Abstract
Premature ovarian insufficiency affects 1% of women under the age of 40 and is associated with a hypoestrogenic state, potentially leading to multiple comorbidities including reduced bone density and fertility. An unpredictable ovarian function is observed in 50% of patients with 5–10% being able to achieve a pregnancy. Longitudinal studies have shown a temporary decline in bone mineral density of up to 5% during pregnancy and lactation in healthy women, with the loss of bone density post-partum being proportional to the period of breastfeeding. Effects of pregnancy in women with premature ovarian insufficiency have not been widely documented. Nevertheless, a lower bone mineral density baseline has been observed pre-conceptually, associated with both the hypoestrogenic state of the condition and the possibility that premature ovarian insufficiency was developed prior to achieving peak bone mass. This may suggest that breastfeeding could cause further deterioration in bone mineral density that may not be easy to recover from due to the reduced baseline levels.
Bone density changes during pregnancy in healthy women have been well documented and shown to be reduced by up to 3–10%, with most calcium mobilization from the mother to the foetus occurring during the last trimester. 1 Further demand for calcium occurs during breastfeeding. 1 The cessation of ovarian function results in bone loss, as sex steroids play an important role in maintaining bone density, posing a significant risk in the development of osteoporosis if left untreated. 2 This is seen in patients with premature ovarian insufficiency (POI), a condition that mimics an endocrine state analogous to the menopause, with hypoestrogenism and high gonadotrophin levels. 3 Despite this, a proportion of women with spontaneous/idiopathic POI has been shown to have residual ovarian follicles with intermittent function that may persist for years, allowing for a natural or assisted pregnancy to potentially occur. 4 Data on bone mineral density (BMD) changes are lacking for this cohort of patients. This review thus aims to explore an area that has minimal literature available, including the fluctuations in BMD measured with dual-energy X-ray absorptiometry in POI patients who achieve a term pregnancy despite their poor ovarian function. 1
POI can be clinically defined as the manifestation of amenorrhoea for a period greater than four months before the age of 40 years, coupled with a rise in serum follicle-stimulating hormone to a menopausal level of over 40 mIU ml−1 shown by two samples taken one month apart, and oestradiol levels below 50 pg. 4 However, this definition varies between different sources. The aetiology is diverse, including iatrogenic, autoimmune and genetic causes, as well as toxins, but in most cases, the cause is unknown. 3 It occurs in 1% of women below the age of 40 and in 0.1% below the age of 35 and leads to a state of hypoestrogenism and anovulation, responsible for the development of vasomotor symptoms, amenorrhoea and infertility. 4 The management of POI patients is complex and should target hormone replacement, fertility issues and bone health, as well as emotional well-being. Hormone replacement therapy (HRT) or the oral contraceptive pill is used long-term not only to relieve vasomotor symptoms but also to prevent the development of problems associated with estrogen deficiency, such as osteoporosis. 4 However, up to 50% of patients have unpredictable ovarian function, and 5–10% are able to achieve a pregnancy, suggesting that the cessation in ovarian function may not be permanent. 4
Fetal maturation and skeletal development in the uterus require a transfer of calcium from the mother amounting up to 30 g throughout pregnancy. 1 The demand for calcium is additionally increased during breastfeeding, requiring compensatory mechanisms to be further upregulated in the mother, to increase intestinal and renal calcium absorption and to mobilize calcium from the maternal bones. However, maternal estrogen concentrations, shown to be high during pregnancy, have a positive effect on BMD, allowing for the supply of calcium to be adequate for both the mother and the foetus. Gestation and lactation thus both play an important role in the changes observed in BMD. 1
Calcium works as a threshold nutrient; intake is important up to a certain level, but supplements above that point will not significantly affect BMD. 5 BMD can be positively influenced by the cessation of smoking and alcohol, as these normally exert a negative effect on it. Lean body mass, fat body content, body weight as well as fluid shifts are other confounding variables that are altered during pregnancy and lactation, and therefore influence BMD, possibly by increasing mechanical load and peripheral estrogen production. 5 Nonetheless, a study carried out by Karlsson et al. 5 which adjusted for discrepancies in the composition of soft tissue, showed a 7.6% reduction in lower lumbar spine BMD and a 3.9% drop in total body BMD in immediately post-partum healthy women as compared to age-matched controls. 5
