That would be an interesting finding, and you should try to confirm it first, before drawing conclusions (sure about the group-assignment of the samples, about the unit, do you have any positive controls (spike), any negative controls?) .  LPS would be expected to be eventually high in the portal venue, and the concentrations you mention are higher compared to other publications, as for example: http://www.biomedcentral.com/content/pdf/1471-230X-11-107.pdf (Benten, et al., 2011: The transhepatic endotoxin gradient is present despite liver cirrhosis and is attenuated after transjugular portosystemic shunt (TIPS). BMC Gastroenterology 11:107.) 

after

transjugular portosystemic shunt (TIPS) 

transjugular portosystemic shunt (TIPS). 

I have few citations for you that confirm your observation:

http://www.sciencedirect.com/science/article/pii/S0021915011002073

(Atherosclerosis. 2011 May;216(1):1-6. Diet, commensals and the intestine as sources of pathogen-associated molecular patterns in atherosclerosis, type 2 diabetes and non-alcoholic fatty liver disease. Erridge C1.)

http://www.ncbi.nlm.nih.gov/pubmed/12073822

(Plasma endotoxin level of healthy donors. Nádházi Z, Takáts A, Offenmüller K, Bertók L. Acta Microbiol Immunol Hung. 2002;49(1):151-7

Thanks Rafal for this. We did come across that LPS levels could be high in type 2 diabetes and individuals with atherosclerosis. So when we chose healthy individuals for our study as controls, we excluded everyone with dyslipidaemia, diabetes, hypertension, obesity, high waist circumference etc... also because of the possible effects of the diet, all subjects were fasting. 

The levels detected in healthy individuals were >1000ng/ml. 

Yes, the level of LPS seems pretty high, especially when you look at this paper http://www.ncbi.nlm.nih.gov/pubmed/10515819 and normal donor have ~6pg/ml of LPS in plasma.