Key words
adolescent - breastfeeding - bone mineral density - postmenopausal osteoporosis -
interpregnancy interval
Schlüsselwörter
Adoleszenz - Stillen - Knochenmineraldichte - postmenopausale Osteoporose - Abstand
zwischen Schwangerschaften
Introduction
Osteoporosis (OP) is the most common metabolic bone disease characterized by low bone
mass. Osteoporotic fractures cause severe morbidity and mortality in women, and postmenopausal
osteoporosis is an important public health problem [1]. Risk factors for osteoporosis (OP) are multifactorial. These include white race,
female sex, low body mass index (BMI), occupation, socioeconomic status, family history,
fracture history, diet, habits such as tea, coffee and smoking, sedentary lifestyle,
and drugs with adverse effects on the bone [1], [2].
Reproductive factors are known to affect bone mineral density (BMD) [3], [4], [5], [6], [7], [8], [9], [10], [11]. Calcium loss from maternal bone and decreased BMD have been observed especially
during pregnancy and lactation [4], although this loss has been reported to recover within 6 – 12 months [5]. There is no consensus on whether the effect of reproductive factors on the bone
is positive or negative.
The term “adolescent” is often used synonymously with “teenager” and refers to the
transition from childhood to adulthood, when there is not only physical development,
but also sexual, and psychosocial development. According to the World Health Organization
(WHO) definition, the adolescent period is between the ages of 10 and 19 years, with
the period of 10 – 14 years considered as early adolescence and the period of 15 to
19 years, late adolescence [12]. The adolescent period is important for total bone mass, and total bone mass is
significant in osteoporosis [6]. There are conflicting study results showing the effect on the bone of a single
birth in adolescence, but there are no studies in literature which have investigated
the effect of multiple pregnancies [3], [7]. The current study was planned from the hypothesis that two births during this period
would have a greater effect on the bone compared to a single birth. Factors such as
diet and lifestyle habits were questioned.
In this study, an investigation was made of the effect on bone density in the postmenopausal
period of first gestational age, adolescent pregnancy, multiple births during this
period, duration of lactation, interpregnancy intervals and dietary habits and lifestyle.
It was aimed with this cross-sectional study to make a contribution to this issue
which has not been clarified in literature.
Materials and Methods
Study design and patient population
This cross-sectional study was conducted at Düzce University Medical Faculty Training
and Research Hospital between September 2017 and February 2018. Patients were included
from those who underwent bone mineral density (BMD) testing as a requirement for the
Gynecological Diseases and Obstetrics Polyclinic or for the Physical Therapy and Rehabilitation
Polyclinic for any reason, with normal references values in blood tests, Ca and 25-OH
D-vit, had delivered at least once, had been through a normal menopause and did not
meet the exclusion criteria. Patients who had not had menstrual bleeding for the previous
12 months were considered postmenopausal.
Patients were excluded if they were aged < 40 years, had cancer and had undergone
chemoradiotherapy, thyroid or parathyroid surgery, were taking medication for parathyroid
diseases, had uncontrolled thyroid disease, were taking medication for bone remodeling,
had received hormone replacement therapy, had bone metabolism disease, were taking
medicine (heparin, corticosteroid) which would affect bone, were surgically menopausal,
or had any autoimmune disease such as rheumatoid arthritis. Patients were also excluded
from the study due to causes of secondary osteoporosis such as chronic liver disease,
type 1 diabetes, malabsorption, chronic malnutrition, and hypogonadism. Therefore,
of the 256 patients initially reviewed, 31 were excluded due to nulliparity, 24 due
to other diseases and medication and 5 as they did not wish to participate, and the
study was completed with 196 patients ([Fig. 1]).
Fig. 1 The progress of all participants through the trial; flow diagram.
The demographic data of the patients were obtained in one-to-one interviews. Detailed
questions were asked in respect of age, height, weight, menarche age, first gestational
age, number of births before the age of 20, the number of pregnancies and deliveries
which completed six months, number of live births, birth dates, the shortest interpregnancy
interval, breastfeeding times for each pregnancy, age of first breastfeeding, age
of last breastfeeding, total breastfeeding time, marital status, level of income,
education level, smoking, exercise habits, black tea consumption, caffeine intake,
comorbid disease (DM, HT, hypothyroidism, asthma), current medication (iron supplements,
Vitamin D, aspirin, thyroid hormone, glucocorticoid), weight gain during the last
30 years, the lifetime maximum weight, shortening of height, and surgical history
(oophorectomy, hysterectomy, parathyroidectomy, etc.). The anamnesis was taken by
a single physician.
