Schlüsselwörter Frühe rheumatoide Arthritis - Osteoporose - Rheumafaktor - Antikörper gegen citrullinierte
Proteine
Key words Early rheumatoid arthritis - osteoporosis - anti-citrullinated protein antibodies
- rheumatoid factor
Introduction
Osteoporosis and its clinical consequence fracture is a well-known extra-articular
manifestation in rheumatoid arthritis (RA) [1 ]. Osteoporosis prevalence among RA patients is approximately double that
found in general population [2 ]. The overall
occurrence of systemic osteoporosis in RA patients ranges from 12.3 to 38.9%
at the site of lumbar spine (LS) and 6.3 to 36.3% at the hip site [2 ]
[3 ]
[4 ]. There is at least a 2-fold to 6-fold
increase in vertebral fracture (VF) risk in RA patients than in those with primary
osteoporosis [4 ]
[5 ]. Moreover, among patients with early RA,
up to 10% had an evidence of systemic bone loss [6 ]. In addition, VFs were reported in the first
year after the onset of RA and around one-third of the females with RA report VF
during five years of follow up [7 ].
It had been long assumed that the increased prevalence of osteoporosis in RA patients
is mainly attributed to an inherent RA factors such as chronic inflammatory process
in long-standing disease, use of glucocorticoids or some disease-modifying
anti-rheumatic drugs (DMARDs), and immobilization. However, these factors do not
explain the higher prevalence bone loss and VFs in patients with early RA
particularly in treatment naïve patients. In contrary, emerging data suggest
that destruction of bone in RA is mainly attributed to: inflammatory and
autoimmunity mechanisms [8 ]
[9 ].
Pro-inflammatory cytokines are considered the key drivers of bone tissue destruction
by increasing the number of mature osteoclasts [8 ]
[10 ]
[11 ]. However, autoantibodies in RA patients,
especially anti-citrullinated protein antibodies (ACPA), increases the risk of
osteoporosis by inducing osteoclast differentiation which increases bone resorption
[12 ]
[13 ]. It had been reported that ACPA-positive
RA patients had increased serum receptor activator of nuclear factor kappa B ligand
(RANKL) and had higher rates of joint damage progression over time independent
pro-inflammatory cytokines [14 ]
[15 ]. In this regard, the clinical features of
RA were found to be preceded by ACPA by about 5–10 years [9 ]
[16 ]. Therefore, it is expected that
ACPA-positive early RA patients may have signs of generalized bone loss besides
joint destruction.
Thus, this study aimed to investigate the risk factors for osteoporosis in patients
with early RA and who was treatment-naïve at inclusion.
Methods
Patients
In this study, 200 consecutive patients with early inflammatory arthritis were
invited to participate in the study. Patients were collected from the outpatient
Clinic of Rheumatology and Rehabilitation Department, from different Hospitals
in Saudi Arabia during the period from February 2015 till May 2017. All patients
were naïve to glucocorticoids and DMARDs, and had short duration of
arthritis symptoms (≤12 weeks).
Twenty-two patients did not provide a written consent and were excluded from the
study. Hence, at inclusion, collection of clinical data, laboratory
investigations and BMD measurement were performed only for patients who had
provided a written informed consent (n=178). Those patients were
followed up. Only data of patients who fulfilled the American College of
Rheumatology (ACR)/European League Against Rheumatism (EULAR) 2010
classification criteria for RA [17 ] were
analyzed (n=135) while patients with diagnoses other than RA
(n=25) or underwent spontaneous remission (n=18) were excluded
from the study ([Fig. 1 ]). This study was
approved by the local ethical committees.
Fig. 1 Flowchart for identification of population with early
rheumatoid arthritis.
Demographic and Clinical Data
During history taking, data known to affect BMD were obtained including age,
gender, body mass index (BMI), menopausal status, current smoking status, risk
factors for secondary osteoporosis other than RA, history of osteoporotic
fractures, use of calcium and vitamin D supplements, hormone replacement therapy
and bisphosphonates. RA duration was defined as the duration since first
self-reported joint symptoms was obtained.
Laboratory Investigations
ACPA & IgM rheumatoid factor (RF) were measured by a second-generation
Phadia Immuno-CAP 250 EliA CCP assay (Phadia, Freiburg, Germany) and by
immuno-nephelometry using a Dimension Vista 1500 system (Siemens, Erlangen,
Germany) respectively, according to the manufacturers’ guidelines.
