Complete and in-depth discussion of contraception during breastfeeding is provided in the Academy of Breastfeeding Medicine Protocol #13, recently revised by two of the Academy’s most knowledgeable members on the subject.
Fertility
The understanding of the underlying mechanisms of infertility and fertility return during lactation has been increasing in the last 30 years, with studies of both animal and human models. Much has been learned from comparing the lactating and nonlactating hormonal physiology and from the study of the associated brain peptides. Although we understand more today, many significant questions remain.
Lactational Infertility
The prolonged postpartum infertility associated with lactation has been attributed to changes in the hypothalamic-pituitary-ovarian axis mediated by gonadotropin secretion. Frequent suckling at the breast causes changes in gonadotropin-releasing hormone (GnRH), which impacts anterior pituitary hormone and disorganizes the pulsatility and levels of luteinizing hormone (LH) and follicle-stimulating hormone (FSH), disallowing the rhythmic patterns that result in ovulation. Frequent suckling also results in high prolactin levels; however, the role of prolactin in fertility suppression is less clear.
Figure 20-1 illustrates the menstrual cycle and gonadotropic control. Key points include the following:
- 1.
Follicular development is initiated by pituitary gonadotropin FSH.
- 2.
Continued growth requires FSH and estradiol from the growing follicle in response to LH, which is released in a pulsatile fashion from the pituitary.
- 3.
At midcycle, an increase in estradiol triggers the release of preovulatory surges of LH and FSH.
- 4.
The follicle secretes predominantly progesterone (luteinization).
- 5.
The oocyte is released 36 hours later.
The pulsatile release of GnRH from the hypothalamus stimulates the release of LH. In the cycling woman, estrogen increases GnRH secretion, and the combination of progesterone and estrogen decreases it.
The postpartum period, however, is characterized hormonally by elevated levels of prolactin and low levels of gonadotropins, resulting in anovulation and amenorrhea. During breastfeeding, this state can persist for an extended period, even though prolactin levels decrease over time. As currently understood, the pulsatile secretion of GnRH is altered by the suckling stimulus, influencing ovarian activity. Although the action of prolactin on multiple target organ sites has frequently been suggested as the cause of this ovarian quiescence, it appears that a suckling-induced alteration in hypothalamic GnRH production is the primary mechanism. Zinaman et al. found that when pulsatility is restored during lactation by administering exogenous pulsatile GnRH, LH values increase and FSH levels decrease while pulsatile organization of both increases, and the ovary responds accordingly.
The period of lactational amenorrhea depends on frequency and intensity of suckling ( Figures 20-2 to 20-4 ). In these figures, open circles are lactating subjects, the solid circles are nonlactators postpartum. Animal studies have shown that the release of FSH and LH is inhibited by intense suckling. In addition, when the nipple is stimulated while the milk ducts have been tied off there is still a suppression of estrous and menstrual cycles. Selye and McKeown concluded that interruption of sexual cyclicity during lactation is a result of the suckling and not of the secretory activity of the mammary gland. This early animal work has stood the test of time. The more the suckling stimulus in frequency and duration, the more consistent is the suppression of ovulation.
The levels of gonadotropin in all postpartum women for the first weeks of the postpartum period are decreased, which substantiates the theory of postpartum ovarian refractoriness. In the first 2 weeks postpartum, low levels of FSH are found in urine and plasma. Estrogen excretion is low with a linear increase during the first 5 to 8 weeks. In a longitudinal study of 48 women, endocrine profiles were assessed with morning blood samples from the first postpartum month until the recovery of ovulation. , Additional samples were drawn throughout 24 hours at the end of the third postpartum month in 10 exclusively nursing amenorrheic women. Prolactin, LH, FSH, estradiol (E 2 ), progesterone, cortisol, and dehydroepiandrosterone sulfate were measured. In response to suckling, there was a smaller increase in prolactin and higher levels of E 2 in women who ovulated within 6 months postpartum compared with those who did not. Diaz et al. , suggest that this may explain some of the variability in duration of lactational amenorrhea. The greater prolactin response to suckling associated with longer amenorrhea may result from higher sensitivity to the breast-hypothalamus-pituitary system.
