Accurate and continuous measurement of oxygen deficit during haemorrhage in pigs.

Objective: Haemorrhagic shock can cause organ failure and high mortality. Uncontrolled bleeding, a predetermined bleeding volume or blood pressure controlled bleeding are traditionally used to study haemorrhagic shock. These models are influenced by compensatory mechanisms preventing accurate knowledge about the severity of cellular insult. We describe the use of a method for continuous measurement of oxygen deficit during haemorrhage in pigs.

Methods: We defined a cumulative oxygen deficit of approximately 100mL/kg as the primary endpoint for severe haemorrhage. For continuous assessment of oxygen deficit a metabolic monitor (Deltatrac II, Datex-Ohmeda Instrumentation Corp., Helsinki, Finland) was used. Data are presented as mean+/-SD; (*)P<0.05 was considered to be significant.

Results: 17 out of 22 anaesthetised male pigs achieved a mean cumulative oxygen deficit of 106+/-3 mL/kg (range: 95-117 mL/kg) by withdrawing an average blood volume of 47+/-6 mL/kg over 1h. Mean arterial blood pressure (MAP) fell from 83+/-19 to 22+/-7mmHg (baseline versus shock), heart rate increased from 83+/-7 to 147+/-37min(-1). Venous base excess changed from 4.8+/-2.4 to -12.5+/-3.4 mmol/L and venous lactate increased from 1.5+/-0.4 to 13.3+/-2.4 mmol/L after haemorrhage. Two pigs (11%) died during the haemorrhagic shock phase. The traditional method of assessing haemorrhage (measuring blood volume lost) showed only a poor correlation with heart rate (r=0.3872; P=0.1540), MAP (r=0.3901; P=0.1505), mixed venous oxygen saturation (svO(2); r=0.0944; P=0.7379) or cardiac index (CI; r=0.2101; P=0.4523). Cumulative oxygen deficit correlated significantly better with heart rate (r=0.7175; P=0.0026), MAP (r=0.5039; P=0.0556), svO(2) (r=0.7084; P=0.0031) or CI (r=0.6260; P=0.0125).

Conclusion: We describe a model to study haemorrhagic shock based on the cumulative oxygen deficit. We believe that the use of a metabolic monitor to measure oxygen deficit in our model represents an improvement on the current available methods to study the effects of haemorrhagic shock.