The pharmacokinetic and pharmacodynamic effects of vaginal rings releasing levonorgestrel (L-NOG) at an initial rate of 27 mu g/24 h were studied in a group of 12 normally menstruating women during 90 days of continuous use (i.e., during three 30-day treatment segments). Blood samples were drawn immediately before insertion, 15 and 30 min, as well as 1, 2, 4, 8, 12 and 24 h after insertion of the rings, and thereafter three times weekly throughout the study for the analysis of L-NOG, estradiol, progesterone and sex hormone-binding globulin (SHBG). Endometrial biopsies were obtained for a morphometric analysis in a pre-treatment (control) cycle and in the 6th and 10th weeks of treatment. The peak of average L-NOG levels was reached within two hours after the insertion of rings. Until 24 h after insertion, the levels did not change significantly. Thereafter, a decrease at a rate of 0.2% per day was initiated. The L-NOG and SHBG levels were highly correlated. This was seen for both the pre-treatment SHBG vs L-NOG (r = 0.96) and the treatment SHBG vs L-NOG levels (r = 0.92). There was a significant (p < 0.001) decrease of SHBG levels due to treatment. During the total of 36 treatment segments, a normal ovarian function was seen in 47% of the segments. The women were anovulatory and had an inadequate lutal function in 28% and 25% of segments,respectively. No correlation between the L-NOG levels and ovarian reaction to treatment was found. The use of L-NOG induced significant changes in the endometrium; the number of glands/mm(2) decreased after 6 (p < 0.02) and 10 weeks of use (p < 0.01). Also, the diameter of glands and the occurrence of vacuolated cells decreased significantly (p < 0.02 and p < 0.005, respectively). None of the endometrial parameters or dating was correlated with the ovarian reaction to treatment, indicating independent endometrial effects of L-NOG.
