Quinidine therapy is one of the most common causes of the acquired long QT syndrome and torsade de pointes. In reviewing clinical data in 24 patients with the quinidine-associated long QT syndrome, 20 of whom had torsade de pointes, we have delineated several heretofore unreported or underemphasized features. (1) This adverse drug reaction occurred either in patients who were being treated for frequent nonsustained ventricular arrhythmias or for atrial fibrillation or flutter. (2) In patients being treated for atrial fibrillation, torsade de pointes occurred only after conversion to sinus rhythm. (3) Although most patients developed the syndrome within days of starting quinidine, four had torsade de pointes during long-term quinidine therapy, usually in association with hypokalemia. (4) Because of the large experience with this entity at our institution, we have been able to estimate the risk as at least 1.5% per year. (5) Twenty of the 24 patients had at least one major, easily identifiable, associated risk factor including serum potassium below 3.5 mEq/L (four); serum potassium between 3.5 and 3.9 mEq/L (nine); high-grade atrioventricular block (four); and marked underlying, (unrecognized) QT prolongation (two). Plasma quinidine concentrations were low, being at or below the lower limit of the therapeutic range in half of patients. The ECG features typically included absence of marked QRS widening, marked QT prolongation (by definition), and a stereotypic series of cycle length changes just prior to the onset of torsade de pointes. Torsade de pointes started after the T wave of a markedly prolonged QT interval that followed a cycle that had been markedly prolonged (usually by a post ectopic pause).(ABSTRACT TRUNCATED AT 250 WORDS)