Arthritis & Rheumatism, Volume 60,
October 2009 Abstract Supplement

The 2009 ACR/ARHP Annual Scientific Meeting
Philadelphia October 16-21, 2009.

Risk of Infection in Rheumatoid Arthritis Associated with Use of Anti-TNF Agents, Dmards and/or Prednisone

Trejos1,  Eduardo Bonilla, Hannon2,  Michael J., Skanderson1,  Melissa, Good1,  Chester, Kwoh2,  C. Kent

Pittsburgh VA Healthcare System, Pittsburgh, Pittsburgh, PA,
University of Pittsburgh, Pittsburgh, PA


To compare the risk of infection associated with different treatment options for RA taking into account the potential combinations of DMARDs, anti-TNF agents and corticosteroids


Data from both the VA National Patient Care Database and the Pharmacy Benefits Management Database dating from Oct. 1, 2000 to Sept. 30, 2007 were utilized to identify cases of RA. The definition of RA was based on ICD-9 codes 714* (either two outpatient codes at least 30 days apart or one inpatient code) and having ever had at least one prescription for a DMARD. A nested case-control study design was utilized to examine 3457 cases of RA patients with a serious infection that were identified based on their corresponding ICD-9 codes from an inpatient admission. Controls with RA but without a serious infection were matched 2:1 by year of initial RA visit, an inpatient or outpatient visit in the same month and year as the inpatient visit for RA case, and location in the same VA region (VISN). Conditional logistic regression was used to examine the risk of serious infection based on DMARD exposure in the 100 days prior to the inpatient admission while controlling for age, gender and the Deyo modification of the Charlson Comorbidity score. The exposures were DMARD1 (i.e., methotrexate/leuflonomide), DMARD2 (i.e., sulfasalazine/hydroxychloroquine/azathiprine), TNF (i.e., infliximab/etanerecept/adalimumab) and Prednisone equivalents (PRED, i.e., categorized as none, low dose (mean of 1–10mg/day) or high dose (mean of >10mg/day)). The reference group consisted of RA patients not on any TNF, DMARDs or PRED.


The results are summarized in the table below.

RA treatmentaOR*95% CI
TNF and DMARD1 and high PRED5.183.66–7.33
TNF and DMARD2 and any PRED4.803.01–7.66
TNF and DMARD1 and low PRED4.232.95–6.08
TNF and DMARD1 and DMARD2 and any PRED4.002.70–5.92
DMARD1 and high PRED3.692.97–4.58
DMARD2 and high PRED3.632.88–4.57
TNF and any PRED3.422.44–4.81
DMARD1 and DMARD2 and high PRED3.252.52–4.19
TNF alone3.062.03–4.60
DMARD2 and low PRED3.022.40–3.80
DMARD1 and low PRED2.962.40–3.67
Any PRED alone2.361.85–3.00
TNF and DMARD12.331.60–3.38
DMARD2 alone2.131.71–2.64
TNF and DMARD22.061.10–3.85
DMARD1 alone2.021.64–2.50
DMARD1 and DMARD22.011.64–2.50
*adjusted for age, gender and Charlson Comorbidity score


The use of any PRED, especially high dose PRED, is associated with an increased odds of infection in RA when added to a TNF or DMARD. The risk of serious infection with TNF alone is similar to that of DMARD1 or DMARD2 combined with low dose PRED. The combination of TNF, any DMARD and any PRED is associated with the highest odds of serious infection and should be used with special caution.

To cite this abstract, please use the following information:
Trejos, Eduardo Bonilla, Hannon, Michael J., Skanderson, Melissa, Good, Chester, Kwoh, C. Kent; Risk of Infection in Rheumatoid Arthritis Associated with Use of Anti-TNF Agents, Dmards and/or Prednisone [abstract]. Arthritis Rheum 2009;60 Suppl 10 :2058
DOI: 10.1002/art.27130

Abstract Supplement

Meeting Menu