Chronic shoulder pain is a common and disabling symptom in patients with rheumatoid arthritis (RA). It has been previously shown that a suprascapular nerve block (SSNB) using the standard mixture of bupivacaine and adrenaline (Ba) plus methylprednisolone (P), which is routinely used in pain clinics, results in a considerable improvement in pain relief and range of movement compared with conventional intra-articular steroid injections in such patients. A double blind study was carried out in 29 patients (58 shoulders) with RA to compare SSNB induced with Ba alone with that induced using the conventional mixture of Ba plus P. Highly significant improvements were noted in measures of pain, stiffness, and range of most movements for both treatments (up to three months) compared with baseline. Results favoured Ba alone; the differences between the two treatments reached statistical significance for stiffness (at 12 weeks) and active abduction (at one week). It is concluded that the addition of P to the SSNB mixture confers no benefit in these patients.