Is bone health of young atopic dermatitis patients a signicant healthcare problem?

Background Atopic dermatitis (AD) incidence has increased over the past decades, especially among young adults. However, their bone health has not been clearly elucidated with discordant results. Objective We aimed to investigate the bone mineral density (BMD) measured by double energy X-ray absorptiometry and Z-score at the lumbar spine, femur neck, and total femur to evaluate the risk of osteoporosis in men aged <50 years and premenopausal women with AD. Methods The Korea National Health and Nutrition Examination Survey 2007-2009 data were used in this case-control cohort study. We included young AD patients (aged 19 ≤ and <50 years) diagnosed by a doctor and compared these patients with 1:5 propensity score weighting controls by age, sex, body mass index (BMI), vitamin D level, and alcohol/smoking status. Results We analyzed 311 (weighted n=817,014) AD patients and 1,555 (weighted n=4,155,855) controls. BMD at the lumbar spine was signicantly lower in the male AD group than in the male control group (mean ± SE, 0.989 ± 0.002 vs. 0.954 ± 0.016, P = 0.03) while BMDs at the femur neck and total femur were comparable. For women, BMDs at the three sites were not signicantly different between the AD and control groups. Low bone mass (dened by a Z-score of ≤ 2.0) was not signicantly different between the AD and control groups in both sexes. Conclusions Bone health, especially BMD and Z-score, in young AD patients were comparable with those without AD. AD was not a risk factor for low BMD.


Introduction
Atopic dermatitis (AD) is one of the most common in ammatory skin diseases affecting 15%-30% of children and 2%-10% of adults. 1 The AD prevalence has been increasing worldwide in recent decades. As AD is a chronic in ammatory disease that usually begins in early childhood or adolescence, various comorbidities have been related to AD. Bone health problems in AD patients have increasingly received attention because AD is associated with risk factors for osteopenia and osteoporosis, including topical/systemic corticosteroid usage, less physical activity, malnutrition from food restriction, and AD with chronic in ammatory disease course. 2 Haeck et al. reported that low bone mineral density (BMD), de ned as a Z-score of ≤−1, was prevalent in adults with moderate-to-severe AD. 5 Another study from the UK showed that patients with atopic eczema are associated with a 10% increased fracture risk compared with those without this condition. 6 However, existing studies have several limitations. First, most results have included the elderly population, the major risk factor for osteoporosis and fracture, which cannot represent the bone health of the majority of young patients with AD. Second, multiple factors that affect bone health, such as sex, vitamin D level, alcohol consumption, smoking, and menopausal status, have not been fully applied in previous studies. 1 Thus, we aimed to compare the BMD and Z-score between young AD patients and a general population with propensity score weighting for age, sex, body mass index (BMI), vitamin D level, alcohol/smoking status, and menopausal status using the data from the Korean National Health and Nutrition Examination Survey (KNHANES) 2007-2009.

Study population and design
The KNHANES is a cross-sectional nationwide survey conducted by the Korea Centers for Disease Control and Prevention (KCDC). 7 It has been conducted periodically since 1998, and data were gathered through household interviews, standardized physical examinations, laboratory tests, and nutritional status assessments. The KNHANES IV, which only performed the BMD test, was conducted between 2007 and 2009, including more than 260,000 primary sampling units each year. After applying the strati ed, multistage clustered systemic sampling method, 19,841 of the 25,250 systemically selected individuals participated in this survey. We excluded (1) individuals aged ≥50 and <19 years; (2) those with incomplete information on the atopic disease, BMD results, BMI, and 25(OH)D level; or (3) postmenopausal women for any reason. We de ned AD based on the questionnaire about the previous history of AD diagnosed by a doctor. All survey protocols were approved by the Institutional Review Board of the KCDC and we con rmed that all methods were performed in accordance with the relevant guidelines and regulation. All participants signed an informed consent form before the start of the survey.
Bone densitometry and de nition BMD at the lumbar spine, femoral neck, and total hip was measured using dual-energy X-ray absorptiometry (DXA, QDR4500A System; Hologic Inc., Waltham, MA, USA). BMD was expressed as the total bone mineral content (g) divided by the area (cm 2 ). We calculated the Z-score of the lumbar spine, femoral neck, and total hip using the age-and sex-speci c mean BMD (g/cm 2 ): (patient BMD-mean BMD of reference group)/(standard deviation [SD] of reference group). The SDs for the complex survey design were calculated using the Taylor series linearization method. 11 The weighted mean and SD of the reference BMD for age and sex used to calculate the Z-score are described in Supplementary Table 1. According to the World Health Organization diagnostic classi cation, we de ned low bone mass based on the expected range for their age group as a Z-score of ≤−2.0 and normal as a Z-score of >−2.0. 12

