Abstract
Introduction:
Evidence on the association between early-life malnutrition exposure at different developmental stages and the subsequent risk of osteoporosis and fractures in adulthood remains sparse and equivocal. This study sought to elucidate the relationship between malnutrition exposure in early-life and the occurrence of osteoporosis and fractures later in life.
Methods:
This research is a cross-sectional analysis carried out within the framework of the China Community-based Cohort of Osteoporosis (CCCO), an ongoing community-based cohort study. Participants were stratified by birthdate into several categories: nonexposed, fetal, early childhood, mid-childhood, late childhood, and adolescence exposure groups. The nonexposure and adolescence exposure groups were consolidated into an ‘age-matched group’ to provide a robust comparative framework for analyzing the probability of developing osteoporosis (defined as a T-score ≤−2.5 in bone mineral density) and the frequency of self-reported fracture. Multiple logistic regression models were utilized to investigate the association between early-life malnutrition exposure and the risks of osteoporosis and fracture. Additionally, our findings were validated in the China Northwest Cohort (CNC).
Results:
A total of 12 789 participants were included into the final analysis. After adjusting for various covariates, individuals exposed to malnutrition during their fetal and childhood stages (early, middle, and late) increased the likelihood of developing osteoporosis in adulthood, compared to their age-matched counterparts. In these four groups, the ORs (95% CI) for osteoporosis risk were 1.223 (1.035 to 1.445), 1.208 (1.052 to 1.386), 1.249 (1.097 to 1.421), and 1.101 (1.001 to 1.210), respectively (all P-values <0.05). Specifically, the late childhood exposure group showed a heightened risk of fracture, with an OR (95% CI) of 1.155 (1.033–1.291) and a P-value of 0.01127. Stratified analyses further found a significant correlation between early-life exposure to malnutrition and an elevated risk of osteoporosis in participants with lower educational attainment, overweight, or obese participants. Additionally, corroborating evidence from the CNC confirmed the influence of malnutrition exposure on osteoporosis risk.
Conclusions:
Early-life exposure to malnutrition had a detrimental impact on bone health. Individuals who had experienced malnutrition during fetal and childhood stages (early, middle, and late) exhibited a high susceptibility to osteoporosis in adulthood, compared to age-matched cohorts. This susceptibility was particularly pronounced in women, and individuals who were overweight or obese, or had lower levels of education.
Keywords: bone mineral density, cross-sectional study, malnutrition Exposure, osteoporosis
Introduction
Highlights
Individuals exposed to malnutrition during their fetal and childhood stages (early, middle, and late) increased the likelihood of developing osteoporosis in adulthood.
Individuals exposed to malnutrition during their late childhood showed a heightened risk of fracture.
Participants with lower educational attainment and those who were overweight or obese, both exhibited a significant correlation between early-life exposure to malnutrition and an elevated risk of osteoporosis.
Osteoporosis, a chronic, systemic metabolic bone disorder in adults, arises from an imbalance between osteoblasts and osteoclasts, as well as subsequent deterioration of bone microstructure and progressive bone loss1,2. Patients with osteoporosis, due to their diminished bone density and strength, face a high-risk of fragility fracture3, as shown by ~9 million fracture worldwide each year4. It is estimated that 30% of women and 20% of men over the age of 50 will experience an osteoporotic fracture in their later lives5. Fractures secondary to osteoporosis are the leading cause of disability and mortality among older adults5. Osteoporosis and secondary fractures also impose a significant financial burden on patients’ families6.
Osteoporosis is routinely diagnosed according to bone mineral density (BMD), typically determined through dual-energy X-ray absorptiometry (DXA)7. According to the guidelines established by the WHO, osteoporosis is defined as a BMD more than 2.5 SD below the mean reference in the young adult population3. Epidemiological research indicates that in China, osteoporosis affects an estimated 6.46% of males and 29.13% of females aged 50 years and above8. Recent survey data indicate that China has the second highest global disability burden attributed to fracture resulting from low BMD9. Advanced age, deficiency in sex hormones, and elevated oxidative stress are significant contributors to the development of osteoporosis. Moreover, lifestyle factors, such as dietary habits and physical activity, also influence both bone density and structural integrity10. However, how these factors influence the risk of osteoporosis remain to be eclucidated. Delving into the intricate mechanisms influencing osteoporosis prevalence across diverse populations, alongside pioneering precise diagnostic, and therapeutic strategies, represents the forefront of osteoporosis research11,12.
Growing research demonstrates a robust correlation between early-life nutritional status and adult bone mineral density. For instance, a study conducted in Hong Kong13 has found that early-life malnutrition is associated with an increased prevalence of osteoporosis in postmenopausal women. However, further research conducted in the urban areas of Chongqing, China, suggests that early-life inadequate nutrition exerts a prominent effect on the skeletal health only in adult males14. Although some previous studies have analyzed the relationship between early-life nutritional status and BMD in adulthood, several limitations remain: small sample sizes; assessment of BMD using quantitative ultrasound (QUS) rather than DXA, which is the gold standard for BMD measurement; and adjustment for only a limited number of risk factors for osteoporosis or fractures in model construction, which may lead to potential bias in the models. Therefore, future research on the relationship between early-life nutritional status and adult BMD needs to include larger sample sizes, utilize DXA for measurement, and adjust for a comprehensive range of potential risk factors to obtain more reliable and generalizable conclusions.
Chinese people were inflicted with a severe food shortage during 1959–1961, suggesting their poor nutritional status in the early-life. Here, utilizing the data from the China Community-based Cohort of Osteoporosis (CCCO), we examined how early-life exposure to malnutrition affects osteoporosis risk later in life.
Materials and methods
Subjects
The work has been reported in line with the strengthening the reporting of cohort, cross-sectional, and case–control studies in surgery (STROCSS) criteria15 (Supplemental Digital Content 1, http://links.lww.com/JS9/D373). This cross-sectional analysis was performed as part of the ongoing prospective CCCO study. All methodologies adhered strictly to the ethical guidelines of the Declaration of Helsinki, including all subsequent amendments or comparable ethical standards. Written informed consent was obtained from all participants prior to their inclusion into the study. The protocol for the CCCO has been detailed in our previously published work16.
We enrolled a total of 22 018 participants from urban residential communities and rural villages across seven regions in China, encompassing Shanghai (East), Guangdong (South), Gansu (West), Beijing (North), Jilin (Northeast), Yunnan (Southwest), and Jiangxi (Southwest). All participants were recruited using a multistage stratified cluster random sampling approach. Initially, individuals lacking recorded birth dates were excluded from the study (n=1985). To minimize age-related biases, participants born before 30 September 1940 (n=805) and after 30 September 1972 (n=432) were also excluded. We further excluded individuals born between 1 October 1958 and 30 September 1959 (n=654), and between 1 October 1961 and 30 September 1962 (n=616), to avoid misclassification of exposure times. Also excluded were those lacking bone mineral density (BMD) data (n=3807), diagnosed with malignant tumors (n=130), under 50 years of age (n=25), and in the premenopausal age (n=731). Finally, 12 789 participants were included.
