Plasma biomarkers of oxidant stress and development of organ failure in severe sepsis

Abstract

We hypothesized that circulating levels of lipid peroxidation products in patients with severe sepsis are associated with the development of pulmonary, renal, hepatic, circulatory, and coagulation failure. Plasma levels of F2-isoprostanes and isofurans were measured by mass spectroscopy on intensive care unit day 2 in 50 critically ill patients with severe sepsis. Plasma F2-isoprostane levels were higher in patients who developed renal failure compared with those who did not (65 pg/mL [interquartile range {IQR} 44-112] vs. 44 pg/mL [IQR 29-54], P = 0.009) as were isofuran levels (1,223 pg/mL [IQR 348-2,531] vs. 329 pg/mL [IQR 156-1,127], P = 0.009). Plasma F2-isoprostane levels were higher in patients who developed hepatic failure compared with those who did not (72 pg/mL [IQR 44-112] vs. 44 pg/mL [IQR 30-65], P = 0.023), and there was also a trend for higher isofuran levels (1,411 pg/mL [IQR 298-1,965] vs. 525 pg/mL [IQR 160-1,223], P = 0.14). Coagulation failure (thrombocytopenia) was associated with higher isofuran levels. Circulatory failure and acute lung injury were not associated with elevated levels of isoprostanes or isofurans. Patients with isoprostane levels above the 25th percentile had higher mortality (42%) compared with patients with levels below the 25th percentile (8%, P = 0.03). Plasma levels of F2-isoprostanes and isofurans are associated with renal, hepatic, and coagulation failure, but not with circulatory or pulmonary failure in severe sepsis, suggesting that lipid peroxidation is a prominent feature of septic multisystem organ failure. Plasma isoprostanes and isofurans may be useful for monitoring oxidative stress in treatment trials of antioxidant therapies in severe sepsis.

Fig. 1. Plasma levels of both F2-isoprostanes and isofurans are elevated in severe sepsis patients with renal failure

A, Boxplot summary of plasma levels of F2-isoprostanes in patients who had or developed renal failure (n = 19) compared with those who did not (n = 31) *P = 0.009. B, Comparison of percent of patients who had or developed renal failure between quartiles of plasma F2-isoprostane level. First quartile is the lowest quartile (0 – 25th percentile). P = 0.013 for trend across quartiles. C, Boxplot summary of plasma levels of isofurans in patients who had or developed renal failure (n = 19) compared with those who did not (n = 31) **P = 0.009. D, Comparison of percent of patients who had or developed renal failure between quartiles of plasma isofuran level, P = 0.026 for trend across quartiles. For boxplots, horizontal line represents median, box encompasses 24th to 75th percentile, and error bars encompass 10th to 90th percentile.

Fig. 2. Plasma levels of both F2-isoprostanes and isofurans are elevated in severe sepsis patients with hepatic failure

A, Boxplot summary of plasma levels of F2-isoprostanes in patients who had or developed hepatic failure (n = 15) compared with those who did not (n = 35, *P = 0.023). B, Comparison of percent of patients who had or developed hepatic failure between quartiles of plasma F2-isoprostane level. First quartile is the lowest quartile (0 – 25th percentile), P = 0.045 for trend across quartiles. C, Boxplot summary of plasma levels of isofurans in patients who had or developed hepatic failure (n = 15) compared with those who did not (n = 35, P = 0.136). D, Comparison of percent of patients who had or developed hepatic failure between quartiles of plasma isofuran level. P value for trend across quartiles is 0.071. For boxplots, horizontal line represents median, box encompasses 24th to 75th percentile, and error bars encompass 10th to 90th percentile.

Fig. 3. Plasma levels of isofurans are elevated in severe sepsis patients with coagulation failure

A, Boxplot summary of plasma levels of isofurans in patients who had or developed coagulation failure (n = 18) compared with those who did not (n = 32, *P = 0.01). B, Comparison of percent of patients who had or developed coagulation failure between quartiles of plasma isofuran level, P = 0.008 for trend across quartiles. For boxplot, horizontal line represents median, box encompasses 24th to 75th percentile, and error bars encompass 10th to 90th percentile.

Fig. 4. Hospital mortality was lower in patients with the lowest quartile of plasma isoprostane levels compared with the second, third, and fourth quartiles

*P = 0.03 compared with first quartile.