Statistical analyses
To obtain the appropriate nationally representative estimates, all analyses in this study were performed considering the complex survey design applied to the KNHANES. The propensity score weighting method was used to balance the variables between patients with and without AD. Individual propensity scores were calculated using multiple logistic regression models, including the presence of atopy as the dependent variable and age, sex, BMI, vitamin D level, alcohol consumption, smoking, and complex survey sampling weight as covariates. The nal weight was obtained by multiplying the sampling weight by the propensity score weight. 13 Continuous data were presented as the weighted mean and standard error of the mean for normally distributed variables and as medians and interquartile ranges (IQRs) for non-normally distributed variables. Categorical data were expressed as weighted frequencies and percentages. To compare the characteristics between patients with AD and those without AD, we performed a weighted t-test for continuous variables and the Rao-Scott chi-square test for categorical variables. Finally, multiple logistic regression models considering the complex survey design were used to investigate the risk factors for low BMD group according to sex. In this model, the low BMD group was included as binary outcome variable and age, atopic dermatitis, body mass index, vitamin D de ciency, education, drinking and age at menarche (female only) were included as dependent variables. The results for logistic regression analysis were presented the odds ratios (ORs) and corresponding 95% con dence intervals (CIs). All reported P-values were two-sided, and statistical signi cance was set at P < 0.05. All statistical analyses were performed using SAS software (version 9.4; SAS Institute Inc., Gary, NC, USA).

Participants' clinical characteristics
This study included 311 (weighted n=817,014) patients with AD and 8,972 (weighted n=20,880,643) without AD. The median ages at AD diagnosis were 18.7 years (IQR: 8.6-31.3 years) and 17.0 years (IQR 8.4-26.2 years) in male and female AD patients, with median durations of AD 9.8 years (IQR: 4.7-19.3 years) and 10.8 years (IQR: 5.5-19.1 years), respectively. Meanwhile, 29.7% of male and 20.5% of female AD patients received atopy treatment at the time of the survey. Among female patients, the median age of menarche was not signi cantly different between AD patients and the control group. Higher family income was more frequently reported in the AD group than in the control group among the female patients. Among female patients, no signi cant difference was observed in the history of oral contraceptive use between AD patients and the control group. The experience of pregnancy and giving birth (55.4 vs. 65.0%; P = 0.040) were signi cantly lower in the AD group than in the control group. Asthma was signi cantly higher in AD patients in male (10.4 vs. 2.0%; P = 0.003) and female groups (5.5 vs. 2.0%; P = 0.022). Depression mood was not signi cantly different in AD patients compared with the control group in both sexes. Sickness at 1 month was signi cantly higher in AD patients than in the control group (16.6 vs. 8.6%; P = 0.003); absence at 1 month was not signi cantly different among the female group (Table 1).
Bone mineral density and Z-score between the AD and control groups BMD at the lumbar spine was lower in male AD patients than in the control group (mean ± SE, 0.954 ± 0.016 vs. 0.989 ± 0.002; P = 0.03). BMD at the total femur was lower in AD patients than in the control group (0.977 ± 0.012 vs. 0.999 ± 0.002; P = 0.08), with no signi cant difference. For female AD patients, BMDs at the lumbar, femur neck, and total femur were not signi cantly different between the AD and control groups ( Table 2). The rates of low bone mass, de ned by the lowest Z-score ≤−2.0 in the three measured sites (lumbar spine, total femur, and femoral neck), were not signi cantly different in the AD group compared with the control group in both male (3.8% vs. 2.7%; P = 0.56) and female patients (6.4% vs. 3.3%; P = 0.40) ( Table 3).

Risk factors for low BMD in all participants strati ed by sex
We further calculated the crude-and adjusted odds ratios (aORs) to determine the risk factors for low BMD among all participants strati ed by sex. Older age was associated with increased odds of low BMD with a greater effect of aging on bone loss in women than in men. A higher BMI was associated with decreased odds of lower BMI in both sexes. In female, having vitamin D de ciency was associated with increased odds of low while in male, vitamin D de ciency was not associated with low BMD. Atopic dermatitis itself was not associated with low BMD in both male and female (Table 4).

Relationship between clinical characteristics and low BMD among AD patients
We compared the clinical characteristics of the normal and low BMD group among AD patients strati ed by sex. Among male AD patients, the low BMD group was diagnosed with AD at younger age, lower BMI, and frequently resided in the rural areas compared with the normal BMD group (Suppl. Table 2).
In female AD patients, the low BMD group were diagnosed with AD at earlier age and longer AD duration. The low BMD group had lower BMI, less educated, lower vitamin D level, later age at menarche, and higher number of pregnancies compared with the normal BMD group (Suppl. Table 3).