As did in previous studies17–20, the participants were categorized into six groups: a nonexposure group (born between 1st October 1962 and 30th September 1972), a fetal exposure group (born between 1st October 1959 and 30th September 1961), an early childhood exposure group (born between 1st October 1956 and 30th September 1958), a mid-childhood exposure group (born between 1st October 1954 and 30th September 1956), a late childhood exposure group (born between 1st October 1949 and 30th September 1954), and an adolescent exposure group (born between 1st October 1940 and 30th September 1949). To mitigate age-related biases in the study outcomes, we combined the nonexposure and adolescent exposure groups into a single, age-matched control group.
Questionnaire and measurements
Each participant was interviewed face-to-face by a well-trained researcher and allowed to complete a validated, structured questionnaire in print format. Demographic data (including sex, age, education level, marital status, household income, and regional residence) and lifestyle factors (such as smoking, alcohol consumption, and physical activity) were collected. Additionally, participants provided data about self-reported medical conditions (hyperlipidemia, hypertension, diabetes, malignancies, and history of fracture) and details on pharmaceutical and supplement usage (such as antiosteoporosis medications and calcium supplements).
Educational level was classified into two categories: junior high school and below, and senior high school and above; marital status into married or other (including unmarried, divorced, or widowed); monthly household income into less than 5000 RMB and 5000 RMB or more; smoking into never smoking, current smoking (including active and long-term passive exposure), and former smoking; alcohol consumption into never drinking, current drinking, and former drinking; residential area into northern and southern China (divided by the Huaihe River-Qinling Mountains Line); physical activity into sedentary or no, light, moderate, and vigorous, according to the International Physical Activity Questionnaire (IPAQ).
Participants’ medical records were reviewed to confirm the history of chronic diseases and fracture. Height and weight were measured with participants standing in indoor clothing and bare feet. BMI was calculated by dividing the weight in kilograms by the square of their height in meter, with a value under 24 indicating underweight or normal and 24 or over indicating overweight or obese.
After overnight fast for a minimum of 8 h, venous blood was sampled from all participants. The serum was then separated and analyzed to determine the concentrations of N-terminal propeptide of type I collagen (PINP), β-C-terminal telopeptide of type I collagen (β-CTX), total 25-hydroxyvitamin D [25(OH)D] [including 25(OH)D3 and 25(OH)D2], and total calcium (Ca). BMD was assessed in the lumbar spine and both hip using the Hologic Discovery CI DXA densitometer. All devices used were of the same model and had passed an annual inspection to ensure their accuracy and consistency.
Definition and diagnostic criteria
Adhering to the criteria established by the WHO21, a T-score of −2.5 or lower was indicative of osteoporosis, a T-score ranging from −2.5 to −1.0 osteopenia, and a T-score of −1.0 or higher normal bone mass.
Statistical analysis
Continuous variables were described as means and SD (mean±SD), while categorical variables were described as counts and percentages to depict the baseline characteristics of the cohort. Statistical differences between exposed and unexposed groups were assessed using one-way analysis of variance (ANOVA), χ 2 tests, and Kruskal–Wallis H tests. Logistic regression models were utilized to calculate the odds ratios (ORs) and 95% CIs for evaluating the association between early-life exposure to malnutrition and the risk of osteoporosis and fracture in adulthood. Linear regression models explored the relationship between early-life malnutrition exposure and BMD. All models were adjusted for potential confounders, including sex, age, education levels, marital status, household income, smoking, alcohol consumption, region of residence, chronic conditions (such as hyperlipidemia, hypertension, and diabetes mellitus), physical activity, and the use of antiosteoporosis medications and calcium supplements. Further stratified analyses were conducted based on sex, BMI, education level, and geographic region. The statistical analyses were performed using the EmpowerStats (http://www.empowerstats.com) and R (http://www.r-project.org) software. Statistical significance was determined at two-sided P-values <0.05.
External validation
The China Northwest Cohort (CNC), part of the CCCO, was employed for external validation. The CNC initially recruited 15 946 participants, of whom 7651 met the inclusion and exclusion criteria specified for this study. In the CNC, participants’ heel BMD was measured using ultrasound bone sonometers. Logistic regression models were applied to calculate the ORs and 95% CIs to assess the association between early-life malnutrition exposure and the subsequent risk of osteoporosis and fracture in adulthood.
Results
Sample description
Table 1 presents the key characteristics of the study sample. After exclusion, a total of 12 789 individuals were included (Fig. 1). Of these, 862 participants had fetal exposure to malnutrition, while 1392, 1413, 3507, and 3280 experienced malnutrition exposure during early childhood, mid-childhood, late childhood, and adolescence, respectively. An additional 2398 participants were not exposed. Compared to their nonexposure participants, individuals exposed to malnutrition were characterized by an older age, a lower educational level, a higher BMI, a greater likelihood of smoking, a reduced physical activity, a higher prevalence of hypertension, diabetes, and hyperlipidemia, and a larger consumption of calcium supplements (P<0.05 for all comparisons). Biochemically, this cohort had lower serum levels of PINP, total 25(OH)D, total calcium, lumbar spine BMD, and hip BMD. Notably, the nonexposure group demonstrated a higher serum β-CTX level.
Table 1.
Baseline characteristics of participants in different malnutrition exposure cohorts.
