Hormonal contraception caused serum levels of vitamin B12 to decrease over a period of 3 years, but the reduction did not appear to be clinically significant or to impact bone mineral density in a large study limited to women of child-bearing age.
"This longitudinal study adds to the literature by demonstrating a cause and effect relationship between OC use and serum B12 levels. In addition, we observed lower B12 levels among DMPA [depot medroxyprogesterone acetate] users, which has not been previously reported in the literature," lead author Dr. Abbey B. Berenson reported online March 28 in the journal Contraception (Contraception 2012 March 28 [doi: 10.1016/j.contraception.2012.02.015]). "However, levels still remained in the normal range for almost all users. By and large, these findings agree with previous cross-sectional and case-control studies on the effect of OCs on serum B12 levels.
Dr. Berenson, of the department of obstetrics and gynecology and the Center for Interdisciplinary Research at The University of Texas Medical Branch, Galveston, and her associate, Dr. Mahbubur Rahman, recruited 703 women aged 16-33 years. The women were asked to choose one of three birth control methods: 245 chose an OC (0.15 mg desogestrel+20 mcg EE taken for 21 days, followed by 2 days of placebo and 5 days of 10 mcg EE); 240 chose DMPA; and 218 chose a nonhormonal method (barrier method, 53%; tubal ligation, 33%; copper T, 10%; and partner vasectomy, 4%). DMPA and the OC were dispensed every 3 months. Nonhormonal contraceptive users also attended a clinic every 3 months and were offered a supply of condoms at each visit.
At each 6-month visit over the course of 3 years, study participants underwent weight and height measurements, BMD measures of the lumbar spine and femoral neck via dual energy X-ray absorptiometry, and serum B12 measures via immunoassay.
The mean age of the study participants was 24 years, 36% were Hispanic, 35% were non-Hispanic white, and 29% were non-Hispanic black.
Women in all three groups had a decrease in mean B12 levels over the 3-year time period. Those taking an OC or DMPA had greater B12 decreases than women using nonhormonal contraception (P less than .001). During the first 6 months of use, for example, women in the OC and DMPA groups experienced B12 decreases of 97 pg/mL and 64 pg/mL, respectively, which represented declines of 20% and 13% from their baseline values. At the same time, women in the nonhormonal group experienced a decrease of 14 pg/mL, or a decline of 3% from the baseline value. Over the remaining 30 months of the study, B12 levels did not further decline among women in the OC group, but dropped another 22 pg/mL in the women in the DMPA group, and fell another 17 pg/mL in the women in using nonhormonal contraception.
After adjusting for age, whole body lean mass, duration of use, contraceptive methods, race/ethnicity, smoking status, and duration of previous DMPA use, the researchers found no significant association between mean B12 levels and mean BMD at the spine (P = .107) or femoral neck (P = .877).
Very few women had B12 deficiency, defined as a B12 level of less than 180 pg/mL: There were four in the nonhormonal group, two in the DMPA group, and nine in the OC group.
The researchers noted that the manner by which hormonal contraception causes a decrease in serum B12 remains unclear. "One possible mechanism is a deficiency in the level of serum B12 binders resulting in a false low B12 level in OC users," they hypothesized. "However, several studies have shown no difference in the level of mean unsaturated B12 binders between OC users and nonusers, suggesting that absorption is not affected and that redistribution of B12 throughout the body could be responsible. However, the mechanism of this redistribution is unknown. Furthermore, the mechanism of suppression of B12 levels among DMPA users, which was less severe than that observed among OC users in our study, has not been determined."
The observation that B12 levels were not associated with BMD, regardless of type of contraceptive use, contrasted with findings from several previous studies. "All but one of these prior studies, however, focused on postmenopausal women, who may react differently," the researchers noted. "The one study that did include reproductive-age women was limited to European adolescents who adhered to a specific diet plan (one group consumed a macrobiotic diet for up to 6 years followed by a vegetarian/omnivorous diet, while the other group ate an omnivorous diet throughout their life). Furthermore, it differed from the current study in that it was cross-sectional in design and included males as well as females. These differences could explain the variation in findings between their study and ours."