In 8 dogs with acute cardiac tamponade, we studied the hemodynamic effects of three modes of positive-pressure ventilation: intermittent positive-pressure ventilation (IPPV) at a ventilatory frequency (f) of 14/min, volume-controlled high frequency positive-pressure ventilation (HFPPV) at f of 60/min without positive end-expiratory pressure (PEEP), and HFPPV at f of 60/min with PEEP. The study was carried out at normocarbia with light thiopental anesthesia at a standardized level of cardiac tam-ponade (about 60% reduction in cardiac output). Ventilator gas outputs were 7,380 ml/min with IPPV, 5,700 ml/min with HFPPV, and 8,800 ml/min with HFPPV/PEEP. Because of its small functional dead space, the system J HFPPV device produced normocarbia efficiently. Without cardiac tamponade, cardiac output and stroke volume were significantly higher with HFPPV than with IPPV or HFPPV/PEEP. When tamponade was present, however, these differences were not found. These findings indicate that in cardiac tamponade, effects induced by ventilatory flow/pressure patterns are overshadowed by the more pronounced cardiac impairment caused by pericardial pressures. The results suggest that sudden hemodynamic col-lapse in acute cardiac tamponade patients may not be caused by the institution of positive-pressure ventilation per se. If mechanical ventilation is required in patients with acute cardiac tamponade, either IPPV, HFPPV, or HFPPV/PEEP may be used to produce normocarbia, and PEEP may be applied if needed to increase arterial oxygenation.