Positive results were considered with values above 10 IU/mL for ACPA and above
20 IU/mL for RF. A low antibody level was defined as values greater than
the defined upper limit of normal and ≤100 IU/ml, and a high
antibody level as a value>100 IU/ml [18 ].
Disease activity was assessed using erythrocyte sedimentation rate (ESR) and
C-reactive protein (CRP). Disease Activity Score 28 using erythrocyte
sedimentation rate (DAS28-ESR) was used to assess RA activity for
patient’s global assessment (0–100) combining the visual
analogue scale, number of tender and swollen joints and ESR [19 ].
BMD Measurements
BMD measurements were performed using dual-energy X-ray absorptiometry (DXA)
equipment (Hologic, Vilvoorde, Belgium) at LS, femoral neck (FN) and total hip
(TH). All procedures were performed by a trained technician in accordance with
the manufacturers’ standardized scan and positioning protocols.
Specifically, in the current study we measured BMD from L2 to L4 (LS.BMD), TH
(TH.BMD) and FN (FN.BMD). BMD is expressed in g/cm2 .
Diagnoses of decreased BMD were made when T-score is<−1.0,
according to the World Health Organization (WHO) [20 ]. Besides, osteoporosis is diagnosed on
the basis of the lowest T-score at the three sites (LS, FN, and TH) according to
recommendations of the International Society for Clinical Densitometry [20 ].
Statistical Analysis
All continuous data were tested for normality of distribution prior to any
calculations. Continuous data with normal distribution were described as mean
and standard deviation (SD) while continuous data not normally distribution was
described as median and interquartile range (IQR). Categorical data were
described as counts and percent. Comparison of the patient characteristic
between patients with reduced BMD and patients with normal BMD were evaluated
using the independent samples t test. Possible confounders that are known to
impact the BMD (including age, BMI and vitamin D status and duration of the
symptoms) were analyzed by regression analyses to evaluate the relationship
between the auto-antibodies and the reduced BMD. Statistical analyses were
performed using SPSS version 20.0, and<0.05 was considered
significant.
Results
Characteristics of the Patients
[Table 1 ] demonstrates the
characteristics of the 135 patients with early RA enrolled in the study. All
patients fulfilled the ACR/ EULAR 2010 classification criteria for RA.
The study population was all premenopausal females with an average age of
46.3±3.6 years. Patients were treatment-naïve either to
glucocorticoids or to DMARDs. The duration of RA ranged from 7–12 weeks
with an average of 9.6±1.7 weeks. All patients had an active disease
with an average DAS28-ESR and -CRP of 4.2±0.8 and 3.5±0.7
respectively. Of these patients, 54.8% were RF-positive, 31.9%
were ACPA-positive for and ACPA and 24.4% were double-positive.
Table 1 Demographic and clinical characteristics of the
patients with ERA.
Age (years), mean±SD
46.3±3.6
BMI (kg/m2 ), mean±SD
24.2±3.0
Vitamin D status, n (%)
Normal
40 (29.6)
Insufficiency
41 (30.4)
Deficiency
54 (40)
Duration of RA symptoms (weeks), mean±SD
9.6±1.7
DAS28-ESR, mean±SD
4.2±0.8
DAS28-CRP, mean±SD
3.5±0.7
ESR (mm), mean±SD
41.9±15.2
CRP (mg/dl), mean±SD
8.1±3.2
RF positivity, n (%)
74 (54.8)
RF high titer, n (%)
42 (31.1)
CCP positivity, n (%)
62 (45.9)
CCP high titer, n (%)
373 (24.4)
Double RF and CCP positivity, n (%)
61 (45.2)
DXA
LS
BMD (gm/cm2 ),
mean±SD
0.927±0.205
T-score<−1.0, n
(%)
22 (16.2)
FN
BMD (gm/cm2 ),
mean±SD
0.782±0.183
T-score<−1.0, n
(%)
30 (22.2)
TH
BMD (gm/cm2 ),
mean±SD
0.766±0.178
T-score<−1.0, n
(%)
32 (23.7)
Either LS, FN or TH
T-score<−1.0, n
(%)
40 (29.6)
Prevalence of Decreased BMD
The average BMD at the LS, FN and the TH were 0.927±0.205,
0.782±0.183 and 0.766±0.178 gm/cm2
respectively. T score was<−1.0 in the LS in 16.2%, in
the FN in 22.2% and in the TH in 23.7%. Among our patients,
29.6% had below normal T score at any site ([Table 1 ]).