Prolactin and Dopamine
There is a relationship between lactational infertility and physiologic hyperprolactinemia; however, its role in fertility suppression is not as clear. In an extensive study of prolactin levels in lactating women, Tay et al. measured the pattern of prolactin secretion in relation to suckling and the return of ovarian activity. Blood samples were drawn at 10-minute intervals for 24 hours at 4 and 8 weeks, when weaning was initiated and suckling reduced, at first menses, and in the follicular phase of the first menstrual cycle after weaning. Mothers fed their infants on their usual pattern, with no restrictions or alterations, in an effort to replicate natural lactation.
These data confirmed that with frequent suckling, prolactin levels do not decline significantly between feeds. When suckling became less frequent, prolactin dropped to baseline levels between feeds but surged when suckling was initiated (see Chapter 3 ). The natural increase in prolactin at night was evident only after weaning. Prolactin also declined greatly in association with suckling after the return of menses. This occurred at 33.6 ± 3.5 weeks postpartum in this study. No relationship was seen between the duration of amenorrhea and plasma prolactin levels throughout a day, at night, or throughout lactation. The timing of the introduction of solids was strongly correlated with the duration of amenorrhea. The authors concluded that no exact link exists between release of prolactin during lactation and the duration of lactational infertility in breastfeeding women.
However, at least one study has identified a differential response to different prolactins.
Another study explored the ability to predict duration of amenorrhea based on parameters during pregnancy. Campino et al. followed 17 women at 34 and 38 weeks’ gestation who fully breastfed for at least 6 months. During pregnancy, prolactin, estrogens (total estradiol, unconjugated estrone, unconjugated estriol), sex hormone binding globulin, dehydroepiandrosterone sulfate, progesterone, and placental lactogen, and during postpartum, prolactin, estrogens, and sex hormone binding globulin, were measured. Free estradiol in pregnancy and postpartum was calculated. They found that the 10 women who experienced long lactational amenorrhea (greater than 6 months) had a different hormonal profile during pregnancy than the seven who experienced a short duration (less than 6 months) of lactational amenorrhea. At 38 weeks’ gestation, the women who experienced a long lactational amenorrhea had twice as much prolactin, approximately half the total estradiol, significantly lower sex hormone binding globulin concentration, but similar free estradiol concentration compared with those who experienced short lactational amenorrhea. They concluded that at 38 weeks’ gestation, the higher ratio of prolactin/estradiol identified all women who would go on to experience a longer duration of their lactational amenorrhea, suggesting that duration of lactational amenorrhea is conditioned during pregnancy.
The inhibition of dopamine secretion from the hypothalamus has been associated with the neural impulses from stimulation of the nipple during lactation. Normally dopamine inhibits the secretion of prolactin, and, conversely, when dopamine is inhibited, prolactin rises. Two pathways of ovulation inhibition are possible as a result of the rise in prolactin. One is a lack of responsiveness to ovarian steroids of the hypothalamic-pituitary axis of a lactating woman, leading to nonpulsatile release of pituitary gonadotropins, FSH, and LH, which in turn results in absent or reduced ovarian activity. FSH may actually be higher at some points; LH is nonpulsatile.
The fact that there are prolactin receptors on the ovary indicates that there may be a second mechanism contributing to the infertility through the impaired ovarian response to gonadotropins (see Figure 20-1 ).
Return of Menses
The transition from amenorrhea to regular menstrual cycles is one of the most challenging times while breastfeeding and wishing to use natural family planning. The uncertainty of the onset of ovulation with the return of menses is especially difficult. The efficacy of a new postpartum transition protocol for avoiding pregnancy is reported by Bouchard et al. The use of an electronic hormonal fertility monitor (Clear Blue Easy Fertility Monitor, Swiss Precision Diagnostics, Geneva, Switzerland) identifies the fertile period. It measures changes in urinary estrone-3-glucuronide from baseline and urinary LH above a specific threshold. The device was developed to assist with conception; it has been shown to be equally effective when utilized to avoid pregnancy. The use of an online teaching and charting protocol has potential for avoiding pregnancy postpartum while lactating.