| Unexposed group | Adolescence exposure group | Late childhood eExposure group | Mid-childhood exposure group | Early childhood exposure group | Fetal exposure group | |
|---|---|---|---|---|---|---|
| N | 2398 | 3280 | 3507 | 1413 | 1329 | 862 |
| Age, mean±SD, years | 53.796±2.878 | 71.774±3.029* | 65.856±2.229* | 62.665±2.084* | 60.706±2.011* | 57.723±2.242* |
| BMI, mean ± SD, m2/kg | 23.927±3.474 | 24.380±3.461* | 24.447±3.368* | 24.076±3.245 | 24.440±3.318* | 24.113±3.376 |
| Sex, n (%) | ||||||
| Male | 447 (18.641%) | 1164 (35.488%) | 1149 (32.763%) | 400 (28.309%) | 325 (24.454%) | 168 (19.490%) |
| Female | 1951 (81.359%) | 2116 (64.512%) | 2358 (67.237%) | 1013 (71.691%) | 1004 (75.546%) | 694 (80.510%) |
| Education levels, n (%) | ||||||
| Junior high school and below | 1036 (43.203%) | 2097 (63.933%) | 2403 (68.520%) | 728 (51.522%) | 468 (35.214%) | 214 (24.826%) |
| Senior high school and above | 1362 (56.797%) | 1183 (36.067%) | 1104 (31.480%) | 685 (48.478%) | 861 (64.786%) | 648 (75.174%) |
| Marital status, n (%) | ||||||
| Married | 2243 (93.536%) | 2802 (85.427%) | 3205 (91.389%) | 1300 (92.003%) | 1219 (91.723%) | 782 (90.719%) |
| Other | 155 (6.464%) | 478 (14.573%) | 302 (8.611%) | 113 (7.997%) | 110 (8.277%) | 80 (9.281%) |
| Income levels, n (%), RMB/month | ||||||
| <5000 | 1169 (48.749%) | 1848 (56.341%) | 1833 (52.267%) | 660 (46.709%) | 670 (50.414%) | 428 (49.652%) |
| ≥5000 | 1229 (51.251%) | 1432 (43.659%) | 1674 (47.733%) | 753 (53.291%) | 659 (49.586%) | 434 (50.348%) |
| Smoking behavior, n (%) | ||||||
| Current | 2149 (89.616%) | 2733 (83.323%) | 2879 (82.093%) | 1185 (83.864%) | 1150 (86.531%) | 759 (88.051%) |
| Former | 166 (6.922%) | 312 (9.512%) | 387 (11.035%) | 153 (10.828%) | 127 (9.556%) | 73 (8.469%) |
| Never | 83 (3.461%) | 235 (7.165%) | 241 (6.872%) | 75 (5.308%) | 52 (3.913%) | 30 (3.480%) |
| Alcohol consumption, n (%) | ||||||
| Current | 2018 (84.153%) | 2649 (80.762%) | 2871 (81.865%) | 1172 (82.944%) | 1116 (83.973%) | 737 (85.499%) |
| Former | 308 (12.844%) | 476 (14.512%) | 501 (14.286%) | 182 (12.880%) | 172 (12.942%) | 106 (12.297%) |
| Never | 72 (3.003%) | 155 (4.726%) | 135 (3.849%) | 59 (4.176%) | 41 (3.085%) | 19 (2.204%) |
| Residence region, n (%) | ||||||
| North | 754 (31.443%) | 1086 (33.110%) | 1203 (34.303%) | 558 (39.490%) | 564 (42.438%) | 346 (40.139%) |
| South | 1644 (68.557%) | 2194 (66.890%) | 2304 (65.697%) | 855 (60.510%) | 765 (57.562%) | 516 (59.861%) |
| Chronic health conditions hyperlipidemia, n (%) | ||||||
| Yes | 2092 (87.239%) | 2559 (78.018%) | 2734 (77.958%) | 1079 (76.362%) | 1057 (79.533%) | 706 (81.903%) |
| No | 306 (12.761%) | 721 (21.982%) | 773 (22.042%) | 334 (23.638%) | 272 (20.467%) | 156 (18.097%) |
| Hypertension, n (%) | ||||||
| Yes | 1893 (78.941%) | 1595 (48.628%) | 2081 (59.338%) | 933 (66.030%) | 909 (68.397%) | 635 (73.666%) |
| No | 505 (21.059%) | 1685 (51.372%) | 1426 (40.662%) | 480 (33.970%) | 420 (31.603%) | 227 (26.334%) |
| Diabetes, n (%) | ||||||
| Yes | 2277 (94.954%) | 2689 (81.982%) | 3035 (86.541%) | 1238 (87.615%) | 1201 (90.369%) | 788 (91.415%) |
| No | 121 (5.046%) | 591 (18.018%) | 472 (13.459%) | 175 (12.385%) | 128 (9.631%) | 74 (8.585%) |
| Physical activity, n (%) | ||||||
| Sedentary or non-exercise | 350 (14.595%) | 1166 (35.549%) | 958 (27.317%) | 233 (16.490%) | 184 (13.845%) | 117 (13.573%) |
| Low | 1043 (43.495%) | 1319 (40.213%) | 1499 (42.743%) | 749 (53.008%) | 697 (52.445%) | 427 (49.536%) |
| Moderate | 822 (34.279%) | 631 (19.238%) | 795 (22.669%) | 330 (23.355%) | 361 (27.163%) | 246 (28.538%) |
| High | 183 (7.631%) | 164 (5.000%) | 255 (7.271%) | 101 (7.148%) | 87 (6.546%) | 72 (8.353%) |
| Calcium supplement use, n (%) | ||||||
| Yes | 2296 (95.746%) | 2695 (82.165%) | 3143 (89.621%) | 1184 (83.793%) | 995 (74.868%) | 786 (91.183%) |
| No | 102 (4.254%) | 585 (17.835%) | 364 (10.379%) | 229 (16.207%) | 334 (25.132%) | 76 (8.817%) |
| Antiosteoporosis drug use, n (%) | ||||||
| Yes | 2397 (99.958%) | 3234 (98.598%) | 3488 (99.458%) | 1410 (99.788%) | 1325 (99.699%) | 861 (99.884%) |
| No | 1 (0.042%) | 46 (1.402%) | 19 (0.542%) | 3 (0.212%) | 4 (0.301%) | 1 (0.116%) |
| Serum-related indices, mean±SD | ||||||
| Serum Ca, mmol/l | 2.363±0.092 | 2.322±0.099* | 2.333±0.095* | 2.334±0.087* | 2.348±0.095* | 2.354±0.093 |
| 25(OH)D (ng/ml) | 21.769±7.769 | 20.758±8.263* | 20.704±7.562* | 21.092±7.709 | 21.268±7.582 | 21.996±8.123 |
| β-CTX (ng/ml) | 0.335±0.169* | 0.319±0.206* | 0.317±0.261* | 0.309±0.265* | 0.312±0.272* | 0.314±0.147* |
| PINPI (ng/ml) | 58.485±23.831* | 49.301±21.169* | 50.303±22.712* | 52.167±23.276* | 53.309±21.391* | 56.626±26.024 |
| Total BMD and T-score | ||||||
| Total LS BMD | 0.863±0.146 | 0.854±0.193* | 0.853±0.175* | 0.842±0.163* | 0.847±0.148* | 0.853±0.144 |
| Total LS T-score | −1.746±1.308 | −1.876±1.696* | −1.877±1.514* | −1.952±1.416* | −1.895±1.307* | −1.822±1.285 |
| Total hip BMD | 0.829±0.132 | 0.780±0.153* | 0.802±0.141* | 0.800±0.140* | 0.807±0.130* | 0.823±0.126 |
| Total hip T-score | −1.032±1.011 | −1.199±1.281* | −1.182±1.096* | −1.306±0.968* | −1.206±0.974* | −1.071±0.948 |
Ca, calcium, 25(OH)D, 25-Hydroxyvitamin D, β-CTX, β-C-terminal telopeptide of type I collagen, OST, osteocalcin, PINP, N-terminal propeptide of type I collagen, BMD, bone mineral density, LS, lumber spine.
