Risk factors for total joint arthroplasty infection in patients receiving tumor necrosis factor α-blockers: a case-control study
1 Rheumatology B Department, Cochin Hospital, AP-HP, 27 rue du faubourg Saint-Jacques, Paris 75014, France; UPRES-EA 4058, Medicine Faculty, Paris Descartes University, 12 rue de l'Ecole de Médecine, Paris 75006, France
2 Department of Rheumatology, Bicêtre Hospital, AP-HP, 78 rue du Général Leclerc, Le Kremlin-Bicêtre 94270, France; INSERM U802, Paris-Sud University, 63 rue Gabriel Péri, Le Kremlin-Bicêtre 94270, France
3 Infectious Diseases - Internal Medicine Department, Cochin Hospital, AP-HP, 27 rue de faubourg Saint-Jacques, Paris 75014, France
4 Department of Rheumatology, Provo Hospital, 25 rue de Barbieux, Roubaix 59100, France
5 Department of Rheumatology, Hotel Dieu Hospital, 1 place Alexis-Ricordeau, Nantes 44000, France
6 Department of Rheumatology, Minjoz Hospital, 3 boulevard Alexandre Fleming, Besançon 25000, France
7 Department of Infectious Diseases, University Hospital, 2 rue de l'Hôtel-Dieu, Rennes 35000, France
8 Department of Infectious Diseases, University Hospital, 4 avenue Reine Victoria, Nice 06000, France
9 Department of Clinical Epidemiology and Biostatistics, Bichat Hospital, AP-HP, 46 rue Henri Huchard, Paris 75018, France; INSERM U738, Medicine Faculty, Paris 7 Denis Diderot University, 16 rue Henri Huchard, Paris 75018, France
Arthritis Research & Therapy 2010, 12:R145 doi:10.1186/ar3087Published: 16 July 2010
The objective of this study was to assess natural microbial agents, history and risk factors for total joint arthroplasty (TJA) infections in patients receiving tumor necrosis factor (TNF)α-blockers, through the French RATIO registry and a case-control study.
Cases were TJA infections during TNFα-blocker treatments. Each case was compared to two controls (with TJA and TNFα-blocker therapy, but without TJA infection) matched on age (±15 years), TJA localization, type of rheumatic disorder and disease duration (±15 years). Statistical analyses included univariate and multivariate analyses with conditional logistic regression.
In the 20 cases (18 rheumatoid arthritis), TJA infection concerned principally the knee (n = 12, 60%) and the hip (n = 5, 25%). Staphylococcus was the more frequent microorganism involved (n = 15, 75%). Four patients (20%) were hospitalized in an intensive care unit and two died from infection. Eight cases (40%) versus 5 controls (13%) had undergone primary TJA or TJA revision for the joint subsequently infected during the last year (P = 0.03). Of these procedures, 5 cases versus 1 control were performed without withdrawing TNFα-blockers (P = 0.08). In multivariate analysis, predictors of infection were primary TJA or TJA revision for the joint subsequently infected within the last year (odds ratio, OR = 88.3; 95%CI 1.1-7,071.6; P = 0.04) and increased daily steroid intake (OR = 5.0 per 5 mg/d increase; 1.1-21.6; P = 0.03). Case-control comparisons showed similar distribution between TNFα-blockers (P = 0.70).
In patients receiving TNFα-blockers, TJA infection is rare but potentially severe. Important risk factors are primary TJA or TJA revision within the last year, particularly when TNFα-blockers are not interrupted before surgery, and the daily steroid intake.