Clinically the proxy for the return of fertility is the onset of menstruation. Return of reproductive function varies depending on the length and degree of lactation. Most studies do not, in fact, report pattern of breastfeeding, that is, whether the infant is fully or exclusively breastfed or is also receiving solid foods or supplemental bottles. By the end of the third month, only 33% of fully lactating women have had a menstrual period, whereas 91% of nonlactating women have had at least one period.
Not all vaginal bleeds are menses; not all bleeds follow ovulation. In 72 fully breastfeeding women studied prospectively from 42 days postpartum, vaginal bleeding was recorded daily if it occurred. Approximately half the women had some bleeding or spotting between 6 and 8 weeks postpartum. Those who experienced this bleeding eventually menstruated and ovulated earlier than those who did not, but differences were not significant. Seven women had ovarian follicular development before day 56, but neither bleeding nor follicular development was associated with ovulation in the first 8 weeks. The authors stated it was unlikely that vaginal bleeding before 8 weeks in a fully breastfeeding woman indicates a return to fertility and, therefore, is not the return of menses.
Perez et al. diagnosed the first postpartum ovulation by endometrial biopsy, basal body temperature, vaginal cytologic evaluation, and cervical mucus in a group of 200 women in a prospective study. The dates of first ovulation, first menses, and nursing status were analyzed. No woman demonstrated signs of ovulation before day 36, whether lactating or not. The intensity of nursing and time postpartum affected ovulation occurrence; 78% of the women ovulated before the first menses, but only 12 pregnancies occurred with first ovulation. Of the 170 women who breastfed, 24 ovulated while completely nursing, 49 while partially nursing, and 97 after weaning.
Possibility of Conception
A nonlactating woman has a return of her period at 25 days at the earliest, a return of ovulation at 25 to 35 days, and a 5% chance of regaining fertility before 6 weeks postpartum.
Risk of ovulation during lactation was studied by Gray et al. in Baltimore and in Manila. During the first 6 months postpartum, amenorrheic women had a low risk of ovulation (less than 10%) with partial breastfeeding and a 1% to 5% risk with exclusive breastfeeding with either frequent short feeds or infrequent longer feeds. This would have resulted in a pregnancy rate of 2% and 1%, respectively.
In a detailed study of 130 women in Chile, Diaz et al. found the cumulative probability of pregnancy at the end of 6 months postpartum in women who were exclusively nursing and amenorrheic to be 1.8%. For exclusively nursing women who had a return of menses it was 27.2%, and for those partially nursing it was 40.5%.
Although many investigators continue to evaluate the impact of lactation on ovulation and menstruation, the fundamental observations remain the same ( Table 20-1 ). Available data on return of ovulation and menstruation can be summarized as follows :
- I.
Nursing mothers
- A.
Ovulation generally occurs before menses return and varies 14% to 75%.
- B.
The longer the first menses is delayed, the more likely the first cycle will be ovulatory.
- C.
Continued suckling and elevated prolactin levels produce inadequate luteal function in first cycles.
- D.
Exclusive breastfeeding: First bleed generally precedes ovulation return, and if an ovulation occurs it is generally inadequate for conception.
- A.
- II.
Nonnursing mothers
- A.
Earliest possible menstruation is 4 weeks postpartum.
- B.
Most women are menstruating by third month postpartum.
- C.
Return of menstruation
- 1.
6 weeks postpartum: 40%
- 2.
12 weeks postpartum: 65%
- 3.
24 weeks postpartum: 90%
- 1.
- D.
Earliest possible ovulation is 3½ to 5 weeks postpartum.
- E.
Ovular cycles occur in 50% with first menstrual period postpartum.