Mean±SD for continuous variables: P-values were calculated using a linear regression model with the unexposed group as the reference. Number (proportion) for categorical variables: P-values were calculated using the χ 2 test with the unexposed group as the reference. Statistical significance is defined as P<0.05 and is denoted by an asterisk (*).
Figure 1.
Flowchart of participants selected in the study.
Prevalence of osteoporosis and fracture
As illustrated in Figure 2, the prevalence of osteoporosis in the nonexposure, fetal exposure, early childhood exposure, mid-childhood exposure, late childhood exposure, and adolescent exposure groups was 31.276% (n=750), 34.803% (n=300), 36.569% (n=486), 40.340% (n=570), 39.378% (n=1381), and 41.585% (n=1364), respectively. The incidence of osteoporosis showed a significant increasing trend from the earlier to later exposure to malnutrition (P<0.001). Similarly, the incidences of fracture among the nonexposure, fetal-exposure, early childhood-exposure, mid-childhood-exposure, late childhood-exposure, and adolescent-exposure groups were 12.469% (n=299), 15.197% (n=131), 17.532% (n=233), 17.197% (n=243), 19.618% (n=688), and 20.518% (n=673), respectively. Trend analysis revealed a significant increasing trend in fracture incidence across these groups (P<0.04).
Figure 2.
Prevalence of osteoporosis or fracture by life different stage when exposed to malnutrition.
Association between early-life malnutrition exposure and BMD, OP, and fracture in adulthood
After adjusting for confounding factors, Table 2 illustrates that the fetal-exposure group (β: −0.016; 95% CI: −0.027 to −0.005; P=0.00457), the early childhood-exposure group (β: −0.021; 95% CI: −0.030 to −0.011; P<0.00001), the mid-childhood exposure group (β: −0.018; 95% CI: −0.027 to −0.009; P<0.0001), and the late childhood-exposure group (β: −0.006; 95% CI: −0.013 to 0.000; P=0.04715) all demonstrated a significantly lower total lumbar spine BMD, compared to the age-matched control group. Furthermore, the early childhood-exposure cohort exhibited a significantly lower BMD, compared to the age-matched cohort (β: −0.008; 95% CI: −0.015 to 0.000; P=0.04109). In contrast, there were no significant differences in total hip BMD between the fetal exposure, mid-childhood exposure, and late-childhood exposure groups (P>0.05).
Table 2.
Association between life stages when exposed to malnutrition and OP, fracture risk, and total BMD.
| OP | Fracture | Total LS BMD | Total hip BMD | |||||
|---|---|---|---|---|---|---|---|---|
| OR (95% CI) | P | OR (95% CI) | P | β (95% CI) | P | β (95% CI) | P | |
| Age-balanced group (Ref.) | 1 | / | 1 | / | 0 | / | 0 | / |
| Late childhood exposure group | 1.101 (1.001–1.210) | 0.04756 | 1.155 (1.033–1.291) | 0.01127 | –0.006 (−0.013, −0.000) | 0.04715 | 0.002 (−0.003, 0.007) | 0.43333 |
| Mid-childhood exposure group | 1.249 (1.097–1.421) | 0.00078 | 1.001 (0.853–1.173) | 0.9942 | –0.018 (−0.027, −0.009) | 0.00008 | –0.005 (−0.013, 0.002) | 0.13045 |
| Early childhood exposure group | 1.208 (1.052–1.386) | 0.0073 | 1.054 (0.893–1.245) | 0.53413 | –0.021 (−0.030, −0.011) | 0.00001 | 0.008 (−0.015, −0.000) | 0.04109 |
| Fetal exposure group | 1.223 (1.035–1.445) | 0.01779 | 0.973 (0.789–1.202) | 0.80268 | –0.016 (−0.027, −0.005) | 0.00457 | 0.004 (−0.005, 0.013) | 0.38375 |
BMD, bone mineral density; LS, lumber spine; OP, osteoporosis; Ref., reference.
The model was adjusted for sex, age, BMI, education levels, marital status, income levels, smoking behavior, alcohol consumption, residence region, hyperlipidemia, hypertension, diabetes, physical activity, calcium supplement use, and antiosteoporosis drug use. Statistical significance was set at P<0.05.
Furthermore, binary logistic regression was utilized to investigate the association between malnutrition exposure and the risks of osteoporosis and fracture. Individuals who were exposed to malnutrition during their early, middle, late childhood, or fetal exhibited an increased risk of developing osteoporosis in adulthood, compared to age-matched controls. This association remained significant even after adjusting for relevant confounding factors, with ORs (95% CIs) of 1.223 (1.035–1.445), 1.208 (1.052–1.386), 1.249 (1.097–1.421), and 1.101 (1.001–1.210) in fetal, early, middle, and late childhood exposure groups, respectively (all P<0.05). Additionally, the late childhood exposure group showed a significantly higher risk of fracture, compared to the age-matched cohort, with an OR of 1.155 (95% CI: 1.033–1.291) (P=0.01127), after adjusting for pertinent confounding variables.
Stratified analysis
As depicted in Figure 3 and Table 3, analyses were stratified by sex, BMI, educational level, and geographical location. Sex-stratified analysis revealed a significant association between malnutrition exposure during mid-childhood and an increased risk of developing osteoporosis in adulthood for females (OR=1.188, 95% CI: 1.025–1.377; P=0.02184), after adjusting for potential confounders; however, no significant association was observed in males. Moreover, there was no correlation between malnutrition exposure and bone fracture in both subgroups.
Figure 3.
Associations between malnutrition exposure and OP, Fracture by sex, BMI, education levels, residence region.
Table 3.
