JORT

Introduction

Psoriatic arthritis (PsA) is a chronic inflammatory rheumatic disease characterised by heterogeneous involvement of peripheral joints, axial skeleton, skin, nails, and entheses [1,2]. The disease is frequently accompanied by comorbidities such as cardiovascular disease and metabolic syndrome, adding to its clinical complexity [3-7].

Management remains challenging due to the disease’s variable course and the need for therapies targeting both joint and skin manifestations [5]. A treat-to-target strategy is recommended to optimize outcomes and prevent structural damage and typically includes a combination of conventional synthetic diseasemodifying antirheumatic drugs (csDMARDs), biologics, and targeted synthetic therapies, with treatment tailored to individual domains of involvement [2,8].

Systemic glucocorticoids (GCs) have historically been avoided in PsA and psoriasis due to concerns over triggering cutaneous flares, including severe forms such as pustular or erythrodermic psoriasis, but this caution originates largely from early anecdotal reports and case series [9-11], and is reflected in treatment guidelines which discourage the use of systemic GCs in psoriasis and PsA unless it is interpreted without alternative [2]. However, GCs are regularly used in clinical practice for PsA patients, particularly during acute flares or as a bridging strategy when initiating csDMARDs [12].

Recent evidence challenges the assumption that systemic GCs are strongly associated with psoriasis flares. For example, a large retrospective study by Gregoire et al. found a low incidence of psoriasis flares (1.42%) following systemic GC use in patients with a history of psoriasis, with only one severe flare recorded in over 500 patients [13].

Vincken et al. conducted a systematic review examining the safety of systemic GC use in both psoriasis and PsA [14]. While the review concluded that psoriatic flares were rarely reported, the authors noted that among the PsA-specific studies included (n=5), only one directly reported on skin flare outcomes [11,1517]. This highlights a major evidence gap: despite frequent GC use in PsA patients, high-quality data specifically addressing the risk of psoriasis flares in PsA populations remains scarce and inconclusive.

There is a need for focused investigation into these common treatment situations. A more precise risk analysis of short-term GC therapy is expected to have significant impact on clinical decision-making. In the present study, we report the largest realworld dataset to date examining the association between systemic glucocorticoid use and the risk of psoriasis flares in individuals with PsA.

Methods

Study Design

This is a single-center retrospective observational cohort analysis conducted at the Department of Rheumatology, Esbjerg University Hospital, Denmark.

Settings and Participants

All patients diagnosed with PsA treated at an outpatient clinic were identified through the electronic medical records system in November 2023. Medical records were reviewed, and if the PsA diagnosis were verified, patients were subsequently contacted via telephone and invited to participate. Before categorizing them as unreachable, up to five contact attempts were made per patient. During structured telephone interviews, patients were asked to recall whether they had experienced any flares of psoriatic skin lesions within the past year. For those who had received GC therapy within the preceding year and reported psoriatic flares, additional information was collected regarding the occurrence of flares within four weeks following the GC administration.

Furthermore, participants were asked to describe any perceived positive or negative effects of GC treatment, irrespective of the route of administration. The study was approved by the local scientific authorities in the region of Southern Denmark (approval number 23/50845).

Data Sources and Variables

Data were collected from medical records and telephone interviews, and included the following variables: age, sex, height, weight, PsA treatment during the observational period, concomitant medications, use of glucocorticoids (GCs) within the past year (via oral, intramuscular, or intra-articular administration), patientreported occurrence of psoriatic flares within the past year, and any occurrence of psoriatic flares within four weeks following GC administration.

Study Size

This is an observational study of all accessible PsA patients from the outpatient clinic of the department of rheumatology at Esbjerg University Hospital. Thus, no formal study size calculations were performed.

Statistics

Categorical variables were summarized using frequencies and percentages, while continuous variables were described by means and standard deviations (SDs). Baseline group differences were assessed using standardised differences, which offer a sample size– independent measure of imbalance. Unlike p-values, standardized differences quantify the magnitude of group differences in standard deviation units, facilitating consistent interpretation across variables. A standardized difference <0.1 suggests negligible imbalance, whereas values >0.8 indicate substantial differences. Additionally, absolute mean differences with 95% confidence intervals (CIs) were reported to provide interpretable effect sizes in original measurement units, with CIs indicating the precision of estimates.

Associations between GC therapy and psoriasis flare were tested using chi-square analysis on 2×2 contingency tables, with significance defined as p < 0.05. Chi-square tests were also used to evaluate differences in psoriasis flare risk by GC administration route. Finally, logistic regression was conducted to identify baseline predictors of psoriasis skin flare. All statistical analyses were conducted using the R 4.4.2 software.

Results

A total of 343 potential PsA patients were identified through the electronic medical records system. Of these, 61 patients did not meet the Classification Criteria for Psoriatic Arthritis (CASPAR)

(21) and were therefore excluded. An additional 61 patients could

not be reached despite multiple contact attempts, resulting in 221 eligible patients for interview. Among these, 208 patients consented to and successfully completed the interview. Within this cohort, glucocorticoid treatment during the observation period was documented in 90 patients, according to the electronic medical records (see Figure 1).

Figure 1: Study Flow.

Table 1 presents the demographic and clinical characteristics of the study groups. Overall, group differences were minimal. Patients receiving GC therapy were more likely to be treated with leflunomide (11.1% vs. 3.4%), whereas TNF inhibitor use was more common in the non-GC group (39% vs. 26.7%). Although these differences were statistically significant based on 95% confidence intervals, the standardized differences indicated that the magnitude of imbalance was small. No notable differences were observed between groups in terms of sex, age, BMI, or other treatment regimens.

Table 1: Patient characteristics across groups.

Incidence of Psoriatic Flares among Patients treated with Glucocorticoids is summarized in Table 2. Within the past year, 47% (56/118) of patients without GC exposure reported a flare, compared with 63.3% (57/90) of those who received GCs (p = 0.0326). This corresponds to an absolute risk difference of 15.9 percentage points (95% CI 2.4–29.3) and a number needed to harm (NNH) of 6.3 (95% CI 3.4–40.9). The unadjusted odds ratio for flare within 1 year was 1.91 (95% CI 1.09–3.35). When flares were assessed within 4 weeks of GC administration (primary endpoint), only 5/90 (6%) GC-treated patients reported a flare; all five cases occurred after intra-articular injection.