Bone resorption and formation markers indicate that bone turnover is increased during lactation, indirectly suggesting that BMD levels are also affected. This response has been shown to be dose dependent, so that a longer lactation period leads to a greater loss in BMD. 5 A cross-sectional case–control study including 65 breastfeeding women carried out by Karlsson et al. 5 analysed BMD changes during the first 12 months post-partum, comparing healthy non-breastfeeding women and healthy breastfeeding women for up to six months. 5 Women lactating for prolonged periods of time show a greater change in femoral neck BMD, which stays below the baseline level at one year post-partum. 5 A similar change is also reported for spine BMD but a recovery is seen after weaning. 5
A study carried out by Uygur et al. 2 looked at bone loss in POI patients. 2 In this study, women with chromosomal abnormalities were excluded, and it was concluded that karyotypically normal POI patients have a statistically significant lower BMD as compared to healthy controls, despite the fact that some women may have taken HRT at least intermittently in the past. 2 Androgen levels were also found to be lower in POI, and this has been associated with bone loss, in pre-, peri- and post-menopausal women. 2 Androgen replacement may, therefore, be required to maintain bone density, alongside with estrogen and progestin. 2 Another study by Anasti et al. 6 that looked at a cohort of 89 women with POI wishing to conceive further suggested that despite 87% having had some form of hormone replacement, bone density scores in two thirds of age-matched POI patients were over one standard deviation below the mean, with an associated risk of hip fractures of 2.6-fold. 6 This was also observed in the women who had their bone density measured within 18 months of diagnosis. 6
Leite-Silva et al. 7 carried out a cross-sectional study including 50 patients with spontaneous POI and age-matched controls, with the aim of determining any differences in bone density at the spine and femur. Their results showed that the age of amenorrhoea and a state of hypoestrogenism were the most important factors affecting bone loss. 7 Trabecular bone in the lumbar vertebrae has been shown to be more metabolically active than cortical bone in the femur, and thus is increasingly susceptible to hypoestrogenism, with a rate of bone loss of twice of the rate of bone loss seen in the hip. 7
Further supporting the protective effect of estrogen on BMD, and reduced incidence of fractures is a population-based study carried out by Nguyen et al. 8 which comprises data from 1091 women who participated in the Dubbin Osteoporosis Epidemiology study. This showed a reduction in the incidence of atraumatic fractures and a higher BMD in women with an increased duration of estrogen exposure either through natural menstruation or hormone replacement, as well as in women with higher parity. 8
Conclusion
Hormone replacement or the oral contraceptive pill has been used as a strategy to prevent osteoporosis in post-menopausal women, and their use has extended to younger patients with POI, where BMD is reported to be significantly lower as compared to age-matched controls. The effect of pregnancy on bone density in healthy women has been described and longitudinal studies have suggested that there is a loss in BMD of up to 5%, which recovers after gestation and lactation. However, the decline in BMD during lactation seems to be directly proportional to the period of breastfeeding. There is a lack of information about the dynamics of BMD in women with POI who become pregnant and the effects seen post-partum and during lactation. This group of patients has been shown to have a lower baseline BMD pre-conceptually, possibly due to the hypoestrogenic state-associated with the condition, and the fact that they may have developed POI prior to achieving peak bone mass. This, therefore, suggests that breastfeeding could cause further deterioration in BMD that may not be easy to recover from due to the reduced baseline levels. Future research should focus on the possible accelerated bone loss caused by calcium transfer to the foetus and gonadal sex steroid deficiency during pregnancy and lactation in this cohort of patients, with the aim of advising them with regards to modifiable factors including timing of commencement of hormone replacement post-partum, optimum breastfeeding period, and other lifestyle changes, to prevent future pregnancy-related osteoporosis.
Relevance in clinical practice
POI patients use hormone replacement therapies in various forms to counteract the hormonal changes associated with the condition. Despite a much reduced or absent ovarian function, pregnancy can occasionally be achieved either naturally or via assisted conception with ovum donation. Pregnancy and breastfeeding are associated with a temporary loss of BMD in healthy women, with lactation causing a proportional decline with the period of breastfeeding. In POI patients who achieve a full-term pregnancy, BMD may deteriorate further than their already reduced baseline levels during pregnancy and be additionally worsened by breastfeeding, requiring additional monitoring to be carried out. This poses a challenging clinical question: can recovery from pregnancy-associated BMD loss be improved with reduced breastfeeding duration and earlier introduction of hormone replacement post-partum?
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest
None declared.