Associations between malnutrition exposure and OP, Fracture by sex, BMI, education levels, and residence region.
| OP | Fracture | |||
|---|---|---|---|---|
| Stratification factors | OR (95% CI) | P | OR (95% CI) | P |
| Sex | ||||
| Male | ||||
| Age-balanced group (Ref.) | 1 | / | 1 | / |
| Late childhood exposure group | 0.906 (0.744–1.102) | 0.32255 | 1.209 (0.980–1.493) | 0.07711 |
| Mid-childhood exposure group | 1.061 (0.806–1.398) | 0.67227 | 1.038 (0.760–1.418) | 0.8125 |
| Early childhood exposure group | 1.216 (0.896–1.650) | 0.21038 | 1.106 (0.787–1.554) | 0.56333 |
| Fetal exposure group | 1.333 (0.897–1.982) | 0.15461 | 1.049 (0.663–1.661) | 0.83727 |
| Female | ||||
| Age-balanced group (Ref.) | 1 | / | 1 | / |
| Late childhood exposure group | 1.086 (0.972–.214) | 0.14425 | 1.111 (0.973–1.268) | 0.12008 |
| Mid-childhood exposure group | 1.188 (1.025–1.377) | 0.02184 | 0.962 (0.798–1.158) | 0.67994 |
| Early childhood exposure group | 1.136 (0.974–1.324) | 0.10426 | 1.003 (0.828–1.215) | 0.97485 |
| Fetal exposure group | 1.163 (0.970–1.395) | 0.10239 | 0.935 (0.737–1.185) | 0.57681 |
| BMI | ||||
| Underweight or normal (<24) | ||||
| Age-balanced group (Ref.) | 1 | / | 1 | / |
| Late childhood exposure group | 1.105 (0.969–1.260) | 0.13521 | 1.220 (1.032–1.444) | 0.02023 |
| Mid-childhood exposure group | 1.086 (0.913–1.291) | 0.34994 | 1.049 (0.830–1.326) | 0.68951 |
| Early childhood exposure group | 1.032 (0.858–1.241) | 0.73669 | 1.229 (0.963–1.568) | 0.09704 |
| Fetal exposure group | 1.140 (0.918–1.414) | 0.23504 | 1.116 (0.830–1.501) | 0.4658 |
| Overweight or obese (≥24) | ||||
| Age-balanced group (Ref.) | 1 | / | 1 | / |
| Late Childhood Exposure Group | 1.069 (0.933–1.225) | 0.33485 | 1.100 (0.948–1.278) | 0.20921 |
| Mid-childhood exposure group | 1.510 (1.250–1.824) | 0.00002 | 0.962 (0.774–1.196) | 0.72564 |
| Early childhood exposure group | 1.347 (1.101–1.650) | 0.00388 | 0.926 (0.737–1.165) | 0.51284 |
| Fetal exposure group | 1.316 (1.018–1.703) | 0.03634 | 0.848 (0.627–1.146) | 0.28202 |
| Education level | ||||
| Junior high school and below (Low) | ||||
| Age-balanced group (Ref.) | 1 | / | 1 | / |
| Late childhood exposure group | 1.173 (1.041–1.321) | 0.00865 | 1.220 (1.058–1.408) | 0.00639 |
| Mid-childhood exposure group | 1.270 (1.059–1.522) | 0.00988 | 0.852 (0.673–1.079) | 0.18261 |
| Early childhood exposure group | 1.354 (1.085–1.689) | 0.00736 | 1.102 (0.837–1.451) | 0.4888 |
| Fetal exposure group | 1.791 (1.320–2.430) | 0.00018 | 1.193 (0.805–1.767) | 0.37925 |
| Senior high school and above (High) | ||||
| Age-balanced group (Ref.) | 1 | / | 1 | / |
| Late childhood exposure group | 0.996 (0.847–1.171) | 0.96042 | 1.021 (0.849–1.229) | 0.82285 |
| Mid-childhood exposure group | 1.203 (0.997–1.451) | 0.05323 | 1.140 (0.916–1.419) | 0.23886 |
| Early childhood exposure group | 1.107 (0.926–1.324) | 0.26413 | 1.003 (0.812–1.239) | 0.97662 |
| Fetal exposure group | 1.057 (0.863–1.295) | 0.59164 | 0.899 (0.699–1.156) | 0.40443 |
| Residence region | ||||
| North | ||||
| Age-balanced group (Ref.) | 1 | / | 1 | / |
| Late childhood exposure group | 1.188 (1.005–1.404) | 0.04307 | 1.019 (0.856–1.214) | 0.83028 |
| Mid-childhood exposure group | 1.242 (1.000–1.543) | 0.04959 | 1.040 (0.826–1.308) | 0.7393 |
| Early childhood exposure group | 1.047 (0.834–1.315) | 0.68977 | 0.931 (0.731–1.187) | 0.56621 |
| Fetal exposure group | 1.239 (0.938–1.636) | 0.13071 | 0.946 (0.697–1.283) | 0.71889 |
| South | ||||
| Age-balanced group (Ref.) | 1 | / | 1 | / |
| Late childhood exposure group | 1.085 (0.965–1.219) | 0.17283 | 1.220 (1.053–1.412) | 0.00792 |
| Mid-childhood exposure group | 1.268 (1.077–1.494) | 0.00445 | 0.909 (0.725–1.139) | 0.40518 |
| Early childhood exposure group | 1.333 (1.117–1.589) | 0.00139 | 1.179 (0.938–1.482) | 0.15885 |
| Fetal exposure group | 1.220 (0.989–1.506) | 0.06361 | 0.931 (0.694–1.249) | 0.63229 |
OP, osteoporosis, Ref., reference, BMD, bone mineral density. All the model was adjusted for sex, age, BMI, education levels, marital status, income levels, smoking behavior, alcohol consumption, residence region, hyperlipidemia, hypertension, diabetes, physical activity, calcium supplement use, and antiosteoporosis drug use. Statistical significance was set at P<0.05.
In overweight or obese individuals (BMI ≥24) adjusted for potential confounders, malnutrition exposure during fetal development (OR=1.316, 95% CI: 1.018–1.703; P=0.03634), early childhood (OR=1.347, 95% CI: 1.101–1.650; P=0.00388), or mid-childhood (OR=1.510, 95% CI: 1.250–1.824; P=0.00002) significantly increased the risk of developing osteoporosis in adulthood, compared to the age-matched group. Conversely, no significant associations were found in individuals with a BMI below 24. However, in the lower BMI cohort, exposure to malnutrition during late childhood was associated with an increased risk of fracture (OR=1.220, 95% CI: 1.032–1.444; P=0.02023).
Among those with lower educational levels (junior high school and below) adjusted for confounders, exposure to malnutrition during fetal life (OR=1.791, 95% CI: 1.320–2.430; P=0.00018), early childhood (OR=1.354, 95% CI: 1.085–1.689; P=0.00736), mid-childhood (OR=1.270, 95% CI: 1.059–1.522; P=0.00988), and late childhood (OR=1.173, 95% CI: 1.041–1.321; P=0.00865) significantly increased the risk of developing osteoporosis in adulthood. No such correlations were found in individuals with higher educational levels (senior high school and above).
