Contraception xx (2014) xxx – xxx
Original research article
Postplacental intrauterine device insertion at a teaching hospital☆ Tara C. Jatlaoui a,⁎, Michele Marcus b , Denise J. Jamieson a , Peggy Goedken a , Carrie Cwiak a a Department of Gynecology and Obstetrics, Emory University School of Medicine, Atlanta, GA, USA Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, GA 30322, USA Received 30 April 2013; revised 11 October 2013; accepted 16 October 2013
b
Abstract Objective: To determine whether postplacental intrauterine device (IUD) insertion can be safely and effectively performed within a teaching program. Study Design: This was a prospective cohort of 177 subjects planning vaginal delivery enrolled antenatally who desired postplacental IUD insertion of either the copper T380A IUD or levonorgestrel IUS. Insertions were performed primarily by resident physicians following a training session. Follow-up included a 4- to 8-week visit and telephone calls at 3 and 6 months. Results: Ninety-nine subjects underwent successful postplacental IUD insertion of 100 attempts. Seventeen expulsions (17%) were noted: 10 complete and 7 partial. The study identified no differences in outcome by training level; however, the study lacked statistical power to evaluate anything other than large differences. Conclusion: Postplacental IUD insertions can be safely and effectively performed within a training program. Implications: A training protocol may safely and feasibly be initiated among physicians, advanced practice clinicians or trainees with no prior experience with postplacental IUD insertion. By initiating this practice, access to highly effective contraception may increase for patients who have difficulty returning for a visit or otherwise receiving effective methods. © 2014 Elsevier Inc. All rights reserved. Keywords: Intrauterine device; Vaginal delivery; Postpartum contraception
1. Introduction The intrauterine device (IUD) has the highest continuation and cost-effectiveness compared to other reversible contraceptive methods [1–3]. Women who choose longacting reversible contraception, like the levonorgestrel intrauterine system (LNG-IUS) or copper T380A IUD, have 20 times less risk of unintended pregnancy compared to women who use pills, patch or the ring [4]. The postpartum period may be an optimal time for IUD uptake, especially for women who otherwise have difficulty with access, motivation or side effects [5]. A Cochrane Review of randomized trials found that postplacental IUD insertion (within 10 min of placental delivery) appeared safe and effective, though expulsion rates were higher compared to interval insertions ☆
Conflicts of interest: The authors have no conflicts of interest to declare. ⁎ Corresponding author. Department of Gynecology and Obstetrics, Emory University School of Medicine, Atlanta, GA 30303, USA. E-mail address: [email protected] (T.C. Jatlaoui). 0010-7824/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.contraception.2013.10.008
(more than 6 weeks after delivery) [6]. One randomized trial has compared postplacental insertion with interval insertion for the LNG-IUS, similarly finding higher expulsion rates compared to interval insertion (24% vs. 4.4%) [7]. Still, an increased risk of expulsion may be outweighed by an increase in access to highly effective reversible contraception in cases where significant barriers exist to both the device and its insertion. It has been suggested that expulsion rates are associated with the experience of the provider, whether across institutions or training levels [8,9]. However, training sessions stressing high fundal placement [10] and cumulative number of insertions have been positively correlated with decreased expulsion rates, suggesting that success rates can be improved significantly with training [11]. The specific aim of this study was to determine whether postplacental IUD insertion can be safely and effectively performed within a teaching program. The primary outcome was IUD continuation rate at 6 months. Secondary outcomes included IUD expulsion, infection, perforation and pregnancy.
2
T.C. Jatlaoui et al. / Contraception xx (2014) xxx–xxx
2. Methods and materials A prospective cohort study was conducted at Grady Memorial Hospital, Atlanta, Georgia, following approval by the Emory University Institutional Review Board. All patients receiving prenatal care and requesting an IUD as their primary method of postpartum contraception were offered participation. Additional inclusion criteria were as follows: age 18 years or older, anticipating a vaginal delivery, able to read and/or understand English or Spanish, able and willing to give informed consent and willing to follow-up over 6 months. Exclusion criteria at the time of placement included the following: contraindications to IUD placement [12], fever in labor, rupture of membranes for greater than 24 hours, need for cesarean section, postpartum hemorrhage, sexually transmitted infection in current pregnancy and leaving the hospital area within 6 months. Subjects could choose either the LNG-IUS or the CuT380A IUD. Participants reaffirmed consent for IUD insertion upon presentation for delivery. Those ineligible for the study were offered IUD insertion at their postpartum visit. All IUDs were provided at no cost regardless of study participation or time of insertion via grant funding that was independent of the study. An initial training session was conducted for all Emory University Gynecology and Obstetrics residents. Refresher training sessions were conducted every 6 weeks throughout the study. Sessions reviewed the following: counseling and consent for the IUD, insertion technique for both IUD types, postplacental IUD insertion technique and abdominal ultrasound guidance for fundal placement. For all eligible subjects, insertions were performed under ultrasound guidance using the LNG-IUS prepackaged inserter or ring forceps for CuT380A IUD, given the shorter inserter. If the IUD could not be placed with the inserter or ring forceps, hand insertion was utilized. The IUD strings were cut at the external os. Subjects returned for a 4- to 8-week postpartum visit. They were asked about satisfaction with the IUD and symptoms of expulsion or infection. Pelvic exams were performed to check for bleeding and infection, to visualize IUD strings and trim strings if necessary. Infection was defined as cervicitis (N 10 white blood cells per high-power field via wet prep) or Pelvic Inflammatory Disease (cervicitis, plus either cervical motion, adnexal or fundal tenderness). Transvaginal ultrasound was performed to ensure IUD intrauterine placement if necessary. Complete expulsion was defined as (1) no IUD seen on ultrasound and patient report of seeing the IUD come out or (2) an abdominal x-ray confirming absence of the IUD. Partial expulsion was defined as an IUD visible within the cervical canal by exam or ultrasound. Questionnaires were repeated at 3 and 6 months via telephone. If any contact was missed, subjects were contacted via telephone and certified mail. Subjects were considered lost to follow-up if they had not had at least one contact (visit or telephone) within the 6month study time.
