Objectives: (a) To investigate the relationship between gut ischemia parameters (gastric intramucosal pH [pHi], mucosal-arterial carbon dioxide difference [PCO2-gap]), and endotoxin or cytokine release during hemorrhagic shock; (b) to compare the predictive value of pHi, PCO2-gap and arterial lactate concentrations.
Design: Prospective study.
Setting: Surgical intensive care unit of a university hospital.
Patients: 20 multiple trauma patients with severe hemorrhagic shock.
Interventions: Intramucosal measurements and blood samples were obtained on admission to the emergency room and repeatedly over 48 h.
Measurements and results: Endotoxin was measured using a chromogenic limulus amoebocyte assay. Cytokine [tumor necrosis factor-alpha (TNF alpha) and interleukin-6 (IL-6)] values were evaluated by immunoradiometric assays. Only 3 patients had positive blood cultures but endotoxins were detected at least once in all patients. Endotoxin levels were similar in survivors and non-survivors over the study period and were not related to pHi or PCO2-gap. Initially, high levels of IL-6 were observed in both nonsurvivors and survivors [median 1778 pg/ml (range 435-44,540) vs 2068 pg/ml (range 996-92,300)]. IL-6 levels progressively decreased in the survivors but not significantly. On admission, TNF alpha concentrations were similar in nonsurvivors and survivors (42 +/- 35 vs 46 +/- 27 pg/ml). From the 24th h, TNF alpha values were higher in the nonsurvivors than in the survivors (24 h: 72 +/- 38 vs 34 +/- 17 pg/ml, p < 0.05). The greatest IL-6 levels were found for a pHi < 7.20 (28.5 +/- 36.5 vs 1.8 +/- 1.3 ng/ml, p < 0.05) or a PCO2-gap > 7.5 mmHg (1 kPa) (32.5 +/- 37.5 vs 1.7 +/- 1.3 ng/ml, p < 0.01). With the same pHi threshold, no difference was found in endotoxin levels. The lactate concentrations were predictive for outcome from the 12th h (9.5 +/- 5.9 vs 3.6 +/- 2.3 mmol/l, p < 0.05).
Conclusions: During severe hemorrhagic shock, endotoxin translocation from the gut was a common phenomenon that seemed independent of both pHi values and outcome. It could not explain IL-6 and TNF alpha release. In severe hemorrhagic shock, neither pHi nor PCO2-gap provides additional information to the lactate measurements.