OBJECTIVE: To present the clinical and imaging characteristics of patients with Axial Psoriatic Arthritis (PsA) and to identify possible subtypes. METHODS: Data were retrieved from the Greek-multicentre PsA study. Axial PsA (axPsA) was defined as PsA (CASPAR criteria) accompanied by inflammatory back pain (present or ever) and positive imaging findings of the sacroiliac joints and/or spine (MRI: active inflammation of sacroiliac joints and/or spine; X-rays: 1984 New York-criteria for radiographic sacroiliitis and/or presence of syndesmophytes in the spine). Demographic and clinical characteristics were compared between axPsA and non-axPsA subsets. Two additional analyses were conducted: a. Isolated sacroiliac joint involvement (sacroiliac axPsA) vs isolated involvement of the rest of the spine (spinal axPsA). b. non-radiographic axPsA (nr-axPsA, positive MRI findings only) vs radiographic axPsA (r-axPsA, positive X-ray findings). RESULTS: Among 922 patients with PsA, 238 (25.8%) had axPsA. Patients with axPsA had less frequently peripheral arthritis at diagnosis, whereas had increased rates of HLA-B27 positivity, enthesitis ever and inflammatory bowel disease. Among patients with axPsA, 42% (n = 101) had isolated sacroiliac axPsA and 32% (n = 75) had isolated spinal axPsA. Sacroiliac axPsA was associated with younger age (OR: 0.97, 95% CI: 0.94-0.99) and enthesitis at diagnosis (OR: 3.37, 95% CI: 1.66-6.82). 35% of patients with axPsA had nr-axPsA and were more commonly females (OR: 2.59, 95% CI: 1.39-4.82) and younger (OR: 0.96, 95% CI: 0.94-0.99) compared with those with r-axPsA. CONCLUSION: Approximately one quarter of patients with PsA exhibit axial involvement, and among them, approximately 30% have isolated spinal axPsA and nr-axSpA, respectively.