Background Marked polyarthritis can be induced in male Lewis rats by injection of either heat-killed Mycobacterium tuberculosis (H37Ra, Difco) in paraffin oil (AdA) or lipoidal amine in complete Freund’s adjuvant (LA/CFA).

Objectives The present study examined whether or not the disease severity in these two models is related to the dose of the inciting agent.

Methods Adjuvant arthritis was induced in male Lewis rats (190 to 240 g) by administration of M. tuberculosis (AdA; 1.25, 2.5, 5, 10, 20 or 40 mg/ml) or lipoidal amine (LA/CFA; 4, 6.25, 12.5, 25, 50, 100, 200, or 400 mg/ml). The clinical evaluation included measurements of body weight and hind paw volume (by water plethysmography) on day 0 and then daily between days 8 to 16. At necropsy on day 16, inflammation (by histopathology) and bone integrity (bone mineral density [BMD] by DEXA; histopathology) were assessed in the tibiotarsal (hock) and femorotibial (knee) joints.

Results Induction of AdA followed a typical dose-response curve for most parameters, including body weight, loss of BMD (hock and knee), inflammation (paw volume and histopathology), and bone erosion (histopathology). In contrast, LA/CFA exhibited an “all-or-nothing” response with a threshold for activity of 50 mg/ml. The severity of arthritis (measured both clinically and morphologically) was comparable at day 16 for AdA engendered by > 1.25 mg/ml of mycobacteria and LA/CFA induced by > 50 mg/ml. These data indicate that AdA and LA/CFA exhibit different dose-response patterns.

Conclusion The practical implication of this finding is that AdA will be the better adjuvant arthritis model for investigating the biology and treatment of inflammatory joint disease over a range of lesion severities.