Key words
autoimmune polyglandular syndrome - vitamin D deficiency - low bone mineral density - Addison’s disease
Introduction
Autoimmune polyglandular syndromes (APS) are defined by the coexistence of at least
two autoimmune-mediated endocrinopathies [1].
Specific clustering of monoglandular autoimmune diseases depends on genetic and
non-genetic environmental factors. It differs considerably at the time of
presentation, distinguishing between two major subtypes of APS. The juvenile APS-1
shows a monogenetic inheritance. In contrast, multiple genes contribute to the
etiopathogenesis of the adult APS [1]. Based
on the various combinations of autoimmune endocrinopathies, the adult APS is
subdivided into types 2–4 [2].
Patients with APS are at high risk for developing non-glandular autoimmune diseases
[3]. APS-2 is a rare condition defined by
the presence of Addison’s disease (AD) along with autoimmune thyroid disease
(ATD) either by hypothyroidism [Hashimoto’s thyroiditis, (HT)], or
hyperthyroidism [Graves’ disease, (GD)] and/or type 1 diabetes
(T1D). With a relative prevalence of approximately 40%, APS type 3 is the
most frequent subtype, encompassing T1D and an ATD [2].
Several epidemiological studies and meta-analyses demonstrated a significant
association between impaired serum 25-hydroxyvitamin D (25-OHD) levels and increased
incidence of several autoimmune diseases [4]
[5]. Among the endocrine
disorders, low 25-OHD levels have been described in ATD [6], T1D [7], and AD [8]. In particular, a
polymorphism of the gene codifying for 1-alphahydroxylase (CYP27B1), the enzyme that
catalyzes the conversion of 25-OHD to 1,25(OH)2D3, is associated with AD,
ATD, and T1D [9]. Furthermore, Bellastella et
al. highlighted that patients with APS present reduced vitamin D circulating levels,
but the vitamin D status is not different between patients with single or multiple
autoimmune diseases [4]. The resulting
endocrinopathies and their treatment may impact bone health. Vitamin D deficiency
and secondary hyperparathyroidism are generally associated with decreased bone
mineral density (BMD) and lead to increased bone turnover, which is catabolic for
both cortical and trabecular bone.
Osteoporosis and decreased BMD are frequently observed in several autoimmune
endocrine disorders, including T1D, AD, growth hormone deficiency, and premature
ovarian failure (POF), all diseases with increased risk of hip and vertebral
fractures and associated morbidity, mortality, and health care costs. Many reports
suggest altered bone and mineral metabolism in T1D and AD, which are the components
of APS [10]
[11]. The relationship between T1D and decreased BMD is well known [12]. The onset of T1D typically occurs at a
young age when bone mass is still increasing. People with T1D may achieve lower peak
bone mass (PBM), the maximum strength and density that bones reach. People usually
reach their peak bone mass in their 20 s. Low PBM can increase one’s
risk of developing osteoporosis later in life. AD requires lifelong glucocorticoid
replacement therapy, which, in excessive doses, may result in impaired bone quality
and reduced bone mass [11]. Although thyroid
dysfunction has been known to represent a risk factor for bone disease, the role of
thyroid hormone excess or deficiency in the pathogenesis of osteoporosis and risk
factors of fractures has been underestimated, and the underlying mechanisms are
still uncertain [13].
However, no study evaluated whether the association of APS further affects BMD.
Therefore, our study aimed to examine the possible adverse effects on vitamin D
levels and BMD in patients with APS and investigate its relation with age, body mass
index (BMI), and disease duration.
Materials and Methods
Study design and participants
In a cross-sectional study, we evaluated patients with APS (n=44) who
were 18–45 years of age and receiving outpatient care at Uludag
University Medical School Endocrinology and Metabolism Clinic in Bursa. Age-,
gender- and ethnicity-matched healthy controls (n=55). They were
volunteer blood donors without personal or a family history of autoimmune
diseases and were randomly recruited from staff personnel or medical students
from the University Hospital. The respective type of APS was included according
to the accepted clinical, endocrinological, and immunological criteria. APS-2 is
characterized by the presence of AD along with ATD and/or T1D [2]. In APS-3, ATD can co-occur with any
autoimmune disorder but not AD [1]. APS-2
was found in 31.8% (n=14) and APS-3 in 68.1%
(n=30). In APS-2, AD coexisted with ATD, T1D, or the combination of
these conditions in 64% (n=9), 18% (n=3), and
13% (n=2) of cases, respectively. In addition, three
non-endocrine organ-specific autoimmune disorders (vitiligo, autoimmune
hepatitis, and myasthenia gravis) were detected. In APS-3, only one autoimmune
endocrinopathy (T1D), five non-endocrine organ-specific (vitiligo, autoimmune
hepatitis, pernicious anemia, myasthenia gravis, and alopecia) autoimmune
endocrinopathies were detected in combination with ATD. All patients with APS
are treated according to standard guidelines. Ages at the diagnosis of disease
manifestations were collected from hospital records.
