• Potential conflict of interest: Nothing to report.


We thank Dr. Fujita for his comments on our article examining the ability of probiotics to prevent the breakdown in intestinal barrier function, bacterial translocation, and hepatic damage in mice with sepsis induced by coinjection of galactosamine (GalN) and lipopolysaccharide (LPS). In his letter, Dr. Fujita refers to previous studies1 showing that mice receiving GalN and LPS go into irreversible septic shock and rapid death within 5-9 hours of injection, and he questions whether this is a suitable model to study the effectiveness of probiotics. In those studies, mice received a very high dose of GalN (600 mg/kg body weight) together with a lower dose of LPS (1-10 μg/kg).1 In our studies, we used a much lower dose of GalN (360 mg/kg) combined with a higher dose of LPS (40 μg/kg).2 These concentrations were chosen on the basis of a dose-response curve that we established prior to beginning these studies. At the concentrations used in our study (360 mg/kg of GalN and 40 μg/kg of LPS), 80% of the animals survived through 12 hours with 60% surviving through 96 hours (Fig. 1). Because we were interested in whether an early increase in permeability occurred before the onset of sepsis, and whether probiotics would be effective in preventing the onset of gut permeability, we did not carry out long-term survival studies. However, in short-term survival studies (12 hours), all animals receiving probiotics survived (100%) compared with 80% of animals not receiving probiotics (n = 16).

Figure 1.

Survival of 129 Sv/ev mice after injection of lipopolysaccharide (40 μg/kg) with increasing doses of D-galactosamine (GalN). Injection of LPS and the highest dose of GalN (720 mg/kg) resulted in the death of 100% of the mice by 12 hrs, while injection of LPS and the lowest doses (36 and 180 mg/kg) resulted in 100% survival up to 96 hrs. Injection with LPS and 360 mg/kg GalN resulted in 50% survival up to 96 hrs.

Regarding the second point of Dr. Fujita's letter, we agree that the relationship between increased permeability and liver damage in sepsis remains to be clarified. However, there have been numerous clinical findings demonstrating that a dysfunctional small intestine has deleterious effects on the liver, implying the existence of a gut-liver axis.3 In addition, a breakdown in barrier function of the intestine may be part of a more generalized tissue dysfunction, with loss of epithelial barrier in other organs, including the liver, kidney, and lung.4 It is clear that although the exact pathogenesis of sepsis and the trigger of inflammation remains a matter of debate, the inflammatory response is a key event and dysregulation of this response may be pivotal.5 An interesting study examining the relationship between nonsteroidal anti-inflammatory drugs (NSAIDs), intestinal permeability, serum levels of inflammatory cytokines, and hepatic gene expression has shown that only those NSAIDs that caused a breakdown in gut permeability induced an acute phase response and increased serum levels of LPS.6 Further, an association was seen between the NSAIDs that eventually induced a systemic inflammatory response with early alterations in hepatic gene expression.6 This data would support the concept that increases in intestinal permeability results in leakage of bacteria and/or bacterial products (including LPS and bacterial DNA) into the circulation, which provokes altered hepatic gene expression and a subsequent systemic inflammatory response.3, 7 Finally, it is possible that the benefits attributed to probiotic bacteria may involve both effects on gut permeability and modulation of innate and systemic immune responses.8, 9

Karen Madsen*, * University of Alberta, Faculty of Medicine, Edmonton, Alberta, Canada.