To the Editor:
Reported data of axial involvement in psoriatic arthritis (PsA) are variable (25–70%). This variability is mainly linked to different ways of defining this feature. Gladman1 established that the prevalence of axial involvement in PsA was close to 50% and that it is associated with HLA-B27. Likewise, psoriasis (PsO) spondylitis, unlike ankylosing spondylitis (AS), is characterized by not having a greater preponderance of the male sex, greater skin involvement, and a less severe course.2
We carried out an observational, cross-sectional, single-center study. The objective of our study was to estimate the frequency of axial involvement in patients with a recent diagnosis of PsA in a rapid diagnostic circuit called Reuma-Check3 and to carry out a comprehensive characterization (clinical, laboratory, and images). We also aimed to analyze the effect of axial involvement on other manifestations, and finally, to compare all features with a group of patients with axial spondyloarthritis (axSpA), diagnosed in the same circuit (with the same evaluators and the same imaging and laboratory techniques) in the same period of time, who did not present current PsO or history of PsO.
This observational study was approved by an institutional ethics committee and was conducted in accordance with the current Declaration of Helsinki, the resolution 1480/11 of the local Health Ministry, and local regulations applicable to this type of study. Patients’ confidentiality was respected according to local law and informed consent was obtained for publication.
Patients included were older than 18 years and were admitted between 2017 and 2019 according to the following criteria: musculoskeletal symptoms such as arthralgia/arthritis, dactylitis, and/or enthesitis associated with PsO or family history. Once diagnosis of PsA was made (according to expert evaluation and the Classification Criteria for Psoriatic Arthritis), the presence of axial symptoms was investigated, which was defined as chronic low back pain (LBP) associated with at least one of the following: Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) > 4, positive sacroiliac clinical tests (flexion, abduction, and external rotation test [FABER]), and/or inflammatory LBP (Assessment of Spondyloarthritis international Society [ASAS] criteria). If patients met these criteria, a second Reuma-Check was performed for the complete study of axial involvement (HLA-B27 and images). The complete description of the Reuma-Check circuit is published.3
A total of 139 consecutive patients with musculoskeletal symptoms plus PsO or family history were admitted (PsO 52%, family history 48%). Seventy-three (53%, 95% CI 44–60) were finally diagnosed with PsA (characteristics in Supplementary Data, available from the authors on request). Twenty-four out of 73 (33%, 95% CI 22–44) patients diagnosed with PsA had clinical axial involvement.
The main characteristics of the patients with axial PsA (axPsA) were as follows: inflammatory LBP (87%), response to nonsteroidal antiinflammatory drugs (NSAIDs; 75%), sacroiliac test (71%), HLA-B27 (17%), positive sacroiliac radiograph (29%), positive sacroiliac magnetic resonance imaging (MRI; 37%), BASDAI (median 4 [IQR 2.8–5]), Bath AS Functional Index (median 4.6 [IQR 3.8–5.3]), morning stiffness (median 40 min [IQR 30–45]), visual analog scale night pain (median 7 [IQR 5–8]), and SpA features (mean 5.6 [SD 1.4]).
Table 1 shows the differential features between PsA patients with and without axial involvement. In the logistic regression model, the variables that were independently associated with axPsA were the presence of nonmusculoskeletal manifestations and Maastricht AS Enthesitis Score. The analysis of the differential features between axPsA and axSpA is shown in Table 2.
Differential features between PsA patients with and without axial involvement.
Differential features between patients with axPsA and axSpA.
In the absence of criteria that define this condition, we relied on a set of clinical data, such as inflammatory LBP, BASDAI activity, and sacroiliac provocation tests. Regarding the first criterion, the classification is the same used for axSpA; in our experience, its prevalence was close to 90% and there was a good response to NSAIDs (75%).4
Although the presence of HLA-B27 in PsA is low (< 40%), we know that the prevalence of HLA-B27 in Latin America is lower than in European cohorts.5 In Argentina, the prevalence of HLA-B27 reported is around 50%. In this study, 17% of patients with PsA and axial involvement were positive for HLA-B27, a fact that reflects its low prevalence in our population.6
Only 29% had positive changes on sacroiliac radiography and 40% had typical changes on MRI. In general, only a few studies have evaluated the frequency of radiographic spinal involvement in patients diagnosed with PsA. Depending on the classification used, between 25% and 70% of patients diagnosed with PsA have been reported to have such participation in combination with the peripheral symptoms.1,2
We found that patients with axPsA had a greater burden of disease, with greater nonmusculoskeletal manifestations such as uveitis, greater involvement of entheses, and lower functional capacity. Other cohorts reported similar data, which also included greater severity of PsO, which was not quantified in our study.7
In literature, axPsA was always compared with AS, showing differential characteristics.8 We have compared these characteristics with a group of patients with axSpA. In our study, we found no difference in sex, disease activity, or functional tests, but we found differences in smoking, age of diagnosis, HLA-B27, and MRI. These findings were expected; however, we highlight the association of inflammatory bowel disease and uveitis in axPsA.9
The strengths of this study include the assessment of axial involvement in PsA in Latin America in a comprehensive manner, as it provides clinical data, imaging, and HLA-B27, and compares the differences between patients with and without axial involvement. It also compares the characteristics of axPsA and axSpA. It would be helpful for groups such as ASAS and the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis to aid in creating a better definition of this condition.10
Footnotes
The authors declare no conflicts of interest.
Partial information of this manuscript was previously presented at the annual meeting of the American College of Rheumatology (ACR 2020; Garcia Salinas R, Ruta S, Torres Chichande J, Sanchez Prado E, Salvatori F, Magri S. Axial involvement in psoriatic arthritis in a comprehensive rapid diagnosis program [Reuma-check PsA]. Analysis of its characteristics [abstract]. Arthritis Rheumatol 2020;72 Suppl 10).
- © 2021 The Journal of Rheumatology