We read with great interest the article by Ruiz-del-Arbol and colleagues.1 They report a lower cardiac index and a higher systemic vascular resistance in patients with cirrhosis and spontaneous bacterial peritonitis (SBP) who developed renal failure when compared with SBP patients who did not develop renal failure. This difference in cardiovascular function was accompanied by a significant increase in plasma norepinephrine (NE) and plasma renin activity (PRA). A combination of cirrhotic and septic cardiomyopathy was proposed as one of the factors leading to circulatory dysfunction, where tumor necrosis factor may play an important pathogenic role.
A decrease in left ventricular function using the Sarnoff curves2 has been found in cirrhotic patients exposed to sepsis.3 Because the systemic inflammatory response syndrome could be found more often in patients with renal failure, the decreased left ventricular stroke work is quite logical. The argument against this concept is the increase in systemic vascular resistance (SVR) found in patients with renal failure; this is in contrast to findings in septic patients with cirrhosis, where an extreme decrease in SVR could be found.4 Increases in SVR, NE, and PRA, as well as a decrease in renal function, are common findings in the setting of elevated intra-abdominal pressures,5 which may have been relevant in the patients who were reported to have renal failure.
In addition, hypovolemia induces an increase in SVR, NE, and PRA, decreases stroke volume, and leads to prerenal failure. Hypovolemia must have played a relevant role in the patients reported, because the right atrial pressure (3 + 3 mm Hg) as well as the pulmonary capillary wedge pressure (PCWP; 5.7 + 4 mm Hg), an indirect parameter of left ventricular filling pressure, were reduced. The relevance of hypovolemia for prognosis in sepsis was recently shown by Rivers and collegues.6 With an early, goal-directed hemodynamic therapy—consisting of infusion of cristalloids and/or colloids to normalize central venous pressure to above 8 mm Hg; application of vasoconstrictors to increase mean arterial pressure above 90 mm Hg; transfusion of red cells until hematocrit is above 30%; and administration of inotropic agents until a central venous oxygen saturation of above 80% is achieved—mortality in septic patients could be significantly reduced.
More confusing is the fact that examination of right atrial pressure and PCWP revealed that patients who were not developing renal failure were also in a hypovolemic state. This might be caused by the high variability in the interpretation of PCWP curves. Trottier and Raylor7 recently demonstrated high intra- and interobserver variabilities in the measurement of PCWP, even in measurements made by physician “experts” and physicians who commonly use the pulmonary artery catheter. Therefore, because of the small numbers of patients studied, hypovolemia with comparable ventricular compliance cannot be excluded totally as the explanation for the hemodynamic findings. In addition, an elevation of intra-abdominal pressure may increase intrathoracic pressure; therefore, measured atrial pressures will not accurately reflect transmural pressures.
In conclusion, the interesting findings of Ruiz-del-Arbol and colleagues may be caused by a combination of septic and cirrhotic cardiomyopathy with hypovolemia combined with systemic inflammatory response syndrome (SIRS) and increased intra-abdominal pressure. The finding of high mortality is a further argument for an early, goal-directed therapy as proposed by Rivers and colleagues6 in patients with spontaneous bacterial peritonitis.