2317: Cardiovascular Risk and Psoriatic Arthritis Features: Dactylitis may Predict Cardiovascular Events

Background/Purpose: Patients with PsA have an elevated risk of cardiovascular (CV) disease. We assessed whether specific PsA features may predict CV events among PsA patients.

Methods: Data are presented from patients seen at the University of California, San Diego’s Healthcare system. Data were collected from 1/1/2012 through 4/5/2025 from all patients with a PsA ICD-10 code, and were procured through data extraction concierge services. All charts were then reviewed and 766 met inclusion criteria: at least one office visit, PsA diagnosis and features confirmed by a rheumatologist.

Participant characteristics were assessed at office visits. Patients were tracked from time of PsA diagnosis until occurrence of a CV event, or their last office visit before the end of the follow-up period. We define a CV event as a new occurrence of one of the following ICD-10 codes: coronary artery disease, peripheral vascular disease, heart failure and transient ischemic attack or stroke. We considered the following PsA features: dactylitis, enthesitis, nail pitting and severe psoriasis (PsO; defined as body surface area ≥%4). Presence/absence of each predictor was ascertained and was assessed as a binary predictor of CV events. Peripheral arthritis was not included as it was inconsistently noted in the charts reviewed.

Unadjusted Kaplan-Meier curves were estimated for each primary predictor. Cox regression models estimated CV risk (Hazard Ratios: HR) and are presented as unadjusted, adjusted for gender, age and smoking status (Model 1), and adjusted for gender, age, smoking status, BMI and race (Model 2).

Results: Mean tracked follow-up time was 4.35 years (SD 3.60) and the median was 3.53. Participant characteristics are shown in Table 1. 56 CV events were reported. Table 2 shows estimated HRs (95% CI) for each predictor and model. Among the variables examined, the presence of dactylitis emerged as a significant predictor of CV events in the adjusted model 2 (HR 1.83; 95% CI [1.04, 3.23]; p=0.036) and neared significance in the adjusted model 1 (HR 1.75; 95% CI [1.00, 3.07]; p=0.05). Kaplan Meier survival curves are presented in Figure 1.

Conclusion: Dactylitis was a significant predictor of CV events in the adjusted model 2. We suggest that confounding by age may have masked this association in the unadjusted model. Individuals with dactylitis were significantly younger than those without, and older age was associated with increased CV risk.

No significant association was observed between CV events and the other PsA features. Some limitations may have affected the strength of our conclusions. We did not have available complete information on PsA treatments, obviating the ability to adjust for this potential confounder. Nor were we able to definitively ascertain prior history of CV events, preventing exclusion of such higher risk patients. Our median tracked follow-up period was shorter than the 7.3 years found in a meta-analysis of 112 cohort studies examining CV outcomes (Magnussen C, et. al. N Engl J Med 2023). Future studies incorporating adjustment for treatment status, more stringent exclusion criteria and a longer follow-up period may help to adjudicate the discrepancy between our findings and literature reporting a link between PsA features and CV risk.