Relationship of the Decreased BMD with the Demographic Data and Vitamin
D
The age and BMI of the patients were not significantly associated with the
decreased BMD as shown in [Table 2 ].
Similarly, the vitamin D status did not differ significantly between the
patients with decreased BMD compared to patients with normal BMD ([Table 2 ]).
Table 2 Comparison of the demographic and clinical
characteristics between ERA patients with and without decreased
BMD.
ERA patients with decreased BMD
ERA patients with normal BMD
Mean±SD
Mean±SD
p
Age (years)
46.1±3.7
46.4±3.5
0.627
BMI (kg/m2 )
23.9±3.0
24.6±2.9
0.168
Vitamin D status, n (%)
Normal
23 (31.9)
17 (27)
Insufficiency
22 (30.6)
19 (30.2)
Deficiency
27 (38.5)
27 (42.9)
0.770
Duration of symptoms (weeks)
11.6±1.8
11.4±1.7
0.614
DAS28-ESR
4.3±0.8
4.1±0.7
0.194
DAS28-CRP
3.6±0.7
3.5±0.6
0.380
ESR (mm)
42.8±15.5
40.9±15.0
0.464
CRP (mg/dl)
8.4±3.2
7.8±3.2
0.290
RF-positive, n (%)
57 (79.2)
17 (27)
<0.001
RF high titer, n (%)
35 (48.6)
7 (11.1)
<0.001
CCP-positive, n (%)
32 (44.4)
11 (17.5)
<0.001
CCP high titer, n (%)
28 (38.9)
9 (14.3)
<0.001
Double RF and CCP positivity, n (%)
27 (37.5)
6 (9.5)
<0.001
BMD in Relation to Duration of early RA and Activity
Early RA duration, ESR level, CRP level, DAS28-ESR and DAS28-CRP did not differ
significantly between the patients with decreased BMD compared to patients with
normal BMD ([Table 2 ]).
Impact of ACPA positivity and levels on BMD
Patients with decreased BMD were more prevalent ACPA-positive than patients with
normal BMD ([Table 2 ]). In addition, to
explore the impact of the ACPA titer on the BMD, at the LS, FN and TH, we had
stratified the patients into 3 subgroups according to ACPA status. The patients
with high ACPA level had significantly lower BMD than patients with low ACPA
level and also than patients who were negative for ACPA at all sites. Also,
patients with low ACPA level had significantly lower BMD than patients who were
negative for ACPA at all sites ([Fig.
2a–c ]).
Fig. 2 The association between the BMD (g/cm2) at the LS, FN and
TH in the subgroups of the ERA patients classified according to the ACPA
titers (a , b and c respectively) and RF titers
(d , e and f respectively). Data are presented
as median and interquartile range. Dots represent the outliers.
Statistical significance was estimated using the Mann-Whitney U
test.
Impact of RF positivity and levels on BMD
Patients with decreased BMD were more prevalent RF-positive than patients with
normal BMD ([Table 2 ]). Our patients were
also stratified according to the RF status to measure the impact of RF status on
BMD. The patients with high RF level had significantly lower BMD than patients
with low RF level and also than patients who were negative for RF at all sites.
Also, patients with low RF level had significantly lower BMD than patients who
were negative for RF at all sites ([Fig.
2d–f ]).
Impact of Auto-antibodies on BMD after Adjustment for Confounding
Factors
To determine whether the relationship between BMD and patient stratification
according to ACPA and RF status were attributed to the effect of the
auto-antibodies or the result of bias related to subgroups’
characteristic differences, we fitted a multivariable analysis model in which we
included in the variables that are known to impact the BMD including age, BMI
and vitamin D status and duration of the symptoms. As shown in [Table 3 ], after adjustment for these
variables, patient stratification according to ACPA status and RF status (into
negative, low-positive and high positive) still a significant independent
variables associated with lower BMD values.
Table 3 Association of ACPA and other variables on BMD at LS,
FN and TH.