The sample size of 100 was based primarily on feasibility; we estimated the number of women who would both desire a postplacental IUD and be eligible for placement at our site during a 2-year recruitment period. Statistical analysis was performed using SAS 9.3 (Cary, NC, USA). Primary analysis sought to determine 6-month continuation rate and identify risk factors for expulsion within 6 months. Bivariate analysis was performed using Fisher's exact test for categorical variables and the Student's t-test for continuous variables. Covariates included the following: age, race, parity, anesthesia, insertion time interval after placental delivery and IUD type. Stratified analysis was performed by the exposure of interest, provider training level, to identify whether covariates were associated with expulsion. Multivariate analysis of the association between provider training level and expulsion was performed with adjustment for age, parity, anesthesia, insertion time interval after placental delivery and IUD type.
3. Results Between March 2009 and March 2011, 177 women were enrolled in the study and 100 underwent attempts at postplacental IUD insertion. Those who were excluded either withdrew consent or were ineligible according to study protocol (Fig. 1). Ninety-nine attempts at insertion were successful. Of the 99 successful placements, 69 women selected LNG-IUS while 30 chose the CuT380A IUD. Demographics were not significantly different between women choosing LNG-IUS and women choosing CuT380A (Table 1). The majority of participants were black, had completed high school or GED, were unemployed, were covered by Medicaid, and were single. Mean time to insertion was 7 min (SD 4.6 min). Most insertions (72.7%) were performed by a post graduate year (PGY) 3 level resident or above, and most were performed under epidural anesthesia (63.6%). Provider assessment of insertion was documented as “very easy” or “somewhat easy” for 65 placements. Of the 99 women who received an IUD, 65 presented for a follow-up visit, and 88 had at least one contact within 6 months. Women who had follow-up were similar to the entire group who underwent postplacental insertion (Table 1). Only 21 subjects had follow-up at the 6-month mark, none of which reported IUD expulsion or removal. Seventeen expulsions (19.3%) occurred among the 88 women who completed one follow-up contact within 6 months. Ten complete expulsions were diagnosed, and all 10 women saw their IUD at the time of expulsion. Eight of these women returned to clinic and three of those had another IUD placed. Seven partial expulsions were diagnosed, and those IUDs were removed. Five of those women chose to have another IUD placed. Replacement IUDs were provided at no cost via independent grant funding.
T.C. Jatlaoui et al. / Contraception xx (2014) xxx–xxx
3
Fig. 1. Enrollments, insertions and exclusions.
No pregnancies or perforations were reported during the follow-up study period. Overall, 10 subjects (11%) were diagnosed with an infection. Seven subjects were empirically treated for cervicitis or PID as an outpatient at the postpartum visit. Three additional subjects reported infection diagnosed at outside clinics. No IUDs were removed for infection or for any other reason. Bivariate analysis found distribution of prior births to be the only statistically significant difference (p=.029) between expulsion and nonexpulsion groups (Table 2). IUDs placed by lower level providers (PGY1, PGY2) did not have a significantly different risk of expulsion compared to those placed by upper level providers (PGY3 and above). The point estimate for expulsion risk among women with anesthesia was half as frequent as the expulsion risk among women without anesthesia, but this was not statistically significant. Stratified analysis found no significant effect modification between provider level and covariates. Multivariate analysis found no significant associations with expulsion, provider level or other factors.
4. Discussion This study demonstrated that postplacental IUD insertion can be performed safely and effectively within a training program, utilizing training sessions and ultrasound guidance. Although the number of women who could be contacted at 6 months was low and precluded a calculation of IUD continuation rate, complication rates among the 88 women who had at least one follow-up contact were comparable to those found in recent studies of postplacental IUD insertion [6–9,13–15]. All 10 complete expulsions were recognized by the patient. Partial expulsions, however, were not always symptomatic and required exam and ultrasound for diagnosis. The number of infections was similar to the high baseline rate in Fulton County, Georgia, and did not require IUD removal [16]. The low 6-month contact rate may be
