Two years' experience with two levonorgestrel-releasing intrauterine devices and one copper-releasing intrauterine device: a randomized comparative performance study.

A randomized comparative performance study of two levonorgestrel intrauterine devices (IUDs) and a copper (Nova-T) IUD (Leiras, Turku, Finland) was performed at two clinics in Finland and one in Brazil. The 24-month cumulative event rates per 100 women are reported. Six thousand woman-months of use were recorded with the levonorgestrel IUD and 3000 with the Nova-T device. Pregnancy rates at 2 years of use were 0.6 and 0 with the two levonorgestrel IUDs and 3.3 with the Nova-T IUD. Removal rates because of bleeding and/or pain were low with all three IUDs: 7.5, 7.6, and 7.1, respectively, at 24 months. Recordings of the number of days of bleeding showed significantly fewer days of bleeding during use of the levonorgestrel IUDs than the Nova-T IUD from the second month of use onward, until the end of the second year. No infections were recorded in association with use of the levonorgestrel IUDs, and only one termination because of infection was recorded for the Nova-T IUD. The continuation rates of use at 24 months were 66.6, 60.7, and 71.9 for the two levonorgestrel IUDs and the Nova-T IUD, respectively; the differences were not statistically significant.