The exercise limit was defined as 30 min/day of moderate intensity activity, and caffeine
consumption was defined as 2 cups/day black tea and > 1 cup/day of coffee. Calcium
consumption was calculated from the self-reported milk and dairy products intake.
The association of the number of births in the adolescent period was the primary focus
of this study, so adolescent pregnancy history and total breastfeeding duration with
the primary outcome of bone mineral density was analyzed taking interpregnancy intervals
into consideration.
In all cases, bone mineral density (BMD) was measured with a GE/LUNAR DPX PRO DEXA
device (Dual Energy X-Ray Absorptiometry). The BMD values of the patients were measured
from the lumbar vertebrae (in the anterior position between L1–L4) and from the proximal
femur (neck, trochanter and Wardʼs triangle), then the femur, lumbar and total BMD
values and T scores were recorded. Osteoporosis was defined as a T score ≤ − 2.5,
osteopenia as a T score from − 1.1 to − 2.4 and normal as a T score ≥ − 1.0. The lower
of either the spine or the hip T-scores was used for classification according to the
ISCD guidelines [13].
The age of 27 years was considered the age for complete peak-bone-mass [14]. Patients who gave birth at the age of 20 years and above were divided again into
2 subgroups of those who gave birth between the age of 20 – 27 years and those who
gave birth aged > 27 years. The interpregnancy interval was calculated as the time
between the month of birth and the date of the conception of the next pregnancy.
Power analysis
As a result of the power analysis, it was decided to include a total of 118 individuals
in order to determine the statistical significance of a 0.8 unit difference (effect
size 0.8) in terms of femoral neck and a 0.7 unit difference (effect size 0.7) in
terms of lumbar spine between the groups in conditions of 90% power and 5% type 1
error. The power analysis was applied with PASS v.11 package software.
Statistical analysis
Continuous data were summarized as mean and standard deviation and categorical data
as frequency and percentage. In the statistical evaluation of the data, the Independent
Samples t-test was used in two group comparisons depending on the distribution pattern
of data. One-way ANOVA or Kruskal-Wallis tests were used in comparisons of three or
more groups. Associations between categorical variables were examined using the Pearson
Chi-Square or Fisherʼs Exact tests. In the univariate analyzes, first, univariate
logistic regression analysis was performed for each of the factors associated with
osteoporosis and then the OR values were calculated using multivariate logistic regression
analysis by taking all as the model. Statistical analyses were performed using SPSS
v.22 packet software and the significance level was taken as 0.05.
Results
Demographic data
Demographic data according to BMD are shown in [Table 1]. When the total 196 postmenopausal patients were analyzed according to bone mineral
densities, three groups were formed; 109 (55.6%) as normal, 63 (32.1%) as osteopenic
and 24 (12.2%) as osteoporotic. The mean age was 60.6 years and no difference was
determined between the groups in respect of age (p = 0.931). The BMI values of the
patients were statistically different, and the group with higher BMI was found to
have better BMD (p = 0.001). There was no significant difference between the bone
mineral density groups in terms of current monthly income, level of education, smoking,
alcohol use, regular caffeine consumption, exercise, current consumption of milk and
dairy products, menarche age and menopause duration ([Table 1]). The menopause age was statistically significantly lower in the osteoporotic group
compared to the normal group (p = 0.004). There was a significant difference in the
rates of osteoporosis according to the style of dressing, and it was seen that 91.7%
of the patients who were diagnosed with osteoporosis wore the chador or traditional
headscarf and 76.2% of patients in the osteopenia group also wore the chador or traditional
headscarf (p = 0.033).
