Background
Osteoporosis is one of the most common chronic metabolic skeletal disease defined by low bone mass and disruption of bone microarchitecture, with increased risk of bone fragility and fracture. Postmenopausal women were considered to be at high risk of developing osteoporosis, as Estrogen deficiency accelerated bone turnover with net bone loss [
1]. Postmenopausal osteoporosis (PMOP) has long been a pervasive public health concern. In China, an approximately 40.1% prevalence of PMOP was reported in a nationwide study [
2]. The most important adverse health outcome of osteoporosis contributes to the occurrence of any bone fractures, which reduced the health-related quality of life in all aspects [
3]. It was estimated that the residual lifetime risk (RLR) of any fracture for 50-year-old Chinese woman was 37.6% [
4]. Therefore, with the aging Chinese population, the prevention, assessment and management of postmenopausal osteoporosis were particularly important. Pregnancy and lactation period were considered to affect bone metabolism and calcium homeostasis. In this period, increased intestinal calcium absorption and breast milk secretion cause the calcium loss in maternal skeleton, but the bone loss could restore within 6–12 months after weaning [
5]. However, it is unclear whether the bone loss is completely compensated. It has been reported that prolonged total breastfeeding time was significantly associated with decreased bone mineral density (BMD) [
6,
7]. But as well known, PMOP was multifactorial, and its related factors also include age, body mass index (BMI), smoking, alcoholic, physical activity, extended calcium and Vitamin D intake, pregnancy interval, the age of pregnancy, multiple births, the age of menarche and menopause status [
8‐
11]. Studies on factors correlated to breastfeeding, as independent risk factors for PMOP remain controversial [
12]. Some studies suggested that the history of breastfeeding can significantly increased the risk of PMOP [
13], while some studies showed that prolonged breastfeeding duration was an independent risk factor for the development of PMOP rather than breastfeeding itself [
14,
15]. Conversely, some studies suggested that the history of breastfeeding may increase the bone mass density, which significantly decreased the incidence of PMOP [
16], whereas other studies demonstrated a non-significant relationship between breast-feeding and PMOP [
9,
17].
As the relationship between breastfeeding and PMOP is different from society to society and is not clear from the literature above, the purpose of this retrospective cross-sectional study was to determine whether breastfeeding and its total breastfeeding duration was an independent factor for the development of PMOP based on Chinese postmenopausal population.
Discussion
The present study examined the association of breastfeeding and the development of PMOP in a population of Chinese postmenopausal women undergoing routine BMD test in a community healthcare center. The main finding of study was that longer breastfeeding duration was not significantly associated with an increased risk for postmenopausal osteoporosis (OR 5.22, 95%CI 0.18–147.76, p = 0.333) when adjusting for the age, BMI, the number of pregnancy and parity. We also observed that advanced age, higher BMI, and more pregnancy were independent risk factors related to PMOP.
The association between PMOP and the total duration of breastfeeding were seen to be complex and differ from society to society. In Korea, a nationwide survey showed that breastfeeding more than 37 months was significantly associated with low BDM and high risk of PMOP [
7,
14], similar result was also found in Israel [
6] and Mexico [
8]. Moreover, an Italy study suggested that long periods of breastfeeding significantly increased the risk of osteoporotic fracture after menopause [
18]. In contrast, another study in US suggested that the history of breastfeeding protected against the development of PMOP, especially in women aged 49–54 [
19], while a research conducted with 93,676 postmenopausal women in US held different point of view. It was reported that breastfeeding more than 1 month was largely unrelated to decreased BMD and any osteoporotic fracture, additionally, breastfeeding more than 1 month could decreased the risk of osteoporotic fracture comparing with never breastfeeding [
20]. In the study of Sri Lanka, prolonged breastfeeding had no detrimental effect on BMD in postmenopausal women [
21]. Similarly, in a latest study in Turkey, no statistically significantly difference was examined among the groups (normal, osteopenia, and osteoporosis) in respect of total breastfeeding time, but whether breastfeeding was a risk factor for osteoporosis compared to other types of feeding was not involved [
17]. These discrepancies above may have been due to the differences in study design and study population. In current study, we supported that breastfeeding and its duration did not associate with the development of PMOP comparing with non-breastfeeding and mixed feeding. The prolonged breastfeeding was not an independent risk factor for PMOP in the population of Chinese postmenopausal women.
