"Uncontrol led bleeding," "a controlled prefixed bleeding volume," or "cont
rolled decrements in blood pressure" are traditional models of experimental
hemorrhagic shock. They are influenced by compensatory mechanisms and do n
ot adequately reflect the severity of the cellular insult as a major target
for therapeutic strategies. The aim of this study was to develop an animal
model that uses oxygen debt (OD) and metabolic acidemia as indicators of h
emorrhage severity. Twenty-five female pigs (mean weight: 23.8 kg) were ane
sthesized and randomized to 1 of 5 groups of increasing OD (< 50 through >
120 mL/kg). The predetermined OD was accrued by hemorrhage uniformly over 6
0 min and followed by retransfusion. The animals were allowed to recover un
der anesthesia for 200 min and were then observed for 3 days. The extent of
metabolic derangements were quantified by arterial base excess (BE) and pl
asma lactate (LAC) measurements. OD, BE, and LAC were shown to be superior
as predictors of outcome in comparison with traditional variables ("bleedin
g volume," "blood pressure," "cardiac output") in correlation and regressio
n. Of the analyzed predictors of outcome, BE and LAC showed the highest cor
relation to levels of OD (r = -0.78, 0.8 respectively; P < 0.0001), and reg
ression models were developed. The LD50 for OD was 95.0 mL/kg, for BE -15.3
mmol/L. and for LAC 7.7 mmol/L. By using the developed regression models,
it is possible to estimate accurately the actual level of OD from BE and LA
C values obtained during hemorrhagic shock. OD, BE, and LAC appear to be op
timal indicators of severity for a pig hemorrhagic shock model.