Table 3 Changes in efficacy parameters.
| Parameter | Mean (SD) at baseline | Mean (SD) at 8 weeks | p Value (95% CI) | Mean (SD) at 24 weeks | p Value (95% CI) |
|---|---|---|---|---|---|
| VAS general health (0–100 mm) | 58 (28) | 47 (29) | 0.137 (−3.7 to 23.9) | 51 (22) | 0.529 (−9.0 to 16.7) |
| VAS dry eyes (0–100 mm) | 53 (32) | 49 (39) | 0.614 (−12.1 to 19.8) | 60 (33) | 0.361 (−24.6 to 9.6) |
| VAS sandy feeling (0–100 mm) | 36 (39) | 45 (43) | 0.287 (−28.9 to 9.2) | 44 (39) | 0.343 (−28.5 to 10.6) |
| VAS dry mouth (0–100 mm) | 67 (31) | 65 (32) | 0.756 (−11.3 to 15.2) | 52 (35) | 0.098 (−3.0 to 32.0) |
| VAS sleep disturbance due to dryness (0–100 mm) | 43 (37) | 39 (31) | 0.304 (−4.9 to 14.7) | 38 (35) | 0.484 (−10.6 to 21.2) |
| MFI general fatigue | 17 (7)* | NP | 11 (5)* | 0.034 | |
| Zung depression score | 37 (17)* | NP | 41.5 (15.7)* | 0.726 | |
| RAND (SF‐36) physical component | 39.8 (16.0)* | NP | 43.8 (10.1)* | 0.026 | |
| RAND (SF‐36) mental component | 48.7 (23.5)* | NP | 52.1 (13.2)* | 0.790 | |
| ESR (mm/1st hour) | 48 (35) | 45 (31) | 0.346 (−3.7 to 9.8) | 42 (34) | 0.200 (−3.6 to 15.6) |
| CRP (g/l) | 8.2 (10.2) | 6.1 (4.1) | 0.287 (−2.1 to 6.4) | 6.3 (4.1) | 0.453 (−4.0 to 8.3) |
| Serum IgA (g/l) | 3.6(1.3) | 3.1 (1.2) | 0.002 (0.2 to 0.7) | 3.1 (1.1) | 0.023 (0.1 to 0.8) |
| Serum IgG (g/l) | 23.6 (8.5) | 20.4 (6.8) | 0.000 (1.7 to 4.6) | 20.2 (6.4) | 0.006 (1.2 to 5.6) |
| Serum IgM (g/l) | 1.6 (0.8) | 1.3 (0.6) | 0.004 (0.1 to 0.5) | 1.2 (0.6) | 0.005 (0.1 to 0.6) |
| RF (U/l) | 481 (664) | NP | 226 (331) | 0.045 (7 to 503) | |
| Schirmer test (mm/5 min) | 7.4 (5.9) | 9.5 (6.8) | 0.253 (−5.7 to 1.63) | 11.1 (10.6) | 0.138 (−4.6 to 0.7) |
| Sialometry (ml/15 min) | 0.8 (1.0) | NP | 0.9 (1.0) | 0.632 (−0.9 to 0.6) |
CRP, C reactive protein; ESR, erythrocyte sedimentation rate; LFS, lymphocytic focus score; LSG, labial salivary gland; MFI, Multidimensional Fatigue Inventory; NP, not performed; RF, rheumatoid factor; SF‐36, 36‐Item Short Form; VAS, visual analogue scale.
*Medians (interquartile ranges) are given because Wilcoxon signed ranks test was used for non‐parametric distributed variables.