indicative of the population most in need of postplacental IUD insertion in that their access to health care outside pregnancy is limited. Importantly, we found that expulsion for an IUD placed by a less clinically experienced provider was not inferior to rates in the literature [6–9,13–15]. This is in contrast to older studies that have found more expulsions among providers with less experience. No studies had involved resident physicians placing the LNG-IUS or copper T380A IUD. In our study, placements were not limited by provider experience and occurred in real-time conditions of a labor and delivery unit. Only two insertions were performed by an attending physician who had prior experience with postplacental insertion. Standardized curricula and models are now available for postplacental IUD insertion training [17]. Refresher training sessions coincided with the short rotation schedule inherent in most training programs. Abdominal ultrasound was used to assure fundal placement and is available in most training hospitals in the United States. This suggests that a training protocol can feasibly be initiated within an institution staffed by physicians, advanced practice clinicians or trainees with no prior experience with postplacental IUD insertion. While we did not find a significant difference in the risk of expulsion among IUDs placed by lower level residents compared to more senior level residents, this study was underpowered to detect a meaningful difference. The expulsion risk of one of the groups would have had to exceed 46% in order to detect a significant difference from 17% at 80% power with a 95% confidence interval. The few years of experience difference among resident levels may not reflect the broader differences observed in previous studies between providers over several years of experience providing postplacental IUDs. In this study, expulsion was significantly associated with low parity. Although no previous studies had a similar finding, it is possible that the IUD is affected more so by uterine involution in primiparous women. The suggestion that lack of anesthesia may be associated with expulsion is
4
T.C. Jatlaoui et al. / Contraception xx (2014) xxx–xxx
Table 1 Characteristics of women receiving a postplacental IUD, by IUD type Total insertions (n=99) Age (y), mean (SD) Race Black Asian Hispanic Unspecified School Less than 8y Some HS HS/GED Some college Other Employment Unemployed Full time Part time Unspecified Annual income $10,000 or less $10,000–30,000 Above $30,000 Unspecified Prior births 0 1 2 3 or greater Unspecified Insurance Yes Medicaid Other Unknown No Unknown Future fertility No Yes Unsure Unknown Relationship status Single Married Unspecified
Completed follow-up* (n=88)
23.7 (4.6)
Mirena (%) (n= 69)
23.7 (4.6)
p-Value a
Paragard (%) (n=30)
23.7 (4.4)
23.7 (5.2)
73 1 19 6
73.7% 1.0% 19.2% 6.1%
67 0 16 5
76.1% 0.0% 18.2% 5.7%
53 0 11 5
76.8% 0.0% 15.9% 7.2%
20 1 8 0
66.7% 3.3% 26.7% 0.0%
6 23 43 17 10
6.1% 23.2% 43.4% 17.2% 10.1%
5 19 40 15 9
5.7% 21.6% 45.5% 17.0% 10.2%
3 16 31 12 7
4.3% 23.2% 44.9% 17.4% 10.1%
3 7 12 5 3
10.0% 23.3% 40.0% 10.0% 10.0%
64 13 12 10
64.6% 13.1% 12.1% 10.1%
57 12 10 4
64.8% 13.6% 11.4% 4.5%
41 10 10 8
59.4% 14.5% 14.5% 11.6%
23 3 2 2
76.7% 10.0% 6.7% 6.7%
50 26 2 21
50.5% 26.3% 2.0% 21.2%
43 24 2 19
48.9% 27.3% 2.3% 21.6%
35 18 2 14
50.7% 26.1% 2.9% 20.3%
15 8 0 7
50.0% 26.7% 0.0% 23.3%
17 34 22 25 1
17.2% 34.3% 22.2% 25.3% 1.0%
14 31 19 24 0
15.9% 35.2% 21.6% 27.3% 0.0%
12 23 16 18 0
17.4% 33.3% 23.2% 26.1% 0.0%
5 11 6 7 1
16.7% 36.7% 20.0% 23.3% 3.3%
78 67 3 8 14 7
78.8% 67.7% 3.0% 8.1% 14.1% 7.1%
68 61 3 4 14 4
77.3% — — — 15.9% 8.0%
54 45 2 7 9 6
78.3% 83.3% 3.7% 13.0% 13.0% 8.7%
24 22 1 1 5 1
80.0% 91.7% 4.2% 4.2% 16.7% 3.3%
57 22 14 6
57.6% 22.2% 14.1% 6.1%
52 19 12 5
59.1% 21.6% 13.6% 5.7%
35 17 12 5
50.7% 24.6% 17.4% 7.2%
22 5 2 1
73.3% 16.7% 6.7% 3.3%
78 15 6
78.8% 15.2% 6.1%
69 14 5
78.4% 15.9% 5.7%
53 11 5
76.8% 15.9% 7.2%
25 4 1
83.3% 13.3% 3.3%
.94 .12
.74
.50
.99
.98
.83
.15
.77
* With at least one follow-up contact in 6 months. a Mirena vs Paragard.
also a new finding. Further studies with adequate power to detect a difference are needed to determine if an association truly exists for parity or anesthesia. This study was limited by sample size and the low percentage of women who were contacted at 6 months. Although 88% of women had study contact after insertion, outcomes such as expulsion, infection, pregnancy and perforation may actually be higher than reported or may change with time since insertion. This study population also represents a particular population, and findings may not apply to all women who desire a postplacental IUD; though
in a population at increased risk for adverse health outcomes, our findings of similar complication rates are reassuring. While the risk of expulsion has been shown to be higher with postplacental insertion compared to interval insertion, certain patient populations may benefit more from the access and delivery of highly effective reversible contraception at the time of vaginal delivery, especially patients who have difficulty returning for postpartum visits. Postplacental IUD placement is therefore supported by the US Medical Eligibility Criteria for Contraceptive Use and the American College of Obstetrics and Gynecology [12,18]. By initiating
T.C. Jatlaoui et al. / Contraception xx (2014) xxx–xxx
5
Table 2 Characteristics of subjects with follow-up* by expulsion versus nonexpulsion (n= 88) Covariate
Expulsion (n=17)
a
Age , mean (SD) Race Black Hispanic Unspecified Prior births 0 1 2 3 or greater Anesthesia None IV pain medicine Epidural Unspecified Insertion time from placenta delivery (min), a mean (SD) IUD type LNG-IUS CuT380A Provider level PGY1 or PGY2 PGY3 or above Infection after discharge No Yes Provider assessment Very easy Somewhat easy Somewhat difficult Very difficult Unspecified
Nonexpulsion (n= 71)
23.5 (5.2)
23.8 (4.5)
p-Value .8 1
14 3 0
82.4% 17.6% 0.0%
53 13 5
74.6% 18.3% 0.1%
1 9 6 1
5.9% 52.9% 35.3% 5.9%
13 22 13 23
18.3% 31.0% 18.3% 32.4%
6 2 8 1
35.3% 11.8% 47.1% 5.9%
11 10 48 2
15.5% 14.1% 67.6% 2.8%
.029
.15
7.5 (5.4)
5 (2.8)
.101 1
12 5
70.6% 29.4%
50 21
70.4% 29.6%
2 15
11.8% 88.2%
22 49
31.0% 69.0%
17 0
100% 0.0%
60 11
84.5% 15.5%
8 5 2 0 2
47.1% 29.4% 11.8% 0.0% 11.8%
25 20 11 3 12
35.2% 28.2% 15.5% 4.2% 16.9%
.14
.11
.932
*With at least one follow-up contact in 6 months. a t-Test.
