Left ventricular catheterization was carried out in 40 patients with acute myocardial infarction. Left ventricular end-diastolic pressure (LVEDP) was elevated in 85% of the patients studied. In 14 patients with apparently uncomplicated infarcts, LVEDP averaged 15 mm Hg, and cardiac index (2.98 liter/min/m(2)), stroke volume (38.3 ml/m(2)), and stroke work (49.2 g-m/m(2)) were within normal limits. In 12 patients with clinical signs of left ventricular failure, LVEDP averaged 29.9 mm Hg, cardiac index was at the lower limit of normal (2.79 liter/min/m(2)), but stroke volume (31.6 ml/m(2)) and stroke work (37.3 g-m/m(2)) were reduced. In 14 patients with clinical signs of shock, LVEDP averaged significantly lower than in the heart failure group (21.1 mm Hg), but cardiac index (1.59 liter/min/m(2)), stroke volume (16.5 ml/m(2)), and stroke work (11.1 g-m/m(2)) were markedly reduced. A large presystolic atrial "kick" (average amplitude 9.5 mm Hg) was an important factor in the high LVEDP in the patients with heart failure but not in those with shock. The first derivative of left ventricular pressure was significantly lower in shock than in the nonshock group. Although right atrial pressure (RAP) and LVEDP were significantly correlated (r = 0.49), wide discrepancies in individual patients rendered the RAP an unreliable indicator of the magnitude of left ventricular filling pressure. THESE DATA SHOW THE FOLLOWING: (a) LVEDP is usually elevated in acute myocardial infarction, even in absence of clinical heart failure; (b) cardiac output apparently is supported by increased LVEDP and compensatory tachycardia; (c) in patients with shock, left ventricular function usually is markedly impaired, but inadequate compensatory cardiac dilatation or tachycardia could contribute to the reduced cardiac output in some individuals; (d) lower LVEDP in shock than in heart failure may represent differences in left ventricular compliance.