Exclusion criteria for both groups were represented by calcium and vitamin D
supplementation, bone active therapy (antiresorptive/bone-forming
therapy), liver disease, moderate and severe chronic kidney disease, history of
parathyroid or rheumatological disease, inflammatory bowel disease, gastric
surgery, isolated malabsorption syndrome, congenital adrenal hyperplasia,
surgical adrenalectomy, severe obesity
(BMI>35 kg/m2) or underweight
(BMI≤19 kg/m2). Postmenopausal women,
smokers, and alcohol users were also excluded from the study.
The study was approved by the Uludag University Clinical Research ethical
committee (decision no: 2013–1/21, dated January 15, 2013) and
was conducted according to the principles of the Declaration of Helsinki. All
participants provided written consent.
Data collection
Anthropometric data, including their height (cm), body weight (kg), and BMI, were
recorded. The BMI was calculated as the ratio of weight and the square of the
height.
Blood serum samples were collected following 10–12 hours of
fasting and obtained to measure serum calcium (Ca), phosphorus (P), alkaline
phosphatase (ALP), 25-OHD, and parathormone (PTH). The laboratory parameters
(Ca, P, ALP) were analyzed photometrically in a Roche c-702 autoanalyzer (Roche
Diagnostics, Ankara, Turkey). Serum concentrations of PTH and 25-OHD were
measured by electroluminescence with a Roche Cobas e-601 (Roche, USA). According
to the Endocrine Society Clinical Guidelines, vitamin D level was classified as
normal (≥30 ng/ml), insufficient (>20
to<30 ng/ml), or deficient
(≤20 ng/ml) [14].
To avoid the interference of seasonal variations of values, we measured 25-OHD
levels on plasma samples of patients and controls drawn from June through early
September, considering that these months usually express the peak of secretion
of vitamin D.
BMD values in the lumbar spine (L1–L4), femur total, and femur neck
regions were measured by the dual X-ray absorptiometry (DXA) method using a
Hologic Delthi-w serial no: 70232 bone densitometer. BMD results were evaluated
according to the Z-score criteria the World Health Organization (WHO)
recommended. The Z-score is the number of standard deviations (SDs) a given BMD
measurement differs from the mean for age- and gender-matched reference
populations. A Z-score of≤–2.0 was defined as low BMD or
“below the expected range for age”; a Z-score between
–1.0 and –2.0 was defined as the low range of normality [15]. An abnormal BMD was determined as low
or low normal BMD (Z-score<–1.0) [16].
Statistical analysis
All data obtained in the study were recorded in the SPSS 21.0 database (SPSS,
Chicago, IL, USA). Statistical analyses were applied. The analysis used
mean±standard deviation, median, minimum, and maximum values for
continuous variables and frequency and percentage values for categorical
variables as descriptive statistics. The chi-square test of independence or
Fisher’s exact was used to test categorical variables between groups.
The Shapiro–Wilk test was used to examine whether the data showed normal
distribution. Student’s t-test was used for data showing normal
distribution in comparisons of two groups, and the Mann–Whitney U-test
was used for those who did not. The associations between continuous variables
were determined by Pearson/Spearman correlation analysis.
ü-Values of<0.05 were considered statistically significant.
Results
For the study, 44 patients with APS and compatible criteria defined for the study and
55 individuals for the healthy control group were included. Of the total subjects,
83 (83.8%) were females, and 16 (16.2%) were males, with a mean age
of 36.4±12.0 years. The demographic and laboratory findings of the study and
control groups are presented in [Table 1].
Demographic variables such as age and gender were similar in both groups
(p>0.05). Only the BMI in the APS patients was lower than in the control
subjects (p=0.01).
