Elevated endotoxin levels in non-alcoholic fatty liver disease

Abstract

Background: Emerging data indicate that gut-derived endotoxin may contribute to low-grade systemic inflammation in insulin resistant states. This study aimed to examine the importance of serum endotoxin and inflammatory markers in non-alcoholic fatty liver disease (NAFLD) patients, with and without type 2 diabetes mellitus (T2DM), and to explore the effect of treatment with a lipase inhibitor, Orlistat, on their inflammatory status.

Methods: Fasted serum from 155 patients with biopsy proven NAFLD and 23 control subjects were analysed for endotoxin, soluble CD14 (sCD14), soluble tumour necrosis factor receptor II (sTNFRII) and various metabolic parameters. A subgroup of NAFLD patients were re-assessed 6 and 12 months after treatment with diet alone (n = 6) or diet plus Orlistat (n = 8).

Results: Endotoxin levels were significantly higher in patients with NAFLD compared with controls (NAFLD: 10.6(7.8, 14.8) EU/mL; controls: 3.9(3.2, 5.2) EU/mL, p < 0.001); NAFLD alone produced comparable endotoxin levels to T2DM (NAFLD: T2DM: 10.6(5.6, 14.2) EU/mL; non-diabetic: 10.6(8.5, 15.2) EU/mL), whilst a significant correlation between insulin resistance and serum endotoxin was observed (r = 0.27, p = 0.008). Both sCD14 (p < 0.01) and sTNFRII (p < 0.001) increased with severity of fibrosis. A positive correlation was also noted between sTNFRII and sCD14 in the NAFLD subjects (r = 0.29, p = 0.004).Sub-cohort treatment with Orlistat in patients with NAFLD showed significant decreases in ALT (p = 0.006), weight (p = 0.005) and endotoxin (p = 0.004) compared with the NAFLD, non-Orlistat treated control cohort at 6 and 12 months post therapy, respectively.

Conclusions: Endotoxin levels were considerably increased in NAFLD patients, with marked increases noted in early stage fibrosis compared with controls. These results suggest elevated endotoxin may serve as an early indicator of potential liver damage, perhaps negating the need for invasive liver biopsy. As endotoxin may promote insulin resistance and inflammation, interventions aimed at reducing endotoxin levels in NAFLD patients may prove beneficial in reducing inflammatory burden.

Figure 1

Levels of endotoxin, sCD14 and TNFRII in NAFLD and NASH subjects compared with controls. The figures show the mean log endotoxin levels (A), mean log sCD14 levels (B) and mean log TNFRII levels (C) in control, NAFLD and NASH subjects for the whole cohort. The last figure (D) shows the mean log of TNFRII levels in NASH versus cirrhosis (* p < 0.05, ** p < 0.01, ***p < 0.001). Endotoxin or sCD14 showed no significant differences between NASH and cirrhosis in these subjects (data not shown).

Figure 2

Serum levels of endotoxin (A), sCD14 (B) and sTNFRII (C) in 23 healthy controls and 155 patients with NAFLD. The horizontal lines represent the median of the data. Statistical analysis compared the log mean serum levels of the inflammatory markers at each fibrosis stage of liver disease against the log mean serum levels of healthy controls subjects, (p < 0.01, p < 0.001).

Figure 3

Correlations between log endotoxin and log fasting insulin, log triglycerides, HOMA-IR, and between sTNFRII and sCD14. The figures show Pearson correlations between log endotoxin (EU/mL) and log fasting insulin (μU/mL) in the whole cohort (A) log triglycerides (mmol/L) in patients with NAFLD (B) log serum levels of endotoxin and HOMA-IR in the whole cohort (C). The lines of best fit are also shown: a) r = 0.31, p = 0.02, b) r = 0.51, p < 0.0001 c) r = 0.27, p = 0.008. A Pearson correlation between log sTNFRII (ng/mL) and sCD14 (μg/mL) in patients with NAFLD is also shown (D). The line of best fit is: d) r = 0.29, p = 0.004.