postplacental IUD insertion protocols in residency training facilities in the United States, access to long-acting reversible contraception could increase for patients who have difficulty returning for a visit or otherwise receiving a highly effective method. Acknowledgments ParaGard Copper T380A IUDs and Mirena Levonorgestrel IUS were provided by an anonymous grant. The grant had no involvement in study design, analysis or reporting. This study was possible with the cooperation and participation of the Emory Department of Gynecology and Obstetrics faculty and residents. References [1] ACOG Practice Bulletin. Clinical management guidelines for obstetrician-gynecologists. Number 59, January 2005. Intrauterine device. Obstet Gynecol 2005;105(1):223–32 [PubMed PMID: 15625179. Epub 2004/12/31. eng]. [2] Trussell J. Contraceptive failure in the United States. Contraception 2004;70(2):89–96 [PubMed PMID: 15288211. Epub 2004/08/04. eng].
[3] Chiou CF, Trussell J, Reyes E, Knight K, Wallace J, Udani J, et al. Economic analysis of contraceptives for women. Contraception 2003;68(1):3–0 [PubMed PMID: 12878280. Epub 2003/07/25. eng]. [4] Winner B, Peipert JF, Zhao Q, Buckel C, Madden T, Allsworth JE, et al. Effectiveness of long-acting reversible contraception. N Engl J Med 2012 May 24;366(21):1998–2007 [PubMed PMID: 22621627. Epub 2012/05/25. eng]. [5] O'Hanley K, Huber DH. Postpartum IUDS: keys for success. Contraception 1992;45(4):351–61 [PubMed PMID: 1516367. Epub 1992/04/01. eng]. [6] Grimes DA, Lopez LM, Schulz KF, Van Vliet HA, Stanwood NL. Immediate post-partum insertion of intrauterine devices. Cochrane Database Syst Rev 2010;5:CD003036 [PubMed PMID: 20464722. Epub 2010/05/14. eng]. [7] Chen BA, Reeves MF, Hayes JL, Hohmann HL, Perriera LK, Creinin MD. Postplacental or delayed insertion of the levonorgestrel intrauterine device after vaginal delivery: a randomized controlled trial. Obstet Gynecol 2010;116(5):1079–87 [PubMed PMID: 20966692. Pubmed Central PMCID: 3104850. Epub 2010/10/23. eng]. [8] Eroglu K, Akkuzu G, Vural G, Dilbaz B, Akin A, Taskin L, et al. Comparison of efficacy and complications of IUD insertion in immediate postplacental/early postpartum period with interval period: 1 year follow-up. Contraception 2006 Nov;74(5):376–81 [PubMed PMID: 17046378. Epub 2006/10/19. eng]. [9] Celen S, Moroy P, Sucak A, Aktulay A, Danisman N. Clinical outcomes of early postplacental insertion of intrauterine contraceptive
6
[10]
[11]
[12]
[13]
T.C. Jatlaoui et al. / Contraception xx (2014) xxx–xxx devices. Contraception 2004;69(4):279–82 [PubMed PMID: 15033401. Epub 2004/03/23. eng]. Stumpf PG, Lenker RM. Insertion technique, not design, affects expulsion rates of postpartum intrauterine device. Contraception 1984;30(4):327–30 [PubMed PMID: 6509985. Epub 1984/10/01. eng]. Thiery M, Laufe L, Parewijck W, van der Pas H, van Kets H, Derom R, et al. Immediate postplacental IUD insertion: a randomized trial of sutured (Lippes Loop and TCu22OC) and non-sutured (TCu22OC) models. Contraception 1983 Oct;28(4):299–313 [PubMed PMID: 6365448. Epub 1983/10/01. eng]. U S. Medical Eligibility Criteria for Contraceptive Use, 2010. MMWR Recomm Rep. 2010;59(RR-4):1–86. [PubMed PMID: 20559203. Epub 2010/06/19. eng]. Hayes JL, Cwiak C, Goedken P, Zieman M. A pilot clinical trial of ultrasound-guided postplacental insertion of a levonorgestrel intrauterine device. Contraception 2007;76(4):292–6 [PubMed PMID: 17900440. Epub 2007/09/29. eng].
[14] Kapp N, Curtis KM. Intrauterine device insertion during the postpartum period: a systematic review. Contraception 2009;80(4):327–36 [PubMed PMID: 19751855. Epub 2009/09/16. eng]. [15] Xu J, Yang X, Gu X, Xu S, Zhou X, Chen Y, et al. Comparison between two techniques used in immediate postplacental insertion of TCu 380A intrauterine device: 36-month follow-up. Reprod Contracept 1999;10(3):156–62 [PubMed PMID: 12349462. Epub 2002/09/28. eng]. [16] Centers for Disease Control and Prevention–Sexually Transmitted Disease Surveillance 2011. In: Services DoHaH, editor. Atlanta, GA2012. [17] The postpartum intrauterine device: a training course for service providers: trainer's manual. New York: Engender Health, 2008. [18] ACOG Practice Bulletin No. 121: Long-acting reversible contraception: implants and intrauterine devices. Obstet Gynecol. 2011 Jul;118(1):184–96. [PubMed PMID: 21691183. Epub 2011/ 06/22. eng].
