P. Mendes

Accuracy in Early Prediction of Prognosis of Patients with Septic Shock by Analysis of Simple Indices: Prospective Study

In 26 consecutive septic shock patients, we analyzed the clinical, hemodynamic, and metabolic data before and during volume infusion to test their circulatory reserve in response to fluid repletion. These patients were investigated to identify early variables that could predict outcome. There were 15 survivors (group A) and 11 nonsurvivors (group B). As a mean, group A patients were hemodynamically evaluated 2.3 h after onset of the sepsis syndrome, whereas group B patients underwent cardiac catheterization after a 12-h interval. At the initial evaluation, both groups demonstrated similarly decreased mean arterial pressure, mean heart rate, and mean cardiac filling pressure. Only group A patients evidenced elevated cardiac index (CI) (>4 L/min·m2) associated with low systemic vascular resistance index (<7400 dyne·sec/cm5·m2), which is generally recognized as hyperdynamic cardiac state. However, none of the initial cardiovascular variables could serve as a predictor for survival. Fluid challenge increased left ventricular preload from 6 to 12.4 and from 7.8 to 12.7 mm Hg in group A and group B, respectively. The increases were associated with significant increases in CI from 4.4 to 6.9 and from 3 to 3.8 L/min·m2. However, at the end of fluid challenge, only group A patients exhibited normal cardiac response, as evidenced by the change in left ventricular stroke work index (LVSWI) for a given increase in the pulmonary capillary wedge pressure (WP) that was referred to as left cardiac preload. Contrary to poor predictive value, on an individual basis, of the initial variables to assess the outcome of septic shock, the multivariate analysis that simultaneously accounted for four variables (LVSWI/WP, WBC count, Po2, and Hct) could separate with accuracy the two groups of patients.

Lung Fluid Dynamics and Supply Dependency of Oxygen Uptake During Experimental Endotoxic Shock and Volume Resuscitation

Background and MethodsWe studied the effect of volume resuscitation on lung fluid balance and systemic oxygen extraction during septic shock in eight anesthetized dogs. Sepsis was induced using a 2-hr continuous infusion of Escherichia coli endotoxin at 0.25 μg/min-kg. Relationships between oxygen uptake (Vo2) and oxygen supply (Do2) were performed acutely during step wise controlled decrements in cardiac output by progressive inflation of an intracardiac balloon. At each stage, Do2 and corresponding Vo2 were measured independently and the individual critical Do2 level was referred to as the point below which the relationship held. The slope of such a constructed relationship was defined as the maximal oxygen extraction ratio. Lung fluid balance was assessed by measurements of extravascular lung water. All values were studied at baseline, after endotoxin insult, and after reversing hypotension by a 10% dextran infusion. ResultsEndotoxin infusion led to a shock state that associated hypotension (from 135 to 63 mm Hg) with increases in blood lactate (from 0.53 to 3.9 mmol/L). The mean critical Do2 and maximal oxygen extraction ratio were significantly altered from 7.9 to 17.8 mL/min-kg and from 0.81 to 0.38, respectively. After reversing hypotension by 28 mL/kg colloid infusion, the critical Do2 (11.4 mL/min-kg) and maximal oxygen extraction ratio (0.48) were significantly improved. However, restoration of normal values required a state of fluid overload by further dextran infusion (8 mL/kg). At the end of the fluid challenge, extravascular lung water significantly increased from 6.4 to 17.4 mL/kg. ConclusionsThese data suggest that volume loading may reverse endotoxin-induced peripheral perfusion abnormalities. However, substantial pulmonary edema may occur, possibly jeopardizing the beneficial effects of fluid expansion. (Crit Care Med 1991; 19:955)

Pressure-Flow Relationships of the Pulmonary Circulation During Endotoxin Infusion in Intact Dogs

Background and MethodsWe aimed to characterize the effects of an endotoxin insult (Escherichia coli 0127:B8) on the relationships between pulmonary vascular pressure and flow in intact dogs. To achieve this goal, multipoint plots of total pressure gradient, arterial pressure gradient, and venous pressure gradient vs. flow were generated by graded inflation of a right atrial balloon, which was used to vary flow. The partitioning of the total pressure decrease across the pulmonary vasculature (total pressure gradient = pulmonary arterial pressure – pulmonary artery occlusion pressure [PAOP]) into gradients across pulmonary arterial (arterial pressure gradient = pulmonary arterial pressure – effective capillary pressure) and pulmonary venous (venous pressure gradient = effective capillary pressure – PAOP) regions was assessed by a waveform mathematical analysis of the pulmonary arterial pressure profile during arterial occlusion, with computation of both PAOP and effective pulmonary capillary pressures. Slopes and extrapolated pressure intercepts from linear regression fits to the pulmonary vascular pressure/flow plots were determined in seven dogs after a 2-hr endotoxic infusion interval and were compared with the corresponding values that characterized a similar group of sham-operated dogs. ResultsUnder normal conditions, the extrapolated pressure intercept for pulmonary arterial pressure gradient was virtually 0 mm Hg; for total pulmonary arterial pressure gradient and pulmonary venous pressure gradient, the mean extrapolated pressure intercepts were substantially positive: 2.4 ± 0.2 and 2.1 ± 0.3 mm Hg, respectively. Endotoxin infusion at 0.25 μg/kg/min significantly increased the pressure intercepts from 2.4 to 8.7 and from 2.1 to 8.3 mm Hg of total pressure gradient and venous pressure gradient vs. flow, respectively. This infusion produced a minor, nonsignificant change in the intercept of arterial pressure gradient vs. flow, whereas it increased its slope significantly (p < .05) from 0.036 to 0.081 mm Hg/mlVmin/kg. ConclusionsThese data suggest that endotoxins effects on vascular resistance are exerted at two different loci such that these effects are additive. These endotoxin-induced effects consisted of increased vascular resistance of the arterial segment and appearance of a Starling resistor at the venous side of the pulmonary circulation, which acted as the relevant back-pressure to flow.

Pathogenesis and management of acute pulmonary embolism

Pulmonary embolism can produce severe cardiopulmonary dysfunction characterized by pulmonary artery hypertension, right ventricular failure, and hypoxemia. The search for the source of a pulmonary embolus, by exploration of the veins of the lower limbs and the inferior vena cava should be systematically carried out in all cases of pulmonary embolus which are not immediately life-threatening to the patient. The treatment of deep vein thrombosis associated with pulmonary embolism with thrombolytic agents has been proposed and utilized for approximately 20 years. Although superior results have been claimed with thrombolytic agents, the use of this type of treatment remains limited to massive or sub-massive pulmonary embolism. Fibrinolytic agents with high specificity for fibrin in the thrombi and little systemic activation of the fibrinolytic system have been developed and tested in preliminary clinical trials of patients with acute pulmonary embolism. The largest published experience available has been with recombinant tissue plasminogen activator (rtPA). The acylated streptokinase-plasminogen complex (APSAC) and pro-urokinase also gave promising results. All these agents were accompanied by unexpectedly high incidence of systemic activation of the fibrinolytic system and by hemorrhagic complications with frequencies similar to those that follows the use of first generation products (urokinase and streptokinase). Hence, their superior clinical efficacy must be clearly proven before they are substituted for a more widely available and less expensive drug, such as streptokinase.