- Email: [email protected]
Immediate placement of intrauterine devices after first and second trimester pregnancy termination
Contraception 83 (2011) 34 – 40
Original research article
Immediate placement of intrauterine devices after first and second trimester pregnancy termination☆ Michelle C. Foxa,⁎, Julia Oat-Judgea , Kathryn Seversona , Roxanne M. Jamshidia , Rameet H. Singhb , Raegan McDonald-Mosleya , Anne E. Burkea a
Department of Gynecology and Obstetrics, Johns Hopkins School of Medicine, Baltimore, MD 21287, USA b Department of Obstetrics and Gynecology, University of New Mexico, Albuquerque, NM 87131, USA Received 17 January 2010; revised 29 June 2010; accepted 30 June 2010
Abstract Background: We reviewed our experience with intrauterine device (IUD) placement after surgical abortion up to 20 weeks' gestation. Study Design: Women presenting for elective abortion between January 2004 and March 2009 who requested an IUD were included in this retrospective review. Results: Of 308 women requesting postabortion IUD placement, 221 (72%) planned insertion at the time of abortion (immediate group) and 87 (28%) planned insertion at their postoperative visit (interval group). IUDs were placed in 96% of the immediate group and in 23% of the interval group (212/221 vs. 20/87; pb.0001). Failure to return for placement was the most common reason for noninsertion in the interval group (60/87=69%). Follow-up information was obtained for 56% of patients and was documented a median of 137 days postabortion (range 3–1594 days). There was no difference in complication rates between groups. Expulsion rates were 3% and 0% in the immediate and interval groups, respectively (6/212 vs. 0/20; p=.4). Considering only those with documented follow-up after immediate insertion (119), there was a nonsignificant trend towards increased expulsion with placement after second vs. first trimester abortion (4/54=7% vs. 2/65=2%; p=.3). When analyzing the 172 subjects with documented follow-up, those planning immediate insertion were more likely to have an IUD in situ at the last contact than those planning later insertion (84/124=68% vs. 20/ 48=42%; p=.002). Conclusion: Immediate postabortion IUD insertion is safe and effective. Given the low rate of return for interval insertion, immediate placement may be preferable. © 2011 Elsevier Inc. All rights reserved. Keywords: Postabortion contraception; Postabortal; Intrauterine contraception; Intrauterine device; IUD; Immediate insertion
1. Introduction Almost half of the 1.2 million abortions performed annually in the United States (US) occur in women who previously have terminated a pregnancy [1]. Long-acting reversible contraceptive methods have the potential to decrease future unplanned pregnancies and repeat abor-
☆
The primary author received compensation from Teva Pharmaceuticals from 2007 to 2009 as a member of its speakers bureau for training clinicians to insert ParaGard® IUDs. No financial support was obtained for this study. ⁎ Corresponding author. Department of Obstetrics and Gynecology, Johns Hopkins Bayview Medical Center, Baltimore, MD 21224, USA. Tel.: +1 410 550 0335; fax: +1 410 550 0245. E-mail address: [email protected] (M.C. Fox). 0010-7824/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.contraception.2010.06.018
tions in women presenting for pregnancy termination. While the intrauterine device (IUD) is a highly effective long-term option, its use for immediate postabortion contraception, especially after second trimester terminations, has been limited. Although IUD insertion is often delayed 2–8 weeks after an abortion (interval insertion), insertion immediately at the time of elective pregnancy termination (immediate insertion) has several advantages [2]. The cervix is already dilated; thus IUD placement may be easier and less painful. Since followup rates after elective abortion are commonly low, immediate IUD insertion assures that the patient has effective contraception without an additional visit [2–4]. Also, postabortal insertion has been associated with a decrease in the rate of repeat abortion [5,6].
M.C. Fox et al. / Contraception 83 (2011) 34–40
Perceived disadvantages of immediate insertion include increased risk of infection, uterine perforation and device expulsion; however, the literature demonstrates that infection and uterine perforation rates are not increased by immediate insertion [2]. Expulsion risk may be theoretically increased secondary to cervical dilation at the time of surgery and subsequent uterine involution. Published expulsion rates of T-shaped IUDs inserted immediately after first trimester abortion vary from 1% to 15% [6–9]. There is minimal data comparing immediate to interval IUD insertion, especially after second trimester dilation and evacuation. There has been only one published randomized trial comparing immediate to interval insertion after suction abortion, and it found a statistically nonsignificant increase in expulsion rate from 3% vs. 15% following interval vs. immediate placement (p=.2). Limited data suggest that IUD placement immediately after second trimester abortion has a higher rate of expulsion than after first trimester procedures. A 1979′s World Health Organization (WHO) trial of immediate postabortal insertion found that the expulsion rate for a copper T-shaped IUD at 120 days postinsertion was 1.9% following first trimester abortion and 19.5% following second trimester procedures [10]. Goodman et al. [6] reported expulsion rates of 1.6% and 7% for IUDs placed immediately after first and second trimester abortions, respectively (p=.02). A recent cohort series published by Drey et al. [8] found a nonsignificant increase in expulsion rates with immediate placement of IUDs after second vs. first trimester abortions (3% vs. 0.8%). We have been placing IUDs immediately after first and second trimester surgical abortions at our institution since 2004. The purpose of this study was to examine the risks and benefits of immediate vs. interval IUD insertion after elective abortion, particularly after second trimester procedures.