- F.
Early postpartum ovulation may occur late in menstrual cycle: Shortening of secretory phase and greater tendency toward irregular menses.
- G.
Return of ovulation
- 1.
6 weeks postpartum: 15%
- 2.
12 weeks postpartum: 40%
- 3.
24 weeks postpartum: 75%
- 1.
- A.
- III.
Amenorrheic nonnursing mothers: Return of ovulation
- A.
12 weeks postpartum: 20%
- B.
16 weeks postpartum: 40%
- A.
| Average Number of Feeds per Day | Relative Risk |
|---|---|
| 0 | 1.0 |
| 1 | 0.62 |
| 2 | 0.43 |
| 3 | 0.28 |
| 4 | 0.19 |
| 5 | 0.12 |
| 6 | 0.08 |
| 7 | 0.05 |
| 8 | 0.04 |
| 9 | 0.02 |
| 10 | 0.01 |
* Breastfeeding episodes per day before ovulation: p < 0.0001.
Milk Composition During the Ovulatory Menstrual Cycle
Acute changes in the composition of milk during the ovulatory menstrual cycle in lactating women were studied by Hartmann and Prosser involving women during lactational amenorrhea, taking oral contraceptives, and during an ovulatory menstrual cycle. Samples of milk were collected from each breast at each feed for each day for 28 days.
During the ovulatory menstrual cycle, two acute changes occurred. For the 5 to 6 days before ovulation and the 6 to 7 days after ovulation, the sodium and chloride values changed from 4.6 mM Na and 11.1 mM Cl to 10.1 and 22.0, respectively, and lactose and potassium decreased. The concentrations of lactose, Cl, K, and Na remained relatively constant during lactational amenorrhea, anovulatory menstrual cycles, and for those women taking oral contraceptives. The authors conclude that an increase in the permeability of the mammary epithelium was effected by changes related to ovulation. Perhaps the first acute change in composition is associated with the final stages of follicle maturation and the second with the regression of the corpus luteum during the ovulatory menstrual cycle ( Figures 20-5 and 20-6 ).
Nutritional status has virtually no effect on amenorrhea, except in the extremes. In a study of Guatemalan women, maternal energy supplements did not shorten length of lactational amenorrhea; however, supplementing their breastfed infants did shorten amenorrhea by reducing suckling. A difference exists between postpartum and nutritional amenorrhea: true nutritional amenorrhea is predictable on the basis of the height/weight ratio; lactational amenorrhea is hormonal, and when it occurs, nutrition has only a trivial role.
Breastfeeding and Birth Interval
Among !Kung hunter-gatherers, long intervals pass between births, which has puzzled investigators because the tribes are well nourished, have low fetal wastage, and do not employ contraceptives or prolonged abstinence. The !Kung eat only what they hunt and gather. They have no agriculture. They are lean, spare people. They have late menarche (approximately 16 years of age), first pregnancy at age 18, and early menopause at approximately 40, leaving 24 reproductive years during which they produce 4.4 children, which, with some perinatal deaths, exactly replaces their society. This compares with industrial society, where productive years begin at 11 and end at 51. Konner and Worthman report that the !Kung have unusual temporal patterns of nursing characterized by highly frequent nursing bouts with short space between nursings. The !Kung nurse several times an hour with only 15 minutes at most between bouts, which last only 15 to 120 seconds each. Serum estradiol and progesterone levels are correspondingly low. Infants are always in the immediate proximity of their mothers until they are weaned, at approximately 3½ years, during a new sibling’s gestation.
In Nigeria, the effect of duration and frequency of breastfeeding on postpartum amenorrhea is comparable in that Nigerians breastfeed for 16.5 months with a frequency of 4.5 times a day. The mean length of amenorrhea is 12.5 months. Amenorrheic mothers who were lactating had lower levels of serum estradiol and lactic dehydrogenase. A significant association was seen between hyperprolactinemia with amenorrhea. The incidence of amenorrhea declined parallel to that of the hyperprolactinemia.