Discussion
This retrospective weighted case-control study demonstrated that BMDs and Z-scores were comparable between the AD and control groups, except that BMD at the lumbar spine was slightly lower in the AD male patients than in the control group. The actual prevalence of low bone mass (Z-score≤−2.0) in the AD group and that in the non-AD group were not signi cant in either sex (male: 3.8% vs. 2.7%, P=0.56; female: 6.4% vs. 3.3%, P=0.40). AD was not a risk factor for low BMD, while older age and lower BMI were signi cantly associated with low BMD in both male and female among all participants. Vitamin D de ciency (de ned as 25-hydroxy vitamin D <20 ng/dL), a well-known risk factor for osteoporosis, 14 was signi cantly associated with low BMD only in female. A few studies have evaluated the prevalence of low BMD and osteoporosis in young patients with AD, particularly for the accurate measurement of BMD using DXA. Our study, using a large sample, provides evidence that BMD and Z-score may not be a signi cant health burden in young AD patients.
In baseline characteristics after propensity score weighting, a lower number of pregnancies and births were reported among women with AD, although oral contraceptive use was comparable. Higher frequency of single or never-married status of AD patients, 16 psychological distress, including depression, anxiety, and lower self-esteem of AD people, 17,18 may affect their partner relationships 19 , which explains the low pregnancy and birth rates among AD patients. AD patients had a higher incidence of asthma consistent with the reports of previous studies, 16,20 while no difference was found in the incidence of depression between the AD and non-AD groups, which was frequently reported previously. 17 Considering that we matched vitamin D levels, which have been signi cantly low in AD patients 21 and negatively correlated with the severity of AD, 22 vitamin D could primarily be attributed to comorbid depression in AD patients.
There is adequate evidence showing a positive relationship between vitamin D de ciency and depression, 23 suggesting its role in the synthesis of dopamine, noradrenaline, and adrenaline that are implicated in mood disorders. 24 Women with AD were more likely to fall ill within 1 month than non-AD women, but no signi cant difference was found within 1 month of absence. According to a meta-analysis on the in uence of AD on work-life, AD affects sick leave, possibly affecting job choice, and leading to change or loss of job. 25 We found a signi cant decrease in BMD in the lumbar spine of male AD patients. It could be considered as an early phenomenon of BMD reduction in AD patients because the trabecular bone of the lumbar spine is more responsive to metabolic changes than the cortical bone of the femoral neck and distal radius. 26,27 However, BMD change could be interpreted as re ecting a "real" change only if the random error of the DXA system is exceeded. The random error or least signi cant change has been described to be between 3.3% and 4.7 % for the total hip. 28, 29 The decrease in BMD in AD patients was 3.5% in our study compared with that in non-AD controls, which is marginal; it appears to be not a relevant decrease clinically.
Previous studies on bone health, including low bone mass, osteopenia, osteoporosis, and fracture or injury risk, have shown con icting results. A study from Finland including 28 adult AD patients suggested that the BMDs of patients with AD did not differ from those of healthy controls 2 . Also, in children with moderate to severe AD, low BMD (de ned as a Z score <−2) did not occur more frequently. 30 According to the previous study, low BMD was not associated with either the use of topical corticosteroids or systemic treatments (oral corticosteroids and/or cyclosporine) in children with moderate-to-severe AD. 30 By contrast, Haeck et al. reported that low BMD, de ned as a Z-score of ≤−1, was found in one-third of adults with moderate-to-severe AD, predominantly male and relatively young individuals, independent of the cumulative dose of topical corticosteroids. 5 In another study of children affected by severe AD, using topical corticosteroids plus oral cyclosporine group had a lower lumbar spine BMD compared with those who only used topical corticosteroids, indicating that low bone mass was primarily mediated by cyclosporine rather than by topical corticosteroids. 27 In addition, previous population-based studies from Taiwan and the United States have shown a signi cant association between adult AD and osteoporosis 4,31 and epidemiologic studies from the British and United States demonstrated increased fracture risk in adult AD patients. 32,33 These inconsistent conclusions are due to the differences in population, sample size, a cut-off value of Z-score, the de nition of osteoporosis, history of AD treatments, and follow-up duration requiring more large-scale and long-term longitudinal studies.