2. Materials and methods The primary objective of this study was to compare IUD insertion rates associated with placement immediately after elective surgical abortion (immediate group) vs. delayed insertion 2–4 weeks later at the time of surgical follow-up (interval group). Secondary objectives were to compare expulsion rates in the immediate group after first and second trimester abortion and to assess continuation, removal and complication rates in both groups. Our clinic's practice of postabortion IUD insertion is described below. The copper-T380A (ParaGard® , Teva Pharmaceuticals, Pomona, NY, USA) and the levonorgestrel intrauterine system (Mirena®, Bayer Healthcare Pharmaceuticals, Inc., Wayne, NJ, USA) are available. Patients desiring an IUD are offered immediate and interval insertion. Immediate insertion is recommended unless the patient is unsure of her contraceptive choice, there are signs of cervicitis or if insurance plan preauthorization is required and cannot be attained prior to the procedure. Timing of insertion is determined by the patient after counseling with
35
the above considerations. Abortions are performed under intravenous conscious sedation and local anesthesia in our hospital-based clinic. The use of misoprostol with and without osmotic dilators for cervical priming varies by clinician and gestational age. Routine cervical screening for gonorrhea and chlamydia by nucleic acid amplification is performed at the preoperative visit unless results from recent testing are available. Since the preoperative visit is scheduled the day prior to the abortion, results from cervical screening performed at that visit are commonly unavailable at the time of surgery. Patients are prescribed doxycycline 100 mg twice daily for 3–7 days according to clinician discretion for procedure prophylaxis. They are subsequently treated with doxycycline, azithromycin and/or ceftriaxone according to Centers for Disease Control guidelines if cervical screening results are positive. IUDs are not removed for positive culture results in asymptomatic patients. If the insertion device is not long enough to reach the fundus at the time of placement, additional instruments (most commonly sponge forceps) are used to assure fundal placement of the IUD using techniques described elsewhere [8]. The use of ultrasound guidance varies amongst our clinicians in the first trimester but is standard for immediate IUD placement following second trimester abortion. Postsurgical follow-up is scheduled in 2–4 weeks. Those in the interval group are asked to return to have their IUD checked 4–6 weeks after insertion. No further routine follow-up is scheduled; however, patients are encouraged to return for any problems. Women presenting for elective abortion between January 2004 and March of 2009 who desired an IUD and intended to follow up within our medical system were included in this retrospective record review. Women with fetal demise or fetal anomalies were excluded since most did not desire long-term contraception and planned to follow up with their primary obstetrician. Records were reviewed in the Spring of 2009, 3 to 60 months postoperatively. Subjects were identified from the procedure ledger in our clinic. All available electronic records dating back to the time of the abortion were reviewed. Gynecology clinic charts for all subjects were reviewed for verification and to obtain missing data. This review was approved by the Institutional Review Board of the Johns Hopkins University School of Medicine. For the purpose of this retrospective analysis, postoperative infection was defined by the prescription of additional broad-spectrum antibiotics. Retained products of conception (RPOC) were defined by clinical suspicion and treatment with misoprostol or reaspiration. Second trimester abortion was defined as those performed at greater than or equal to 14 weeks' gestation. Data were abstracted onto coded forms and were entered into Microsoft Access and Excel files, and were then analyzed using Stata 10 (College Station, TX, USA). Statistical tests included Fisher's Exact Test for categorical data, Student's t test for continuous data and nonparametric statistics where appropriate. A two-sided p value of .05 was considered statistically significant.
36
M.C. Fox et al. / Contraception 83 (2011) 34–40
3. Results Of the 964 women presenting for elective abortion during the study time period, 308 (32%) elected IUD for contraception and were eligible for inclusion. Of these, 221 were in the immediate group and 87 were in the interval group. Those with insurance plans requiring prior authorization for IUD placement that could not be obtained prior to the abortion and those who were undecided regarding IUD placement, and thus did not want immediate insertion, comprised the interval group. No patients desiring immediate insertion had suspected cervicitis. The demographics of women included in this review are presented in Table 1. Most subjects were African American, single and enrolled in a medical assistance program. Over half previously had undergone an abortion and also had at least one child. Those in the immediate group were younger (24 vs. 26.5; p=.002), more likely to have medical assistance (155/221=70% vs. 36/87=41%; pb.0001) and more likely to have planned a copper than a levonorgestrel IUD (72/221=33% vs. 15/ 87=17%; pb.0001) than those in the interval group. Two hundred thirty-two IUDs were inserted, of which approximately two-thirds were levonorgestrel IUDs (157/232). Over half of IUDs placed immediately (111/212) were inserted after a second trimester abortion. Table 1 Subject demographics according to planned timing of IUD placement Variable
Immediate (n=221)
Interval (n=87)
Age, years (mean±SD) Race White Black Asian Hispanic Unknown Other Gravidity, median (range) Parity, median (range) Prior terminations, median (range) Gestational age, weeks (mean±SD) 5–13 6/7 weeks 14–20 weeks IUD type Levonorgestrel Copper-T380A Unknown Insurance type Private plan Medical assistance Uninsured Unknown Marital status Single Married Separated Divorced Other Unknown
24.0±6.2
26.5±7.2
14.0% 83.3% 0.5% 0.5% 1.4% 0.5% 3 (1–15) 1 (0–7) 1 (0–7) 13.9±3.9 47.5% 52.5%
20.7% 74.7% 2.3% 2.3% 0 0 4 (1–12) 1 (0–6) 1 (0–5) 14.1±4.0 42.5% 57.5%
66.5% 32.6% 1%
75.9% 17.2% 6.9%
26.2% 70.1% 2.7% 1%
56.3% 41.4% 1.2% 1%
86.6% 6.9% 1.8% 0.5% 0.5% 3.7%
75% 10% 2.5% 2.5% 0 10%
p=.002 p=.1
p=.06 p=.4 p=.5 p=.8 p=.4 pb.0001
pb.0001
p=.1
Overall, follow-up information was available for 172 subjects, 124 in the immediate group and 48 in the interval group (124/221=56% vs. 48/87=55%; p=.9). Follow-up information was documented a median of 137 days postabortion (range 3–1594 days). Only 117 subjects (38%) attended their scheduled follow-up visit 2–4 weeks postoperatively; data were obtained from the electronic record for an additional 55 subjects (18%). Those who had a prior abortion, had medical assistance or underwent a second trimester termination were less likely to attend their followup visit in our clinic. However, when all follow-up data from electronic record review were included, the only statistically significant difference between those lost to follow-up and those with documented outcome data was gestational age. Follow-up data were available for 63% of those who had first trimester abortions vs. 50% of those who had second trimester abortions (89/142 vs. 83/166; p=.03). Despite counseling regarding the possibility of postprocedural IUD expulsion, only 40% (85/212) returned to clinic for postoperative follow-up after immediate insertion. This percentage was the same as those returning for string checks after interval insertion (8/20=40%). IUD use for postabortion contraception has increased significantly in our clinic since initiating this service. The percentage of patients selecting IUDs for postabortion contraception increased annually from 12% in 2004 to 62% in the first quarter of 2009 (20/169 vs. 48/77; pb.00001). The percentage of planned IUDs that were actually placed increased annually from 55% in 2004 to 92% in 2009 (p=.008). Additionally, the proportion of subjects planning immediate IUD insertion increased from 55% in 2004 to 90% in 2009 (p=.004). Figs. 1 and 2 demonstrate the outcomes of subjects in both groups. Ninety-six percent (212/221) of the immediate group vs. 23% (20/87) of the interval group actually had an IUD placed (pb.0001). Nine of the planned immediate insertions were deferred for the following reasons: intraoperative bleeding (3), subject changed her mind (2), planned surgery for excision of cervical dysplasia (1), failure to obtain insurance approval (1) and unknown (2). Interval placement did not occur for any of these nine, largely due to failure to return for insertion. In the interval group, the reasons IUD placement was not performed include failure to return for IUD insertion (60), no longer desired an IUD (4) and miscellaneous medical concerns (3). There was one failed insertion in the interval group which was successful at a later visit. There were no failed insertions in the immediate group. There were six documented IUD expulsions in the immediate group and none in the interval group (6/ 212=2.8% vs. 0/20=0%, p=.4). When data analysis is limited to subjects with documented follow-up, the expulsion rate in the immediate group increases to 5%; however, this is not statistically higher than in the interval group (6/119=5% vs. 0/ 20=0%, p=.3). Expulsion was reported 2, 91, 139 and 286 days after insertion and occurred at an unknown time for the remaining two subjects. Considering only those with docu-
M.C. Fox et al. / Contraception 83 (2011) 34–40
Fig. 1. Immediate group outcome.
