Table 2. Association between presence of IFNα serum marker and CV events in RA patients.
| Cardiovascular events | Univariate Analysis | Multivariate Analysis* | ||||
| Absent (n = 93) | Present (n = 27) | OR [95% CI] | p | OR [95% CI] | p | |
| IFNα | ||||||
| IFNlow | 68 (73.1) | 12 (44.4) | 1 | 1 | ||
| IFNhigh | 25 (26.9) | 15 (55.6) | 3.400 [1.401–8.253] | 0.007 | 4.816 [1.254–18.488] | 0.022 |
| Sex | ||||||
| Women | 83 (89.2) | 18 (66.6) | 1 | |||
| Men | 10 (10.8) | 9 (33.6) | 4.150 [1.475–11.680] | 0.007 | ||
| Age at diagnosis | 51.00 (17.00) | 56.00 (23.00) | 1.050 [1.008–1.093] | 0.019 | 1.038 [1.015–1.084] | 0.021 |
| DAS28 score | 3.94 (2.66) | 3.26 (2.09) | 0.814 [0.580–1.143] | 0.235 | ||
| HTA | ||||||
| Normotensive | 72 (77.4) | 15 (55.5) | 1 | |||
| Hypertensive | 21 (22.6) | 12 (44.4) | 3.032 [1.218–7.547] | 0.017 | ||
| DM | ||||||
| Non-diabetic | 84 (91.3) | 20 (74.1) | 1 | |||
| Diabetic | 9 (8.7) | 7 (25.9) | 3.675 [1.192–11.326] | 0.023 | ||
| Smoking habit | ||||||
| Non-smoker | 50 (53.7) | 19 (70.3) | 1 | |||
| Smoker | 43 (46.3) | 8 (29.6) | 1.947 [0.773–4.904] | 0.157 | ||
| Hypercholesterolemia | ||||||
| Normocholesterolemic | 83 (89.2) | 26 (96.2) | 1 | |||
| Hypercholesterolemic | 10 (10.7) | 1 (3.70) | 0.311 [0.038–2.559] | 0.278 | ||
| RF | ||||||
| Negative | 34 (36.5) | 8 (30.8) | 1 | |||
| Positive | 60 (64.5) | 19 (69.2) | 1.530 [0.598–3.916] | 0.375 | ||
| Anti-CCP | ||||||
| Negative | 33 (35.4) | 7 (25.9) | 1 | |||
| Positive | 50 (53.7) | 20 (74.1) | 1.791 [0.678–4.734] | 0.415 | ||
| ANA | ||||||
| Negative | 56 (60.2) | 18 (66.6) | 1 | |||
| Positive | 37 (47.3) | 9 (33.3) | 0.730 [0.296–1.800] | 0.499 | ||
Associations were evaluated by logistic regression analysis using the presence of CV events (ischemic heart disease, n = 10; cerebrovascular accidents, n = 4; heart failure, n = 14; peripheral arteriopathy, n = 1) as dependent variable. Associations that reached statistic significance in multivariate analyses are highlighted in bold.
*
Multivariate analysis adjusted by sex, age at diagnosis, disease activity, smoking habits and presence or absence of HTA, DM and hypercholesterolemia. Accuracy of prediction of the final model was 76.7%.