Amna I. Dermish

Copper T380 intrauterine device for emergency contraception: highly effective at any time in the menstrual cycle

STUDY QUESTION Does the efficacy of placing a copper intrauterine device (IUD) for emergency contraception (EC) to prevent pregnancy depend on menstrual cycle timing and timing of unprotected intercourse (UPI)? SUMMARY ANSWER If the urine pregnancy test is negative prior to IUD placement, the copper IUD is highly effective for EC at any point in the menstrual cycle. WHAT IS KNOWN ALREADY The use of the Copper T380A for EC has been encouraged by the failure of oral EC methods to decrease rates of unintended pregnancy and the documented success of the IUD in reducing unintended pregnancies. However scant data exist regarding the efficacy and safety of IUD insertion for EC when accounting for menstrual cycle timing and time since UPI. STUDY DESIGN, SIZE, DURATION This is a secondary analysis of data obtained from a previously published prospective cohort study of women who received the Copper T380A IUD for EC between July 1997 and January 2000. We included 1840 participants according to the study inclusion criteria of a known last menstrual period (LMP) and cycle lengths of 25-35 days. PARTICIPANTS/MATERIALS, SETTING, METHODS The original study included women aged between 18 and 44 years who presented for EC at 18 sites throughout China and who had regular menstrual cycles between 24 and 42 days, a known LMP, UPI within 120 h (5 days) and a negative urine pregnancy test (cutoff <25 IU/ml). Women with uncertain LMP dates were excluded. This study included only participants with cycle lengths of 25-35 days. MAIN RESULTS AND THE ROLE OF CHANCE Among the 1840 participants with usual cycle lengths of 25-35 days, 850 (46.2%) had their IUD inserted following UPI in the expected fertile window and 84 (4.6%) had the insertion >5 days after the predicted ovulation day and 52 (2.8%) had the insertion >5 days after UPI. There were no pregnancies in the first month among the 1771 women who had information available regarding their 1-month follow-up pregnancy test. LIMITATIONS, REASONS FOR CAUTION This was a secondary analysis of an observational study, and thus participants were not randomized to an alternative postcoital method. There were a small number of women who had UPI >5 days after their predicted ovulation day thus limiting the confidence of assuring a low risk of pregnancy in this situation. The ovulation day was calculated based on the LMP prior to IUD insertion and not on the subsequent first day of menses following IUD insertion. WIDER IMPLICATIONS OF THE FINDINGS If the urine pregnancy test is negative prior to IUD placement, the copper IUD is likely to be effective for EC at almost any point in the menstrual cycle. STUDY FUNDING/COMPETING INTEREST(S) The original study was funded by the UNDP/UNPFA/WHO/World Bank Special Programme of Research, Development and Research Training in Human Reproduction. The donors and sponsors of the study had no role in the study design, data collection, data analysis, data interpretation, writing of the report or the decision to submit the paper for publication.

Self-administered misoprostol or placebo before intrauterine device insertion in nulliparous women: A randomized controlled trial

OBJECTIVES: To estimate the effects of self-administered misoprostol compared with placebo before intrauterine device (IUD) insertion in nulliparous women. METHOD: Nulliparous women requesting either the copper T380A or levonorgestrel IUD were randomized to self-administer either 400 &mgr;g of misoprostol or placebo (vaginally or buccally) 3–4 hours before the IUD insertion appointment. The primary outcome was health care provider–perceived ease of insertion recorded on a visual analog scale (anchors: 0 extremely easy, 100 impossible). Patients completed questionnaires addressing pain using a validated visual analog scale (anchors: 0 none, 100 worst imaginable) before insertion, immediately postinsertion, and before clinic discharge. RESULTS: Of the 108 women enrolled in the study, 54 received misoprostol and 54 received placebo. There was no significant difference in perceived ease of insertion between the two groups (25.0 mm [standard error 3.5] compared with 27.4 mm [standard error 3.5], P=.64). Patients who received misoprostol before IUD insertion had significantly higher pain scores before placement (17.1 mm [standard error 3.5] compared with 4.7 mm [standard error 2.0], P=.003). Groups did not differ in perception of pain during IUD insertion (58.4 mm [standard error 3.3] compared with 56.9 mm [standard error 3.0], P=.74). There were two expulsions in the misoprostol group and none in the placebo group. Failed insertions, need for adjuvant pain medication, and need for cervical dilation or ultrasonographic guidance did not differ between the two groups. CONCLUSION: Self-administered misoprostol before IUD insertion does not ease IUD insertion or reduce patient-perceived pain in nulliparous women. These data do not support the routine use of misoprostol before IUD insertion in nulliparous women. CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov, www.clinicaltrials.gov, NCT00886834. LEVEL OF EVIDENCE: I