Fig. 2. Interval group outcome.
37
38
M.C. Fox et al. / Contraception 83 (2011) 34–40
mented follow-up after immediate IUD insertion (119), there was a nonsignificant increase in expulsion rate following immediate placement after first vs. second trimester abortion from 3.1% to 7.4%, respectively (2/65 vs. 4/54; p=.3). Twenty-nine early removals were documented in the immediate group vs. none in the interval group (29/ 212=14% vs. 0/20=0%; p=.08). Two IUDs were removed during reaspiration procedures within the first week, and two were removed in the first month for suspected endometritis or pelvic inflammatory disease. An additional 10 early removals occurred in the first year after placement. Twelve were removed after a year or more. The dates of removal for three subjects were unknown. The most common reasons for removal remote from insertion were pain and irregular bleeding. There was no difference in early removal rates between subjects in their first or second trimester at the time of termination (15/114=13% vs. 14/118=12%; p=.8). Those receiving the copper-IUD were almost twice as likely to have it removed early compared to the levonorgestrel IUD (14/ 75=19% vs. 15/157=10%; p=.05). As shown in Figs. 1 and 2, those in the immediate group were significantly more likely to have an IUD at the last
documented follow-up at the time of the chart review. Including only the 172 subjects with documented followup, 68% of the immediate group vs. 42% of the interval group had an IUD in situ at their last known visit (84/124 vs. 20/48; p=.002). Three-fourths (157/212) of IUDs in the immediate group were placed without available cervical culture results, although prophylactic doxycyline was given as previously described. Twenty-two subjects planning IUD insertion had asymptomatic gonorrhea or chlamydia infections at the time of their abortion, representing 5% of the immediate group and 11% of the interval group (12/221 vs. 10/87; p=.06). Eleven had their IUD placed immediately at the time of their abortion in the presence of unknown gonorrhea or chlamydia infection. One woman had her IUD removed 3 weeks later for mild endometritis in the setting of an untreated chlamydia infection. This subject received 1 week of oral antibiotic treatment and had no further known complications. No cases of pelvic inflammatory disease occurred; however, four of the 11 were lost to follow-up after IUD insertion in the setting of undiagnosed chlamydia despite extensive efforts to contact them. The remaining 10 women with asymptomatic infection
Table 2 Postoperative complications: details of subjects presenting with complications within 6 weeks of their abortion Age, G/P
Estimated gestational Group age (weeks)
IUD placed IUD removed Narrative
17 G2 P0010 8 6/7
Immediate Y
Y
23 G3 P2002 11 6/7
Immediate Y
N
22 G4 P3003 14 1/7
Immediate Y
Expelled
15 G1 P0000 7 6/7
Immediate Y
Y
24 G3 P1011 13 2/7
Immediate Y
Y
40 G1 P0000 8 6/7
Immediate Y
Y
29 G6 P1132 14 1/7
Interval
N
N/A
34 G5 P1031 14 6/7
Interval
N
N/A
29 G4 P3003 17 3/7
Interval
Y
Unknown
Undiagnosed chlamydia at time of abortion. Did not comply with prescribed treatment. Presented on POD 26 with mild endometritis. Treated with 7 days doxycycline after IUD removal. Presented POD 8 with pain and cramping. Treated with 14 days PO antibiotics for presumed PID. Symptoms resolved. IUD expelled on POD 2 and replaced on POD 6. Seen in ER on POD 16 for bleeding. Sonogram showed IUD in place and possible RPOC. Misoprostol prescribed in ER. IUD in place on POD 22. Expulsed second IUD on POD 30. Switched to DMPA for contraception. Normal postoperative visit and string check on POD 13. Presented POD 25 with PID. IUD removed and PO antibiotics prescribed. Admitted same day to outside hospital due to worsening pain, vomiting and inability to tolerate PO antibiotics. She received IV antibiotics and was subsequently LTFU. Presented POD 5 with heavy bleeding and pain. Sonogram showed RPOC. IUD removed and reaspiration performed without complications. IUD replaced 2 months later. Presented POD 6 with cramping and heavy bleeding. Sonogram showed RPOC and hematometra. IUD removed and reaspiration performed. Diagnosed with endometritis on POD 20. Treated with 2 weeks of PO antibiotics. IUD replacement deferred initially due to infection and then LTFU. Multiple visits elsewhere for pain beginning 3 weeks after abortion. Imaging inconclusive for RPOC. Returned to clinic for persistent pain (10 weeks after abortion) and was treated with PO antibiotics for 2 weeks for suspected infection with resolution in pain. Reaspiration on POD 23 for hematometra and RPOC. Hospitalized overnight. Subsequently LTFU. Admitted to outside hospital on POD 4 with fevers and RPOC and had reaspiration. Treated with IV antibiotics and discharged with 2-week course of PO antibiotics. Interval IUD insertion on POD 20. Subsequently LTFU.
G: Gravidity; P: parity; Immediate: immediate IUD insertion; Interval: interval IUD insertion; Y: yes; N: no; RPOC: retained products of conception; POD: postoperative day; PO: oral; IV: intravenous; ER: emergency room; PID: pelvic inflammatory disease; N/A: not applicable; LTFU: lost to follow-up.
