“Our study suggests that axial PsA and AS with psoriasis seem to be two different diseases with different genetics, demographics, and disease expression,” wrote Joy Feld, MD, of the University of Toronto and coauthors. Their findings were published in.
To investigate the similarities and differences between axPsA and AS patients, the researchers compared two adult cohorts recruited from Toronto clinics. The first was made up of AS patients and divided into two groups: with psoriasis (n = 91) and without psoriasis (n = 675). The second was made up of PsA patients and divided into two groups: axPsA (n = 477) and peripheral PsA (n = 826).
In comparing AS patients with and without psoriasis to axPsA patients, AS patients had a younger age at diagnosis (28.7 years and 30.4 years vs. 35.6 years; P less than .001), were more often male (76% and 72% vs. 64%; P less than .001), and were more likely to be HLA-B27 positive (82% and 75% vs. 19%; P less than .001).
At baseline, AS patients had more back pain (90% and 92% vs. 21%; P less than .001) and worse back metrology (Bath Ankylosing Spondylitis Metrology Index [BASMI] of 3.1 and 2.3 vs. 1.9; P less than .001).
The mean follow-up periods in the axial and peripheral PsA groups were 12.6 and 6.7 years, respectively, whereas in the AS groups with and without psoriasis the periods were 5.4 and 3.5 years. Over time and after longitudinal analysis, axPsA patients had more tender and swollen joints than AS patients with and without psoriasis (5.2 vs. 1.5 and 0.9; P less than .001) while AS patients with and without psoriasis had a higher BASMI (2.9 and 2.2 vs. 1.8; P less than .001) and worse axial disease activity scores (4.1 and 3.9 vs. 3.5; P = .02) as measured by the Bath Ankylosing Spondylitis Disease Activity Index.
After univariate analysis, AS with psoriasis was found to be more associated with HLA-B27 (odds ratio, 16.37; 95% confidence interval, 8.89-30.13; P less than .0001), a higher adjusted mean BASMI (OR, 1.41; 95% CI, 1.21-1.63; P less than .0001), worse sacroiliitis (OR, 7.58; 95% CI, 3.68-15.59; P less than .0001), and greater use of biologics (OR, 1.25; 95% CI, 0.77-1; P = .37), compared with axPsA. A multivariate analysis produced similar findings, including the lack of association between AS and active arthritis (OR, 0.75; 95% CI, 0.64-0.86; P less than .0001).
The authors acknowledged their study’s limitations, including the fact that symptoms often dictate which of the two clinics patients will be referred to, which can ultimately define the diagnosis. “Patients with significant back symptoms are more likely to be referred to the AS clinic,” they wrote, “while patients with more prominent peripheral symptoms are more likely to be referred to the PsA clinic.” Patients with AS in the study were also required to have back pain or limitations in spinal range of motion, while PsA patients were accepted even if they were asymptomatic.
Finally, they noted that some milder cases of the two diseases may have been missed in the cohort recruiting process, although they added that mild cases were, in fact, “present in the cohort, which might improve the generalizability of this study to primary rheumatology clinics.”
The University of Toronto Psoriatic Arthritis Program is supported by a grant from the Krembil Foundation, but this study received no specific funding to carry out the research. Dr. Feld reported being supported by a grant from Novartis. The authors reported no conflicts of interest.
SOURCE: Feld J et al. Rheumatology. 2019 Oct 8.