Characteristics of patients from DESIR cohort according to presence or not of MRI inflammatory lesions
| Inflammatory MRI | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Overall | Thoracic | Lumbar | Sacroiliac | |||||||
| N=648 | Abnormal n=125 | Normal n=523 | p* | Abnormal n=140 | Normal n=508 | p* | Abnormal n=303 | Normal n=345 | p* | |
| Gender (women) | 338 (53) | 39 (31.5) | 299 (58.1) | <0.0001 | 55 (3.,3) | 286 (56.9) | <0.0001 | 132 (44.2) | 209 (61) | <0.0001 |
| Age (years) | 33.8 (8.7) | 34.4 (8.5) | 33.6 (8.7) | 0.35 | 35.4 (8.5) | 33.4 (8.7) | 0.01 | 32.5 | 35 | 0.0003 |
| HLA B27 status | 374 (58.2) | 92 (74) | 54.6 (282) | <0.0001 | 89 (63.5) | 287 (5.7) | 0.15 | 207 (68.1) | 169 (49) | <0.0001 |
| BMI (m/kg²) | 23.9 (4.6) | 24.7 (6.9) | 23.7 (3.8) | 0.04 | 24.2 (6.5) | 23.8 (3.9) | 0.34 | 23.9 | 23.9 | 0.91 |
| DMARD† | 57 (8.8) | 9 (7.1) | 48 (9.3) | 0.21 | 11 (7.6) | 46 (9.1) | 0.51 | 24 (7.7) | 33 (9.7) | 0.49 |
| NSAID intake† (Y/N) | 597 (92.8) | 117 (93.6) | 480 (92.7) | 0.72 | 131 (93.6) | 471 (9.7) | 0.73 | 288 (95.1) | 314 (91) | 0.05 |
| NSAID intake† (% of daily intake) | 4.3 (14.7–78.3) | 44.8 (17.4–71.4) | 40.4 (15–78.8) | 0.97 | 37.6 (14.6–71.3) | 43.7 (15.7–79.8) | 0.42 | 48.2 (19.1–80.5) | 38 (13.3–76.4) | 0.04 |
| CRP | 3.1 (1–7.1) | 5 (1.6–13) | 3 (1–6) | <0.0001 | 4 (1–10.5) | 3 (1–6.1) | <0.0001 | 4 (1.5–9.6) | 2.5 (1–5) | <0.0001 |
| Thoracic pain | 393 (61.5) | 86 (68) | 307 (59.7) | 0.06 | 87 (63) | 310 (61.3) | 0.7 | 176 (58.7) | 221 (64.2) | 0.14 |
| Lumbar pain | 589 (91.7) | 113 (90.4) | 476 (92.1) | 0.54 | 134 (95.7) | 460 (90.7) | 0.06 | 275 (91.1) | 319 (9.5) | 0.51 |
| Buttock pain | 513 (79.6) | 96 (76.8) | 412 (80.2) | 0.4 | 105 (76.6) | 408 (80.5) | 0.32 | 257 (85.7) | 256 (74.4) | 0.0001 |
BMI, Body Mass Index; CRP, C-reactive protein; DMARD, Disease Modifying AntiRheumatic Drug; ; NSAID, Non-Steroidal Anti-inflammatory Drugs; SIJ, Sacro-Iliac Joints.
Categorical variables were expressed as percentage (%), continuous variables as mean and SD except for CRP as median (25–75) interquartile space.
*Continuous variables were compared using student t-test or Wilcoxon-Mann-Whitney as appropriate, categorical variables using Pearson χ2 test.
†Regarding treatments, comparisons were systematically adjusted for propensity score using logistic regression models.
Hs-CRP was log-transformed for statistical test.