Background A previous MRI study showed a high prevalence of subclinical peripheral arthritis in patients with skin psoriasis (1). However, there are no controlled data regarding subclinical axial inflammation in this disorder.
Objectives To assess the prevalence of axial skeleton changes by whole-body MRI (wbMRI) in skin psoriasis patients without clinical evidence of arthritis and in age- and sex-matched healthy controls.
Methods Twenty-five patients (age median 52; range 20-69) with psoriasis vulgaris and no history or clinical evidence of arthritis and twenty-five age- and sex-matched healthy controls with no history of inflammatory back pain or skin psoriasis were recruited by the Nordic questionnaire. All patients and controls were clinically examined by the same dermatologist and rheumatologist according to a standardized protocol including the psoriasis area severity index (PASI), the body surface area (BSA), Menell, Ott and Schober tests, chest expansion and tender/swollen joint count.
The patients and controls underwent a standardized unenhanced wbMRI at 1.5 T including coronal and sagittal T1w and STIR sequences. Image sets including both the sacroiliac joints (SIJ) and the entire spine were read in a random order and independently by a radiologist and a rheumatologist blinded to the clinical and demographic parameters.
First the readers recorded the presence or absence of spondyloarthritis (SpA) by a global assessment of the SIJ and spine images in both MRI sequences. The confidence with this diagnosis was scored on a scale of 0-10 (0 = definitely no SpA, 10 = definite SpA).
Bone marrow edema (BME), fatty marrow infiltration (FMI) and erosive changes in each SIJ quadrant were recorded by the Morpho module. Spinal BME and FMI of all 23 discovertebral units from C2/3 to L5/S1 were assessed according to the CanDen module.
Lesion prevalence was analyzed descriptively in each group as mean number/percentage of subjects according to the two readers. Fisher’s exact test was used to test for a significant difference in prevalence between the 2 groups, with a desired level of significance p ≤ 0.05.
Results 24% of healthy controls and 30% of skin psoriasis patients were assessed as having axial SpA by global MRI assessment. A high confidence of 8-10 with this diagnosis was recorded in 12% of controls and 18% of patients. The differences between the 2 groups were not statistically significant, both for the global assessment and the lesion analysis in SIJ and spine.
Percentage of subjects (mean of 2 readers) with ≥ 2 affected SIJ quadrants or ≥ 2 spinal changes for each lesion type:
Conclusions Every fourth healthy control was falsely classified by wbMRI as axial SpA. Skin psoriasis patients without clinical evidence of axial or peripheral arthritis showed a similar frequency for SIJ and spinal changes as healthy controls.
Offidani A et al. Subclinical joint involvement in psoriasis: magnetic resonance imaging and Xray findings. Acta Derm Venereol (Stockh) 1998, 78:463-465
Disclosure of Interest None Declared