Table 1 Demographic and biochemical parameters of the study
groups.
|
APS-2 (n=14)
|
APS-3 (n=30)
|
Adult APS (n=44)
|
Controls (n=55)
|
pa
|
pb
|
|
Demography
|
|
Age (years)
|
33.8±8.0
|
36.2±11.2
|
35.2±10.0
|
36.8±10.4
|
0.72
|
0.61
|
|
Gender(male/female)
|
2/12
|
4/26
|
6/38
|
10/45
|
0.82
|
0.45
|
|
BMI (kg/m²)
|
21.2±5.3
|
24.2±7.5
|
23.2±3.5
|
25.4±3.3
|
0.09
|
0.01
|
|
Disease duration (months)
(min-max)
|
38 (3.9–137)
|
25 (2.7–105)
|
34 (4.5–109)
|
–
|
0.08
|
–
|
|
Ca metabolism parameters
|
|
Ca (mg/dl)
|
9.0±0.9
|
9.8±1.6
|
9.5±0.6
|
9.6±0.3
|
0.64
|
0.25
|
|
P (mg/dl)
|
3.3±0.4
|
3.7±0.1
|
3.5±0.5
|
3.3±0.60
|
0.56
|
0.17
|
|
ALP (IU/l)
|
65.5±22.8
|
69.4±31.1
|
68.5±37.1
|
73.8±27.3
|
0.08
|
0.35
|
|
PTH (pg/ml)
|
69.2±39.9
|
73.2±31.4
|
71.2±36.4
|
57.5±22.4
|
0.12
|
0.04
|
|
25-OHD (ng/ml)
|
16.4±2.80
|
18.7±8.9
|
17.2±7.2
|
25.4±7.1
|
0.62
|
<0.001
|
|
<20 ng/ml, n (%)
|
9 (64.2)
|
14 (46.6)
|
23 (52.2)
|
9 (16.3)
|
0.09
|
<0.001
|
|
20–29.9 ng/ml,n (%)
|
4 (28.5)
|
11 (36.6)
|
15 (34.1)
|
33 (60)
|
0.42
|
0.85
|
|
≥30 ng/ml,n (%)
|
1 (7.1)
|
5 (16.6)
|
6 (13.6)
|
13 (23.7)
|
0.52
|
0.63
|
Data are expressed as mean±SD, median (min: minimum-max: maximum), or
n, (%). Ca: Calcium; P: Phosphorus; ALP: Alkaline phosphatase;
25-OHD: 25-Hydroxyvitamin D; PTH: Parathormone. pa: Comparisons
between patients with APS-2 and patients with APS-3. pb:
Comparisons between controls and patients with adult APS.
The differences between APS and controls regarding several laboratory findings, such
as Ca, P, and ALP, were insignificant (p>0.05). In contrast, the differences
between the two groups regarding the PTH and 25-OHD values were significant
(p<0.05) ([Table 1]). Compared with
controls, APS patients showed significantly lower 25-OHD levels (25.4±7.1,
17.2±7.2 ng/ml, respectively; p<0.001). Moreover,
there was no significant difference between APS subpopulations (16.4±2.80 in
APS-2 and 18.7±8.9 ng/ml in APS-3; p=0.62). Among
the 55 controls, 9 showed vitamin D deficiency (16.3%), and 33 showed
vitamin D insufficiency (60%). The prevalence of vitamin D deficiency was
significantly higher in patients with APS (52.2%) than in controls
(p<0.001) ([Table 1]). There was no
significant correlation between serum Ca and vitamin D levels (p>0.05). In
contrast, a statistically significant negative correlation existed between serum
vitamin D levels and PTH values (r=–0.23, p=0.01).
The BMD of the patients was evaluated in terms of osteoporosis risk. The BMD
measurement results are summarized in [Table
2]. Lumbar BMD (LBMD), femur neck BMD (FnBMD), femur total BMD (FtBMD)
values, and Z-scores in patients with APS patients were significantly lower than
those of controls.
