Reply:

Abstract

We thank Vlachogiannakos et al. for their interest in our recent article.1 One of the main questions raised in their letter is whether bacterial DNA detection is a more accurate marker of bacterial translocation than endotoxin. Considering that the detection of endotoxin, which is almost exclusively present in gram-negative microorganisms, or its soluble receptor lipopolysaccharide binding protein may severely underestimate the translocation of gram-positive microorganisms, we have proposed bacterial DNA detection as a broad-range marker of bacterial translocation in the setting of cirrhosis. In our experience, bacterial DNA is present in approximately 35% of patients with cirrhosis and ascites, and similar findings have recently been reported by Angeli et al.2 with the same methodology. Our results show that baseline portal pressures are similar for patients with ascites independently of the presence of bacterial DNA. This finding is in line with previous clinical studies showing that elevated serum levels of lipopolysaccharide binding protein, which is a more sensitive marker of endotoxemia than endotoxin itself because of its longer biological half-life, are not associated with a greater degree of portal hypertension.3 In our study,1 which included a large series of uninfected patients with cirrhosis, we did not find any correlation between the bacterial DNA concentration and systemic hemodynamic parameters (mean arterial pressure and systemic vascular resistance). This does not entirely rule out a role for bacterial translocation because other bacterial products besides bacterial DNA can hypothetically play a role in the inflammatory response. Indeed, monocytes from patients with cirrhosis seem to have an enhanced response to bacterial products. Therefore, small or even undetectable amounts of bacterial products may cause an enhanced release of cytokines by monocytes and thus further impair systemic hemodynamics. We cannot offer a clear explanation for the discrepancies between our results and Vlachogiannakos et al.'s results with respect to the detection of bacterial DNA. Methodological variations are likely responsible for the differences because we would expect to find bacterial DNA in 30% to 35% of the baseline samples from Vlachogiannakos et al.'s patients (they all had ascites, and some had measurable levels of endotoxin). The inability to detect bacterial DNA after a 28-day course of rifaximin does not seem unexpected, although we do not have our own data for this. Molecular techniques are in continuous development. The detection of small amounts of bacterial DNA in blood samples from patients is always challenging, and we must take advantage of more powerful methodologies such as pyrosequencing for studying different issues that still remain controversial with respect to bacterial translocation in patients with cirrhosis. Pablo Bellot M.D.*, Jose Such M.D. , Jaime Bosch M.D.*, * Hepatic Hemodynamics Laboratory, Hospital Clinic, Barcelona, Spain, Liver Unit, University General Hospital, Alicante, Spain.