When fertility postpartum during lactation was studied in Edinburgh, suckling was the most important factor inhibiting the return to ovulation. Suckling duration was the first factor to discriminate the mothers who experienced early ovulation. Those mothers who ovulated while breastfeeding had introduced two or more supplementary feeds per day and had reduced suckling to less than six times per day, with 60 minutes or less suckling time per day. The basal prolactin levels were less than 600 mU/L. The mothers who did not ovulate until after 40 weeks postpartum breastfed longest, suckled most intensely, maintained night feeds longest, and introduced supplementary feeds most slowly. The prolactin levels remained substantially greater than 600 mU/L.
Another review of the effects of hormonal contraceptives on lactation by Hull concludes that a significant number of reports indicate decrease in milk yield. The description of severe growth failure in the nursling, even leading to “contraceptive marasmus,” in Egypt and Tunisia is cause for concern.
Most large studies of birth interval and its relationship to method of feeding have been conducted in developing countries. However, Rosner and Schulman reported on 112 Orthodox Jewish women from metropolitan New York with 266 birth experiences. The women strictly adhered to biblical and Rabbinic law that prohibits birth control. They were well-nourished, middle-class, educated women who breastfed on demand (210 infants) for a mean duration of 10.7 months, with 177 of the infants receiving formula less than once a week. Significant positive correlations were found with duration of lactational amenorrhea, which increased as duration of breastfeeding increased. Delay in starting solids, continuation of night feedings, and postponement of other liquid feeds all were associated with prolongation of birth interval. The investigators found a longer mean duration of lactational amenorrhea (8.6 months) and mean birth interval (22 months) than other studies because of the more intensive feeding patterns ( Table 20-2 ).
| Study | Mean Duration of Breastfeeding (mo) | Mean Lactational Amenorrhea (mo) | Mean Birth Interval (mo) |
|---|---|---|---|
| Bonte and van Balen | N/A | 15.2 | 25.2 |
| Berman et al. | 7.0 | 7.1 | 21.6 |
| Prema et al. | 19.8 | 11.1 | 23.8 |
| Perez , * | 4.0 | 3.03 † | N/A |
| Gioiosa | 10.27 | N/A | 21.92 |
| Rosner and Schulman | 10.74 | 8.56 | 21.95 |
| Adnan and Bakr | 36.0 | 12.0 | N/A |
| Howie et al. | 10.0 | 8.1 | N/A |
| Ojofeitimi | 16.5 | 12.5 | N/A |
* Women did introduce family planning after menses return or after supplementation after 6 months.
Contraception During Lactation
Medical Eligibility Criteria for Contraceptive Use
WHO’s Medical Eligibility Criteria for Contraceptive Use provides recommendations for policy makers to help rationalize the provision of various contraceptives in relation to the most up-to-date information available on the safety of the methods for people with certain health conditions. The document covers the following family planning methods: low-dose combined oral contraceptives (COCs), combined patch (P), combined vaginal ring (R), combined injectable contraceptives (CICs), progestin-only pills (POPs), depot medroxyprogesterone acetate (DMPA), norethindrone enanthate (NET-EN), levonorgestrel (LNG) and etonogestrel (ETG) implants, emergency contraceptive pills (ECPs), copper-bearing intrauterine devices (Cu-IUDs), levonorgestrel-releasing IUDs (LNG-IUDs), copper IUD for emergency contraception (E-IUD), barrier methods (BARR), fertility awareness-based methods (FAB), lactational amenorrhea method (LAM), coitus interruptus (CI), and female and male sterilization (STER).
The goal of this document is to provide policy- and decision-makers and the scientific community with a set of recommendations that can be used for developing or revising national guidelines on medical eligibility criteria for contraceptive use. ,
Lactational Amenorrhea Method
The Bellagio Consensus Conference on breastfeeding as a family planning method established that a mother who is fully or nearly fully breastfeeding her infant and remains amenorrheic will have more than 98% protection from pregnancy in the first 6 months postpartum. , , This was codified as a method of family planning the following year at a meeting at an international conference at Georgetown University. , The first study of the method per se found that only one woman in a study of 422 middle-class urban women in Chile became pregnant using the LAM as the only method of pregnancy avoidance in the first 6 months, , a protective rate of 99.5% (see Figures 20-5 and 20-6 ).
