Epinephrine is often added to local anesthetic solutions to minimize and slow the systemic absorption of local anesthetics, and thus reduce the possibility of adverse effects of these drugs. In an earlier study we found that the injection of 5 mg/kg of lidocaine via the paravertebral route depressed myocardial contractility by up to 30%, with practically no changes in heart rate or blood pressure. In the present study we investigated whether these alterations are due to systemic absorption of the local anesthetic, and whether such absorption can be minimized by adding epinephrine to the local anesthetic solution. A prospective, blind, and randomized study was made of 50 patients subjected to lung resection surgery. The subjects were divided into two groups: Lid group (5 mg/kg bolus dose of lidocaine in the thoracic paravertebral space) and Lid+E group (addition of 5 mcg/mL of epinephrine to the local anesthetic). The anesthetic solution was administered through a paravertebral catheter ipsilateral to the operative side. In addition to routine hemodynamic monitoring (heart rate and radial artery blood pressure), an aortic transpulmonary thermodilution catheter was inserted into the femoral artery for recording of the following variables: cardiac index, cardiac function index, maximum pressure derivative, global end diastolic volume, and intrathoracic total blood volume index. Data collection was carried out immediately before administration of the anesthetic solution and 15, 30, and 45 minutes after administration. Measurements were made of the plasma lidocaine levels at those same postparavertebral injection time points. Prior to paravertebral dosing there were no differences in terms of the hemodynamic variables studied. However, 15 minutes after dosing in the Lid+E group, lesser reductions in contractility, cardiac function index, and cardiac index were recorded, compared with the Lid group, with a significant reduction in cardiac filling volumes. Blood lidocaine levels were 53% and 34% lower in Lid+E group, as recorded 15 and 30 minutes after injection. The patients who, 15 minutes after paravertebral injection, had blood lidocaine levels greater than 3 mcg/mL (independently of the type of anesthetic solution used) had a significant reduction in mean blood pressure, cardiac function index, cardiac index, and maximum pressure derivative, compared with the patients with lower blood lidocaine concentrations. Addition of epinephrine to lidocaine when performing thoracic paravertebral block, attenuates the cardiodepressive effects associated with the systemic absorption of lidocaine and also, as a result of the beta-adrenergic consequences of epinephrine, systemic absorption from the paravertebral space.