Emergency contraception with a copper IUD or oral levonorgestrel: an observational study of 1-year pregnancy rates.

OBJECTIVE We investigated the 1-year pregnancy rates for emergency contraception (EC) users who selected the copper T380 intrauterine device (IUD) or oral levonorgestrel (LNG) for EC. STUDY DESIGN This prospective study followed women for 1 year after choosing either the copper T380 IUD or oral LNG for EC. The study was powered to detect a 6% difference in pregnancy rates within the year after presenting for EC. RESULTS Of the 542 women who presented for EC, agreed to participate in the trial and met the inclusion criteria, 215 (40%) chose the copper IUD and 327 (60%) chose oral LNG. In the IUD group, 127 (59%) were nulligravid. IUD insertion failed in 42 women (19%). The 1-year follow-up rate was 443/542 (82%); 64% of IUD users contacted at 1 year still had their IUDs in place. The 1-year cumulative pregnancy rate in women choosing the IUD was 6.5% vs. 12.2% in those choosing oral LNG [hazard ratio (HR) 0.53, 95% confidence interval (CI): 0.29-0.97, p=.041]. By type of EC method actually received, corresponding values were 5.2% for copper IUD users vs. 12.3% for oral LNG users (HR 0.42, 95% CI: 0.20-0.85, p=.017). A multivariable logistic regression model controlling for demographic variables demonstrates that women who chose the IUD for EC had fewer pregnancies in the following year than those who chose oral LNG (HR 0.50, 95% CI: 0.26-0.96, p=.037). CONCLUSION One year after presenting for EC, women choosing the copper IUD for EC were half as likely to have a pregnancy compared to those choosing oral LNG. IMPLICATIONS Compared to EC users who choose oral levonorgestrel, those who select the copper IUD have lower rates of pregnancy in the next year. Greater use of the copper IUD for EC may lower rates of unintended pregnancy in high-risk women.

Vaginal microbiome changes with levonorgestrel intrauterine system placement

OBJECTIVE To investigate changes in female genital tract bacterial ecology associated with levonorgestrel intrauterine system (LNG IUS) use. STUDY DESIGN Study participants were regularly cycling Caucasian women without recent history of pregnancy or exogenous hormone use. Vaginal, cervical and uterine samples were obtained at nine time intervals, from 1 week before to 12 weeks after LNG IUS placement. Replicate paired vaginal and cervical specimens were collected to determine the consistency of specimen collection and processing. RESULTS A total of 406 samples from 11 women were available for analysis after DNA extraction, amplification and species identification. Of the 355 bacterial species or genera detected, Lactobacillus crispatus was most prevalent representing 48.9% of over 6 million total reads. L. crispatus reads of replicate vaginal samples were not significantly different [odds ratio (OR) 0.79, 0.36-1.73]. In addition, L. crispatus reads of vaginal and cervical samples from the same visit were not significantly different (OR 0.69, 0.31-1.51). Compared to sampling visits prior to LNG IUS placement, sampling visits after LNG IUS insertion were more likely to have L. crispatus reads greater than 50% of total reads (OR 2.13, 1.01-4.48). Of the 63 uterine samples demonstrating bacteria, Burkholderia genus proteobacteria, a common environmental contaminant, were most prevalent both before and after LNG IUS insertion accounting for 48.0% of all uterine sample reads. CONCLUSION The vaginal microbiome changes very little in response to LNG IUS placement. Further study is needed to place this finding in context with clinical outcomes. IMPLICATIONS There do not appear to be any clinically important differences in the vaginal microbiome in response to insertion of the LNG IUS.