β co-efficient
95% CI
p
Dependent variable: LS.BMD
Age
0.046
−0.004; 0.010
0.461
BMI
0.019
−0.007; 0.010
0.755
Vitamin D
−0.042
−0.019; 0.010
0.500
Duration of symptoms
−0.057
−0.044; 0.016
0.360
RF status
−0.782
−0.231; −0.140
<0.001
ACPA status
−0.518
0.164; −0.090
<0.001
Dependent variable: FN.BMD
Age
0.047
−0.004; 0.008
0.442
BMI
0.018
−0.006; 0.009
0.768
Vitamin D
−0.023
−0.015; 0.010
0.704
Duration of symptoms
−0.070
−0.042; 0.011
0.252
RF status
−0.820
−0.212; −0.133
<0.001
ACPA status
−0.501
0.142; −0.076
<0.001
Dependent variable: FN.BMD
Age
0.039
−0.004; 0.008
0.442
BMI
0.014
−0.007; 0.008
0.768
Vitamin D
−0.030
−0.016; 0.010
0.704
Duration of symptoms
−0.054
−0.039; 0.015
0.252
RF status
−0.746
−0.196; −0.114
<0.001
ACPA status
−0.515
0.143; −0.078
<0.001
Discussion
The main findings of our study, is that reduced systemic BMD is prevalent in patients
with early RA naïve to treatment, and classic osteoporosis related risk
factors and disease related risk factors had negligible impact on BMD. However,
ACPA-positivity and RF-positivity appear to be associated with significant reduction
of the systemic BMD in the patents with early RA.
Our results had shown that 29.6% of the patients with early RA had evidence
of low BMD. Consistent with our results, earlier reports showed evidence of low BMD
in a small proportion of patients. In patients with untreated early RA, the reduced
BMD were observed in 35.5% [21 ],
24.7% [22 ]. In an earlier study,
Forslind et al. [23 ] reported reduced bone
mass in 46% of the women with recent RA. The age of the women in the study
of Forslind et al. ranged from 55 to 61 years which may explain for the higher
proportion of the reduced bone mass than in our study.
The inclusion of premenopausal females, the short duration of RA symptoms, all
patients had an active disease and lack of treatment could exclude the association
of traditional osteoporosis risk factors or the disease related factors to the
reduced BMD in our patients with early RA in agreement with earlier studies [21 ] and in contrast to findings of other
studies that enrolled patients with early RA but with longer disease duration or
variable treatment [22 ]
[23 ]
[24 ].
The presence of reduced systemic BMD in early stages of the RA indicated that while
synovial inflammation may take a long time to produce systemic effects such as
significant generalized BMD changes [25 ],
reduced BMD in early stages of RA appears to be coupled with autoimmunity. Relevant
to this concept, our results had confirmed the recent evidence that link the ACPA
and RF to reduced BMD in patients with early RA [21 ]
[24 ].
The study of Bugatti et al. [21 ] had shown that
high levels of RF would enhance the association between ACPA and reduced BMD.
Sensitive imaging modalities showed that the size of bone erosions in RA patients
was influenced by positive ACPA and higher RF levels [18 ], and also showed that pro-inflammatory
cytokine production increased with combined presence of ACPA and RF [26 ]
[27 ]. It had been suggested that the RF
activity might enhance an inflammatory background sub clinically that could augment
the ACPA-mediated activation of osteoclasts.
The results of this study support assumption that the presence of ACPA and RF may
lead systemic osteoporosis through the activation of osteoclasts independent on the
chronic inflammation. This concept is based on the following 3 major findings: (a)
evidence of reduced systemic BMD in the early stages of RA; (b) The impact of the
disease on BMD was not directly related to the disease activity or disease duration
and (c) the regression analysis had shown the significant association of ACPA and
RF
with the reduced systemic BMD after adjustment for possible confounding factors.
It is important, in this context, to recall the key role of vitamin-D on bone
metabolism, as serum vitamin D was significantly lower in RA patients than that in
controls [28 ]. Based on the finding that
patients with positive ACPA might have higher prevalence of vitamin D insufficiency
[29 ], it had been suggested that the
association between auto-antibodies and reduced BMD is possibly mediated by low
vitamin D levels. The results of the regression analysis had revealed that the
effect of auto-antibodies on BMD is not associated with low vitamin D levels.
Our study had some limitations. We evaluated the association between the presence
of
ACPA and high levels of RF with systemic BMD, but not evaluated at the
juxta-articular level, in early RA patients’ naïve for treatment.
Besides the impermanence of evaluating other bone quality parameters, including
microarchitecture, mineralization and turnover that may better help to define the
impact of auto-antibodies on bone loss.
Conclusion
In conclusion, our data suggest that patients with positive RF or ACPA have increased
risk for development of reduced systemic BMD since the earliest RA stages.
Furthermore, the risk for osteoporosis increases by presence of high levels of RF
or
ACPA. Measurement of BMD should be performed for ACPA- or RF-positive patients with
early RA.