Table 2 Absolute BMD levels and Z-scores of lumbar and femur
areas of the study groups.
|
APS-2 (n=14)
|
APS-3 (n=30)
|
Adult APS (n=44)
|
Controls (n=55)
|
pa
|
pb
|
|
BMD (g/cm
2
)
|
|
Lumbar L1–L4
|
1.01±0.55
|
1.17±0.23
|
1.05±0.11
|
1.23±0.18
|
0.29
|
0.02
|
|
Femur neck
|
0.57±0.01
|
0.81±0.33
|
0.79±0.06
|
0.95±0.02
|
0.11
|
0.03
|
|
Femur total
|
0.91±0.04
|
0.12±0.35
|
0.95±0.07
|
1.11±0.12
|
0.31
|
<0.001
|
|
Z-score
|
|
Lumbar L1–L4
|
–0.54±1.27
|
–0.59±1.78
|
–0.58±1.42
|
–0.03±1.13
|
0.81
|
0.04
|
|
Femur neck
|
–1.56±1.18
|
–1.13±1.09
|
–1.23±1.20
|
0.11±0.90
|
0.60
|
<0.001
|
|
Femur total
|
–1.33±1.41
|
–1.15±1.87
|
–1.21±1.12
|
0.23±0.81
|
0.77
|
<0.001
|
|
Low BMD, n (%)
*
|
8 (57.1)
|
3 (10)
|
11 (25)
|
2 (3.6)
|
0.04
|
0.03
|
Data are expressed as mean±SD; BMD: Bone mineral density.
*Low BMD was defined as a Z-score of ≤
–2 in at least one of three sites. pa : Comparisons
between patients with APS-2 and patients with APS-3. pb :
Comparisons between controls and patients with adult APS.
The prevalence rates of low BMD, defined by the lowest
Z-score+≤+–2.0 in one or more sites (lumbar spine,
total femur, and femoral neck), were significantly lower in the APS patients
compared with the control group (25% vs. 3.6%;
p+=+0.04). In total, 24 (54.5%) APS patients and 4
(7.2%) controls had abnormal BMD (Z-score≤–1.0). Also, these
were found in patients with APS type 2 and type 3 at (13/14) 92.6%
and (11/30) 36.6%, respectively. The low BMD rate in type 2 and type
3 APS and the control group was (8/14) 57.1, (3/30) 10, and
(2/55) 3.6%, respectively. Among the 24 (54.5%) APS patients
with abnormal BMD Z-scores, eight patients (33.3%) had deficient, and 5
(20.8%) patients had insufficient vitamin D values. Serum vitamin D levels
and abnormal BMD values were not found to be related.
Associations of age, disease duration, BMI, PTH, and 25-OH vitamin D values were
analyzed with the subjects’ lumbar total, femur neck, and femur total BMD
([Table 3]). This study did not find any
significant correlation between the 25-OH vitamin D, PTH values, and the BMD values
of the lumbar total, femur total, and femur neck (p>0.05) ([Table 3]). It was determined that there was a
significant positive correlation between the age and BMI values and the
patient’s lumbar total BMD values and that the duration of the disease was
negatively correlated with the lumbar total, femur neck, and femur total BMD values
(p<0.05) ([Table 3]).
Table 3 The correlation between demographic and some
laboratory values with bone mineral density values.
|
Lumbar L1–L4 (g/cm2)
|
Femur neck BMD (g/cm2)
|
Femur total BMD (g/cm2)
|
|
p
|
r
|
p
|
r
|
p
|
r
|
|
Adult APS (n=44)
|
|
Age
|
0.037
|
0.362
|
0.201
|
0.191
|
0.225
|
0.186
|
|
BMI
|
0.047
|
0.255
|
0.166
|
0.11
|
0.069
|
0.298
|
|
Disease duration
|
0.042
|
–0.212
|
0.034
|
–0.257
|
0.021
|
–0.388
|
|
PTH
|
0.228
|
–0.082
|
0.249
|
–0.17
|
0.152
|
–0.164
|
|
25-OH D
|
0.322
|
–0.193
|
0.873
|
0.003
|
0.522
|
0.08
|
|
Controls (n=55)
|
|
Age
|
0.541
|
–0.08
|
0.496
|
–0.264
|
0.844
|
0.023
|
|
BMI
|
0.181
|
0.243
|
0.258
|
0.369
|
0.142
|
0.19
|
|
PTH
|
0.234
|
–0.283
|
0.062
|
–0.268
|
0.235
|
–0.39
|
|
25-OH D
|
0.895
|
–0.001
|
0.622
|
–0.058
|
0.836
|
–0.006
|
BMD: Bone mineral density; BMI: Body mass index; 25-OHD: 25-Hydroxyvitamin D;
PTH: Parathormone.