In the analysis stratified according to geographical regions, exposure to malnutrition during early (OR=1.333, 95% CI: 1.117–1.589; P=0.00139) and mid-childhood (OR=1.268, 95% CI: 1.077–1.494; P=0.00445) significantly increased the risk of osteoporosis in adulthood in southern China. In northern China, exposure during middle (OR=1.242, 95% CI: 1.000–1.543; P=0.04959) and late childhood (OR = 1.188, 95% CI: 1.005–1.404; P=0.04307) similarly increased the risk of osteoporosis in adulthood. Furthermore, in southern China, exposure to malnutrition in late childhood also significantly increased the risk of fracture (OR=1.220, 95%CI: 1.053–1.412; P=0.00792), compared with the age-matched group.
Validation
A total of 7651 participants from the CNC were included for validation, as depicted in Figure S1 (Supplemental Digital Content 2, http://links.lww.com/JS9/D374). Baseline data are detailed in Table S1 (Supplemental Digital Content 2, http://links.lww.com/JS9/D374). As illustrated in Figure S2 (Supplemental Digital Content 2, http://links.lww.com/JS9/D374), the prevalence of osteoporosis among the groups was as follows: 14.411% (n=475) in the nonexposed group, 21.098% (n=169) in the fetal exposure group, 25.347% (n=146) in the early childhood exposure group, 25.748% (n=172) in the mid-childhood exposure group, 28.581% (n=411) in the late childhood exposure group, and 35.550% (n=310) in the adolescent exposure group. A statistically increasing trend was observed (P<0.001), demonstrating a higher risk of osteoporosis associated with an earlier exposure to malnutrition. Linear regression analysis revealed that a late childhood exposure to malnutrition was significantly associated with a reduced BMD, as evidenced by T-scores (β = −0.23; 95% CI: −0.36 to −0.10; P<0.001), when compared to the age-matched control group. Detailed data can be found in Table S2 (Supplemental Digital Content 2, http://links.lww.com/JS9/D374).
Discussion
In this cross-sectional cohort study, after adjusting for confounding factors, early-life exposure to malnutrition was associated with reduced BMD and an increased risk of osteoporosis in adulthood, compared to an age-matched control group. This association was particularly pronounced among individuals with a lower educational level (junior high school and below) or with a higher BMI (≥24). Our findings not only lend support to the Developmental Origins of Health and Disease (DOHaD) theory22, which suggests that adverse early-life conditions, such as poor nutrition or environmental factors, increase the risk of chronic diseases later in life, but also provide new insights into the etiology of osteoporosis. Stratifying osteoporosis risk based on one’s history of malnutrition exposure could enable more effective targeted prevention and treatment strategies.
The precise mechanisms through which early-life exposure to malnutrition increases the risk of osteoporosis in adulthood remain to be elucidated. Various pathways may be involved, starting with the critical need for diverse nutrients during the phases of bone growth and formation23. Protracted protein-energy malnutrition or sustained nutritional deficiencies can lead to linear growth retardation24. Inadequate dietary intake or impaired absorption of bone-building minerals, particularly calcium and zinc, may delay in linear growth25. During adolescence, bone accrual accelerates and bone mass peaks. This peak is a critical determinant of future bone health and susceptibility to fracture25. A peak bone mass 10% higher than the average has the potential to delay the onset of osteoporosis by up to 13 years26. Peak bone mass and subsequent fracture risk in later life are closely associated with nutritional exposures during fetal period, infancy, childhood, and adolescence23. Here, we suppose that participants who endured a food shortage in 1959–1961 were likely to experience early-life malnutrition, which may have led to reduced peak bone mass and an increased risk of osteoporosis in adulthood.
Additionally, research has shown that malnutrition during childhood significantly increases the likelihood of developing type 2 diabetes in adulthood27, a known risk factor for osteoporosis28. Moreover, early-life malnutrition may predispose adults to an elevated risk of developing sarcopenia29, which is an independent risk factor for osteoporosis30. Furthermore, independent of diabetes, early-life malnutrition exposure may increase the likelihood of insulin resistance and β-cell dysfunction in adulthood31. Insulin resistance enhances inflammation-driven osteoclastic activity to promote osteolysis in individuals with metabolic syndrome, ultimately increasing the risk of developing osteoporosis32. β-cell dysfunction is a key pathophysiological characteristic of type 2 diabetes mellitus33. Furthermore, a significant reduction in caloric intake over a short period decreases serum levels of insulin-like growth factor (IGF)-1 in rodents34. IGF-1 plays a direct role in the development and maintenance of skeletal muscles and bones35.
We also observed that education level and BMI mediated the relationship between early-life malnutrition exposure and the risk of osteoporosis. Specifically, exposure to malnutrition during fetal early childhood and middle childhood significantly increased the risk of osteoporosis among participants with a high BMI (≥24), compared to age-matched controls. In contrast, participants with a BMI below 24 showed no significant difference in osteoporosis risk, relative to their age-matched peers, suggesting that maintaining a normal body weight may alleviate the detrimental effects of early-life malnutrition exposure on bone health. Moreover, our findings indicated that individuals with lower educational levels faced a significantly higher risk of developing osteoporosis, compared to their age-matched counterparts. Conversely, those with a higher educational level did not exhibit an elevated risk, likely due to their richer health knowledge and healthier lifestyle and dietary habits. This highlights the significance of targeted community education programs in preventing osteoporosis.
Previous research has suggested that early-life malnutrition exposure adversely affects women’s bone health, though similar impacts on men have not been observed20,36. However, after combining the nonexposure and adolescent-exposure groups to create an age-matched sample for large-sample analyses, we found that early-life malnutrition exposure did not significantly affect bone health across sex groups. The discrepancy in results may be attributed to differences in sample size or geographical variation. Prior studies have primarily focused on subjects in one region, such as Lanzhou (a city in Gansu Province) and Henan (a province in northern China). In contrast, our study included seven provinces and cities across China, enhancing the representativeness of our findings.
This study possesses several notable strengths. Firstly, the use of DXA, a gold standard for bone densitometry, ensures precise measurement of BMD. Notably, measurements from different models of DXA bone densitometers cannot be directly merged but require conversion, a detail often overlooked in previous studies. In our study, all DXA equipment used was of the same model and underwent annual inspections to ensure accuracy and consistency. Additionally, the incorporation of a large-sample size bolsters the statistical power and generalizability of the findings. Noteworthy is the construction of age-matched groups through the amalgamation of both nonexposed and adolescent-exposed cohorts, effectively minimizing potential age-related biases in the results. Furthermore, our research indicates that interventions targeting health education and body weight management may alleviate the detrimental effects of early-life malnutrition on bone health. The validation of our findings through the CNC further reinforces our findings.
However, certain limitations warrant consideration. The food-shortage period from 1959 to 1961 has no precise initiation and terminal dates, potentially leading to errors in classifying individuals with malnutrition exposure. Moreover, an age-similar cohort unaffected by food-shortages may still contain age-related biases.
