OP22 A significant reduction in "pro-inflammatory" CD14lo/CD16hi monocytes in paracetamol-induced acute liver failure: association with outcome and CCR2 expression. (16th November 2010)

• Record Type: Journal Article
• Authors: Abeles, D; Antoniades, C G; Longhi, M S; Taylor, N J; Shawcross, D L; Auzinger, G; Bernal, W; Heaton, N; Wendo, J A; Vergani, D
• Is Part Of: Gut. Volume 59(2010)Supplement 2
• Page Start: A9
• Page End: A10
• Publication Date: 2010-11-16
• Subjects: Gastroenterology -- Periodicals; 616.33
• DOI: 10.1136/gut.2010.223362.22
• Languages: English
• ISSNs: 0017-5749

Abstract

Abstract : Introduction: Acute liver failure (ALF) shares many clinical and immunological features with severe sepsis. In severe sepsis the relative proportion of "pro-inflammatory" CD14lo/CD16hi monocytes (MØ) is reduced, possibly due to recruitment into inflamed tissue. No data exist on monocyte subsets in paracetamol-induced ALF (PALF). Animal models of ALF suggest the importance of CC chemokine receptor (CCR) 2 mediated MØ recruitment and in human PALF high circulating levels and hepatic expression of CC chemokine ligand 2 have been observed. Aim: Investigate the relative composition of "pro-inflammatory" CD14hi/CD16lo, "intermediate" CD14hi/CD16hi and "classical" CD14lo/CD16hi MØ subsets and the association of CCR2 expression, in patients with PALF. Method: Blood was taken from five healthy controls and 20 patients with PALF within 24–48 hours of admission. Monoclonal antibodies against CD14, CD16 and CCR2 were used to determine MØ subsets and CCR expression. Results are expressed as the percentage proportion of total MØ and mean fluorescence intensity of CCR2 expression. Results: Compared to healthy controls, a lower proportion of CD14lo/CD16hi MØ was observed in PALF (2% vs 10%, p=0.01), CD14hi/CD16lo MØ tended to be more highly represented (95% vs 82%, p<0.08) but there were no differences in the proportion of CD14hi/CD16hi MØ (3% vs 8%). In PALF, CCR2 expression was up-regulated on CD14hi/CD16lo and CD14hi/CD16hi MØ subsets compared to controls (CD16lo: 811 vs 454; CD16hi: 3376 vs 855, p<0.05) but remained low at control levels on CD14lo/CD16hi M&Oslash. 11 patients survived spontaneously (PALF-S) and nine died or required liver transplantation (PALF-NS). PALF-NS had higher peak INR (13.9 vs 7.4, p<0.001), SOFA score (13 vs 6, p<0.001), MELD (46 vs 35, p<0.05) and APACHE II score (20 vs 6.5, p<0.001). There were no differences in the total monocyte counts between PALF-NS and PALF-S (0.2 vs 0.43 X10 000 cells/dl). However, a significant reduction in the proportion of CD14lo/CD16hi MØ was observed in PALF-NS compared to PALF-S (0.5% vs 2.7%, p=0.01) and this was predictive of outcome (AUROC 0.838; values greater than 2% giving a sensitivity of 81%, specificity 89% for survival with medical management). Conclusion: These are the first reported data defining the relative proportions of MØ subsets in PALF. They show that the proportion of CD14lo/CD16hi MØ is significantly reduced in PALF, the degree of which is predictive for outcome. This reduction may be due to early sequestration of these MØ within the liver or other organs through CCR2 independent pathways.