Discussion
The first aim of our study was to evaluate whether there was a change in BMD and
vitamin D levels in patients with adult APS. To the best of our knowledge, this is
the first study examining BMD exclusively in these patients. As a second aim, we
also investigated the correlations between age, BMI, disease duration, PTH, 25-OHD,
and BMD in patients with adult APS.
We found only one study in the literature that evaluated vitamin D status in patients
with adult APS [4]. This study demonstrated a
high prevalence of vitamin D deficiency in adult APS compared to the control
subjects; among the APS subgroups, all patients with vitamin D deficiency consisted
of APS-3. This finding suggested that it is likely linked to an impairment of this
vitamin’s absorption or metabolic steps at the skin, liver, or kidney level
[4].
Our study found that circulating 25-OHD levels were significantly reduced in both
groups of APS patients compared to control subjects, but there was no significant
difference in 25-OHD levels between APS subgroups. However, it was observed that the
prevalence of vitamin D deficiency was higher in those with APS-2. This can be
explained by the fact that AD only exists in patients with APS-2 who use long-term
high-dose glucocorticoids. Only a few observational studies investigated the link
between vitamin D plasma levels and AD [17]. A
record-linkage study by Ramagopalan et al. showed that a large cohort of patients
admitted to a UK hospital for vitamin D deficiency significantly elevated rates of
AD (rate ratio=7.0% CI: 3.6–12.3) and other autoimmune
diseases [8]. Another recent study by Zawadzka
et al. retrospectively analyzed medical records of 31 adult patients diagnosed with
AD, in whom serum vitamin D was measured. A total of 90.3% of AD patients
had inadequate vitamin D concentrations (<30 ng/ml), and
19.3% of them were found to be severely vitamin D deficient
(<10 ng/ml). They also found only serum calcium
concentrations significantly correlated with VD status (r=0.53,
p=0.006) [17]. This and similar
studies have shown that vitamin D could play a protective role in the pathogenesis
of AD [5]. Also, some studies have shown that
glucocorticoid replacement therapy may frequently be administered in doses more than
necessary in patients with AD [18].
Glucocorticoids decrease the production of active vitamin D by inhibiting vitamin D
hydroxylation in the liver and Ca absorption in intestinal mucosal cells when
glucocorticoids are administered more than necessary. In addition, glucocorticoids
have been known to increase Ca and P excretion by inhibiting renal tubular
reabsorption. PTH secretion rises to compensate for the decreased intestinal
absorption and renal Ca excretion. The developed secondary hyperparathyroidism
condition provides serum Ca to be in balance [19]. In our study, depending on this condition, the patient’s
serum Ca level is in the normal range. PTH levels are significantly higher in
patients with all types of APS than in healthy controls.
In this study, we also evaluated the BMD of the patients in terms of the risk of
osteoporosis that may develop due to secondary hyperparathyroidism arising from
vitamin D and Ca absorption problems. Regarding BMD, patients were compared with
their peers using Z scores. It was observed that the BMD of those with APS was
significantly lower statistically than the healthy control group. BMD reduction was
detected in both the femur and the lumbar vertebrae measurements. Among the APS
subgroups, the low BMD rate was approximately six times higher in APS-2 than in
APS-3 (57.1% vs. 10%).
In the literature, there is no data about BMD in APS. It has been widely analyzed
with monoglandular autoimmune endocrine diseases, but most of these studies were
performed on patients with T1D, AD, and ATD [11]
[13].
We found a low BMD rate in the patients with APS-2, in line with previous studies on
patients with AD [19]
[20]
[21]
[22]. This can be explained by
the fact that AD is only present in patients with APS-2. The most extensive
cross-sectional study comprising 292 Addison patients from Norway, Britain, and New
Zealand demonstrated a reduced BMD at the femoral neck and lumbar spine [22]. The decreased BMD in patients with AD is
believed to be reduced because glucocorticoid therapy used for replacement purposes
is generally given more than physiological needs. Steroids directly inhibit bone
formation. Since they shorten the active formation period in the bone remodeling
cycle, they also reduce the number and activity of osteoblasts. Thus, bone matrix
formation is reduced. The serum osteocalcin and bone ALP levels decrease within the
first 24 hours, depending on the dose. Systemic steroids also accelerate
bone resorption by reducing androgen and estrogen and increasing PTH secretion. As a
result, the bone resorbs. These effects are dose and duration-dependent, and the
effect increases in high doses and long-term use [20]
[21]
[22]. Although it has been suggested in some
studies that the osteopenia detected in Addison’s patients is not related to
the dose and duration of glucocorticoid replacement therapy, many studies are
showing that BMD in Addison’s patients who use very low doses close to
physiological needs are not different from healthy controls [23]
[24].