Original research article
Postplacental intrauterine device insertion at a teaching hospital☆ Tara C. Jatlaoui a,⁎, Michele Marcus b , Denise J. Jamieson a , Peggy Goedken a , Carrie Cwiak a a Department of Gynecology and Obstetrics, Emory University School of Medicine, Atlanta, GA, USA Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, GA 30322, USA Received 30 April 2013; revised 11 October 2013; accepted 16 October 2013
b
Abstract Objective: To determine whether postplacental intrauterine device (IUD) insertion can be safely and effectively performed within a teaching program. Study Design: This was a prospective cohort of 177 subjects planning vaginal delivery enrolled antenatally who desired postplacental IUD insertion of either the copper T380A IUD or levonorgestrel IUS. Insertions were performed primarily by resident physicians following a training session. Follow-up included a 4- to 8-week visit and telephone calls at 3 and 6 months. Results: Ninety-nine subjects underwent successful postplacental IUD insertion of 100 attempts. Seventeen expulsions (17%) were noted: 10 complete and 7 partial. The study identified no differences in outcome by training level; however, the study lacked statistical power to evaluate anything other than large differences. Conclusion: Postplacental IUD insertions can be safely and effectively performed within a training program. Implications: A training protocol may safely and feasibly be initiated among physicians, advanced practice clinicians or trainees with no prior experience with postplacental IUD insertion. By initiating this practice, access to highly effective contraception may increase for patients who have difficulty returning for a visit or otherwise receiving effective methods. © 2014 Elsevier Inc. All rights reserved. Keywords: Intrauterine device; Vaginal delivery; Postpartum contraception
1. Introduction The intrauterine device (IUD) has the highest continuation and cost-effectiveness compared to other reversible contraceptive methods [1–3]. Women who choose longacting reversible contraception, like the levonorgestrel intrauterine system (LNG-IUS) or copper T380A IUD, have 20 times less risk of unintended pregnancy compared to women who use pills, patch or the ring [4]. The postpartum period may be an optimal time for IUD uptake, especially for women who otherwise have difficulty with access, motivation or side effects [5]. A Cochrane Review of randomized trials found that postplacental IUD insertion (within 10 min of placental delivery) appeared safe and effective, though expulsion rates were higher compared to interval insertions ☆
Conflicts of interest: The authors have no conflicts of interest to declare. ⁎ Corresponding author. Department of Gynecology and Obstetrics, Emory University School of Medicine, Atlanta, GA 30303, USA. E-mail address: [email protected] (T.C. Jatlaoui). 0010-7824/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.contraception.2013.10.008
(more than 6 weeks after delivery) [6]. One randomized trial has compared postplacental insertion with interval insertion for the LNG-IUS, similarly finding higher expulsion rates compared to interval insertion (24% vs. 4.4%) [7]. Still, an increased risk of expulsion may be outweighed by an increase in access to highly effective reversible contraception in cases where significant barriers exist to both the device and its insertion. It has been suggested that expulsion rates are associated with the experience of the provider, whether across institutions or training levels [8,9]. However, training sessions stressing high fundal placement [10] and cumulative number of insertions have been positively correlated with decreased expulsion rates, suggesting that success rates can be improved significantly with training [11]. The specific aim of this study was to determine whether postplacental IUD insertion can be safely and effectively performed within a teaching program. The primary outcome was IUD continuation rate at 6 months. Secondary outcomes included IUD expulsion, infection, perforation and pregnancy.
2
T.C. Jatlaoui et al. / Contraception xx (2014) xxx–xxx
2. Methods and materials A prospective cohort study was conducted at Grady Memorial Hospital, Atlanta, Georgia, following approval by the Emory University Institutional Review Board. All patients receiving prenatal care and requesting an IUD as their primary method of postpartum contraception were offered participation. Additional inclusion criteria were as follows: age 18 years or older, anticipating a vaginal delivery, able to read and/or understand English or Spanish, able and willing to give informed consent and willing to follow-up over 6 months. Exclusion criteria at the time of placement included the following: contraindications to IUD placement [12], fever in labor, rupture of membranes for greater than 24 hours, need for cesarean section, postpartum hemorrhage, sexually transmitted infection in current pregnancy and leaving the hospital area within 6 months. Subjects could choose either the LNG-IUS or the CuT380A IUD. Participants reaffirmed consent for IUD insertion upon presentation for delivery. Those ineligible for the study were offered IUD insertion at their postpartum visit. All IUDs were provided at no cost regardless of study participation or time of insertion via grant funding that was independent of the study. An initial training session was conducted for all Emory University Gynecology and Obstetrics residents. Refresher training sessions were conducted every 6 weeks throughout the study. Sessions reviewed the following: counseling and consent for the IUD, insertion technique for both IUD types, postplacental IUD insertion technique and abdominal ultrasound guidance for fundal placement. For all eligible subjects, insertions were performed under ultrasound guidance using the LNG-IUS prepackaged inserter or ring forceps for CuT380A IUD, given the shorter inserter. If the IUD could not be placed with the inserter or ring forceps, hand insertion was utilized. The IUD strings were cut at the external os. Subjects returned for a 4- to 8-week postpartum visit. They were asked about satisfaction with the IUD and symptoms of expulsion or infection. Pelvic exams were performed to check for bleeding and infection, to visualize IUD strings and trim strings if necessary. Infection was defined as cervicitis (N 10 white blood cells per high-power field via wet prep) or Pelvic Inflammatory Disease (cervicitis, plus either cervical motion, adnexal or fundal tenderness). Transvaginal ultrasound was performed to ensure IUD intrauterine placement if necessary. Complete expulsion was defined as (1) no IUD seen on ultrasound and patient report of seeing the IUD come out or (2) an abdominal x-ray confirming absence of the IUD. Partial expulsion was defined as an IUD visible within the cervical canal by exam or ultrasound. Questionnaires were repeated at 3 and 6 months via telephone. If any contact was missed, subjects were contacted via telephone and certified mail. Subjects were considered lost to follow-up if they had not had at least one contact (visit or telephone) within the 6month study time.
