Mortality studies in psoriatic arthritis: Results from a single outpatient center. II. Prognostic indicators for death

OBJECTIVE: To investigate prognostic factors associated with mortality in patients with psoriatic arthritis (PsA). METHODS: Patients followed up at the Toronto PsA Clinic between 1978 and 1994 were included. Patients were reviewed at initial clinic entry and at 6-month intervals using a standard protocol. Data on deaths were collected in a prospective manner, and death certificates were used to identify the primary and antecedent cause(s) of death. All death information was recorded in the clinic's computerized database. Only factors that represented standard clinical measures of disease activity and progression were studied. The relationship between potential prognostic factors recorded at the time of the first clinic visit and the mortality rate was determined using the Cox relative risk regression model. RESULTS: There were 428 patients (234 men and 194 women), of whom 68% were known to be alive on September 1, 1994, 20% were lost to followup but assumed to be alive, and 12% had died. Multivariate analysis revealed that an erythrocyte sedimentation rate (ESR) > 15 mm/hour, medications used prior to initial clinic visit, radiologic damage, and the absence of nail lesions were associated with an increased overall mortality rate. There is some suggestion that prior medication use was least important for deaths associated with the circulatory system, while radiologic damage was particularly important for such deaths. A marked sex-associated effect was noted among deaths caused by injuries/poisoning, since 6 of the deaths occurred in men and only 1 was in a woman. CONCLUSION: Patients with PsA are at an increased risk of death compared with the general population. Evidence of previously active and severe disease, as manifested by the prior use of medications and by radiologic changes as well as an elevated ESR at presentation, are prognostic indicators for death. The presence of nail lesions appears to be a protective factor that has the most clinical importance in the context of previously active and severe disease.