Of note, we revealed that AD itself was not an independent risk factor for low bone mass. Instead, it seemed AD patients have risk factors for osteoporosis and fracture, such as low vitamin D level, lower activity, and systemic corticosteroid or immunosuppressant use, which may have a greater effect on bone health. Vitamin D plays a central role in calcium metabolism and skeletal structure. At the same time, vitamin D has been implicated in the pathogenesis of AD by regulating the synthesis of structural proteins of the epidermal barrier and regulating immune response. 34 Choi et al. demonstrated that the prevalence rates of vitamin D insu ciency was 47.3% in men and 64.5% in women in South Korea using the KNHANES IV. More strikingly, vitamin D insu ciency was a great threat to the younger generation, and those in their twenties had the highest prevalence of vitamin D insu ciency in both men and women. Therefore, we performed propensity score weighting not only for age and sex, but also for vitamin D level, alcohol consumption, and smoking habit, which affect osteoporosis. Our study design had a signi cant strength as it evaluates the impact of AD on BMD and osteoporosis.
There are several potential reasons for the lack of signi cant association between AD and low BMD, and AD itself was not an independent risk factor in our study. Type 2 in ammation associated with AD might actually protect against osteoporosis because Th2 cytokines are recognized to have an osteoprotective effect, enhancing the anabolic effects of parathyroid hormone, and decreasing the receptor activator of nuclear factor-kB ligand/osteoprotegerin ratio, leading to inhibition of bone resorption. 35 Oral corticosteroids are a well-recognized risk factor for osteoporosis; however, in a population study by Lowe et al 6 ., no signi cant association was found between oral steroid use and fracture risk, possibly because of the small number of participants who had been receiving long-term steroids as treatment for AD. There is little or no evidence to suggest that topical steroids, even potent topical steroids, are associated with osteoporosis. 5,27,28,30 In the further analyses comparing the low BMD and normal BMD groups among AD patients strati ed by sex, those who live in rural areas (for male AD) and less educated (for female AD) were prone to have low bone mass. In addition, low vitamin D level and higher vitamin D de ciency frequency was presented in female AD patients while inverse association was found in male AD patients, suggesting that the vitamin D de ciency had greater effects especially in women AD patients. The late age of menarche and a higher number of pregnancies/giving births were associated with low bone mass among women with AD. Some studies have suggested that possibly due to the shorter exposure to circulating estrogen, a vital hormonal factor for bone formation, late age at menarche is associated with an increased risk of osteoporosis and osteoporotic fractures. 33,[37][38][39] It also has been shown that during pregnancy, bone absorption is increased. 40 Importantly, we found that early onset and longer duration of AD were associated with low bone mass. Taken together, physicians would be able to identify those who are at risk of low bone mass among AD patients and can offer tailored counseling according to demographics, age of onset and duration of AD, and obstetric history in women with AD.
Nevertheless, several limitations of this study need to be mentioned. First, the KNHANES is a retrospective cross-sectional study, limiting the interpretation of causality between AD and osteoporosis. In addition, there was no long-term follow-up data on future osteoporosis and fracture risk in this young AD patient. Therefore, a more careful interpretation is required to obtain a de nite conclusion. Another major limitation of this study was the lack of information on the AD treatment that may affect BMD in AD patients. Lastly, we did not include biochemical information, such as calcium, phosphorus, parathyroid hormone, and thyroid function tests. This limitation may lead to the potential pitfall of passing over other secondary causes of osteoporosis in the young population.
In conclusion, this study assessed the bone health status of young adult patients with AD and investigated the associated risk factors, compared with a general population with propensity score weighting for age, sex, BMI, vitamin D level, and alcohol/smoking status. The BMDs at all sites were comparable between the AD and control groups. Z-scores below the expected range were not signi cantly different between the two groups. Although the increased risk of osteoporosis and fracture in AD patients is still debatable, our study using a large and meticulous case-control propensity score weighting design provides evidence that supports further long-term and large-scale investigation in AD patients.

Declarations
Funding source: None.
Con ict of interest: The authors declare they have no con icts of interest.
Availability of data and material (data transparency): The dataset generated during the current study is available upon reasonable request from the corresponding author, Kyeong Jin Kim, (jins0707@korea.ac.kr).
Code availability (software application or custom code): The code generated during the current study is available upon reasonable request from the corresponding author, Kyeong Jin Kim, (jins0707@korea.ac.kr).

Tables
Vitamin D de ciency de ned by 25-hydroxy vitamin D < 20ng/dL. * Propensity scores were calculated by two separated multiple logistic regression models according to sex.
The models included the presence of atopy as dependent variable and age, body mass index, vitamin D de ciency, smoking, drinking, and sampling weight as independent variables.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. BMDinAtopySupplementalTableSciRep.docx