There were several evidence support our findings. First, during the pregnancy and lactation, the mechanism of bone calcium protection may be activated. On the one hand, increased intestinal absorption and renal conservation of calcium compensated for the loss of calcium into the breast milk to reduce the extent of bone loss [
22]. On the other hand, the secretion of parathyroid hormone related protein (PTHrP) counterbalances bone loss. PTHrP suppresses parathyroid hormone (PTH) to regulate placental calcium transport and protect maternal skeleton. Besides, prolonged breastfeeding maintains the ovarian in an inactive level. During the breastfeeding, the estrogen level increased, which may keep the balance between bone resorption and formation [
5,
12]. Some researchers considered that the ovarian with low activity may contribute to the contraceptive effect of breastfeeding [
17]. Actually, the pregnancy interval and the number of pregnancy are independent factors for bone recovery instead of breastfeeding [
9]. In our study, we also found that more pregnancies were associated with the risk of PMOP (OR 1.80, 95%CI 1.08–2.98,
p = 0.024), which confirmed the above hypothesis. Secondly, an animal experiment in rats suggested that breastfeeding had no significant association with PMOP, because pre-reproductive females had higher trabecular bone mass, greater microarchitecture and bone strength than males, which minimized the effect of bone loss on the development of osteoporosis. In addition, when normalized for their body size, post-reproductive females had no skeletal deficits compared to males. This animal experiment showed that females may initially had more trabecular bone than necessary to compensate for possible bone loss due to pregnancy and longer breastfeeding [
23], therefore, it helps to explain the result of our study that the prolonged breastfeeding will not increase the risk of PMOP. However, this hypothesis still need to be validated by empirical studies based on larger population.
As breastfeeding and its duration were not risk factors for PMOP, the present study recommends breastfeeding for newborn. It has been reported that breastfeeding had potential benefits to the women’s health [
24]. For short-term effect, breastfeeding may decrease the prevalence of postpartum depression [
25]. For long-term effect, some studies showed that breastfeeding had protective effect on the prevention of carcinoma, such as breast cancer [
26,
27], and ovarian carcinoma [
28]. However, the frequency and intensity of breastfeeding implementation in different periods of infants need to be explored in detail, because the process of bone formation during the lactation is still not very clear.
The study is the first to explore the effect of breastfeeding and its duration on PMOP in representative Chinese postmenopausal women based on the community healthcare center. We also considered factors that could affect BMD, such as the age, BMI, smoking, alcohol intake, the history of fracture, calcium and Vitamin D intake, menarche age, menopause age, the number of pregnancy and parity, feeding pattern were included for univariate and multivariate analysis. To minimize the confounding effect of nulliparity on cumulative breastfeeding duration and PMOP, women without the history of pregnancy were excluded from this study. Since the bone mineral are primarily lost to infants during lactation, we hypothesized the amount of breastfeeding can affect the development of PMOP. Although the frequency and duration of daily breastfeeding were not calculated in this study, the feeding pattern and the total breastfeeding duration were considered to indirectly reflect the cumulative amount of breastfeeding, which is the strength of the study.
Limited by facility condition, BMD was measured using the technique of QUS instead of a dual X-ray absorptiometer (DXA). Although QUS has been considered as effective as DXA in determining BMD without using ionizing radiation and non-invasive [
29], QUS in community healthcare center failed to measure BMD in multi-site of the body. Indeed, the inconsistent result of the relationship between breastfeeding and PMOP may due to different testing region. In the 2010 Korea National Health and Nutrition Examination Survey, a deleterious effect of prolonged breastfeeding duration on lumbar spine BMD was found, but this conclusion was discrepant in femur neck and total femur [
13]. Moreover, animal experiments also showed that the capability of recovery of bone mass may differ from site to site [
30,
31]. Therefore, further studies were necessary to explore the effect of breastfeeding on bone mass in which skeletal sites may be irreversibly. Another limitation is because of limited sample size and retrospective cross-sectional study design. It was possible that participants were too old to recall accurate information, which may cause recall bias, and we did not know whether participants had had osteoporosis before menopause. The next step is to conduct a multi-center prospective cohort study with large sample size to confirm the result of our study. Moreover, this study only recorded the amount of additional calcium and vitamins D taken by participants without considering dietary intake. However, it was reported that dietary Calcium may not related to postmenopausal bone loss [
32], so this bias had little impact on the results of our study.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.