Article Text

OP0186 Analysis of Anti Porphyromonas Gingivalis (PG) Antibodies in a Large Cohort of Patients with an Early Arthritis (ESPOIR)
  1. R. Seror1,
  2. S. Legall2,
  3. M. Bonnaure-Mallet2,
  4. T. Schaeverbeke3,
  5. A. Cantagrel4,
  6. X. Mariette1
  1. 1Rheumatology, Hopital Bicêtre, Le Kremlin-Bicetre
  2. 2Microbiology, University, Rennes
  3. 3Rheumatology, Pellegrin university Hospital, Bordeaux
  4. 4Rheumatology, University Hospital, Toulouse, France


Background Chronic Porphyromonas Gingivalis (PG) infection is associated with parodonditis and the later is associated with rheumatism arthritis (RA). PG contains a bacterial deiminase able to citrullinate bacterial and human antigens. Thus it is logical to suspect an association with chronic PG infection and anti-CCP + RA.

Objectives To investigate a possible link between PG infection and RA antibody profile and structural damage.

Methods For assessing chronic PG infection we look at serum anti PG antibodies (titers were log-transformed) at baseline in the ESPOIR cohort consisting of 813 patients with early arthritis (more than 6 weeks and less than 6 months) not treated with steroid or DMARD. RA was defined according to ACR/EULAR 2010 criteria or ACR 1987 criteria. RA patients were divided between anti-CCP+ and -. Sex and age-matched control were also studied.

Results Among the 812 patients included in the ESPOIR cohort, 694 (78% female, mean age 48.5±12.3, mean DAS28 5.3±1.2) had RA according to ACR/EULAR 2010 criteria and 641 according to ACR 1987 criteria. Whatever the used criteria set, the level of anti-PG antibodies did not significantly differ between RA and non-RA patients. Likewise, the mean level of anti-PG antibodies in the 694 RA patients did not differ from that of 70 healthy controls (1.53±0.50 vs 1.47±0.43; p=0.20). The mean level of anti-PG antibody did not differ between anti-CCP positive and negative RA patients (1.48±0.43 vs 1.46±0.43; p=0.63), and did not correlate with anti-CCP levels.

Anti-PG antibody level did not correlate with DAS28 (rho=-0.01; p=0.76) or other parameters of disease activity or serum level of inflammatory cytokines (IL.6, IL1b, INFg or TNFa). Anti-PG antibody levels were higher in the 361 never smoker patients compared to the 333 ever smokers (1.51±0.42 vs 1.43±0.43; p=0.01). In sensitivity analyses, significantly more never smoker had high anti-PG level (>mean+2SD of healthy controls) than ever smoker (35.7% vs 27.6%; p=0.02). There was a trend for a higher frequency of baseline erosive change (mSharp score ≥1) in patients with high anti-PG levels compared to those with low levels (46.7% vs 32.4%; p=0.008). But the level of anti- PG antibodies was neither associated with baseline modified Sharp score (mSharp) nor with structural progression (defined by a >1 point increase of mSharp).

Conclusions Like in a much smaller study recently published, we could not detect in this large cohort of patients with early arthritis any association between anti PG antibodies with RA or anti CCP+ RA. However, we found an association between anti-PG levels and baseline erosive change in never smokers patients, suggesting a possible association in this population not exposed to this major pathogen.

Disclosure of Interest None Declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.