Amenorrhea rates and predictors during 1 year of levonorgestrel 52 mg intrauterine system use

Abstract

Objective

The objective was to evaluate amenorrhea patterns and predictors of amenorrhea during the first year after levonorgestrel 52 mg intrauterine system (IUS) placement.

Study design

This cohort analysis includes 1714 nulliparous and parous women who received a Liletta® levonorgestrel 52 mg IUS in a multicenter trial to evaluate efficacy and safety for up to 8 years. Participants maintained a daily diary with bleeding information. We assessed bleeding patterns in 90-day intervals; amenorrhea was defined as no bleeding or spotting in the preceding 90 days. We employed multivariable regression to identify predictors of amenorrhea at 12 months. The predictor analysis only included women not using a levonorgestrel IUS in the month prior to study enrollment.

Results

In the month before enrollment, 148 and 1566 women, respectively, had used and not used a levonorgestrel IUS. Prior users averaged 50±19 months of use before IUS placement; 38.4% of these women reported amenorrhea at 12 months. Amenorrhea rates for non-prior-users at 3, 6, 9 and 12 months were 0.2%, 9.1%, 17.2% and 16.9%, respectively. During the first 12 months, 29 (1.7%) women discontinued for bleeding irregularities; no women discontinued for amenorrhea. The only significant predictor of amenorrhea at 12 months was self-reported baseline duration of menstrual flow of fewer than 7 days vs. 7 or more days (18.2% vs. 5.2%, adjusted odds ratio 3.70 [1.69, 8.07]). We found no relationships between 12-month amenorrhea rates and age, parity, race, body mass index, baseline flow intensity or hormonal contraception use immediately prior to IUS placement.

Conclusions

Amenorrhea rates during the first year of levonorgestrel 52 mg IUS use are similar at 9 and 12 months. Amenorrhea at 12 months is most common among women with shorter baseline duration of menstrual flow.

Implications statement

This information provides more data for clinicians when counseling women about amenorrhea expectations, especially since women seeking a levonorgestrel 52 mg IUS for contraception are different than women desiring treatment for heavy menstrual bleeding. Amenorrhea at 12 months is most common among women with shorter baseline duration of menstrual flow.

Introduction

The introduction of the levonorgestrel (LNG) 52 mg intrauterine system (IUS) in Europe in 1990 and the United States in 2000 changed acceptance of intrauterine contraception. Using an intrauterine contraceptive became more appealing due to the potential for less menstrual bleeding and dysmenorrhea with this novel option. For many women, amenorrhea became a desirable effect of using a hormonal contraceptive. The first LNG 52 mg IUS introduced to the market reported a 1-year amenorrhea rate of 20% but lacked information about when amenorrhea occurred during the first year or who becomes amenorrheic [1], [2]. The data for the 20% amenorrhea rate were obtained primarily from Scandinavian multiparous women, 75% of whom had used intrauterine contraception previously [3].

The significant reduction in overall menstrual bleeding with the LNG 52 mg IUS encouraged further studies which led to its approval for treatment of heavy menstrual bleeding (HMB) [4], [5]. Much has been published about decrease in menstrual flow and amenorrhea in women using a LNG 52 mg IUS for HMB [6] but little about changes in women using the product primarily for contraception. Because HMB significantly increases with age and peaks in the perimenopausal period [7], those women who use a LNG IUS primarily for HMB likely differ in age and other characteristics from those who use it primarily for contraception. de Jonge et al. [8] reported predictors of 12-month oligoamenorrhea rates in 141 mostly nonobese women using a LNG 52 mg IUS for contraception (n=98) or HMB (n=52). The investigators reported that women in the contraception group were significantly younger (34 vs. 39 years) and had fewer subjective bleeding abnormalities at baseline (20% vs. 96%). In univariate analyses, duration of menses less than 5 days and absence of HMB at baseline predicted oligoamenorrhea at 12 months.

This report describes amenorrhea rates and predictors over the first year of use among participants in “A Comprehensive Contraceptive Efficacy and Safety Study of an IUS” (ACCESS IUS). This US study of women primarily using the LNG 52 mg IUS for contraception had significant proportions of obese and nulliparous women, which allowed for a broad, generalizable assessment of the characteristics of women who develop amenorrhea [9].