We reviewed a series of 75 pericardiocenteses, using an atraumatic indwelling catheter, comprising all those performed in the lntensive Care Unit from Jan. 1981 to Dec. 1984 and assessed the method, benefits and risks of pericardiocentesis. Pericardial fluid was obtained in 92 of the patients. Underlying diseases were as follow: neoplastic disease (52.0%), tuberculosis (21.4%), pyopericardium (8.0%), hemopericardium (5.3%), congestive heart failure (4.0%), primary idiopathic pericarditis (4.0%), rheumatic disease, chronic renal failure, postpericardiotomy syndrome, in order. Specific etiologic diagnosis of neoplasm, tuberculosis and pyoperioardium were made in 31 patients (41.3%) by the examination of the pericardial fluid. Chemotherapeutic agents were instillated to pericardial cavity through indwelling pericardial catheter in 24 patients. On the basis of the operative results, the patients were separated into three groups for comparison: successful pericardiocentesis (88.0%)-successful and sufficient pericardiocentesis (64,0%), successful and insufficient pericardiocentesis (24.0%); failed pericardiocentesis (8.0%); complicated pericardiocentesis (4.0%). Complications consisted of one death after right ventricular punture, one right ventricular punture without adverse sequelae, and one induced hemopericardium. Pericardiocentesis using indwelling catheter permits continuous pericardial drainage, obviating repeated aspirations; provides a convenient route for intrapericardial instillation of chemotherapeutic agents; and enables one to await the relults of diagnostic studies without sujecting a patient to surgery.