Conclusion
Early-life exposure to malnutrition is linked with an increased risk of osteoporosis in adulthood. The adverse effects of early-life malnutrition exposure can be mitigated by maintaining a healthy body weight and enhancing health education. In light of the findings of this study, it is recommended that public health education be strengthened in order to enhance awareness about osteoporosis and its risk factors, with particular emphasis placed on the significance of early-life nutrition. It is recommended that governments and relevant agencies promote appropriate nutritional intervention strategies to ensure adequate nutrition intake during infancy, thereby reducing the risk of osteoporosis in adulthood. Additionally, an early screening and monitoring system for high-risk populations should be established to facilitate the timely identification and intervention in potential osteoporosis patients.
Ethical approval
The study protocol received approval from the Institutional Review Board of Longhua Hospital Affiliated with Shanghai University of Traditional Chinese Medicine (No. 2016LCSY065).
Consent
Yes. The patients/participants provided their written informed consent to participate in this study. This has been documented in the paper. This study is not a case report study and does not involve patient cases or related images. This study does not involve the names, initials, or hospital numbers of patients or volunteers. This study does not involve any identifying characteristics of the participants.
Source of funding
This work was partially supported by the Inheritance and Innovation Team Project of National Traditional Chinese Medicine (ZYYCXTD-C-202202), National Natural Science Foundation of China (82274548, 81973883, 82360934), Project from Shanghai Collaborative Innovation Center of Industrial Transformation of Hospital TCM Preparation, the Program for Innovative Research Team of Ministry of Education of China (IRT1270), the Program for Innovative Research Team of Ministry of Science and Technology of China (2015RA4002), the Key Program of Natural Science Foundation of Xinjiang Uygur Autonomous Region (No. 2021B03006), the Technical Innovation Leading Talent Program of Xinjiang Uygur Autonomous Region (No. 2022TSYCLJ0007), the Key grant Program of Xinjiang Uygur Autonomous Region (No. 2023A03007-3).
Author contribution
Y.W., D.T., and R.F.: conceptualization; H.X., H.Z., R.A., C.Y., F.C., H.W., J.J., J.L., J.C., J.W., B.S., H.X., Q.L., Q. Shi, Q. Sun, R.F., D.T., and Y.W.: data curation; H.X., H.Z., and R.A.: formal analysis; Y.W. and R.F.: funding acquisition; H.X., H.Z., R.A., and C.Y.: investigation; H.X., H.Z., R.A., and C.Y.: methodology; Y.W., Q. Sun, and R.F.: project administration; Y.W., D.T., Q. Sun, J.C., and R.F.: resources; H.X., H.Z., R.A., and C.Y.: software; Y.W., D.T., Q. Sun, and R.F.: supervision; H.X. and R.A.: validation; H.X. and H.Z.: visualization; H.X., H.Z., R.A.: writing – original draft; Y.W., D.T., H.Z., and R.F.: writing – review and editing.
Conflicts of interest disclosures
The authors have no relevant financial or nonfinancial interests to disclose.
Research registration unique identifying number (UIN)
Guarantor
Prof. Yongjun Wang, Longhua Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai 200032, People’s Republic of China. E-mail: wangyongjun@shutcm.edu.cn; Prof. Dezhi Tang, Longhua Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai 200032, People’s Republic of China. E-mail: dztang@shutcm.edu.cn; Prof. Rui Fang, Affiliated Hospital of Traditional Chinese Medicine, Xinjiang Medical University, Urumqi 830054, People’s Republic of China. E-mail: xjfr@163.com; Prof. Qi Sun, Affiliated Hospital of Traditional Chinese Medicine, Xinjiang Medical University, Urumqi 830054, People’s Republic of China. E-mail: cums6140@hotmail.com.
Data availability statement
The original contributions presented in this study are included in the article/Supplementary material, further inquiries can be directed to the corresponding authors.
Provenance and peer review
Not commissioned, externally peer-reviewed.
Supplementary Material
Acknowledgements
The authors extend our gratitude to the members of the China Community-based Cohort of Osteoporosis (CCCO) collaborative group. Our deepest appreciation goes to the study participants and the survey teams at each center. The authors also acknowledge the efforts of the project development and management teams located in Shanghai, Beijing, and other participating centers.
Footnotes
Hongbin Xu, Haitao Zhang, Remila Aimaiti, Chunchun Yuan, and Feihong Cai contributed equally to this article.
Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.
Supplemental Digital Content is available for this article. Direct URL citations are provided in the HTML and PDF versions of this article on the journal's website, www.lww.com/international-journal-of-surgery.
Contributor Information
Hongbin Xu, Email: xuhongbin98@126.com.
Haitao Zhang, Email: zht1997zht@163.com.
Remila Aimaiti, Email: remila2018@163.com.
Chunchun Yuan, Email: ccyuan0831@shutcm.edu.cn.
Feihong Cai, Email: caifeihong2017@126.com.
Hongyu Wang, Email: 2934786842@qq.com.
Jiangxun Ji, Email: jijiangxun@163.com.
Junhao Liang, Email: 1357056629@qq.com.
Jiarui Cui, Email: jrfffairybabi@163.com.
Jing Wang, Email: 13671683145@qq.com.
Bing Shu, Email: siren17721101@163.com.
Hao Xu, Email: hoxu@163.com.
Qianqian Liang, Email: liangqianqian@shutcm.edu.cn.
Qi Shi, Email: shiqish@hotmail.com.
Qi Sun, Email: cums6140@hotmail.com.
Rui Fang, Email: xjfr@163.com.
Dezhi Tang, Email: dztang@shutcm.edu.cn.
Yongjun Wang, Email: wangyongjun@shutcm.edu.cn.
