目標：類風濕性關節炎（Rheumatoid Arthritis，簡稱RA）使用抗腫瘤壞死因子（anti-tumor necrosis factor alpha，簡稱anti-TNF α）在全人工膝關節置換術（Total Knee Arthroplasty， 簡稱TKA）後發生感染的風險較高，但仍未有定論。本研究欲探討RA患者術前使用anti- TNF α在TKA術後發生感染的風險。方法：我們使用2012至2015年全民健康保險資料庫之全人口檔，納入1,126位RA患者並接受TKA手術，追蹤術後三個月內發生感染情形。研究樣本TKA術前三個月RA用藥分為三組：疾病調節抗風濕藥物（Disease-modifying anti-rheumatic drugs，簡稱DMARDs）組（n=854）、anti-TNF α組（n=178）與皮質類固醇組（n=94），並以多重邏輯斯迴歸分析RA用藥與術後感染之勝算比（Odds Ratio，簡稱OR）及95%信賴區間（confidence interval，簡稱CI）。結果：三組TKA術後感染發生率（人月）依序為DMARDs 組0.011（95%CI: 0.006~0.014），anti-TNF α組0.008 （95% CI: 0.0002~0.0176）與皮質類固醇組為0.005 （95%CI: 0.0002~0.0157）。校正干擾因子後，anti-TNF α組發生術後感染的OR為DMARDs組之0.862倍（95%CI: 0.291~2.548）；若與皮質類固醇組相比，發生術後感染的OR為2.662倍（95%CI: 0.25~28.342），皆無統計顯著。結論：RA患者TKA術前三個月使用anti-TNF α，與DMARDs或皮質類固醇相比，術後三個月之感染風險並無顯著不同。

Objectives: Patients with rheumatoid arthritis (RA) receiving anti-tumor necrosis factor alpha (anti-TNFα) therapy may have an increased risk of infection after total knee arthroplasty (TKA). However, conflicting results have been provided for the association of anti-TNFα and postoperative infection. In this study, we examined whether RA patients receiving anti- TNFα therapy have an elevated risk of postoperative infection after TKA. Methods: We used the Taiwan National Health Insurance Research Database of the whole population from 2012 to 2015. The medical records of patients with RA who had undergone TKA were obtained (n = 1126). We divided patients into three groups based on RA medications: disease-modifying anti-rheumatic drugs (DMARDs) (n = 854), anti-TNFα (n = 178), and corticosteroids (n = 94). Multivariate logistic regression was used to estimate the odds ratio (OR) of infection as well as the corresponding 95% confidence interval (CI). Results: The incidence rate (per person-month) of postoperative infection for each group was as follows: DMARDs: 0.011 (95% CI: 0.006–0.014), anti-TNFα: 0.008 (95% CI: 0.0002–0.0176), and corticosteroids: 0.005 (95% CI: 0.0002–0.0157). After adjustment for potential confounders, RA patients with anti-TNFα use did not have increased risk of infection compared to those with DMARDs (OR = 0.862, 95% CI: 0.291–2.548). Similar finding was found for comparison with corticosteroids (OR = 2.662, 95% CI: 0.25–28.342). Conclusions: The risk of postoperative infection in RA patients receiving anti- TNFα is similar to the risk in those treated with DMARDs or corticosteroids.