M.C. Fox et al. / Contraception 83 (2011) 34–40
were in the interval group. Only two of these women returned for their postoperative visit and neither received an IUD. Overall, nine subjects (3%) had complications within a month of their abortion, six of whom had IUDs inserted immediately. Details of these complications are reviewed in Table 2. No uterine perforations were documented. There was no statistically significant difference in infection (1.8% vs. 2.3%; p=.8), RPOC (1.4% vs. 2.3%; p=.6) or need for reaspiration (0.9% vs. 2.3%; p=.4) between the immediate and interval groups, respectively. The IUD was removed in four cases: two who required reaspiration and two with infection. Immediate postabortion IUD insertion was associated with decreased chance of documented subsequent pregnancy during the follow-up period compared to interval insertion (16/221=7% vs. 19/87=22%; p=.0002). Of the subsequent pregnancies in the immediate group, two occurred in women who did not ever have an IUD inserted with the rest occurring after IUD expulsion or early removal. All of the subsequent pregnancies in the interval group occurred in those who never received an IUD. 4. Discussion This retrospective study adds to the growing evidence supporting immediate postabortion IUD insertion and contributes to the limited information on placement at the time of second trimester abortion. Our results agree with previous studies supporting the use of IUDs immediately after elective pregnancy termination including those procedures performed in the second trimester. This study, unlike the prior report of Drey et al. [8], compares outcomes of interval to immediate insertion. Studies show short-term removal, infection and complication rates after immediate postabortion IUD insertion are not significantly increased when compared to interval insertion a few weeks later [5,6]. In our study, only four of the 212 IUDs placed at the time of abortion were removed within the first month for complications and only one known expulsion occurred during this time period. Two IUDs were removed secondary to infection; however, these removals may not have been clinically necessary [11]. Further education regarding the potential for treating pelvic infection with an IUD in situ is needed for our clinicians. Complications of abortion are low; thus, our study is underpowered to detect a difference in these rates. The low complication rates associated with immediate postabortal insertion noted in more recent literature [8] and in this study may partially be due to increased clinical experience and to the use of ultrasound guidance with second trimester procedures to assure fundal IUD placement. However, these findings may be biased by limited follow-up. Expulsion rates found in our study are low, even after second trimester placement, and are similar to those reported elsewhere [8]. The expulsion rates increased nonsignificantly for immediate insertion after second trimester abortion; however, due to the overall low expulsion rate, this study
39
lacks sufficient power to detect a statistically significant difference. When considering outcomes, one can argue that the high rate of women failing to return for interval IUD placement outweighs the increased expulsion rate after second trimester postabortal insertion. Our study reaffirms past literature [3,5,6,8] showing that the majority of women do not return for follow-up after elective abortion. This limitation is common in most studies in abortion care and our follow-up rates are equivalent to or better than other studies. Poor follow-up rates underscore the value of providing long-term reversible contraception at the time of pregnancy termination. Retrospective chart reviews are subject to several types of bias. Outcome data may be biased by the motivations of those returning for follow-up and may miss complications or early IUD removals in those who sought care elsewhere. Alternatively, those with complications may be more likely to return to the clinic for follow-up care. Most of our followup occurred within a month after IUD insertion and may have missed later complications. Follow-up greater than a year after insertion was only obtained for 12% of subjects. To evaluate the cost-effectiveness and long-term outcomes of our practice, further study is needed to determine whether IUDs placed in our clinic have been removed by providers outside our medical system. Immediate postabortal IUD placement has been shown to decrease the rate of repeat abortions in a large cohort study (6% vs. 15%; hazard ratio 0.37, 95% CI 0.26–0.52) [5]. Because women who received IUDs at the time of their abortion are more likely to be using IUDs than those scheduled for interval placement, they are at lower risk for a subsequent unplanned pregnancy or termination. Since only 20 IUDs were placed in our interval group, accurate comparison of long-term outcomes, continuation rates and subsequent unplanned pregnancies was not feasible. Furthermore, this analysis may be subject to bias as those ambivalent regarding future pregnancies would be less likely to select immediate placement and more likely to fail to return for interval insertion. Many clinicians require recent negative screening for gonorrhea and chlamydia prior to IUD placement. This is a major barrier to IUD insertion, especially at the time of pregnancy termination, as many clinics strive to provide same-day service. According to the American College of Obstetricians and Gynecologists and the WHO, IUDs may be placed without culture results in the absence of significant risk factors or signs of infection [11,12]. The duration of antibiotic prophylaxis for immediate postabortal IUD insertion has varied by provider in our clinic with some preferring a week of doxycycline 100 mg twice daily when chlamydia test results are unavailable. Given that antibiotic prophylaxis for routine IUD insertion is not needed [12], a prolonged course of antibiotics for postabortal IUD placement may be unnecessary. Whether regimens for postabortal infection prevention should be modified in the setting of immediate IUD placement remains unknown.
40
M.C. Fox et al. / Contraception 83 (2011) 34–40
Although immediate postabortal IUD insertion is shown to be both safe and effective, it is not always practical. The cost of the IUD may be prohibitive, especially for women struggling to afford the abortion itself. Government-funded clinics that must keep abortion and contraceptive services separate may be logistically unable to provide this service. Many insurance plans require prior authorization which prevents immediate insertion. The immediate insertion rate rose during our 5-year study as we became more familiar with the authorization process. Clinics can further increase eligibility for immediate IUD insertion by obtaining authorizations for IUD insertion in advance when the patient schedules her pregnancy termination. Furthermore, our data should encourage insurers to lift prior authorization requirements for this valued service. In conclusion, our study shows that postabortion IUD insertion should not be restricted to the first trimester. Future prospective long-term randomized studies are needed comparing interval to immediate insertion to assess continuation, cost-effectiveness and patient satisfaction. References [1] Henshaw SK, Kost K. Trends in the characteristics of women obtaining abortions, 1977 to 2004. New York: Guttmacher Institute; 2008 http:// www.guttmacher.org/pubs/2008/09/18/Report_Trends_Women_ Obtaining_Abortions.pdf. Accessed 1/3/2010. [2] Grimes D, Schulz K, Stanwood N. Immediate postabortal insertion of intrauterine devices. Cochrane Database Syst Rev 2004:CD001777.
[3] Stanek AM, Bednarek PH, Nichols MD, Jensen JT, Edelman AB. Barriers associated with failure to return for intrauterine device insertion following first-trimester abortion. Contraception 2009;79: 216–20. [4] Gillett PG, Lee NH, Yuzpe AA, Cerskus I. A comparison of the efficacy and acceptability of the Copper-7 intrauterine device following immediate or delayed insertion after first-trimester therapeutic abortion. Fertil Steril 1980;34:121–4. [5] Goodman S, Hendlish SK, Reeves MF, Foster-Rosales A. Impact of immediate postabortal insertion of intrauterine contraception on repeat abortion. Contraception 2008;78:143–8. [6] Goodman S, Hendlish SK, Benedict C, Reeves MF, Pera-Floyd M, Foster-Rosales A. Increasing intrauterine contraception use by reducing barriers to post-abortal and interval insertion. Contraception 2008;78:136–42. [7] Pakarinen P, Toivonen J, Luukkainen T. Randomized comparison of levonorgestrel- and copper-releasing intrauterine systems immediately after abortion, with 5 years' follow-up. Contraception 2003;68:31–4. [8] Drey EA, Reeves MF, Ogawa DD, Sokoloff A, Darney PD, Steinhauer JE. Insertion of intrauterine contraceptives immediately following firstand second-trimester abortions. Contraception 2009;79:397–402. [9] Luukkainen T, Allonen H, Haukkamaa M, et al. Effective contraception with the levonorgestrel-releasing intrauterine device: 12-month report of a European multicenter study. Contraception 1987;36:169–79. [10] IUD insertion following termination of pregnancy: a clinical trial of the TCu 220C, Lippes loop D, and copper 7. Stud Fam Plann 1983;14:99–108. [11] World Health Organization Department of Reproductive Health and Research (WHO/RHR) and Johns Hopkins Bloomberg School of Public Health/Center for Communication Programs (CCP), INFO Project. Family planning: a global handbook for providers. Baltimore and Geneva: CCP and WHO, 2007. [12] Intrauterine device. ACOG Practice Bulletin #59. Obstet Gynecol 2005;105:223–32.
