Surgical sympathectomy can relieve symptoms of angina in patients with refractory angina. However, in these high-risk patients this thoracic surgery may result in significant morbidity and mortality rates. Similar sympathetic blockade can now be produced with high thoracic epidural analgesia (HTEA). From September 1995 to August 1996, we treated 10 consecutive patients with HTEA. These eight men and two women, aged 58 ± 5 years, with extensive three-vessel coronary disease and ejection fractions of 40% ± 5%, had New York Heart Association (NYHA) class IV angina despite medical therapy, including nitrates, β-blockade, calcium channel blockade, and narcotics. HTEA was performed at the T1 through T4 levels with a catheter placed either percutaneously or surgically, with radiographic confirmation of catheter placement with an epidurogram or computed tomography scan. Bupivacaine (0.25% to 0.5%), an amide local anesthetic, was given as a bolus through the epidural catheter and then maintained either as a continuous infusion or an intermittent rebolus. The epidural catheter remained in place for 7 days in four patients, 14 days in three patients, and ≥90 days in three patients. Before consideration for HTEA, each patient was deemed unsuitable for or refused coronary bypass surgery and percutaneous coronary angioplasty and had NYHA class IV symptoms of angina. Seven of 10 patients required intravenous nitroglycerin and heparin and were unable to be discharged from the intensive care unit because of anginal symptoms. Two of these seven patients also required an intraaortic balloon pump for symptom control. After HTEA, all 10 patients had improved symptoms, with five patients improving to NYHA class II symptoms and five improving to NYHA class III. All seven patients receiving intravenous nitroglycerin, heparin, or intraaortic balloon pump support had these modalities discontinued. Six of these seven patients were subsequently discharged from the hospital. One patient died from a non-HTEA related cause. There were no HTEA-related deaths. There were three catheter-related complications necessitating catheter removal during 12 months of HTEA use. Local infection developed in one patient, one had catheter occlusion caused by fibrosis, and one patient had chronic back pain exacerbation from a paraspinous muscle spasm. No patient had a myocardial infarction or a significant arrhythmia. In patients with otherwise intractable angina pectoris, HTEA is an effective modality that produces symptomatic relief of angina pectoris and allows increased activity level. (Am Heart J 1997; 134:648-55.)