Table 1 Demographic data according to BMD.
|
Total (n = 196)
|
Normal (n = 109)
|
Osteopenia (n = 63)
|
Osteoporosis (n = 24)
|
p
|
|
Data are presented n (%) or as mean ± SD. Abbreviations: y: years; BMI: body mass
index; SD: standard deviation
|
|
Age (y)
|
60.6 ± 8.02
|
60.5 ± 8.77
|
60.8 ± 8.72
|
60.04 ± 7.87
|
0.931
|
|
BMI
|
32.9 ± 6.2
|
35.13 ± 6.18
|
30.12 ± 4.29
|
30.70 ± 6.88
|
0.001
|
|
Marital status
|
|
|
|
|
0.451
|
|
|
34 (17.3%)
|
16 (14.7%)
|
14 (22.2%)
|
4 (16.7%)
|
|
|
162 (82.7%)
|
93 (85.3%)
|
49 (77.8%)-
|
20 (83.3%)
|
|
|
–
|
–
|
–
|
–
|
|
Income levels
|
|
|
|
|
0.289
|
|
|
72 (36.7%)
|
39 (35.8%)
|
23 (36.5%)
|
10 (41.7%)
|
|
|
82 (41.8%)
|
41 (37.6%)
|
31 (49.2%)
|
10 (41.7%)
|
|
|
36 (18.4%)
|
26 (23.9%)
|
6 (9.5%)
|
4 (16.7%)
|
|
|
6 (3.1%)
|
3 (2.8%)
|
3 (4.8%)
|
0 (0.0%)
|
|
Education level
|
|
|
|
|
0.654
|
|
|
44 (22.4%)
|
24 (22.0%)
|
13 (20.6%)
|
7 (29.2%)
|
|
|
114 (58.2%)
|
66 (60.6%)
|
35 (55.6%)
|
13 (54.2%)
|
|
|
7 (3.6%)
|
5 (4.6%)
|
1 (1.6%)
|
1 (4.2%)
|
|
|
20 (10.2%)
|
9 (8.3%)
|
8 (12.7%)
|
3 (12.5%)
|
|
|
11 (5.6%)
|
5 (4.6%)
|
6 (9.5%)
|
0 (0.0%)
|
|
Smoking
|
42 (21.4%)
|
23 (21.1%)
|
14 (22.2%)
|
5 (20.8%)
|
0.982
|
|
Alcohol
|
2 (1.0%)
|
1 (0.9%)
|
0 (0.0%)
|
1 (4.2%)
|
0.333
|
|
Caffeine
|
160 (81.6%)
|
90 (82.6%)
|
52 (82.5%)
|
18 (75.0%)
|
0.670
|
|
Tea
|
140 (71.4%)
|
81 (74.3%)
|
44 (69.8%)
|
15 (62.5%)
|
0.482
|
|
Exercise
|
26 (13.3%)
|
16 (14.7%)
|
10 (15.9%)
|
0 (0.0%)
|
0.120
|
|
Dressing style
|
|
|
|
|
0.033
|
|
|
3 (1.5%)
|
0 (0.0%)
|
2 (3.2%)
|
1 (4.2%)
|
|
|
162 (82.7%)
|
95 (87.2%)
|
46 (73.0%)
|
21 (87.5%)
|
|
|
31 (15.8%)
|
14 (12.8%)
|
15 (23.8%)
|
2 (8.3%)
|
|
Sufficient calcium consumption
|
74 (37.8%)
|
44 (40.4%)
|
21 (33.3%)
|
9 (37.5%)
|
0.657
|
|
Height reduction
|
108 (55.1%)
|
66 (60.6%)
|
29 (46.0%)
|
13 (54.2%)
|
0.182
|
|
Parity (median [min–max])
|
3 (1 – 11)
|
3 (1 – 9)
|
3 (1 – 11)
|
3 (1 – 9)
|
0.763
|
|
Menarche age (y)
|
13.6 ± 1.54
|
13.56 ± 1.42
|
13.79 ± 1.69
|
13.96 ± 1.68
|
0.421
|
|
Menopause age (y)
|
46.13 ± 5.40
|
47.18 ± 5.13
|
45.25 ± 5.08
|
43.67 ± 6.36
|
0.004
|
|
Menopause duration (y)
|
14.46 ± 9.59
|
13.40 ± 9.38
|
15.57 ± 10.06
|
16.38 ± 9.04
|
0.210
|
Reproductive history
The reproductive histories of the patients are shown in [Table 2]. There was no difference in the first and last breastfeeding age of the patients
(p = 0.761, p = 0.103). There was no statistically significant difference in the duration
of total breastfeeding (p = 0.596).