Original research article
Immediate placement of intrauterine devices after first and second trimester pregnancy termination☆ Michelle C. Foxa,⁎, Julia Oat-Judgea , Kathryn Seversona , Roxanne M. Jamshidia , Rameet H. Singhb , Raegan McDonald-Mosleya , Anne E. Burkea a
Department of Gynecology and Obstetrics, Johns Hopkins School of Medicine, Baltimore, MD 21287, USA b Department of Obstetrics and Gynecology, University of New Mexico, Albuquerque, NM 87131, USA Received 17 January 2010; revised 29 June 2010; accepted 30 June 2010
Abstract Background: We reviewed our experience with intrauterine device (IUD) placement after surgical abortion up to 20 weeks' gestation. Study Design: Women presenting for elective abortion between January 2004 and March 2009 who requested an IUD were included in this retrospective review. Results: Of 308 women requesting postabortion IUD placement, 221 (72%) planned insertion at the time of abortion (immediate group) and 87 (28%) planned insertion at their postoperative visit (interval group). IUDs were placed in 96% of the immediate group and in 23% of the interval group (212/221 vs. 20/87; pb.0001). Failure to return for placement was the most common reason for noninsertion in the interval group (60/87=69%). Follow-up information was obtained for 56% of patients and was documented a median of 137 days postabortion (range 3–1594 days). There was no difference in complication rates between groups. Expulsion rates were 3% and 0% in the immediate and interval groups, respectively (6/212 vs. 0/20; p=.4). Considering only those with documented follow-up after immediate insertion (119), there was a nonsignificant trend towards increased expulsion with placement after second vs. first trimester abortion (4/54=7% vs. 2/65=2%; p=.3). When analyzing the 172 subjects with documented follow-up, those planning immediate insertion were more likely to have an IUD in situ at the last contact than those planning later insertion (84/124=68% vs. 20/ 48=42%; p=.002). Conclusion: Immediate postabortion IUD insertion is safe and effective. Given the low rate of return for interval insertion, immediate placement may be preferable. © 2011 Elsevier Inc. All rights reserved. Keywords: Postabortion contraception; Postabortal; Intrauterine contraception; Intrauterine device; IUD; Immediate insertion
1. Introduction Almost half of the 1.2 million abortions performed annually in the United States (US) occur in women who previously have terminated a pregnancy [1]. Long-acting reversible contraceptive methods have the potential to decrease future unplanned pregnancies and repeat abor-
☆
The primary author received compensation from Teva Pharmaceuticals from 2007 to 2009 as a member of its speakers bureau for training clinicians to insert ParaGard® IUDs. No financial support was obtained for this study. ⁎ Corresponding author. Department of Obstetrics and Gynecology, Johns Hopkins Bayview Medical Center, Baltimore, MD 21224, USA. Tel.: +1 410 550 0335; fax: +1 410 550 0245. E-mail address: [email protected] (M.C. Fox). 0010-7824/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.contraception.2010.06.018
tions in women presenting for pregnancy termination. While the intrauterine device (IUD) is a highly effective long-term option, its use for immediate postabortion contraception, especially after second trimester terminations, has been limited. Although IUD insertion is often delayed 2–8 weeks after an abortion (interval insertion), insertion immediately at the time of elective pregnancy termination (immediate insertion) has several advantages [2]. The cervix is already dilated; thus IUD placement may be easier and less painful. Since followup rates after elective abortion are commonly low, immediate IUD insertion assures that the patient has effective contraception without an additional visit [2–4]. Also, postabortal insertion has been associated with a decrease in the rate of repeat abortion [5,6].
M.C. Fox et al. / Contraception 83 (2011) 34–40
Perceived disadvantages of immediate insertion include increased risk of infection, uterine perforation and device expulsion; however, the literature demonstrates that infection and uterine perforation rates are not increased by immediate insertion [2]. Expulsion risk may be theoretically increased secondary to cervical dilation at the time of surgery and subsequent uterine involution. Published expulsion rates of T-shaped IUDs inserted immediately after first trimester abortion vary from 1% to 15% [6–9]. There is minimal data comparing immediate to interval IUD insertion, especially after second trimester dilation and evacuation. There has been only one published randomized trial comparing immediate to interval insertion after suction abortion, and it found a statistically nonsignificant increase in expulsion rate from 3% vs. 15% following interval vs. immediate placement (p=.2). Limited data suggest that IUD placement immediately after second trimester abortion has a higher rate of expulsion than after first trimester procedures. A 1979′s World Health Organization (WHO) trial of immediate postabortal insertion found that the expulsion rate for a copper T-shaped IUD at 120 days postinsertion was 1.9% following first trimester abortion and 19.5% following second trimester procedures [10]. Goodman et al. [6] reported expulsion rates of 1.6% and 7% for IUDs placed immediately after first and second trimester abortions, respectively (p=.02). A recent cohort series published by Drey et al. [8] found a nonsignificant increase in expulsion rates with immediate placement of IUDs after second vs. first trimester abortions (3% vs. 0.8%). We have been placing IUDs immediately after first and second trimester surgical abortions at our institution since 2004. The purpose of this study was to examine the risks and benefits of immediate vs. interval IUD insertion after elective abortion, particularly after second trimester procedures.
