Shah P, Griffith S, Shadforth M, Fisher J, Dawes P, Poulton K, Thomson W, Ollier WER,
J Rheumatol. 2004;31( 10):1903-5.
OBJECTIVE:. To investigate whether features associated with severe rheumatoid arthritis
(RA) are predictive of adverse drug reactions (ADR) to gold salts, independent of
HLA-DR3 status. METHODS: A cohort of patients with RA (n = 41) who developed thrombocytopenia
(platelets < 100 10(6)/l) or proteinuria (> 1.0 g/24 h) upon treatment with gold sodium
thiomalate was identified from patient records and matched for age, sex, and disease
duration with 41 RA controls treated with gold without development of ADR. A second
group of 161 random RA patients that had received gold therapy for at least as long
without development of an ADR was also compared. All patients were typed for HLA-DRB1,
and the presence of rheumatoid factor (RF), antinuclear antibodies (ANA), and nodules
before initiation of therapy was recorded. Association of clinical or genetic factors
with ADR was investigated using the McNemar test and logistic regression analysis.
RESULTS: Patients with ADR were more likely to have nodular disease than their matched
controls (51.3% vs 25.6%; odds ratio, OR = 3.0, p = 0.02) and more likely to be HLA-DR3
positive (41.2% vs 17.6%; OR = 3.0, p = 0.045). No difference between the groups was
found for RF or ANA. Nodular disease was associated with development of ADR independently
of HLA-DR3, although a combination of both factors significantly increased the likelihood
of an ADR. CONCLUSION: Our data suggest that nodular disease may be a predictor of
gold-induced ADR independent of HLA-DR3.