Table 2 Reproductive history according to BMD.
|
Total (n = 196)
|
Normal (n = 109)
|
Osteopenia (n = 63)
|
Osteoporosis (n = 24)
|
p
|
|
Data are presented as n (%) or as mean ± SD. Abbreviations: m: months; y: years; BMI:
body mass index; SD: standard deviation
|
|
First breastfeeding age (y)
|
21.37 ± 3.87
|
21.56 ± 3.93
|
21.09 ± 4.17
|
21.29 ± 2.85
|
0.761
|
|
Last breastfeeding age (y)
|
30.46 ± 5.08
|
31.02 ± 5.05
|
30.30 ± 5.29
|
28.58 ± 4.28
|
0.103
|
|
Age of first pregnancy (y)
|
20.89 ± 3.98
|
21.06 ± 4.14
|
20.68 ± 4.11
|
20.75 ± 2.82
|
0.825
|
|
Age of first pregnancy (y)
|
|
|
|
|
0.936
|
|
|
82 (41.8%)
|
45 (41.3%)
|
27 (42.9%)
|
10 (41.7%)
|
|
|
97 (49.5%)
|
54 (49.5%)
|
30 (47.6%)
|
13 (54.2%)
|
|
|
17 (8.7%)
|
10 (9.2%)
|
6 (9.5%)
|
1 (4.2%)
|
|
Adolescence pregnancy
|
82 (41.8%)
|
45 (41.3%)
|
27 (42.9%)
|
10 (41.7%)
|
0.980
|
|
Interpregnancy interval (y)
|
2.14 ± 1.54
|
2.45 ± 1.77
|
1.92 ± 1.20
|
1.40 ± 0.73
|
0.005
|
|
Interpregnancy interval (y; n = 169)
|
|
|
|
|
0.020
|
|
|
26/169 (15.4%)
|
10/94 (10.6%)
|
10/51 (19.6%)
|
6/24 (25.0%)
|
|
|
89/169 (52.7%)
|
44/94 (46.8%)
|
30/51 (58.8%)
|
15/24 (62.5%)
|
|
|
29/169 (17.2%)
|
21/94 (22.3%)
|
5/51 (9.8%)
|
3/24 (12.5%)
|
|
|
25/169 (14.8%)
|
19/94 (20.2%)
|
6/51 (11.8%)
|
0/24 (0.0%)
|
|
Repeated births during the adolescent period
|
14 (7.1%)
|
2 (1.8%)
|
6 (9.5%)
|
6 (25.0%)
|
0.001
|
|
Total duration of breastfeeding (m)
|
38.51 ± 33.73
|
37.52 ± 31.71
|
41.96 ± 40.17
|
34.38 ± 24.19
|
0.596
|
|
Total duration of breastfeeding (m)
|
|
|
|
|
0.689
|
|
|
33 (16.8%)
|
21 (19.3%)
|
11 (17.5%)
|
1 (4.2%)
|
|
|
24 (12.2%)
|
12 (11.0%)
|
9 (14.3%)
|
3 (12.5%)
|
|
|
34 (17.3%)
|
21 (19.3%)
|
7 (11.1%)
|
6 (25.0%)
|
|
|
42 (21.4%)
|
21 (19.3%)
|
14 (22.2%)
|
7 (29.2%)
|
|
|
16 (8.2%)
|
8 (7.3%)
|
6 (9.5%)
|
2 (8.3%)
|
|
|
47 (24.0%)
|
26 (23.9%)
|
16 (25.4%)
|
5 (20.8%)
|
It was observed that the onset of menopause at a later age decreased the risk of osteoporosis.
In patients with osteoporosis, the mean menopausal age was 43.67 ± 6.36 years, while
it was 45.25 ± 5.08 years in the osteopenic group and 47.18 ± 5.13 years in the normal
group (p = 0.004). In the results of the univariate logistic regression analysis,
the increase in menopausal age was seen to be associated with a decrease in the risk
of osteoporosis (OR = 0.911, 95% CI = 0.843 – 0.985; p = 0.019).
