Comparative Contraceptive Effectiveness of Levonorgestrel-Releasing and Copper Intrauterine Devices

Abstract

Contraceptive failure rates are low among users of long-acting reversible methods of contraception. Previous studies have reported failure rates for copper and levonorgestrel-releasing intrauterine devices (LNG-IUSs) of 0.1 to 2.2 per 100 women-years and 0.1 to 0.6 per 100 women-years, respectively. The European Active Surveillance Study for Intrauterine Devices (EURAS IUD) was a multinational, prospective, noninterventional cohort study designed to determine the incidence of uterine perforation and to compare contraceptive effectiveness and pregnancy-related outcomes, including ectopic pregnancies in users of LNG-IUS and copper IUDs. The present study was part of the EURAS IUD and designed to measure the rate of unintended pregnancies among a population of new users of LNG-IUS (releasing 20 μg LNG daily) and copper IUD who were seen in routine clinical practice and to describe associated complications. Study participants were recruited from 6 European countries between from 2006 to 2012 by a network of health care professionals (HCPs), such as gynecologists and midwives who regularly insert IUDs. Following a baseline survey of study participants at the time of IUD insertion, both participants and the HCP who inserted the device completed 1 follow-up questionnaire 12 months after enrollment. All outcomes reported by study participants were validated by the treating physicians. A total of 61,448 women between 18 and 50 years of age with a newly inserted IUD were enrolled by more than 1200 participating HCPs. Less than 1235 women (2.0%), 1.7% for LNG-IUS and 2.8% for copper IUD, were lost to follow-up. More than 30 different types of copper IUDs were used by participants. The analysis included validated follow-up data for 58,324 women (LNG-IUS, 70%; copper IUD, 30%). Contraceptive failures occurred in 118 women (26 LNG, 92 copper). Both types of IUD were highly effective contraceptive devices, but the LNG-IUS was more effective: overall Pearl indices were 0.06 (95% confidence interval [CI], 0.04–0.09) for LNG-IUS and 0.52 (95% CI, 0.42–0.64) for copper IUD. Cox regression analysis showed an adjusted hazard ratio of 0.16 (95% CI, 0.10–0.25) for LNG-IUS versus copper IUD. A total of 21 pregnancies (7 LNG IUS, 14 copper IUD) were ectopic. The adjusted hazard ratio for ectopic pregnancy was significantly lower for LNG-IUS than copper IUD (0.26; 95% CI, 0.10–0.66). The contraceptive failure rate was very low among both LNG-IUS and copper IUD users, but the risk of pregnancy, including ectopic pregnancy, was significantly lower for LNG devices. Clinicians should discuss differences between IUDs in rates of unintended pregnancies and associated complications with women seeking contraception.