Description of features of patients with polymyalgia rheumatica (PMR) with and without characteristic extracapsular pattern of inflammation
| Extracapsular pattern (n=14) | Non-extracapsular pattern (n=8) | p Value | |
|---|---|---|---|
| Demographics | |||
| Age, median (range) | 75 (55, 85) | 78 (70, 84) | 0.22 |
| Male, n (%) | 8 (57%) | 0 (0%) | 0.02 |
| Acute-phase markers | |||
| ESR, median (range), mm/h | 46 (9, 119) | 38 (4, 81) | 0.63 |
| CRP, median (range), mg/L | 36 (5, 118) | 5.25 (5, 76) | 0.03 |
| PV, median (range), mPa s | 1.93 (1.78, 2.12) | 1.81 (1.57, 2.04) | 0.36 |
| IL-6, median (range), pg/mL | 25.8 (0.3, 87.6) | 6.0 (0.2, 131.5)* | 0.04 |
| Composite disease activity scores | |||
| PMR-AS (median, IQR) | 78.6 (53.8, 103.0) | 70.2 (34.4, 106.1) | 0.73 |
| Pretreatment patient-reported outcomes | |||
| Pain VAS (median, IQR) | 7.7 (5.0, 8.1) | 7.9 (4.3, 8.7) | 0.63 |
| Stiffness VAS (median, IQR) | 6.3 (4.5, 8.0) | 8.2 (5.2, 8.9) | 0.29 |
| Fatigue VAS (median, IQR) | 7.1 (5.2, 7.6) | 8.4 (7.0, 9.7) | 0.03 |
| HAQ-DI (median, IQR) | 1.25 (1.09, 1.50) | 1.56 (1.28, 2.09) | 0.07 |
| Assessment of glucocorticoid responsiveness at first follow-up | |||
| ‘I feel back to normal since taking steroids’. n (%) | 11/13 (85%) | 1/6 (17%) | 0.01 |
| ‘I feel [or felt] back to normal since taking steroids’. n (%) | 12/14 (86%) | 1/8 (13%) | 0.001 |
| Fatigue VAS at follow-up (median, IQR) | 1.3 (0.2, 3.8) | 7.1 (3.6, 9.8) | 0.02 |
| HAQ-DI at follow-up (median, IQR) | 0 (0, 0.625) | 1.0 (0.76, 2.07) | 0.003 |
| Prognosis | |||
| Stopped glucocorticoids permanently after <1 year† | 1/14 | 4/7 | 0.03 |
| Relapse-free† | 7/14 | 2/8 | 1.00 |
| Relapsed when on 5 mg or more† | 3/14 | 3/7 | 0.35 |
| Required initial dose increase >15 mg | 2/14 | 1/8 | 1.00 |
The PMR-AS is the PMR Activity Score as described by Bird and Leeb. Either Mann–Whitney U test or Fisher's exact test was used for non-normally distributed values; unpaired t test for normally distributed variables. All tests were two-tailed. Apart from glucocorticoid responsiveness (the a priori hypothesis), p values should be interpreted in the light of multiple testing. Bonferroni correction for all the variables reported here (likely over-stringent because of strong correlation between ESR/CRP/PV/IL-6 and between patient-reported VAS scores) would require a threshold of 0.05/19, or p<0.0026.
*Excludes one patient who did not have IL-6 measured.
†Excludes one patient who was lost to follow-up at 4 months.
CRP, C-reactive protein; IL-6, interleukin-6; PV, plasma viscosity; VAS, visual analogue score.