The sample size of 100 was based primarily on feasibility; we estimated the number of women who would both desire a postplacental IUD and be eligible for placement at our site during a 2-year recruitment period. Statistical analysis was performed using SAS 9.3 (Cary, NC, USA). Primary analysis sought to determine 6-month continuation rate and identify risk factors for expulsion within 6 months. Bivariate analysis was performed using Fisher's exact test for categorical variables and the Student's t-test for continuous variables. Covariates included the following: age, race, parity, anesthesia, insertion time interval after placental delivery and IUD type. Stratified analysis was performed by the exposure of interest, provider training level, to identify whether covariates were associated with expulsion. Multivariate analysis of the association between provider training level and expulsion was performed with adjustment for age, parity, anesthesia, insertion time interval after placental delivery and IUD type.
3. Results Between March 2009 and March 2011, 177 women were enrolled in the study and 100 underwent attempts at postplacental IUD insertion. Those who were excluded either withdrew consent or were ineligible according to study protocol (Fig. 1). Ninety-nine attempts at insertion were successful. Of the 99 successful placements, 69 women selected LNG-IUS while 30 chose the CuT380A IUD. Demographics were not significantly different between women choosing LNG-IUS and women choosing CuT380A (Table 1). The majority of participants were black, had completed high school or GED, were unemployed, were covered by Medicaid, and were single. Mean time to insertion was 7 min (SD 4.6 min). Most insertions (72.7%) were performed by a post graduate year (PGY) 3 level resident or above, and most were performed under epidural anesthesia (63.6%). Provider assessment of insertion was documented as “very easy” or “somewhat easy” for 65 placements. Of the 99 women who received an IUD, 65 presented for a follow-up visit, and 88 had at least one contact within 6 months. Women who had follow-up were similar to the entire group who underwent postplacental insertion (Table 1). Only 21 subjects had follow-up at the 6-month mark, none of which reported IUD expulsion or removal. Seventeen expulsions (19.3%) occurred among the 88 women who completed one follow-up contact within 6 months. Ten complete expulsions were diagnosed, and all 10 women saw their IUD at the time of expulsion. Eight of these women returned to clinic and three of those had another IUD placed. Seven partial expulsions were diagnosed, and those IUDs were removed. Five of those women chose to have another IUD placed. Replacement IUDs were provided at no cost via independent grant funding.
T.C. Jatlaoui et al. / Contraception xx (2014) xxx–xxx
3
Fig. 1. Enrollments, insertions and exclusions.
No pregnancies or perforations were reported during the follow-up study period. Overall, 10 subjects (11%) were diagnosed with an infection. Seven subjects were empirically treated for cervicitis or PID as an outpatient at the postpartum visit. Three additional subjects reported infection diagnosed at outside clinics. No IUDs were removed for infection or for any other reason. Bivariate analysis found distribution of prior births to be the only statistically significant difference (p=.029) between expulsion and nonexpulsion groups (Table 2). IUDs placed by lower level providers (PGY1, PGY2) did not have a significantly different risk of expulsion compared to those placed by upper level providers (PGY3 and above). The point estimate for expulsion risk among women with anesthesia was half as frequent as the expulsion risk among women without anesthesia, but this was not statistically significant. Stratified analysis found no significant effect modification between provider level and covariates. Multivariate analysis found no significant associations with expulsion, provider level or other factors.
4. Discussion This study demonstrated that postplacental IUD insertion can be performed safely and effectively within a training program, utilizing training sessions and ultrasound guidance. Although the number of women who could be contacted at 6 months was low and precluded a calculation of IUD continuation rate, complication rates among the 88 women who had at least one follow-up contact were comparable to those found in recent studies of postplacental IUD insertion [6–9,13–15]. All 10 complete expulsions were recognized by the patient. Partial expulsions, however, were not always symptomatic and required exam and ultrasound for diagnosis. The number of infections was similar to the high baseline rate in Fulton County, Georgia, and did not require IUD removal [16]. The low 6-month contact rate may be
