Very recently Durez et al convincingly demonstrated that tumour necrosis factor blockade is, in comparison with glucocorticoid pulse therapy, promising, improving not only clinical measures of disease activity but also biological inflammatory indices in a subset of patients with severe refractory rheumatoid arthritis (RA).¹ We would like to comment on the small effect that was seen with intravenous pulse methylprednisolone (MP) in this study.

We recently suggested that glucocorticoid pulse therapy should be defined as treatment with “⩾250 mg prednisone equivalent per day for one or a few days”.² According to this proposed nomenclature pulse therapy was indeed applied. However, the data obtained for this subset of patients with severe refractory RA show only small, if any, effects on the joint scores, Health Assessment Questionnaire, morning stiffness, or serum C reactive protein (CRP). These results are surprising at first glance. But what can we conclude by trying to interpret these interesting data?

Firstly, a single glucocorticoid application even at a high dose may have a strong but only short lived effect. We assume that 1 g MP produces 100% saturation of cytosolic glucocorticoid receptors with 100% of genomic effects exerted.^2,3 …