Risk of Pregnancy With Levonorgestrel-Releasing Intrauterine System Placement 6–14 Days After Unprotected Sexual Intercourse

Précis

Pregnancy is unlikely when a levonorgestrel-releasing intrauterine system (52 mg) is placed 6–14 days after unprotected intercourse.

Introduction

In this report, we evaluate the chance of pregnancy with levonorgestrel 52 mg intrauterine system (LNG-IUS) placement 6–14 days after unprotected intercourse. Initiation of contraception after unprotected intercourse in the previous 6–14 days [1–4] increases contraceptive use and decreases risk of unplanned pregnancy [5–9]. However, for patients reporting recent unprotected intercourse, a negative urine pregnancy test on day of placement, may not identify a very early pregnancy. Studies of copper IUD placement six or more days following unprotected intercourse reported no pregnancies [1, 10, 11].

Methods

We combined data from three prospective observational studies and one randomized controlled trial. All studies enrolled participants from June 2013 to September 2019, who presented for contraception and reported unprotected intercourse occurring 6–14 days prior to presentation. All participants underwent LNG-IUS placement on the day of presentation. All protocols required negative point of care urine pregnancy testing and all collected demographic information and reporting of days with unprotected intercourse in the two weeks prior to LNG 52mg IUS placement. Two to 4 weeks after LNG-IUS placement, participants reported results of home pregnancy testing. All protocols followed participants for up to 6 months to assess method continuation and the possibility of pregnancy. The four study protocols were conducted at seven family planning clinics in Utah, Pennsylvania and Minnesota as previously described by Thompson et al. [1]. The two clinical research sites in Minnesota and Pennsylvania shared a protocol and enrolled participants desiring an LNG-IUS who reported unprotected intercourse 6–14 days previously. All other data came from participants reporting unprotected intercourse 6–14 days prior to LNG-IUS placement in three other research protocols in Utah [12–14]. Two of these protocols were observational studies - 65 participants had a same day placement of LNG-IUS and an additional 24 participants received the LNG-IUS simultaneously with oral LNG for emergency contraception and were included in this analysis. The lone randomized controlled trial contributing data to this analysis allocated emergency contraception users 1:1 to either an LNG 52 mg IUS or a copper T380A IUD. In that trial 54, participants used the LNG-IUS alone as emergency contraception. We performed statistical analysis using Stata version 15 (StataCorp LP, College Station, TX) using one-way analysis of variance, Chi-square tests, and the exact binomial method when appropriate.

Results

This study includes 187 individuals across the four combined protocols (Appendix 1, available online at http://links.lww.com/xxx, Table 1). The majority of participants reported regular menstrual cycles and multiple episodes of unprotected sex in the 6–14 days prior to enrollment. Follow up home pregnancy test results were reported by 123 participants (66%) 2–4 weeks after LNG-IUS placement; 64 participants (34%) self-reported pregnancy status without a pregnancy test. Only 1 participant reported a pregnancy within four weeks of LNG-IUS placement (0.5%; 95% confidence interval 0.01%−2.9%). We performed subanalyses (Appendix 2, available online at http://links.lww.com/xxx).

Table 1.

Characteristics of participants with a negative pregnancy test who initiated a 52 mg levonorgestrel-releasing intrauterine system 6–14 days after unprotected intercourse

CHARACTERISTIC	n (%)
Total participants	187 (100)
Age in years*
18–19	53 (28)
20–24	54 (29)
25–29	38 (20)
30–34	28 (15)
35+	14 (7)
Previously pregnant*	129 (69)
Relationship status
Single	84 (45)
Married or cohabitating	97 (52)
Missing	6 (3)
Education
High school or less	91 (49)
College or vocational training	96 (51)
Body Mass Index (kg/m2)
Underweight (<18.5)	10 (5)
Normal (18.5–24.9)	51 (27)
Overweight (25–29.9)	33 (18)
Obese (≥30)	27 (14)
Missing†	66 (35)
Cycle length 25–35 days	159 (85)
Menstrual cycle day at placement‡	17 ± 11
Unprotected sex in fertile window§	68 (36)
Multiple episodes of unprotected sex in 6–14 days	116 (62)

Group percentages may not total 100% due to rounding.

^*Distributions of age, race, and parity differed across study protocols (p<0.05) using chi2; other variables statistically similar.

^†Body mass index has large number of missing values as it was not collected in the HER protocol.

^‡Data presented as mean ± standard deviation.

^§Fertile window is defined as 5 days prior to and one day after calculated ovulation date [11].

Discussion

Pregnancy is unlikely with LNG-IUS placement 6–14 days following unprotected intercourse, even for women reporting multiple episodes of unprotected intercourse. Further, pregnancy rates appear similar to those among patients receiving a copper IUD 6–14 days after unprotected intercourse, which have been reported to be 0–2.7% [1, 10, 11].