A recent prospective two-year study on adult patients with AD demonstrated a
significant increase in femoral neck and lumbar spine BMD Z-scores in the patients
with a cautious reduction in hydrocortisone equivalent dose [25]. In our study, the leading cause of the low
BMD rate of patients with APS-2 could be glucocorticoid therapy, likely to be
long-term and supra-physiological doses. In addition, the sudden decrease in BMD
might be related to the lack of protective role played by adrenal gland androgens in
the body.
Most APS patients with low BMD measurements may have secondary multifactorial causes
such as malabsorption disorders, low BMI, smoking, and alterations in vitamin D
metabolism or drug exposure that inhibits PBM gain during adolescence [26]. Considering these factors, patients with
these characteristics were excluded from the study. One of the other primary
etiology that comes to mind in the case of low BMD in patients with APS is celiac
disease (CD), which can be seen with autoimmune comorbidities and malabsorption
evolving because of it [27]
[28]. It has been known for a long time that
osteoporosis and bone deformities secondary to osteoporosis are more common in
celiac patients. The reported prevalence of osteopenia or osteoporosis in CD is
variable, ranging from 38–72% of newly diagnosed patients [29]. However, in our study, no patient had CD
and follow-up malabsorption findings.
Our results showed that age, BMI, and disease duration are important factors
affecting BMD, and this has been widely recognized. However, we have demonstrated no
correlation between BMD values and PTH and 25-OH vitamin D levels.
This study found a positive correlation between age and lumbar total BMD values. As
known in the literature, advancing age is a risk factor for low BMD. It is important
to note that bone mass is accrued throughout adolescence and early adulthood, and
also stated that BMD could reach peak bone mass until the age of 35 at the latest
[30]. In our study, an average age of 35
might correspond to PBM formation. Therefore, our positive correlation might be
related to PBM.
One of the strongest predictors of BMD is body weight. Many studies have shown that
obesity is important in maintaining BMD and quality [31]. Body weight depends on fat and lean mass, and a positive correlation
was found between BMD and fat mass [31]. Our
result in controls concord with this finding, indicating that controls with
significantly higher BMI values tended to have higher BMD at all sites of the body
than patients.
Some reports also indicate that disease duration is negatively associated with BMD at
all sites in patients with AD treated with glucocorticoid, and bone loss occurs
mainly in patients with longer disease duration [20]
[21]. Therefore, we investigated
the relationships between disease duration and BMD at all sites. In our study, in
accordance with the literature, we found that the disease duration was negatively
correlated with the femur total BMD; it was determined that with increased duration
of the disease, a significant reduction of femur total BMD values might have
occurred.
Limitations
The first limitation of our study is the cross-sectional design, which restricts
the assessment of causal relationships. Although there was no significant
difference between the groups concerning gender, the lack of subgroup analyses
regarding gender is another limitation because we could not administer a
subgroup analysis since the number of male patients was insufficient for
statistical analysis. Second, we have excluded the patients with malabsorption
as the study needs to include exact data regarding dietary habits or daily
calcium intake. Finally, there needs to be more data on drugs used by patients
in our study (as glucocorticoid replacement regimens have different effects on
BMD). Despite these limitations, our study has contributed to the knowledge of
vitamin D status and bone diseases in adult APS patients and the results of BMD
as real-life data.
Conclusion
In this study, vitamin D and BMD values (lumbar total, femur neck, and femur total)
are significantly lower in patients with all types of APS compared to healthy
gender- and age-matched controls. These data suggest that osteoporosis screening and
prevention among patients with APS need to be prioritized. It is also important for
all of these patients to receive treatment to increase bone mass and reduce the risk
of fractures. In addition, the effects of altered osteoimmunology on the function of
bone cells and reduction of BMD need to be elucidated in future prospective studies
on patients with APS.