indicative of the population most in need of postplacental IUD insertion in that their access to health care outside pregnancy is limited. Importantly, we found that expulsion for an IUD placed by a less clinically experienced provider was not inferior to rates in the literature [6–9,13–15]. This is in contrast to older studies that have found more expulsions among providers with less experience. No studies had involved resident physicians placing the LNG-IUS or copper T380A IUD. In our study, placements were not limited by provider experience and occurred in real-time conditions of a labor and delivery unit. Only two insertions were performed by an attending physician who had prior experience with postplacental insertion. Standardized curricula and models are now available for postplacental IUD insertion training [17]. Refresher training sessions coincided with the short rotation schedule inherent in most training programs. Abdominal ultrasound was used to assure fundal placement and is available in most training hospitals in the United States. This suggests that a training protocol can feasibly be initiated within an institution staffed by physicians, advanced practice clinicians or trainees with no prior experience with postplacental IUD insertion. While we did not find a significant difference in the risk of expulsion among IUDs placed by lower level residents compared to more senior level residents, this study was underpowered to detect a meaningful difference. The expulsion risk of one of the groups would have had to exceed 46% in order to detect a significant difference from 17% at 80% power with a 95% confidence interval. The few years of experience difference among resident levels may not reflect the broader differences observed in previous studies between providers over several years of experience providing postplacental IUDs. In this study, expulsion was significantly associated with low parity. Although no previous studies had a similar finding, it is possible that the IUD is affected more so by uterine involution in primiparous women. The suggestion that lack of anesthesia may be associated with expulsion is
4
T.C. Jatlaoui et al. / Contraception xx (2014) xxx–xxx
Table 1 Characteristics of women receiving a postplacental IUD, by IUD type Total insertions (n=99) Age (y), mean (SD) Race Black Asian Hispanic Unspecified School Less than 8y Some HS HS/GED Some college Other Employment Unemployed Full time Part time Unspecified Annual income $10,000 or less $10,000–30,000 Above $30,000 Unspecified Prior births 0 1 2 3 or greater Unspecified Insurance Yes Medicaid Other Unknown No Unknown Future fertility No Yes Unsure Unknown Relationship status Single Married Unspecified
Completed follow-up* (n=88)
23.7 (4.6)
Mirena (%) (n= 69)
23.7 (4.6)
p-Value a
Paragard (%) (n=30)
23.7 (4.4)
23.7 (5.2)
73 1 19 6
73.7% 1.0% 19.2% 6.1%
67 0 16 5
76.1% 0.0% 18.2% 5.7%
53 0 11 5
76.8% 0.0% 15.9% 7.2%
20 1 8 0
66.7% 3.3% 26.7% 0.0%
6 23 43 17 10
6.1% 23.2% 43.4% 17.2% 10.1%
5 19 40 15 9
5.7% 21.6% 45.5% 17.0% 10.2%
3 16 31 12 7
4.3% 23.2% 44.9% 17.4% 10.1%
3 7 12 5 3
10.0% 23.3% 40.0% 10.0% 10.0%
64 13 12 10
64.6% 13.1% 12.1% 10.1%
57 12 10 4
64.8% 13.6% 11.4% 4.5%
41 10 10 8
59.4% 14.5% 14.5% 11.6%
23 3 2 2
76.7% 10.0% 6.7% 6.7%
50 26 2 21
50.5% 26.3% 2.0% 21.2%
43 24 2 19
48.9% 27.3% 2.3% 21.6%
35 18 2 14
50.7% 26.1% 2.9% 20.3%
15 8 0 7
50.0% 26.7% 0.0% 23.3%
17 34 22 25 1
17.2% 34.3% 22.2% 25.3% 1.0%
14 31 19 24 0
15.9% 35.2% 21.6% 27.3% 0.0%
12 23 16 18 0
17.4% 33.3% 23.2% 26.1% 0.0%
5 11 6 7 1
16.7% 36.7% 20.0% 23.3% 3.3%
78 67 3 8 14 7
78.8% 67.7% 3.0% 8.1% 14.1% 7.1%
68 61 3 4 14 4
77.3% — — — 15.9% 8.0%
54 45 2 7 9 6
78.3% 83.3% 3.7% 13.0% 13.0% 8.7%
24 22 1 1 5 1
80.0% 91.7% 4.2% 4.2% 16.7% 3.3%
57 22 14 6
57.6% 22.2% 14.1% 6.1%
52 19 12 5
59.1% 21.6% 13.6% 5.7%
35 17 12 5
50.7% 24.6% 17.4% 7.2%
22 5 2 1
73.3% 16.7% 6.7% 3.3%
78 15 6
78.8% 15.2% 6.1%
69 14 5
78.4% 15.9% 5.7%
53 11 5
76.8% 15.9% 7.2%
25 4 1
83.3% 13.3% 3.3%
.94 .12
.74
.50
.99
.98
.83
.15
.77
* With at least one follow-up contact in 6 months. a Mirena vs Paragard.
also a new finding. Further studies with adequate power to detect a difference are needed to determine if an association truly exists for parity or anesthesia. This study was limited by sample size and the low percentage of women who were contacted at 6 months. Although 88% of women had study contact after insertion, outcomes such as expulsion, infection, pregnancy and perforation may actually be higher than reported or may change with time since insertion. This study population also represents a particular population, and findings may not apply to all women who desire a postplacental IUD; though
in a population at increased risk for adverse health outcomes, our findings of similar complication rates are reassuring. While the risk of expulsion has been shown to be higher with postplacental insertion compared to interval insertion, certain patient populations may benefit more from the access and delivery of highly effective reversible contraception at the time of vaginal delivery, especially patients who have difficulty returning for postpartum visits. Postplacental IUD placement is therefore supported by the US Medical Eligibility Criteria for Contraceptive Use and the American College of Obstetrics and Gynecology [12,18]. By initiating
T.C. Jatlaoui et al. / Contraception xx (2014) xxx–xxx
5
Table 2 Characteristics of subjects with follow-up* by expulsion versus nonexpulsion (n= 88) Covariate
Expulsion (n=17)
a
Age , mean (SD) Race Black Hispanic Unspecified Prior births 0 1 2 3 or greater Anesthesia None IV pain medicine Epidural Unspecified Insertion time from placenta delivery (min), a mean (SD) IUD type LNG-IUS CuT380A Provider level PGY1 or PGY2 PGY3 or above Infection after discharge No Yes Provider assessment Very easy Somewhat easy Somewhat difficult Very difficult Unspecified
Nonexpulsion (n= 71)
23.5 (5.2)
23.8 (4.5)
p-Value .8 1
14 3 0
82.4% 17.6% 0.0%
53 13 5
74.6% 18.3% 0.1%
1 9 6 1
5.9% 52.9% 35.3% 5.9%
13 22 13 23
18.3% 31.0% 18.3% 32.4%
6 2 8 1
35.3% 11.8% 47.1% 5.9%
11 10 48 2
15.5% 14.1% 67.6% 2.8%
.029
.15
7.5 (5.4)
5 (2.8)
.101 1
12 5
70.6% 29.4%
50 21
70.4% 29.6%
2 15
11.8% 88.2%
22 49
31.0% 69.0%
17 0
100% 0.0%
60 11
84.5% 15.5%
8 5 2 0 2
47.1% 29.4% 11.8% 0.0% 11.8%
25 20 11 3 12
35.2% 28.2% 15.5% 4.2% 16.9%
.14
.11
.932
*With at least one follow-up contact in 6 months. a t-Test.