2. Materials and methods The primary objective of this study was to compare IUD insertion rates associated with placement immediately after elective surgical abortion (immediate group) vs. delayed insertion 2–4 weeks later at the time of surgical follow-up (interval group). Secondary objectives were to compare expulsion rates in the immediate group after first and second trimester abortion and to assess continuation, removal and complication rates in both groups. Our clinic's practice of postabortion IUD insertion is described below. The copper-T380A (ParaGard® , Teva Pharmaceuticals, Pomona, NY, USA) and the levonorgestrel intrauterine system (Mirena®, Bayer Healthcare Pharmaceuticals, Inc., Wayne, NJ, USA) are available. Patients desiring an IUD are offered immediate and interval insertion. Immediate insertion is recommended unless the patient is unsure of her contraceptive choice, there are signs of cervicitis or if insurance plan preauthorization is required and cannot be attained prior to the procedure. Timing of insertion is determined by the patient after counseling with
35
the above considerations. Abortions are performed under intravenous conscious sedation and local anesthesia in our hospital-based clinic. The use of misoprostol with and without osmotic dilators for cervical priming varies by clinician and gestational age. Routine cervical screening for gonorrhea and chlamydia by nucleic acid amplification is performed at the preoperative visit unless results from recent testing are available. Since the preoperative visit is scheduled the day prior to the abortion, results from cervical screening performed at that visit are commonly unavailable at the time of surgery. Patients are prescribed doxycycline 100 mg twice daily for 3–7 days according to clinician discretion for procedure prophylaxis. They are subsequently treated with doxycycline, azithromycin and/or ceftriaxone according to Centers for Disease Control guidelines if cervical screening results are positive. IUDs are not removed for positive culture results in asymptomatic patients. If the insertion device is not long enough to reach the fundus at the time of placement, additional instruments (most commonly sponge forceps) are used to assure fundal placement of the IUD using techniques described elsewhere [8]. The use of ultrasound guidance varies amongst our clinicians in the first trimester but is standard for immediate IUD placement following second trimester abortion. Postsurgical follow-up is scheduled in 2–4 weeks. Those in the interval group are asked to return to have their IUD checked 4–6 weeks after insertion. No further routine follow-up is scheduled; however, patients are encouraged to return for any problems. Women presenting for elective abortion between January 2004 and March of 2009 who desired an IUD and intended to follow up within our medical system were included in this retrospective record review. Women with fetal demise or fetal anomalies were excluded since most did not desire long-term contraception and planned to follow up with their primary obstetrician. Records were reviewed in the Spring of 2009, 3 to 60 months postoperatively. Subjects were identified from the procedure ledger in our clinic. All available electronic records dating back to the time of the abortion were reviewed. Gynecology clinic charts for all subjects were reviewed for verification and to obtain missing data. This review was approved by the Institutional Review Board of the Johns Hopkins University School of Medicine. For the purpose of this retrospective analysis, postoperative infection was defined by the prescription of additional broad-spectrum antibiotics. Retained products of conception (RPOC) were defined by clinical suspicion and treatment with misoprostol or reaspiration. Second trimester abortion was defined as those performed at greater than or equal to 14 weeks' gestation. Data were abstracted onto coded forms and were entered into Microsoft Access and Excel files, and were then analyzed using Stata 10 (College Station, TX, USA). Statistical tests included Fisher's Exact Test for categorical data, Student's t test for continuous data and nonparametric statistics where appropriate. A two-sided p value of .05 was considered statistically significant.
36
M.C. Fox et al. / Contraception 83 (2011) 34–40
3. Results Of the 964 women presenting for elective abortion during the study time period, 308 (32%) elected IUD for contraception and were eligible for inclusion. Of these, 221 were in the immediate group and 87 were in the interval group. Those with insurance plans requiring prior authorization for IUD placement that could not be obtained prior to the abortion and those who were undecided regarding IUD placement, and thus did not want immediate insertion, comprised the interval group. No patients desiring immediate insertion had suspected cervicitis. The demographics of women included in this review are presented in Table 1. Most subjects were African American, single and enrolled in a medical assistance program. Over half previously had undergone an abortion and also had at least one child. Those in the immediate group were younger (24 vs. 26.5; p=.002), more likely to have medical assistance (155/221=70% vs. 36/87=41%; pb.0001) and more likely to have planned a copper than a levonorgestrel IUD (72/221=33% vs. 15/ 87=17%; pb.0001) than those in the interval group. Two hundred thirty-two IUDs were inserted, of which approximately two-thirds were levonorgestrel IUDs (157/232). Over half of IUDs placed immediately (111/212) were inserted after a second trimester abortion. Table 1 Subject demographics according to planned timing of IUD placement Variable
Immediate (n=221)
Interval (n=87)
Age, years (mean±SD) Race White Black Asian Hispanic Unknown Other Gravidity, median (range) Parity, median (range) Prior terminations, median (range) Gestational age, weeks (mean±SD) 5–13 6/7 weeks 14–20 weeks IUD type Levonorgestrel Copper-T380A Unknown Insurance type Private plan Medical assistance Uninsured Unknown Marital status Single Married Separated Divorced Other Unknown
24.0±6.2
26.5±7.2
14.0% 83.3% 0.5% 0.5% 1.4% 0.5% 3 (1–15) 1 (0–7) 1 (0–7) 13.9±3.9 47.5% 52.5%
20.7% 74.7% 2.3% 2.3% 0 0 4 (1–12) 1 (0–6) 1 (0–5) 14.1±4.0 42.5% 57.5%
66.5% 32.6% 1%
75.9% 17.2% 6.9%
26.2% 70.1% 2.7% 1%
56.3% 41.4% 1.2% 1%
86.6% 6.9% 1.8% 0.5% 0.5% 3.7%
75% 10% 2.5% 2.5% 0 10%
p=.002 p=.1
p=.06 p=.4 p=.5 p=.8 p=.4 pb.0001
pb.0001
p=.1
Overall, follow-up information was available for 172 subjects, 124 in the immediate group and 48 in the interval group (124/221=56% vs. 48/87=55%; p=.9). Follow-up information was documented a median of 137 days postabortion (range 3–1594 days). Only 117 subjects (38%) attended their scheduled follow-up visit 2–4 weeks postoperatively; data were obtained from the electronic record for an additional 55 subjects (18%). Those who had a prior abortion, had medical assistance or underwent a second trimester termination were less likely to attend their followup visit in our clinic. However, when all follow-up data from electronic record review were included, the only statistically significant difference between those lost to follow-up and those with documented outcome data was gestational age. Follow-up data were available for 63% of those who had first trimester abortions vs. 50% of those who had second trimester abortions (89/142 vs. 83/166; p=.03). Despite counseling regarding the possibility of postprocedural IUD expulsion, only 40% (85/212) returned to clinic for postoperative follow-up after immediate insertion. This percentage was the same as those returning for string checks after interval insertion (8/20=40%). IUD use for postabortion contraception has increased significantly in our clinic since initiating this service. The percentage of patients selecting IUDs for postabortion contraception increased annually from 12% in 2004 to 62% in the first quarter of 2009 (20/169 vs. 48/77; pb.00001). The percentage of planned IUDs that were actually placed increased annually from 55% in 2004 to 92% in 2009 (p=.008). Additionally, the proportion of subjects planning immediate IUD insertion increased from 55% in 2004 to 90% in 2009 (p=.004). Figs. 1 and 2 demonstrate the outcomes of subjects in both groups. Ninety-six percent (212/221) of the immediate group vs. 23% (20/87) of the interval group actually had an IUD placed (pb.0001). Nine of the planned immediate insertions were deferred for the following reasons: intraoperative bleeding (3), subject changed her mind (2), planned surgery for excision of cervical dysplasia (1), failure to obtain insurance approval (1) and unknown (2). Interval placement did not occur for any of these nine, largely due to failure to return for insertion. In the interval group, the reasons IUD placement was not performed include failure to return for IUD insertion (60), no longer desired an IUD (4) and miscellaneous medical concerns (3). There was one failed insertion in the interval group which was successful at a later visit. There were no failed insertions in the immediate group. There were six documented IUD expulsions in the immediate group and none in the interval group (6/ 212=2.8% vs. 0/20=0%, p=.4). When data analysis is limited to subjects with documented follow-up, the expulsion rate in the immediate group increases to 5%; however, this is not statistically higher than in the interval group (6/119=5% vs. 0/ 20=0%, p=.3). Expulsion was reported 2, 91, 139 and 286 days after insertion and occurred at an unknown time for the remaining two subjects. Considering only those with docu-
M.C. Fox et al. / Contraception 83 (2011) 34–40
Fig. 1. Immediate group outcome.