The increase in interpregnancy interval was found to decrease osteoporosis risk. The
mean interpregnancy interval was 1.4 ± 0.73 years in patients with osteoporosis, whereas
it was longer in patients with osteopenia (1.92 ± 1.20 years) and in the normal BMD
group (2.45 ± 1.77 years) (p = 0.005). Each year increase in interpregnancy interval
was determined to decrease the risk of osteoporosis 2-fold (OR = 0.490, 95% CI = 0.288 – 0.835;
p = 0.009). In the results of the multivariate analysis, it was seen that for each
year increase in interpregnancy interval, there could be as much as a 2.6-fold decrease
in osteoporosis risk (OR = 0.386, 95% CI = 0.211 – 0.707; p = 0.002) ([Table 3], [Fig. 2]).
Table 3 Logistic regression analysis for osteoporosis.
|
Univariate
|
Multivariate
|
|
OR (95% CI)
|
p value
|
OR (95% CI)
|
p value
|
|
BMI
|
0.930 (0.864 – 1.002)
|
0.056
|
0.924 (0.855 – 0.998)
|
0.045
|
|
Menopause age
|
0.911 (0.843 – 0.985)
|
0.019
|
–
|
–
|
|
Interpregnancy interval
|
0.490 (0.288 – 0.835)
|
0.009
|
0.386 (0.211 – 0.707)
|
0.002
|
|
Last breastfeeding age
|
0.916 (0.838 – 1.001)
|
0.054
|
0.849 (0.764 – 0.944)
|
0.002
|
|
Repeated births during adolescence
|
6.833 (2.131 – 21.908)
|
0.001
|
–
|
–
|
Fig. 2 Interpregnancy intervals for groups.
An older last breastfeeding age was not determined to be statistically significant,
but it was seen to decrease the risk of osteoporosis. The mean age of the last breastfeeding
was 28.58 ± 4.28 years in the patients with osteoporosis, 30.30 ± 5.29 years in the
osteopenic group and 31.02 ± 5.05 years in the normal group. According to the univariate
logistic regression analysis, the increase in the last breastfeeding age led to a
decrease in the limit of the risk of osteoporosis (OR = 0.916, 95% CI = 0.838 – 1.001;
p = 0.054) and the multivariate analysis showed that this reduction was more noticeable
(OR = 0.849, 95% CI= 0.764 – 0.944; p = 0.002) ([Table 3]).
The risk of osteoporosis was determined to increase with a history of more than one
birth in late adolescence. None of the patients in this study gave birth more than
twice in this period. Of the osteoporosis group with pregnancies in adolescence, 25%
had two deliveries. Univariate logistic regression analysis revealed that multiple
births in the late adolescent period increased the risk of osteoporosis 6.833-fold
(OR = 6.833, 95% CI= 2.131 – 21.908; p = 0.001).
Discussion
The aim of this study was to examine the effects on bone mineral density in postmenopausal
patients of parity, age at first pregnancy, the number of pregnancies and births in
adolescence, the interpregnancy interval throughout the entire reproductive age, total
breastfeeding period, menarche age, menopausal age and lifestyle habits.
The exclusion of many confounding factors, questioning of living conditions such as
diet features, exercise, clothing, and income status can be considered to be the strengths
of the study. The power of the study could have been reduced as the number of osteoporotic
patients in the whole study group of patients was low. Risk factors for osteoporosis
are known to be smoking, income level, calcium consumption, and low levels of exercise
and no significant differences thought to be connected to these factors were determined
between the groups. In accordance with previous findings in literature, low BMI and
modest Islamic clothing were found to increase the development of osteoporosis [15], [16]. Since the amount of breastfeeding can affect the bone as well as the duration of
breastfeeding, the fact that the number of daily breastfeedings and duration were
not considered could constitute a limitation of the study. Another limitation is that
there was no evaluation of differences between single and multiple pregnancies. It
is also possible that individuals do not give accurate information in the face-to-face
interviews as they cannot recall such old events.
Bone mineral densities may be affected by endocrine changes during pregnancy. In a
large-scale study, it was found that the risk of postpartum hip fracture declined
by 10% in the following years [17]. Similar studies show that a greater number of births is protective against osteoporotic
fracture risk [18], [19]. In a large-scale observational study investigating the effect of parity and breastfeeding
on BMD in postmenopausal women, it was observed that parity and breastfeeding had
no effect on bone density [6]. In this regard, the results of the current study support the studies in literature
showing that parity has no effect on bone mineral density.
