Background Most incident cases of psoriatic arthritis (PsA) present among patients with pre-existing psoriasis, and because of the high absolute prevalence of PsA among psoriasis patients this group offers a unique opportunity for risk modification. Several environmental and genetic factors have been associated with PsA. Trauma has been found to be associated with PsA in case series and cross sectional studies. This observation has led to the idea of a “deep Koebner” phenomenon playing a role in PsA, mirroring the Koebner phenomenon in skin psoriasis. We have performed a population based cohort study to determine the risk of PsA following trauma in psoriasis patients.
Objectives To evaluate the risk of incident PsA among psoriasis patients exposed to physical trauma.
Methods We performed a matched cohort study within the Health Improvement Network (THIN) database among psoriasis patients with data available between 1995 and 2013. Patients exposed to trauma were randomly matched to up to five unexposed controls based on gender, age, and the date of entry into THIN. Trauma exposure was stratified into subgroups of joint, bone, nerve, and skin trauma. Cox proportional hazard models were used to calculate the hazard ratio of developing psoriatic arthritis among those exposed to physical trauma, and the full model included adjustment for age, gender, date of entry into THIN, duration of psoriasis, BMI, smoking, alcohol consumption and the number of visits to the general practitioner. Missing values for smoking, alcohol use, and BMI were imputed by a sequential regression method. For comparison an identical analysis was performed in the non-psoriasis THIN population to evaluate the risk of developing rheumatoid arthritis (RA) following physical trauma.
Results Psoriasis patients exposed to trauma (N=15,416) and unexposed controls (N=55,230) were identified and followed for a total of 425,120 person-years (py) during which 1,010 incident PsA cases were recorded. The incidence rate of PsA among unexposed psoriasis patients was 22 (95% CI 21 to 24) per 10,000 py and 30 (95% CI 26 to 34) per 10,000 py in the exposed group. The results of the fully adjusted cox model analysis showed that psoriasis patients exposed to trauma had an increased risk of PsA compared to controls, with a hazard ratio (HR) of 1.32 (95% CI 1.13 to 1.54). In our subset analysis, bone and joint traumas were associated with multivariate HRs of 1.46 (95% CI 1.04 to 2.04), and 1.50 (95% CI 1.19 to 1.90), respectively, while nerve trauma and skin trauma were not associated with a statistically significant difference in risk compared to controls. Patients without psoriasis exposed to trauma did not have an increased multivariate adjusted risk of developing RA: HR 1.04 (95% CI 0.99 to 1.10) based on an analysis of 551,723 exposed individuals and 2,672,836 unexposed individuals followed for 19,479,771 py.
Conclusions We found an increased risk of PsA among psoriasis patients exposed to physical trauma, particularly when trauma to bone and joints was recorded.
Acknowledgements Partial funding by RANNIS 120433021 (TJL) and NIH/NIAMS (JMG)
Disclosure of Interest None declared