postplacental IUD insertion protocols in residency training facilities in the United States, access to long-acting reversible contraception could increase for patients who have difficulty returning for a visit or otherwise receiving a highly effective method. Acknowledgments ParaGard Copper T380A IUDs and Mirena Levonorgestrel IUS were provided by an anonymous grant. The grant had no involvement in study design, analysis or reporting. This study was possible with the cooperation and participation of the Emory Department of Gynecology and Obstetrics faculty and residents. References [1] ACOG Practice Bulletin. Clinical management guidelines for obstetrician-gynecologists. Number 59, January 2005. Intrauterine device. Obstet Gynecol 2005;105(1):223–32 [PubMed PMID: 15625179. Epub 2004/12/31. eng]. [2] Trussell J. Contraceptive failure in the United States. Contraception 2004;70(2):89–96 [PubMed PMID: 15288211. Epub 2004/08/04. eng].
[3] Chiou CF, Trussell J, Reyes E, Knight K, Wallace J, Udani J, et al. Economic analysis of contraceptives for women. Contraception 2003;68(1):3–0 [PubMed PMID: 12878280. Epub 2003/07/25. eng]. [4] Winner B, Peipert JF, Zhao Q, Buckel C, Madden T, Allsworth JE, et al. Effectiveness of long-acting reversible contraception. N Engl J Med 2012 May 24;366(21):1998–2007 [PubMed PMID: 22621627. Epub 2012/05/25. eng]. [5] O'Hanley K, Huber DH. Postpartum IUDS: keys for success. Contraception 1992;45(4):351–61 [PubMed PMID: 1516367. Epub 1992/04/01. eng]. [6] Grimes DA, Lopez LM, Schulz KF, Van Vliet HA, Stanwood NL. Immediate post-partum insertion of intrauterine devices. Cochrane Database Syst Rev 2010;5:CD003036 [PubMed PMID: 20464722. Epub 2010/05/14. eng]. [7] Chen BA, Reeves MF, Hayes JL, Hohmann HL, Perriera LK, Creinin MD. Postplacental or delayed insertion of the levonorgestrel intrauterine device after vaginal delivery: a randomized controlled trial. Obstet Gynecol 2010;116(5):1079–87 [PubMed PMID: 20966692. Pubmed Central PMCID: 3104850. Epub 2010/10/23. eng]. [8] Eroglu K, Akkuzu G, Vural G, Dilbaz B, Akin A, Taskin L, et al. Comparison of efficacy and complications of IUD insertion in immediate postplacental/early postpartum period with interval period: 1 year follow-up. Contraception 2006 Nov;74(5):376–81 [PubMed PMID: 17046378. Epub 2006/10/19. eng]. [9] Celen S, Moroy P, Sucak A, Aktulay A, Danisman N. Clinical outcomes of early postplacental insertion of intrauterine contraceptive
6
[10]
[11]
[12]
[13]
T.C. Jatlaoui et al. / Contraception xx (2014) xxx–xxx devices. Contraception 2004;69(4):279–82 [PubMed PMID: 15033401. Epub 2004/03/23. eng]. Stumpf PG, Lenker RM. Insertion technique, not design, affects expulsion rates of postpartum intrauterine device. Contraception 1984;30(4):327–30 [PubMed PMID: 6509985. Epub 1984/10/01. eng]. Thiery M, Laufe L, Parewijck W, van der Pas H, van Kets H, Derom R, et al. Immediate postplacental IUD insertion: a randomized trial of sutured (Lippes Loop and TCu22OC) and non-sutured (TCu22OC) models. Contraception 1983 Oct;28(4):299–313 [PubMed PMID: 6365448. Epub 1983/10/01. eng]. U S. Medical Eligibility Criteria for Contraceptive Use, 2010. MMWR Recomm Rep. 2010;59(RR-4):1–86. [PubMed PMID: 20559203. Epub 2010/06/19. eng]. Hayes JL, Cwiak C, Goedken P, Zieman M. A pilot clinical trial of ultrasound-guided postplacental insertion of a levonorgestrel intrauterine device. Contraception 2007;76(4):292–6 [PubMed PMID: 17900440. Epub 2007/09/29. eng].
[14] Kapp N, Curtis KM. Intrauterine device insertion during the postpartum period: a systematic review. Contraception 2009;80(4):327–36 [PubMed PMID: 19751855. Epub 2009/09/16. eng]. [15] Xu J, Yang X, Gu X, Xu S, Zhou X, Chen Y, et al. Comparison between two techniques used in immediate postplacental insertion of TCu 380A intrauterine device: 36-month follow-up. Reprod Contracept 1999;10(3):156–62 [PubMed PMID: 12349462. Epub 2002/09/28. eng]. [16] Centers for Disease Control and Prevention–Sexually Transmitted Disease Surveillance 2011. In: Services DoHaH, editor. Atlanta, GA2012. [17] The postpartum intrauterine device: a training course for service providers: trainer's manual. New York: Engender Health, 2008. [18] ACOG Practice Bulletin No. 121: Long-acting reversible contraception: implants and intrauterine devices. Obstet Gynecol. 2011 Jul;118(1):184–96. [PubMed PMID: 21691183. Epub 2011/ 06/22. eng].
