Complication rate after termination of pregnancy for fetal defects

To assess the risk of complications in women undergoing termination of pregnancy (TOP) for fetal defects and to examine the impact of gestational age on the complication rate.

17.4 weeks, respectively. In the first, second and third trimesters, respectively, 84 (20.2%), 278 (66.8%) and 54 (13.0%) pregnancies were terminated, for which D&C or D&E was used in 80 (95.2%), 21 (7.6%) and 0 (0.0%) cases. Seventy-seven (18.5%) women had at least one previous Cesarean section and 169 (40.6%) had at least one previous spontaneous delivery. Overall,95 (22.8%) women had complications during or after TOP. A significantly higher complication rate was noted for terminations performed later in pregnancy. The median gestational age at termination was 16.6 weeks in women who did not experience complications and 20.7 weeks in those with complications (P < 0.001). The respective complication rates in the first, second and third trimesters were 6.0%, 27.0% and 27.8%.

INTRODUCTION
One of the benefits of first-trimester screening is early detection of fetal defects [1][2][3][4] . The meta-analysis by Karim et al. 5 examined the detection rate of first-trimester screening for structural defects. The detection rate for major fetal defects was 46% in low-risk pregnancies and 61% in the high-risk population. Similar results were found for first-trimester screening for cardiac defects 6 .
The increasing availability of cell-free DNA (cfDNA) screening for trisomy 21 has introduced alternative pregnancy care models that challenge the need for a detailed first-trimester scan. The most discussed concept combines cfDNA analysis at or even before 10 weeks' Complication rate after termination of pregnancy 89 gestation and only a basic first-trimester ultrasound examination 3 . This scan is confined to measurement of crown-rump length, documentation of viability, and determination of chorionicity in multiple gestations. Consequently, an early anatomical assessment is bypassed in favor of a detailed second-trimester anomaly scan. Most healthcare providers that offer such an approach argue that the cost of two anomaly scans per pregnancy is too high and that major defects will be identified in the second trimester anyway. By their logic, the gestational age at the time of diagnosis of the fetal defect plays a minor role, as all management options, including, in many countries, termination of pregnancy (TOP), are available throughout the whole course of the pregnancy.
The validity of this approach could be undermined if the risk of complications associated with TOP is shown to increase with gestational age. Unfortunately, there are only a few studies available that focus on the morbidity and mortality associated with terminations. Bartlett et al. 7 examined the mortality rate of legal terminations in the USA between 1988 and 1997. The authors reported an overall maternal death rate of 0.7 per 100 000 terminations. The risk of death increased exponentially by 38% for each additional week of gestation. Compared with women whose pregnancies were terminated at or before 8 weeks of gestation, women whose terminations were performed in the second trimester were significantly more likely to die of TOP-related causes (relative risk was 14.7 at 13-15 weeks and 76.6 at or after 21 weeks). Bryant et al. 8 examined the morbidity associated with terminations performed between 13 and 24 weeks of gestation either by induction of labor or dilatation and evacuation (D&E). They reported that 24% of women who underwent induction and 3% of those who underwent D&E experienced at least one complication, including: fever requiring antibiotics; injury to the uterus or cervix warranting surgical repair; admission or readmission to hospital; retained tissue requiring dilatation and curettage (D&C) or manual removal of the placenta; failure to terminate by the primary method; or emergency department visit after the procedure.
In many countries, such as Germany, TOP in the first trimester and the beginning of the second trimester is usually performed using either D&C or D&E. Later in pregnancy, induction of labor is employed more commonly. The aims of this study were to determine the overall risk of complications associated with TOP and to establish whether the rate of procedure-related complications changes with gestational age.

