Objectives: To measure and compare the incidence of adverse events during use of two medicated intrauterine devices (IUDs).
Design: A multicenter prospective 7-year randomized study.
Setting: Family planning clinics, primarily in developing countries.
Subjects: Women age 18 to 38 years at admission, desiring contraception and without contraindications to IUDs.
Main outcome measures: Incidence of complaints, conditions, and rates of specific termination for each IUD.
Methods: Subjects recorded menstrual events, and clinical staff registered all complaints and conditions found on examination at four first-year clinic visits and at semiannual visits thereafter. Difference in rates were analyzed by chi 2 statistics.
Results: Annual pregnancy rates for each IUD averaged 0.2/100 women whereas upper genital tract infection occurred at rates of 0.6 to 0.7 per 100 years of use. The levonorgestrel-releasing IUD significantly decreased bleeding and spotting days in comparison with historical data for noncontraceptors and with the copper-medicated IUD. Dysmenorrhea, vaginitis, and myoma in women with the levonorgestrel IUD were markedly decreased in comparison with the experience of copper IUD users. Significantly higher rates of amenorrhea, delayed ovarian follicular atresia, skin and hair conditions, and headache were observed with the steroid IUD than with the copper-releasing IUD. Rates of reported adverse effects for either IUD were highest in the first 2 years of use and among women under age 25.
Conclusions: Long-term use of copper or levonorgestrel IUDs is characterized by very low rates of pregnancy and by a low and declining annual incidence of side effects, including pelvic infection and borderline anemia. The levonorgestrel-releasing IUD reduced the incidence of bleeding and, in the long term, of myoma and myoma-related surgery in comparison with the copper T IUD. Both IUDs proved highly acceptable and had few unanticipated side effects.