Failed IUD insertions in community practice: an under-recognized problem?

BACKGROUND The data analysis was conducted to describe the rate of unsuccessful copper T380A intrauterine device (IUD) insertions among women using the IUD for emergency contraception (EC) at community family planning clinics in Utah. METHODS These data were obtained from a prospective observational trial of women choosing the copper T380A IUD for EC. Insertions were performed by nurse practitioners at two family planning clinics in order to generalize findings to the type of service setting most likely to employ this intervention. Adjuvant measures to facilitate difficult IUD insertions (cervical anesthesia, dilation, pain medication, and use of ultrasound guidance) were not utilized. The effect of parity on IUD insertion success was determined using exact logistic regression models adjusted for individual practitioner failure rates. RESULTS Six providers performed 197 IUD insertion attempts. These providers had a mean of 14.1 years of experience (range 1-27, S.D. ±12.5). Among nulliparous women, 27 of 138 (19.6%) IUD insertions were unsuccessful. In parous women, 8 of 59 IUD insertions were unsuccessful (13.6%). The adjusted odds ratio (aOR) showed that IUD insertion failure was more likely in nulliparous women compared to parous women (aOR=2.31, 95% CI 0.90-6.52, p=.09). CONCLUSION The high rate of unsuccessful IUD insertions reported here, particularly for nulliparous women, suggests that the true insertion failure rate of providers who are not employing additional tools for difficult insertions may be much higher than reported in clinical trials. Further investigation is necessary to determine if this is a common problem and, if so, to assess if the use of adjuvant measures will reduce the number of unsuccessful IUD insertions.

The copper intrauterine device for emergency contraception: an opportunity to provide the optimal emergency contraception method and transition to highly effective contraception

Worldwide, 40% of all pregnancies are unintended. Widespread, over-the-counter availability of oral emergency contraception (EC) has not reduced unintended pregnancy rates. The EC visit presents an opportunity to initiate a highly effective method of contraception in a population at high risk of unintended pregnancy who are actively seeking to avoid pregnancy. The copper intrauterine device (IUD), the most effective method of EC, continues to provide contraception as effective as sterilization for up to 12 years, and it should be offered as the first-line method of EC wherever possible. Increased demand for and supply of the copper IUD for EC may have an important role in reducing rates of unintended pregnancy. The EC visit should include access to the copper IUD as optimal care but should ideally include access to all highly effective methods of contraception.

Evaluation of an intervention designed to improve the management of difficult IUD insertions by advanced practice clinicians

OBJECTIVES To assess whether clinical skills training in cervical dilation and paracervical anesthesia, as well as the introduction of a clinical protocol for difficult insertions, increased intrauterine device (IUD) insertion success rates among advanced practice clinicians (APCs) including womens health and family practice nurse practitioners, physician assistants and certified nurse midwives. STUDY DESIGN This prospective study assessed an intervention to improve IUD insertions among APCs at six family planning clinics in Utah. We collected data on IUD insertions performed by participating clinicians during two observation phases: preintervention (Phase 1) and postintervention (Phase 2). Using electronic medical records, we tracked patient characteristics, IUD insertion success, difficulties and failures. We constructed a mixed-effects logistic regression model to control for provider and patient demographics influencing insertion success rates. RESULTS Nine clinicians participated in both preintervention and postintervention phases. The analysis included 428 IUD insertion procedures (242 preintervention and 186 postintervention). During Phase 1, 31/242 (12.8%) insertions failed. Insertion rates in Phase 2 improved with only 8/186 (4.3%) failures. Using mixed-effects logistic regression, the odds of a successful insertion postintervention was 4.8 times greater than preintervention (aOR=4.8.95% CI 1.8-12.7) when controlling for provider and patient characteristics. Increased risk of insertion failure was associated with nulliparity and younger age during Phase 1, but not during Phase 2. CONCLUSIONS A brief training for APCs and the use of a clinical protocol for difficult insertions may be able to improve IUD insertion rates. IMPLICATIONS Clinics with high rates of IUD insertion failure can improve care with a simple intervention, which may yield significant benefits in IUD service delivery.