METHODS
This was a retrospective study of singleton pregnancies that were terminated at the Department of Women's Health, University Hospital of Tübingen, Tübingen, Germany, between January 2018 and December 2021. The University Hospital is a tertiary referral center with special expertise in prenatal medicine and a catchment area of about 11 million people. This study was approved by the ethical committee of the University Hospital of Tübingen (440/2022BO2).
In Germany, standard pregnancy care involves three ultrasound examinations performed at around 10, 20 and 30 weeks' gestation. Screening for major fetal defects is generally carried out by local obstetricians. In the case of a suspected anomaly, the pregnant woman is referred to a fetal medicine expert for further management. Most anomalies are detected during the second-trimester anomaly scan. A detailed first-trimester anomaly scan is not part of standard care, but is available on a private basis.
All TOP procedures were performed by experienced operators in women requesting termination due to the presence of fetal abnormalities (German criminal law §218a, second paragraph 9 ). In Germany, TOP can be carried out at any stage of the pregnancy until the onset of labor. The method employed for the procedure (D&C, D&E or induction of labor) was selected primarily based on gestational age.
The standard protocol for induction of labor includes administration of 200 μg mifepristone, followed by 400 μg misoprostol (200 μg vaginally and 200 μg orally, at the same time) 24 h later [10][11][12][13][14] . In the third trimester, the dose of misoprostol is reduced to 200 μg. Administration of misoprostol is repeated every 4-6 h until adequate contractions are produced. If there is a history of Cesarean section, 200 μg of mifepristone is still administered initially, but the induction is carried out with intravaginal dinoprostone (maximum 3 mg/day). In the few cases in which adequate contractions do not develop after 2 days of induction, the dose is increased or a cervical-ripening balloon is used (prolonged induction). In cases in which postmortem examination is indicated, induction is attempted irrespective of gestational age. If a liveborn neonate is expected, predominantly in pregnancies after 22 weeks' gestation, feticide using intracardiac potassium chloride is performed prior to induction. Blood loss of more than 500 mL (assessed by the indirect weighing method, involving the weighing and measurement of blood and blood-soaked materials following the cessation of bleeding 15,16 ) is treated according to national guidelines 17 . If D&C or D&E is planned, 400 μg of misoprostol is administered intravaginally 4 h prior to the procedure. All patients are required to consent to the off-label use of induction medication.
In each case, maternal and pregnancy characteristics, the method of termination, the induction protocol and the time interval between first application of misoprostol and delivery were recorded in our database. Furthermore, we noted in the database if mifepristone was given or if feticide was performed.
We considered the following to be significant procedure-related complications: blood loss of more than 500 mL; uterine perforation; need for blood transfusion; allergic reaction; creation of a false passage (via falsa); systemic infection; readmission to hospital; any unplanned surgical procedure, such as repeat D&C/D&E or hysterectomy; and maternal death 8,18,19 . In order 90 Spingler et al.
to be included as a complication, a systemic infection had to meet the Global Alignment of Immunization Safety in Pregnancy (GAIA) group's diagnostic criteria for infection following incomplete or complete abortion (Level 3B: Report of fever and evidence of uterine infection based on clinical exam) 20 . To be consistent with the World Health Organization (WHO) definition of maternal death, we included only those complications that occurred within 42 days after the termination. All complications were recorded but, in cases in which more than one complication occurred, for the purpose of our analysis, the patient was counted only once.

Statistical analysis
The cohort was stratified according to the presence or absence of complications. Data are given as median (interquartile range) or n (%). Continuous maternal and pregnancy characteristics were compared between groups using Student's t-test or the Mann-Whitney U-test, according to the distribution of the variable, and frequencies were compared using the chi-square test. We used univariate and multivariate logistic regression analysis to determine significant covariates that were associated with the presence or absence of complications. The Clopper-Pearson method was used to compute 95% CI. The significance level was set at a P-value of 0.05.

RESULTS
We identified 416 pregnancies in our hospital database that met the inclusion criteria. Characteristics of the study population stratified according to the presence or absence of complications are detailed in Table 1. Median maternal and gestational age at termination were 34.1 years and 17.4 weeks, respectively. Of the included pregnancies, 84 (20.2%), 278 (66.8%) and 54 (13.0%) were terminated in the first, second and third trimesters, respectively. D&C/D&E and induction were performed, respectively, in 80 (95.2%) and 4 (4.8%) cases in the first trimester, 21 (7.6%) and 257 (92.4%) cases in the second trimester, and 0 (0.0%) and 54 (100%) cases in the third trimester. Seventy-seven (18.5%) women had at least one previous Cesarean section and 169 (40.6%) had at least one previous spontaneous delivery.
A total of 95 (22.8%) women suffered complications during or after termination, of whom six underwent D&C/D&E (6/101 (5.9%)) and 89 underwent induction (89/315 (28.3%)). Observed complications are summarized in Table 2. The majority of cases (76/95) had blood loss of more than 500 mL. The next most common complication, occurring in 16 women, was a clinically significant maternal infection. An emergency hysterectomy was performed in two women whose pregnancies were terminated, respectively, at 16 and 21 weeks' gestation.
We also examined separately whether increasing gestational age affected the rate of complications depending on the method of termination employed (D&C/D&E or   Spingler et al.

