Session Information
Session Type: Abstract Submissions (ACR)
Background/Purpose: Rheumatoid arthritis (RA) is associated with increased mortality due to cardiovascular disease (CVD). Select antigen-specific anti-citrullinated protein antibodies (ACPA) are associated with atherosclerotic burden in RA. Furthermore, citrullinated proteins have been localized in atherosclerotic plaque. Vascular calcifications (VC) may be found incidentally on hand and wrist radiographs in RA. This study examined the relationship of VC with CVD risk factors, ACPA subtypes, and all-cause mortality in RA.
Methods: Hand and wrist radiographs from 906 RA patients were scored for the presence of VC as either “positive” or “negative”. Patient characteristics associated with VC were assessed using univariate and multivariable logistic regression. ACPAs were measured using an enzyme-linked immunosorbent assay for second generation anti-cyclic citrullinated peptide (CCP2) antibodies, then 19 distinct ACPA subtypes were measured by a bead-based immunoassay and sorted based on q-values calculated by Significance Analysis of Microarrays (SAM). ACPA associations with VC were further examined using multivariable quantile regression. VC and all-cause mortality were examined using Cox proportional hazards regression.
Results: Ninety-nine (11%) patients demonstrated VC on hand and wrist radiographs (Table 1). In multivariable analyses, factors associated with VC included diabetes (OR 2.85; 95% CI 1.43, 5.66, p=0.003), history of CVD (OR 2.48; 95% CI 1.01, 6.09, p=0.047), prednisone use (OR 1.90; 95% CI 1.25, 2.91, p=0.003), current vs. never smoking (OR 0.06; 95% CI 0.01, 0.23, p=0.001) and former vs. never smoking (OR 0.36; 95% CI 0.27, 0.48, p=0.001). In the ACPA subtype analyses using SAM, antibodies to citrullinated forms of Apolipoprotein E (anti-Cit-ApoE), fibrinogen, and vimentin, but not anti-CCP antibody, were differentially expressed in patients with VC (Table 2). The association of anti-Cit-ApoE with VC remained significant following all multivariate adjustments, as well as adjustment for known CVD. After adjusting for significant covariates and stratifying by age and gender, VC were associated with increased all-cause mortality (HR = 1.41; 95% CI 1.12, 1.78, p=0.004).
Conclusion: In this cohort, ~1 in 10 RA patients had VC on hand and wrist radiographs. VC were associated with traditional CVD risk factors and prednisone use and yielded an independent association with all-cause mortality. ACPA targeting Cit-ApoE were increased among patients with VC. Mechanisms underpinning the association of select ACPA with CVD in RA warrant further investigation.
|
Table 1. Hazard ratios for association of vascular calcification all-cause mortality (top). Demographics of rheumatoid arthritis subjects with and without radiographic vascular calcifications with unadjusted comparisons (bottom). |
|||
|
|
Age-adjusted and sex-stratified univariate hazard ratio (95% CI) |
Multivariable hazard ratio* (95% CI) |
|
|
Vascular calcifications |
1.36 (1.06, 1.74) p = 0.016 |
1.41 (1.12, 1.78) p = 0.004 |
|
|
|
|
|
|
|
Demographics & Health Behaviors |
Vascular Calcifications Present (n = 99) |
Vascular Calcifications Absent (n = 807) |
p-value** |
|
Age, years (SD) |
72 (9) |
64 (11) |
<0.001 |
|
Men (%) |
96 (97) |
725 (90) |
0.02 |
|
Race Caucasian (%) African American (%) |
79 (80) 13 (13) |
615 (76) 142 (18) |
0.53 |
|
Smoking Never (%) Former (%) Current (%) |
38 (38) 53 (54) 8 (8) |
146 (18) 427 (53) 234 (29) |
<0.001 |
|
BMI, kg/m2 (SD) |
27 (5) |
28 (5) |
0.02 |
|
Diabetes (%) |
38 (38) |
139 (17) |
<0.001 |
|
Hyperlipidemia (%) |
45 (45) |
343 (43) |
0.58 |
|
Cardiovascular disease (%) |
43 (43) |
164 (20) |
<0.001 |
|
RA-Related Factors |
|
|
|
|
Disease duration, years (SD) |
16 (13) |
13 (11) |
0.004 |
|
Anti-CCP positive (%) |
74 (77) |
620 (78) |
0.79 |
|
RAPID-3 score (SD) |
2.9 (1.3) |
2.5 (1.4) |
0.02 |
|
DAS-28 score (SD) |
3.9 (1.5) |
3.8 (1.5) |
0.29 |
|
CRP, mg/L (SD) |
14 (24) |
11 (18) |
0.21 |
|
Prednisone use (%) |
53 (55) |
313 (41) |
0.01 |
|
*Adjusted for age, gender, diabetes, cardiovascular disease, smoking, and prednisone use. CI: confidence intervals, SD: standard deviation, BMI: body mass index; anti-CCP: cyclic citrullinated peptide 2; RAPID-3: routine assessment of patient index data; CRP: C-reactive protein. **T-test or chi-square as indicated, p<0.05 significant. |
|||
|
Table 2. ACPA subtypes generated from SAM analysis followed by multivariable quantile regression* examining ACPA associations with radiographic vascular calcifications in rheumatoid arthritis. |
||||||
|
ACPA subtype |
Fold Change in ACPA VC+ vs. VC- |
q-value (%) |
Model 1 p-value |
Model 2 p-value |
Model 3 p-value |
Model 4 p-value |
|
Citrullinated Vimentin |
1.35 |
0 |
0.031 |
0.021 |
0.025 |
0.084 |
|
Citrullinated Fibrinogen |
1.30 |
0 |
0.41 |
0.49 |
0.49 |
0.60 |
|
Citrullinated Apolipoprotein E |
1.28 |
0 |
0.014 |
0.013 |
0.018 |
0.034 |
|
*Model 1: Unadjusted. Model 2: Adjusted for age and gender. Model 3: Model 2 plus disease duration, RAPID3, prednisone. Model 4: Model 3 plus a history of hypertension, hyperlipidemia, and diabetes, body mass index, C-reactive protein, and smoking status (current, former vs never). ACPA: anti-citrullinated protein antibodies; VC: vascular calcifications; SAM: significance analysis of microarrays. |
||||||
Disclosure:
E. B. Solow,
None;
F. Yu,
None;
G. M. Thiele,
None;
J. Sokolove,
None;
W. H. Robinson,
None;
Z. M. Pruhs,
None;
K. Michaud,
None;
A. R. Erickson,
None;
H. Sayles,
None;
G. S. Kerr,
Bristol Myers Squibb,
2,
Pfizer,
2,
Genentech,
2;
A. L. Gaffo,
None;
L. Caplan,
None;
L. A. Davis,
None;
G. W. Cannon,
None;
A. M. Reimold,
None;
J. Baker,
None;
P. Schwab,
None;
D. Anderson,
None;
T. R. Mikuls,
None.
« Back to 2014 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/vascular-calcifications-on-hand-and-wrist-radiographs-are-associated-with-cardiovascular-risk-factors-antigen-specific-anti-citrullinated-protein-antibodies-and-mortality-in-rheumatoid-arthritis/