Since osteoblastic activity continues during adolescence, the effect of pregnancy
in this period on bone mass may be different from that of pregnancy in later reproductive
years. Adolescence is an important period for total bone mass formation and total
bone mass is a determinant of osteoporosis. 90% of the total bone mass is achieved
at the age of 18 and 99% at the age of 27 [20]. The issue of how the adolescent pregnancy history affects bone mineral densities
in later life remains a matter of current debate. There are studies in literature
showing varying results ranging from harmful effects to protective effects [3], [21], [22]. In one study, adolescent pregnancy was reported to be a predictor of postmenopausal
osteoporosis and increase the risk 2.2-fold [21]. In another study, while no difference was found in the bone mineral densities of
the lumbar region of adolescent pregnancy patients, bone mineral densities of the
femur neck were found to be better than those of the control group [3]. In another study, it was suggested that pregnancy during adolescence may damage
physiological processes and thus have harmful effects on peak bone mass [22]. In the current study, there was no significant difference in both the lumbar and
femoral bone mineral densities of the patients who gave birth and those who did not
give birth in the adolescent period. Patients who gave birth twice in adolescence
were determined to have a 6.8 times higher risk of osteoporosis. This was a finding
that showed that late adolescent pregnancy had a negative effect on bone mineral density.
The difference between the single child and two children may be due to the small sample
size.
Early onset of menopause has been shown to increase postmenopausal osteoporosis [9], [11] and the results of the current study are consistent with previous findings in literature.
The association between BMD and the duration of breastfeeding was investigated, and
the results were seen to be as complex as the correlation between parity and late
adolescent pregnancy. While the protective effect of breastfeeding on the lumbar vertebrae
has been reported, it has been stated that every 6 months of breastfeeding reduced
the risk of hip fracture by 13%, and as a result, the risk for hip fracture decreased
as the duration of breastfeeding increased. However, in another study, it was shown
that a six-month lactation period was accompanied by a loss of 5 – 7% in the spine
and hip bone mass [23], [24], [25]. It was reported that even if calcium intake in the diet is sufficient during the
lactation period, an important part of the calcium requirement is provided by the
maternal skeleton, which is why long-term lactation causes more bone loss in the mother
[26]. The frequency of postmenopausal osteoporosis was observed to be high in the group
of patients who breastfed for over one year and whose first breastfeeding age was
below 27 years [27]. In the same study, the protective effect of parity was mentioned. Similarly, in
the current study, no relationship was found between the first breastfeeding age and
postmenopausal osteoporosis. It has been reported that breastfeeding had no negative
effect on peak bone mass or risk of hip fracture and that BMD values immediately returned
to their previous values when breastfeeding was stopped [5], [14], [28], [29]. This improvement requires a certain interval between two pregnancies. The risk
of osteoporosis in the postmenopausal period has been found to be higher in patients
with an interpregnancy interval shorter than one year [30]. In the current study, in line with literature, a longer interpregnancy interval
was found to result in a decrease in osteoporosis risk. No statistically significant
difference was determined between the total breastfeeding period and BMD. In studies
which have shown that the breastfeeding period has a positive effect on bone mineral
density, this could be due to the association of the contraceptive effect of breastfeeding
and the longer interpregnancy interval.
Conclusion
While no effect of breastfeeding on BMD was determined in this study, a shorter interpregnancy
interval and having more than one child in late adolescence was seen to have a negative
effect on bone mineral density in the postmenopausal period. The adolescent period
is an important period for total bone mass formation and more than one pregnancy at
this age can trigger postmenopausal osteoporosis. Postmenopausal bone density is multifactorial
throughout the life of a woman, so there is conflicting information in literature.
More prospective randomized studies are required to clarify this issue.
Funding
None of the authors have received funding for this article.
Ethical approval
Written consent documents were obtained from all the patients who agreed to participate
in the study. Before starting the study, approval was granted by the Ethics Committee
of Duzce University Medical Faculty (2017/155). All procedures were carried out in
line with the Helsinki Declaration.