Menses as an indicator of ovulation has been studied with data collected not only on onset of menses but on urinary hormone assays. Among women who menstruated before 6 months postpartum, 67% of cycles were anovulatory, and the lag between anovular first menses and subsequent ovulation was 15.7 weeks. On the other hand, after 6 months postpartum, the proportion of anovular first menses declined to 22%, and the lag to ovulation declined to 7.3 weeks. Comparing all menstrual episodes, the mean interval between first observed menses and ovulation was 8.4 weeks in the first 6 months and only 0.1 week after 6 months postpartum.
A significant distinction should be made between token breastfeeding with early solids and more rigid feeding schedules and the ad lib breastfeeding around the clock with no solids until the infant is 6 months old. The amount and frequency of sucking are closely related to the continued amenorrhea in most women. When a totally breastfed infant sleeps through the night at an early age, requiring no suckling for 6 hours or so at night, the suppressive effect on menses diminishes. It has also been shown that if the infant uses a pacifier rather than receiving nonnutritive sucking at the breast, the suppression of ovulation is diminished.
The degree of fertility inhibition associated with breastfeeding has decreased remarkably since the time of hunter-gatherers, cautions Diaz et al. She points out that fertility rates vary; population and socioeconomic factors, urbanization, and nutrition influence not only breastfeeding patterns but associated ovarian quiescence. Lactational amenorrhea can provide protection against pregnancy for the first 6 months even in well-nourished women who are giving the infant some supplemental foods ( Figure 20-7 ). For women who practice LAM, the efficacy is remarkably good ( Table 20-3 ). Table 20-4 details the reestablishment of menses in breastfeeding women using LAM. Bellagio and Beyond: Breastfeeding and LAM in Reproductive Health was published as a final report in 1997 of the many years of work worldwide involving the use of LAM. It was concluded that the efficacy of LAM is well established in prospective studies. Policy support is still needed to institute an additional method that increases the family planning choices of postpartum women.
| Month | No. of Pregnancies | WM | WMAC | R × 100 | P × 100 |
|---|---|---|---|---|---|
| 1 | 0 | 384 | 384 | 0.00 | 0.00 |
| 2 | 0 | 327 | 711 | 0.00 | 0.00 |
| 3 | 0 | 272 | 983 | 0.00 | 0.00 |
| 4 | 0 | 243 | 1226 | 0.00 | 0.00 |
| 5 | 0 | 224 | 1450 | 0.00 | 0.00 |
| 6 | 1 | 221 | 1671 | 0.45 | 0.45 |
* Characteristics of women: Mean age (SEM; range) was 27.1 years (5.0; 18 to 39). Mean parity (SEM; range) was 2.0 (1.0; 1 to 5). 23.6% of the women had primary education, and only 5.4% had completed university studies.
| Month | Bleeding | WM | WMAC | R × 100 | P × 100 |
|---|---|---|---|---|---|
| 1 | 0 | 384 | 384 | 0.00 | 0.00 |
| 2 | 8 | 327 | 711 | 2.45 | 2.42 |
| 3 | 18 | 272 | 983 | 6.62 | 8.67 |
| 4 | 11 | 243 | 1226 | 4.55 | 12.72 |
| 5 | 10 | 224 | 1450 | 4.46 | 16.53 |
| 6 | 6 | 221 | 1671 | 2.73 | 18.78 |
* Characteristics of women: mean age (SEM; range) was 27.1 years (5.0; 18 to 39); mean parity (SEM; range) was 2.0 (1.0; 1 to 5); 23.6% of women had primary education, and only 5.4% had completed university studies.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
