The treatment of polymyalgia rheumatica (PMR) and temporal arteritis (TA) is still controversial. To assess the influence on the course of these diseases of the clinical symptoms at initial presentation and of the starting dosage of corticosteroid (CS) treatment the data for 210 patients, who were diagnosed as having PMR or TA from 1976 to 1986 and were followed up closely, were reviewed. One hundred and thirty two patients were diagnosed as having 'clinically pure' PMR; prednisone starting doses of over 15 mg daily provided more CS related adverse effects without any advantage. The mean duration of treatment was 25.7 months. Nine patients later developed symptoms of TA, and there were no predictive features for this. None experienced visual or neurological complications. Seventy eight patients were diagnosed as having clinical TA. Twenty five patients treated with low starting doses of prednisone, ranging from 10 to 20 mg/d (mean 16.2 mg/d), developed less CS related adverse effects and did not have more visual or neurological complications than 53 patients treated with higher doses. The mean duration of treatment was 30.9 months. Fifteen patients experienced visual or neurological complications and men (10/30) developed these complications more frequently than women (5/48) (p less than 0.02). These results suggest that (a) clinically pure PMR is a benign disease requiring low doses of CS treatment; (b) low doses of CS seem an adequate treatment for most cases of TA; (c) a worse prognosis seems attached to the male sex in TA.
