作者: Michele F. Doran , Cynthia S. Crowson , Gregory R. Pond , W. Michael O'Fallon , Sherine E. Gabriel
DOI: 10.1002/ART.10529
关键词:
摘要: Objective. Patients with rheumatoid arthritis (RA) have been shown to an increased susceptibility the development of infections. The exact causes this risk are unknown, but may relate immunologic disturbances associated disease or immunosuppressive effects agents used in its treatment. This study was undertaken identify predictors serious infections among patients RA. Identification such factors is necessary first step reducing excess infection Methods. Members a population-based incidence cohort Rochester, Minnesota residents ages >18 years, who had diagnosed RA between 1955 and 1994, were followed up longitudinally through their complete medical records until January 1, 2000. We examined potential for all objectively confirmed (by microbiology radiology) requiring hospitalization. Potential included severity measures (rheumatoid factor positivity, elevated erythrocyte sedimentation rate, extraarticular manifestations RA, functional status), comorbidities (diabetes mellitus, alcoholism, chronic lung disease), other (presence leukopenia, smoking). Predictors identified using multivariate timedependent Cox proportional hazards modeling. Results. 609 total followup time 7,729.7 person-years (mean 12.7 years per patient). A 389 (64%) at least 1 objective confirmation, 290 (48%) Increasing age, presence (chronic disease, organic brain diabetes mellitus), as well use corticosteroids, strong (P < 0.004) both univariate analyses. Notably, disease-modifying antirheumatic drugs not analyses, after adjustment demographic characteristics, comorbidities, disease-related variables. Conclusion. number These results can be prospectively high-risk patients, benefit from closer implementation preventive strategies. Concern about has become heightened recent reports describing severe opportunistic treated new biologic (1–4). Although increase septic reported (5), little known rates various these patients. In elsewhere issue Arthritis & Rheumatism (6), we found higher rate virtually types compared age- sex-matched group subjects without companion that predictive observed identification determinants toward burden