Postabortion contraceptive use and continuation when long-acting reversible contraception is free

OBJECTIVE To compare preference for long-acting contraception (LARC) and subsequent use, year-long continuation, and pregnancy among women after induced abortion who were and were not eligible to participate in a specialized funding program that provided LARC at no cost. METHODS Between October 2014 and March 2016, we conducted a prospective study of abortion patients at Planned Parenthood in Austin, Texas (located in Travis County). We compared our primary outcome of interest, postabortion LARC use, among women who were eligible for the specialized funding program (low-income, uninsured, Travis County residents) and two groups who were ineligible (low-income, uninsured, non-Travis County residents, and higher income or insured women). Secondary outcomes of interest included preabortion preference for LARC and 1-year continuation and pregnancy rates among the three groups. RESULTS Among 518 women, preabortion preference for LARC was high among all three groups (low-income eligible: 64% [91/143]; low-income ineligible: 44% [49/112]; and higher income 55% [146/263]). However, low-income eligible participants were more likely to receive LARC (65% [93/143] compared with 5% [6/112] and 24% [62/263], respectively, P<.05). Specifically, after adjusting for age, race-ethnicity, and education, low-income eligible participants had a 10-fold greater incidence of receiving postabortion LARC compared with low-income ineligible participants (incidence rate ratio 10.13, 95% confidence interval [CI] 4.68-21.91). Among low-income eligible and higher income women who received postabortion LARC, 1-year continuation was 90% (95% CI 82-97%) and 86% (95% CI 76-97%), respectively. One-year pregnancy risk was higher among low-income ineligible than low-income eligible women (hazard ratio 3.28, 95% CI 1.15-9.31). CONCLUSION Preference for postabortion LARC was high among all three eligibility groups, yet women with access to no-cost LARC were more likely to use and continue these methods. Low-income ineligible women were far more likely to use less effective contraception and become pregnant. Specialized funding programs can play an important role in immediate postabortion contraceptive provision, particularly in settings where state funding is limited.

We should really keep in touch: predictors of the ability to maintain contact with contraception clinical trial participants over 12 months

OBJECTIVES This study assesses the ability to maintain contact with participants enrolled in an emergency contraception (EC) trial with 12 months of follow-up based on the modes of contact they provided at enrollment. STUDY DESIGN Data came from a clinical trial offering women the copper intrauterine device or oral levonorgestrel for EC. A modified Poisson regression was used to assess predictors associated with the ability to contact study participants 12 months after enrollment. RESULTS Data were available for 542 participants; 443 (82%) could be contacted at 12 months. Contact at 12 months was greatest for those whose preferred the method of contact was text messaging, e-mail or any (62/68; 91% contacted) and worst for the 18 who had a landline phone (only 7 contacted; 39%). After controlling for age, having an e-mail address, text messaging, language preference, type of EC chosen and insurance, preferred contact other than phone increased the likelihood of follow-up by 10% [risk ratio (RR) 1.1 95% confidence interval (CI) 1.0-1.2], while having a landline reduced a womans likelihood of being contacted at 12 months by 50% compared to women with a contract cell (RR 0.5, 95% CI 0.3-1.0). CONCLUSION The few women with a landline for contact had poor follow-up at 1 year, while women who preferred e-mail or text had the highest rate of follow-up. IMPLICATIONS Understanding how best to reduce loss to follow-up is an essential component of conducting a contraceptive clinical trial. Improved participant retention maximizes internal validity and allows for important clinical outcomes, such as pregnancy, to be assessed.