Fig. 2. Interval group outcome.
37
38
M.C. Fox et al. / Contraception 83 (2011) 34–40
mented follow-up after immediate IUD insertion (119), there was a nonsignificant increase in expulsion rate following immediate placement after first vs. second trimester abortion from 3.1% to 7.4%, respectively (2/65 vs. 4/54; p=.3). Twenty-nine early removals were documented in the immediate group vs. none in the interval group (29/ 212=14% vs. 0/20=0%; p=.08). Two IUDs were removed during reaspiration procedures within the first week, and two were removed in the first month for suspected endometritis or pelvic inflammatory disease. An additional 10 early removals occurred in the first year after placement. Twelve were removed after a year or more. The dates of removal for three subjects were unknown. The most common reasons for removal remote from insertion were pain and irregular bleeding. There was no difference in early removal rates between subjects in their first or second trimester at the time of termination (15/114=13% vs. 14/118=12%; p=.8). Those receiving the copper-IUD were almost twice as likely to have it removed early compared to the levonorgestrel IUD (14/ 75=19% vs. 15/157=10%; p=.05). As shown in Figs. 1 and 2, those in the immediate group were significantly more likely to have an IUD at the last
documented follow-up at the time of the chart review. Including only the 172 subjects with documented followup, 68% of the immediate group vs. 42% of the interval group had an IUD in situ at their last known visit (84/124 vs. 20/48; p=.002). Three-fourths (157/212) of IUDs in the immediate group were placed without available cervical culture results, although prophylactic doxycyline was given as previously described. Twenty-two subjects planning IUD insertion had asymptomatic gonorrhea or chlamydia infections at the time of their abortion, representing 5% of the immediate group and 11% of the interval group (12/221 vs. 10/87; p=.06). Eleven had their IUD placed immediately at the time of their abortion in the presence of unknown gonorrhea or chlamydia infection. One woman had her IUD removed 3 weeks later for mild endometritis in the setting of an untreated chlamydia infection. This subject received 1 week of oral antibiotic treatment and had no further known complications. No cases of pelvic inflammatory disease occurred; however, four of the 11 were lost to follow-up after IUD insertion in the setting of undiagnosed chlamydia despite extensive efforts to contact them. The remaining 10 women with asymptomatic infection
Table 2 Postoperative complications: details of subjects presenting with complications within 6 weeks of their abortion Age, G/P
Estimated gestational Group age (weeks)
IUD placed IUD removed Narrative
17 G2 P0010 8 6/7
Immediate Y
Y
23 G3 P2002 11 6/7
Immediate Y
N
22 G4 P3003 14 1/7
Immediate Y
Expelled
15 G1 P0000 7 6/7
Immediate Y
Y
24 G3 P1011 13 2/7
Immediate Y
Y
40 G1 P0000 8 6/7
Immediate Y
Y
29 G6 P1132 14 1/7
Interval
N
N/A
34 G5 P1031 14 6/7
Interval
N
N/A
29 G4 P3003 17 3/7
Interval
Y
Unknown
Undiagnosed chlamydia at time of abortion. Did not comply with prescribed treatment. Presented on POD 26 with mild endometritis. Treated with 7 days doxycycline after IUD removal. Presented POD 8 with pain and cramping. Treated with 14 days PO antibiotics for presumed PID. Symptoms resolved. IUD expelled on POD 2 and replaced on POD 6. Seen in ER on POD 16 for bleeding. Sonogram showed IUD in place and possible RPOC. Misoprostol prescribed in ER. IUD in place on POD 22. Expulsed second IUD on POD 30. Switched to DMPA for contraception. Normal postoperative visit and string check on POD 13. Presented POD 25 with PID. IUD removed and PO antibiotics prescribed. Admitted same day to outside hospital due to worsening pain, vomiting and inability to tolerate PO antibiotics. She received IV antibiotics and was subsequently LTFU. Presented POD 5 with heavy bleeding and pain. Sonogram showed RPOC. IUD removed and reaspiration performed without complications. IUD replaced 2 months later. Presented POD 6 with cramping and heavy bleeding. Sonogram showed RPOC and hematometra. IUD removed and reaspiration performed. Diagnosed with endometritis on POD 20. Treated with 2 weeks of PO antibiotics. IUD replacement deferred initially due to infection and then LTFU. Multiple visits elsewhere for pain beginning 3 weeks after abortion. Imaging inconclusive for RPOC. Returned to clinic for persistent pain (10 weeks after abortion) and was treated with PO antibiotics for 2 weeks for suspected infection with resolution in pain. Reaspiration on POD 23 for hematometra and RPOC. Hospitalized overnight. Subsequently LTFU. Admitted to outside hospital on POD 4 with fevers and RPOC and had reaspiration. Treated with IV antibiotics and discharged with 2-week course of PO antibiotics. Interval IUD insertion on POD 20. Subsequently LTFU.
G: Gravidity; P: parity; Immediate: immediate IUD insertion; Interval: interval IUD insertion; Y: yes; N: no; RPOC: retained products of conception; POD: postoperative day; PO: oral; IV: intravenous; ER: emergency room; PID: pelvic inflammatory disease; N/A: not applicable; LTFU: lost to follow-up.
