceration (OR 7.3, 95% CI 3.8–16.5 and OR 1.7, 95% CI 1.1–2.6, respectively), a history of induced abortion had the opposite effect (OR 0.6, 95% CI 0.4–0.9).
Of the 78 cases with uterine atony, 32 (41%) were treated successfully with uterotonics alone. The remainder required additional interventions such as a Foley balloon (n=37 [35%]), reaspiration (n=22 [21%]), hospitalization (n=16 [15%]), transfusion (n=8 [7%]), or uterine artery embolization (n=5 [5%]). Multivariable regression analysis revealed that increased patient age (OR 1.05, 95% CI 1.01–1.08) and a gestational age at or beyond 20 weeks (OR 4.5, 95% CI 2.4–9.4) were each independent predictors of uterine atony. “Other” complications included concern for perforation or retained products resulting in an additional procedure with normal findings (n=5), nongynecologic issues requiring consultation (hematemesis [n=1], tachycardia [n=1], chest pain [n=1], back spasms [n=1]), allergy to laminaria (n=1), laryngospasm interrupting the abortion (n=1), coughing necessitating general anesthesia (n=1), and follow-up at an outside hospital for pain or bleeding with no additional information (n=4).
Within this ethnically diverse group of women seeking second-trimester abortions, a major complication was encountered in 1.3% of cases. A history of more than one cesarean delivery was associated with the greatest risk of major complication in this cohort, whereas advanced gestational age and insufficient initial cervical preparation requiring additional dilation also were associated with a significantly increased risk of major complication. Women with a history of only one cesarean delivery were not found to have an increased risk of a major complication, although it is possible this would be observed among a larger cohort.
Cervical laceration was the most common complication in our study population, with risk increasing dramatically with gestational duration. This incidence is comparable to a recent publication in which the observed rate of cervical lacerations was also 3.3%.16 The long-term clinical implications of cervical lacerations in this context have not been investigated.
One of the major strengths of this study is the cohort size and large number of D&Es performed beyond 20 weeks of gestation. The data also reflect contemporary techniques for the second trimester. A weakness of the study is the omission of body mass index data, which were not available for this cohort. Obesity can increase the technical difficulty of a D&E and thus may put the patient at increased risk of complications. In addition, the complication rates noted here may have limited generalizability, because the Women's Options Center is a referral center for a high-risk population with complicated medical and obstetric issues. It also is located in a training hospital with 24-hour direct access to anesthesia, operating suites, and interventional radiology that may not be readily available in other outpatient settings. Finally, the data are limited to short-term complications and do not include delayed complications such as infection. We are aware of 10 women who were evaluated at outside institutions for delayed complications. However, given our clinic's extensive geographic catchment area, there likely are others who experienced a complication but did not have further communication with our clinic.
The findings presented here suggest that second-trimester abortion procedures are associated with an approximately 1% risk of major complication. However, a history of more than one cesarean delivery, advanced gestational duration, and insufficient initial cervical dilation significantly increase this risk. Thus, clinicians should consider cesarean delivery a risk factor for future pregnancy complications, even when termination is chosen. In addition, researchers should continue to work to identify techniques that decrease the risk of complications associated with second-trimester abortion.
1. Altman AM, Stubblefield PG, Schlam JF, Loberfeld R, Osathanondh R. Midtrimester abortion with laminaria and vacuum evacuation on a teaching service. J Reprod Med 1985;30:601–6.
2. Woodward G. Intraoperative blood loss in midtrimester dilatation and extraction. Obstet Gynecol 1983;62:69–72.
3. Schneider D, Halperin R, Langer R, Caspi E, Bukovsky I. Abortion at 18–22 weeks by laminaria dilation and evacuation. Obstet Gynecol 1996;88:412–4.
4. Jacot FR, Poulin C, Bilodeau AP, Morin M, Moreau S, Gendron F, et al. A five-year experience with second-trimester induced abortions: no increase in complication rate as compared to the first trimester. Am J Obstet Gynecol 1993;168:633–7.
5. Chasen ST, Kalish RB, Gupta M, Kaufman JE, Rashbaum WK, Chervenak FA. Dilation and evacuation at > or =20 weeks: comparison of operative techniques. Am J Obstet Gynecol 2004;190:1180–3.
6. Autry AM, Hayes EC, Jacobson GF, Kirby RS. A comparison of medical induction and dilation and evacuation for second-trimester abortion. Am J Obstet Gynecol 2002;187:393–7.
7. Grimes DA, Schulz KF, Cates W Jr, Tyler CW Jr. Mid-trimester abortion by dilatation and evacuation: a safe and practical alternative. N Engl J Med 1977;296:1141–5.
8. Peterson WF, Berry FN, Grace MR, Gulbranson CL. Second-trimester abortion by dilatation and evacuation: an analysis of 11,747 cases. Obstet Gynecol 1983;62:185–90.
9. Bartlett LA, Berg CJ, Shulman HB, Zane SB, Green CA, Whitehead S, et al. Risk factors for legal induced abortion-related mortality in the United States. Obstet Gynecol 2004;103:729–37.
10. Hamilton BE, Martin JA, Ventura SJ. Births: preliminary data for 2007. Natl Vital Stat Rep 2009;57:1–23.
11. Wu S, Kocherginsky M, Hibbard JU. Abnormal placentation: twenty-year analysis. Am J Obstet Gynecol 2005;192:1458–61.
12. Smith JG, Mertz HL, Merrill DC. Identifying risk factors for uterine rupture. Clin Perinatol 2008;35:85–99, viii.
13. Lu MC, Korst LM, Fridman M, Muthengi E, Gregory KD. Identifying women most likely to benefit from prevention strategies for postpartum hemorrhage. J Perinatol 2009;29:422–7.
14. Lalitkumar S, Bygdeman M, Gemzell-Danielsson K. Mid-trimester induced abortion: a review. Hum Reprod Update 2007;13:37–52.
15. Schneider D, Bukovsky I, Caspi E. Safety of midtrimester pregnancy termination by laminaria and evacuation in patients with previous cesarean section. Am J Obstet Gynecol 1994;171:554–7.
16. Ben-Ami I, Schneider D, Svirsky R, Smorgick N, Pansky M, Halperin R. Safety of late second-trimester pregnancy termination by laminaria dilatation and evacuation in patients with previous multiple cesarean sections. Am J Obstet Gynecol 2009;201:154.e1–5.