Monocyte anergy is present in patients with severe acute pancreatitis and is significantly alleviated by granulocyte-macrophage colony-stimulating factor and interferon-γ in vitro

Objectives: Severe acute pancreatitis (AP) is frequently associated with immune suppression, which increases the risk of infections, organ failure, and death. Our aims were to measure monocyte function (ie, HLA-DR expression and tumor necrosis factor-alpha [TNF-alpha] production as markers of immune suppression) in patients with severe AP and to determine whether treatment of blood samples with granulocyte-macrophage colony-stimulating factor (GM-CSF) and/or interferon-gamma (IFN-gamma) corrected the functional defects of monocytes in vitro. Methods: The study consisted of 28 patients with severe AP who were treated at intensive care unit and in whom the proportion of HLA-DR-positive monocytes in the circulation was less than 70%, and 28 matched control subjects who were selected from healthy laboratory personnel. HLA-DR density was determined by whole blood flow cytometry. Monocyte TNF-alpha production in response to bacterial lipopolysaccharides (LPSs) was studied in a whole blood assay. Aliquots of blood were supplemented with IFN-gamma (all 28 patients), GM-CSF (the last 24 patients), or both (the last 12 patients). Results: The median proportion of HLA-DR-positive monocytes was 45% in patients (range, 18%-73%) and was 98% in controls (range, 86%-100%; P < 0.001). TNF-alpha levels in response to LPSs were lower in patients (545 pg/mL; range, 84-1990 pg/mL) than in controls (1415 pg/mL; range, 660-5490 pg/mL; P < 0.001). The proportion of HLA-DR-positive cells correlated positively with TNF-alpha levels (r = 0.56; P < 0.01). Both GM-CSF and IFN-gamma increased HLA-DR expression of monocytes in patients (98%; range, 74%-100% for GM-CSF; 99%; range, 86%-100% for IFN-gamma; both P < 0.001). The combination restored monocyte HLA-DR expression (99%; range, 96%-100%; P = 0.002). Compared with basal levels, GM-CSF increased TNF-alpha production of monocytes both in blood samples from patients (median, 1320 pg/mL; range, 35-8015 pg/mL) and controls (median, 3450 pg/mL; range, 1040-9835 pg/mL; both P < 0.001). IFN-gamma increased TNF-alpha production by monocytes in patients (683 pg/mL; range, 186-2705 pg/mL; P < 0.05) but not in controls (1658 pg/mL; range, 765-4755 pg/mL; P = 0.31). With the combination of GM-CSF and IFN-gamma, the TNF-alpha levels of monocytes in patients (3185 pg/mL; range, 545-8280 pg/mL) and in controls (2800 pg/mL; range, 1080-6860 pg/mL) were comparable. Conclusions: The proportion of HLA-DR-positive monocytes correlates with TNF-alpha production, and they both reflect the degree of immune suppression. The low proportion of HLA-DR-positive monocytes in AP can be reversed in vitro by GM-CSF and/or IFN-gamma. The GM-CSF and IFN-gamma treatments also increase LPS-induced TNF-alpha production. By the combination of GM-CSF and IFN-gamma, but not by either agent alone, LPS-induced TNF-alpha production of monocytes was equally high in patients and in controls.