M.C. Fox et al. / Contraception 83 (2011) 34–40
were in the interval group. Only two of these women returned for their postoperative visit and neither received an IUD. Overall, nine subjects (3%) had complications within a month of their abortion, six of whom had IUDs inserted immediately. Details of these complications are reviewed in Table 2. No uterine perforations were documented. There was no statistically significant difference in infection (1.8% vs. 2.3%; p=.8), RPOC (1.4% vs. 2.3%; p=.6) or need for reaspiration (0.9% vs. 2.3%; p=.4) between the immediate and interval groups, respectively. The IUD was removed in four cases: two who required reaspiration and two with infection. Immediate postabortion IUD insertion was associated with decreased chance of documented subsequent pregnancy during the follow-up period compared to interval insertion (16/221=7% vs. 19/87=22%; p=.0002). Of the subsequent pregnancies in the immediate group, two occurred in women who did not ever have an IUD inserted with the rest occurring after IUD expulsion or early removal. All of the subsequent pregnancies in the interval group occurred in those who never received an IUD. 4. Discussion This retrospective study adds to the growing evidence supporting immediate postabortion IUD insertion and contributes to the limited information on placement at the time of second trimester abortion. Our results agree with previous studies supporting the use of IUDs immediately after elective pregnancy termination including those procedures performed in the second trimester. This study, unlike the prior report of Drey et al. [8], compares outcomes of interval to immediate insertion. Studies show short-term removal, infection and complication rates after immediate postabortion IUD insertion are not significantly increased when compared to interval insertion a few weeks later [5,6]. In our study, only four of the 212 IUDs placed at the time of abortion were removed within the first month for complications and only one known expulsion occurred during this time period. Two IUDs were removed secondary to infection; however, these removals may not have been clinically necessary [11]. Further education regarding the potential for treating pelvic infection with an IUD in situ is needed for our clinicians. Complications of abortion are low; thus, our study is underpowered to detect a difference in these rates. The low complication rates associated with immediate postabortal insertion noted in more recent literature [8] and in this study may partially be due to increased clinical experience and to the use of ultrasound guidance with second trimester procedures to assure fundal IUD placement. However, these findings may be biased by limited follow-up. Expulsion rates found in our study are low, even after second trimester placement, and are similar to those reported elsewhere [8]. The expulsion rates increased nonsignificantly for immediate insertion after second trimester abortion; however, due to the overall low expulsion rate, this study
39
lacks sufficient power to detect a statistically significant difference. When considering outcomes, one can argue that the high rate of women failing to return for interval IUD placement outweighs the increased expulsion rate after second trimester postabortal insertion. Our study reaffirms past literature [3,5,6,8] showing that the majority of women do not return for follow-up after elective abortion. This limitation is common in most studies in abortion care and our follow-up rates are equivalent to or better than other studies. Poor follow-up rates underscore the value of providing long-term reversible contraception at the time of pregnancy termination. Retrospective chart reviews are subject to several types of bias. Outcome data may be biased by the motivations of those returning for follow-up and may miss complications or early IUD removals in those who sought care elsewhere. Alternatively, those with complications may be more likely to return to the clinic for follow-up care. Most of our followup occurred within a month after IUD insertion and may have missed later complications. Follow-up greater than a year after insertion was only obtained for 12% of subjects. To evaluate the cost-effectiveness and long-term outcomes of our practice, further study is needed to determine whether IUDs placed in our clinic have been removed by providers outside our medical system. Immediate postabortal IUD placement has been shown to decrease the rate of repeat abortions in a large cohort study (6% vs. 15%; hazard ratio 0.37, 95% CI 0.26–0.52) [5]. Because women who received IUDs at the time of their abortion are more likely to be using IUDs than those scheduled for interval placement, they are at lower risk for a subsequent unplanned pregnancy or termination. Since only 20 IUDs were placed in our interval group, accurate comparison of long-term outcomes, continuation rates and subsequent unplanned pregnancies was not feasible. Furthermore, this analysis may be subject to bias as those ambivalent regarding future pregnancies would be less likely to select immediate placement and more likely to fail to return for interval insertion. Many clinicians require recent negative screening for gonorrhea and chlamydia prior to IUD placement. This is a major barrier to IUD insertion, especially at the time of pregnancy termination, as many clinics strive to provide same-day service. According to the American College of Obstetricians and Gynecologists and the WHO, IUDs may be placed without culture results in the absence of significant risk factors or signs of infection [11,12]. The duration of antibiotic prophylaxis for immediate postabortal IUD insertion has varied by provider in our clinic with some preferring a week of doxycycline 100 mg twice daily when chlamydia test results are unavailable. Given that antibiotic prophylaxis for routine IUD insertion is not needed [12], a prolonged course of antibiotics for postabortal IUD placement may be unnecessary. Whether regimens for postabortal infection prevention should be modified in the setting of immediate IUD placement remains unknown.
40
M.C. Fox et al. / Contraception 83 (2011) 34–40
Although immediate postabortal IUD insertion is shown to be both safe and effective, it is not always practical. The cost of the IUD may be prohibitive, especially for women struggling to afford the abortion itself. Government-funded clinics that must keep abortion and contraceptive services separate may be logistically unable to provide this service. Many insurance plans require prior authorization which prevents immediate insertion. The immediate insertion rate rose during our 5-year study as we became more familiar with the authorization process. Clinics can further increase eligibility for immediate IUD insertion by obtaining authorizations for IUD insertion in advance when the patient schedules her pregnancy termination. Furthermore, our data should encourage insurers to lift prior authorization requirements for this valued service. In conclusion, our study shows that postabortion IUD insertion should not be restricted to the first trimester. Future prospective long-term randomized studies are needed comparing interval to immediate insertion to assess continuation, cost-effectiveness and patient satisfaction. References [1] Henshaw SK, Kost K. Trends in the characteristics of women obtaining abortions, 1977 to 2004. New York: Guttmacher Institute; 2008 http:// www.guttmacher.org/pubs/2008/09/18/Report_Trends_Women_ Obtaining_Abortions.pdf. Accessed 1/3/2010. [2] Grimes D, Schulz K, Stanwood N. Immediate postabortal insertion of intrauterine devices. Cochrane Database Syst Rev 2004:CD001777.
[3] Stanek AM, Bednarek PH, Nichols MD, Jensen JT, Edelman AB. Barriers associated with failure to return for intrauterine device insertion following first-trimester abortion. Contraception 2009;79: 216–20. [4] Gillett PG, Lee NH, Yuzpe AA, Cerskus I. A comparison of the efficacy and acceptability of the Copper-7 intrauterine device following immediate or delayed insertion after first-trimester therapeutic abortion. Fertil Steril 1980;34:121–4. [5] Goodman S, Hendlish SK, Reeves MF, Foster-Rosales A. Impact of immediate postabortal insertion of intrauterine contraception on repeat abortion. Contraception 2008;78:143–8. [6] Goodman S, Hendlish SK, Benedict C, Reeves MF, Pera-Floyd M, Foster-Rosales A. Increasing intrauterine contraception use by reducing barriers to post-abortal and interval insertion. Contraception 2008;78:136–42. [7] Pakarinen P, Toivonen J, Luukkainen T. Randomized comparison of levonorgestrel- and copper-releasing intrauterine systems immediately after abortion, with 5 years' follow-up. Contraception 2003;68:31–4. [8] Drey EA, Reeves MF, Ogawa DD, Sokoloff A, Darney PD, Steinhauer JE. Insertion of intrauterine contraceptives immediately following firstand second-trimester abortions. Contraception 2009;79:397–402. [9] Luukkainen T, Allonen H, Haukkamaa M, et al. Effective contraception with the levonorgestrel-releasing intrauterine device: 12-month report of a European multicenter study. Contraception 1987;36:169–79. [10] IUD insertion following termination of pregnancy: a clinical trial of the TCu 220C, Lippes loop D, and copper 7. Stud Fam Plann 1983;14:99–108. [11] World Health Organization Department of Reproductive Health and Research (WHO/RHR) and Johns Hopkins Bloomberg School of Public Health/Center for Communication Programs (CCP), INFO Project. Family planning: a global handbook for providers. Baltimore and Geneva: CCP and WHO, 2007. [12] Intrauterine device. ACOG Practice Bulletin #59. Obstet Gynecol 2005;105:223–32.