References
- 1.Foessl I, Dimai HP, Obermayer-Pietsch B. Long-term and sequential treatment for osteoporosis. Nat Rev Endocrinol 2023;19:520–533. [DOI] [PubMed] [Google Scholar]
- 2.Marini F, Cianferotti L, Brandi M. epigenetic mechanisms in bone biology and osteoporosis: can they drive therapeutic choices? Int J Mol Sci 2016;17:1329. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Pouresmaeili F, Kamalidehghan B, Kamarehei M, et al. A comprehensive overview on osteoporosis and its risk factors. Ther Clin Risk Manag 2018;14:2029–2049. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Zhou Q, Zhu L, Zhang D, et al. Oxidative stress-related biomarkers in postmenopausal osteoporosis: a systematic review and meta-analyses. Dis Markers 2016;2016:7067984. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Wright NC, Looker AC, Saag KG, et al. The recent prevalence of osteoporosis and low bone mass in the United States based on bone mineral density at the femoral neck or lumbar spine. J Bone Miner Res 2014;29:2520–2526. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Compston JE, McClung MR, Leslie WD. Osteoporosis. Lancet 2019;393:364–376. [DOI] [PubMed] [Google Scholar]
- 7.Roubenoff R, Kehayias JJ. The meaning and measurement of lean body mass. Nutr Rev 1991;49:163–175. [DOI] [PubMed] [Google Scholar]
- 8.Zeng Q, Li N, Wang Q, et al. The prevalence of osteoporosis in China, a nationwide, multicenter DXA survey. J Bone Miner Res 2019;34:1789–1797. [DOI] [PubMed] [Google Scholar]
- 9.Shen Y, Huang X, Wu J, et al. The global burden of osteoporosis, low bone mass, and its related fracture in 204 countries and territories, 1990-2019. Front Endocrinol (Lausanne) 2022;13:882241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hendrickx G, Boudin E, Van Hul W. A look behind the scenes: the risk and pathogenesis of primary osteoporosis. Nat Rev Rheumatol 2015;11:462–474. [DOI] [PubMed] [Google Scholar]
- 11.Yuan C, Yu XT, Wang J, et al. Multi-modal molecular determinants of clinically relevant osteoporosis subtypes. Cell Discov 2024;10:28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yuan C, Zhang W, Wang J, et al. Chinese medicine phenomics (Chinmedphenomics): personalized, precise and promising. Phenomics 2022;2:383–388. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kin CF, Shan WS, Shun LJ, et al. Experience of famine and bone health in post-menopausal women. Int J Epidemiol 2007;36:1143–1150. [DOI] [PubMed] [Google Scholar]
- 14.Chen J, Peng B, Que P, et al. The impact of early life exposure to famine on osteoporosis in adulthood. Chin J Osteoporos 2016;22:492–496. [Google Scholar]
- 15.Mathew G, Agha R, Albrecht J, et al. STROCSS 2021: strengthening the reporting of cohort, cross-sectional and case-control studies in surgery. Int J Surg 2021;96:106165. [DOI] [PubMed] [Google Scholar]
- 16.Wang J, Shu B, Tang DZ, et al. The prevalence of osteoporosis in China, a community based cohort study of osteoporosis. Front Public Health 2023;11:1084005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Rong H, Lai X, Mahmoudi E, et al. Early-life exposure to the Chinese famine and risk of cognitive decline. J Clin Med 2019;8:484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Liu L, Pang ZC, Sun JP, et al. Exposure to famine in early life and the risk of obesity in adulthood in Qingdao: Evidence from the 1959-1961 Chinese famine. Nutr Metab Cardiovasc Dis 2017;27:154–160. [DOI] [PubMed] [Google Scholar]
- 19.Huang LN, Wang HJ, Wang ZH, et al. Association between Chinese famine exposure and the risk of overweight/obesity and abdominal obesity in laterlife: a cross-sectional study. Biomed Environ Sci 2020;33:133–137. [DOI] [PubMed] [Google Scholar]
- 20.Yang M, Yin H, Zhen D, et al. Exposure to famine in every stage of life and the risk of osteoporosis and fractures later in life: a cross-sectional study. Bone 2023;168:116644. [DOI] [PubMed] [Google Scholar]
- 21.Lewiecki EM, Kendler DL, Kiebzak GM, et al. Special report on the official positions of the International Society for Clinical Densitometry. Osteoporos Int 2004;15:779–784. [DOI] [PubMed] [Google Scholar]
- 22.Hoffman DJ, Powell TL, Barrett ES, et al. Developmental origins of metabolic diseases. Physiol Rev 2021;101:739–795. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Prentice A, Schoenmakers I, Laskey MA, et al. Nutrition and bone growth and development. Proc Nutr Soc 2006;65:348–360. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.de Onis M, Blossner M. The World Health Organization Global database on child growth and malnutrition: methodology and applications. Int J Epidemiol 2003;32:518–526. [DOI] [PubMed] [Google Scholar]
- 25.Prentice A, Bates CJ. Adequacy of dietary mineral supply for human bone growth and mineralisation. Eur J Clin Nutr 1994;48 Suppl 1:S161–S176. [PubMed] [Google Scholar]
- 26.Hernandez CJ, Beaupre GS, Carter DR. A theoretical analysis of the relative influences of peak BMD, age-related bone loss and menopause on the development of osteoporosis. Osteoporos Int 2003;14:843–847. [DOI] [PubMed] [Google Scholar]
- 27.van Abeelen AF, Elias SG, Bossuyt PM, et al. Famine exposure in the young and the risk of type 2 diabetes in adulthood. Diabetes 2012;61:2255–2260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.An Y, Zhang H, Wang C, et al. Activation of ROS/MAPKs/NF-kappaB/NLRP3 and inhibition of efferocytosis in osteoclast-mediated diabetic osteoporosis. FASEB J 2019;33:12515–12527. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Wu T, Yan X, Liu Y, et al. Association between early life exposure to the great famine and possible sarcopenia in older Chinese adults: a national cross-sectional study. BMJ Open 2023;13:e065240. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Wu CH, Yang KC, Chang HH, et al. Sarcopenia is related to increased risk for low bone mineral density. J Clin Densitom 2013;16:98–103. [DOI] [PubMed] [Google Scholar]
- 31.Wang Y, Wan H, Chen C, et al. Association between famine exposure in early life with insulin resistance and beta cell dysfunction in adulthood. Nutr Diabetes 2020;10:18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Rhee EJ, Kim YC, Lee WY, et al. Comparison of insulin resistance and serum high-sensitivity C-reactive protein levels according to the fasting blood glucose subgroups divided by the newly recommended criteria for fasting hyperglycemia in 10059 healthy Koreans. Metabolism 2006;55:183–187. [DOI] [PubMed] [Google Scholar]
- 33.Pasquier A, Vivot K, Erbs E, et al. Lysosomal degradation of newly formed insulin granules contributes to beta cell failure in diabetes. Nat Commun 2019;10:3312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Elias SG, Keinan-Boker L, Peeters PH, et al. Long term consequences of the 1944-1945 Dutch famine on the insulin-like growth factor axis. Int J Cancer 2004;108:628–630. [DOI] [PubMed] [Google Scholar]
- 35.Bikle DD, Tahimic C, Chang W, et al. Role of IGF-I signaling in muscle bone interactions. Bone 2015;80:79–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Qi W, Cui L, Yin X, et al. Association of early-life famine exposure with low bone mass in adulthood. Arch Osteoporos 2023;18:32. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The original contributions presented in this study are included in the article/Supplementary material, further inquiries can be directed to the corresponding authors.