DISCUSSION
This study has shown that the complication rate in women undergoing TOP for fetal defects increases with gestational age. The complication rates in the first, second and third trimesters were 6%, 27% and 28%, respectively. Overall, we observed at least one complication in 23% of cases.
Our findings are in agreement with the published literature. Several studies have highlighted the low complication rate of TOP in the first trimester. In a systematic review, White et al. 21 found a low rate of uterine perforation (0.1%), cervical injury (0.1-0.6%) and retained pregnancy tissue (0.7-3.0%) after first-trimester surgical termination. Cleland et al. 22 reviewed the outcome of 233 000 first-trimester medical terminations in the USA between 2009 and 2010 and reported a similarly low complication rate. In a Cochrane review, Say et al. 23 compared surgical and medical approaches in the first trimester. The study showed that first-trimester vacuum aspiration was more effective, shorter in duration and resulted in less bleeding and pain compared to medical termination with prostaglandin. Costescu and Mui 11 focused on first-and second-trimester surgical termination and the respective complication rates. They reported an increase in the frequency of complications with advancing gestational age. This is consistent with the findings of Lederle et al. 24 . In contrast, we did not find an association between gestational age and complications in either group.
Bryant et al. 8 compared the safety and effectiveness of labor induction and D&E in the second trimester. The complication rate for induction of labor between 13 and 24 weeks of gestation was 24%, which is consistent with the findings of this study. In the first trimester, they observed complications in only 3% of cases that underwent D&E, which is also in agreement with this study. A review of 46 000 patients who underwent medical termination up to 26 weeks of gestation found that the rate of infection ranged from 0% to 6% 19 . This in line with the present series, in which the infection rate was 4%. Mark et al. 18 examined whether body mass index (BMI) influences the complication rate after surgical TOP. They showed that the complication rate was higher in women with Class-3 obesity (BMI > 40 kg/m 2 ) compared to that in women with a lower BMI. In our data, this correlation was not found.
There are several arguments in favor of a detailed ultrasound examination at the end of the first trimester. First, roughly 2-3% of all fetuses have a major defect and the overall detection rate throughout the whole course of pregnancy is only about 40% 25 . In contrast, several meta-analyses have shown that the detection rate for major defects in the first trimester is about 40-60% 5,6 . Therefore, inclusion of the first-trimester anomaly scan may increase the overall detection of fetal defects. Second, some anatomical structures are easier to assess in the first trimester compared with later in pregnancy. Examples include evaluation of the abdominal wall and examination of the fetal profile for the presence of retrognathia. This is particularly true in women who are obese; ultrasound examination in the first trimester in these cases can be performed transvaginally, resulting in better visualization of fetal structures compared with later in pregnancy. Third, in cases with an inconclusive cfDNA test result for trisomy, the combined test at 11-13 weeks' gestation can be used to assess the risk of trisomy, rather than invasive testing, which is generally recommended in these cases. It should be stressed that at this gestational age, the nuchal translucency may be measured. It is well established that nuchal translucency thickening is associated with not only fetal aneuploidy, but also a number of fetal structural defects and genetic and non-genetic syndromes. Fourth, if a fetal anomaly is observed, intensive genetic follow-up is often necessary. These tests require up to 4-6 weeks to yield results. If the defect is found in the first trimester, follow-up examinations can be conducted in a timely fashion and the final result is usually available before 20 weeks' gestation. However, if an anomaly is detected in the second trimester, the final result of the genetic evaluation may not be available until the third trimester. In several countries, TOP is not available to women at this late gestational age. Fifth, most pregnant women prefer to find out about potential fetal anomalies as early in pregnancy as possible 26 . Sixth, research has shown that the frequency of psychological distress following termination increases with gestational age 27 . Finally, even though this study was not designed to compare procedure-related costs, it is likely that using induction for TOP, and accommodating for the higher number of associated complications, is costlier than using D&C or D&E.
The results of this study add an additional benefit of detailed ultrasound examination in the first trimester: the complication rate of TOP increases with advancing gestational age. As such, for the sake of the safety of the mother, TOP for fetal defects should be performed as early in pregnancy as possible, preferably in the first or early second trimester. This can be accomplished only if careful fetal evaluation is performed in the first trimester.
We acknowledge that this study has some limitations. First, this is a retrospective study and second, we report on the experience of a single center. However, our management is in line with national guidelines and our cohort is sufficiently large to draw relevant conclusions.
In conclusion, this study highlights that the risk of complications in women undergoing TOP for fetal defects increases with increasing gestational age. This finding adds to the mounting evidence that an early fetal anomaly scan is an essential part of pregnancy management. It not only increases the amount of time available for detailed follow-up examination and carefully thought-out decision-making, but also allows for termination to be performed at the safest point in pregnancy.