Potential conflict of interest: Nothing to report.
Article first published online: 27 FEB 2008
Copyright © 2008 American Association for the Study of Liver Diseases
Volume 47, Issue 3, page 1100, March 2008
How to Cite
Thabut, D., Tazi, K., Bonnefont-Rousselot, D., Moreau, R. and Lebrec, D. (2008), Reply:. Hepatology, 47: 1100. doi: 10.1002/hep.22234
- Issue published online: 27 FEB 2008
- Article first published online: 27 FEB 2008
We thank Dr. Fujita and colleagues for their useful comments about our article. Indeed, we performed two different types of experiments. We first studied the neutralization of lipopolysaccharide (LPS) by high-density lipoprotein (HDL). The aim of that first experiment was to compare LPS tissue distribution and the effects of HDL on this tissue distribution independently of tumor necrosis factor-alpha production. We used FITC-LPS at 0.17 mg/kg, that is, the same dose as previously used for the study of LPS clearance in normal rat.1 In the second protocol, LPS was used at a dose which may induce tumor necrosis factor-alpha hyperproduction in cirrhotic rats, that is 0.5 mg/kg.
Dr. Fujita asked why serum HDL levels increased during invasive experiments under anesthesia. In control rats, HDL levels did not significantly rise after the experiment (0.46 versus <0.44 g/L). In our cirrhotic rats, plasma HDL levels rose from 0.25 ± 0.09 to 0.47 ± 0.02 g/L (mean, standard error of the mean), but this value was not statistically significant (P = 0.11).
The authors wonder why we administered reconstituted HDL (rHDL) before LPS challenge. This is a pertinent issue. As underlined in the discussion of our article, the best way to study the potential curative effects of rHDL would be to administer rHDL after LPS injection, as Fujita et al. previously did with ApoA1 intraperitoneal administration.2 However, we aimed to study the possible preventive action of rHDL. Indeed, we know well that in patients with cirrhosis without infection, serum bacterial DNA can be found in ascitic fluid,3 suggesting a “preinfected state” and a potential beneficial effect of rHDL even before infection.
We agree with the authors that testing different doses of HDL is necessary to determine the optimal dosage requested in cirrhotic rats. However, we chose the dosage of 80 mg/kg because it was that used in previous studies after LPS exposure in rats infected with E. coli–derived LPS, and that Gram-negative infections are the most common infections in patients with cirrhosis.
Finally, the authors express their concerning regarding the administration of rHDL in patients with primary biliary cirrhosis. They refer to the article by Longo et al. where patients with primary biliary cirrhosis were prospectively followed during several years. In this article, in fact, the patients with moderate elevations of serum cholesterol levels (between 200 and 250 mg/dL) were significantly more likely to suffer from cardiovascular events than patients with high hypercholesterolemia, who had more than 250 mg/dL.4 One possible explanation could be that the former patients, who do not have severe cholestasis, are exposed to the same cardiovascular risk as their counterparts and have hypercholesterolemia of familial and nutritional origin. On the contrary, the latter patients have severe cholestasis due to biliary obstruction, which is characterized by a different molecular pattern, with higher HDL levels and modified LDL composition. Those patients are probably protected against atherosclerosis. For that reason, we do believe that increasing HDL levels in the subset of patients with primary biliary cirrhosis could have beneficial effects on infection and maybe protect them from cardiovascular diseases.
Dominique Thabut* , Khalid Tazi* , Dominique Bonnefont-Rousselot§ ¶, Richard Moreau* , Didier Lebrec* , * INSERM, U773, Centre de Recherche Biomédicale Bichat-Beaujon CRB3, Hôpital Beaujon, Clichy, France, Service d'Hépato-Gastroenterologie, Hôpital de la Pitié-Salpêtrière (AP-HP), Paris, France, Université Denis Diderot-Paris 7, site Bichat, Paris, France, § Laboratoire des Lipides, Hôpital de la Pitié-Salpêtrière (AP-HP), Paris, France, ¶ Laboratoire de Biochimie Métabolique (EA 3617), Faculté de Pharmacie, Paris, France.