Our approach to the clinical management of uremic and dialysis-associated pericarditis has been presented previously and is outlined in Figure 1. In hemodynamically stable patients with no effusion and in those with small to medium effusions, we recommend initial therapy with intensified dialysis. Close monitoring, perhaps every third day, with echocardiography should be carried out. If pericardial effusion progressively increases or if a large pericardial effusion fails to resolve after 7 to 10 days of intensive dialysis, the pericardial effusion may be drained by subxiphoid pericardiotomy or by pericardiectomy. Similarly, if hemodynamic evidence of cardiac pretamponade or tamponade appears, surgical drainage also should be carried out. If the echocardiogram is inadequate for interpretation but tamponade physiology is present, we recommend confirmation by cardiac catheterization before surgical drainage is attempted, recognizing that there may be circumstances such as left ventricular failure and pulmonary hypertension that may complicate the interpretation of the catheterization data. The type of invasive pericardial procedure chosen is determined by local experience. As stated, we prefer not to perform pericardiocentesis before surgery unless tamponade-induced hypotension is so severe that an adequate blood pressure cannot be maintained by means of plasma volume expansion. Under these circumstances, we prefer that pericardiocentesis be performed in the operating room immediately before the induction of anesthesia for the definitive surgical procedure. Although pericardiectomy is a definitive procedure for pericarditis with effusion in the uremic patient, the procedure has substantial morbidity. The results of subxiphoid pericardiotomy are encouraging, and it is clear that it can be carried out safely in patients who are debilitated or who are at increased risk from general anesthesia and major surgery.(ABSTRACT TRUNCATED AT 250 WORDS)