Background: Normally, end-tidal CO(2) is within 2 mm Hg of arterial PO(2) (PaCO(2)). However, if dead space in the lungs increases owing to shock with poor lung perfusion, the arterial-end tidal PCO(2) difference [P(a-ET)CO(2)] increases. We have found that in severely injured patients, P(a-ET)CO(2) of less than 10 mm Hg is associated with survival and P(a-ET)CO(2) of greater than 16 mm Hg is usually fatal. Our initial studies suggested that intravenously administered bicarbonate increases P(a-ET)CO(2).
Methods: This retrospective therapeutic study evaluated the effects of intravenously administered bicarbonate in a cohort of 225 severely acidotic (arterial pH ≤ 7.10) trauma patients who underwent emergency surgery from 1989 through 2011. Patients were divided into groups: early deaths (<48 hours), deaths in the operating room, deaths within 48 hours, and survivors. Winter's formula was defined as PaCO(2) = (HCO(3)) (1.5) + 8 ± 4.
Results: Of the 225 patients, the mean (SD) initial arterial pH was 6.92 (0.16) with HCO(3) of 11.0 (3.5) mEq/L. According to the Winter's formula, PaCO(2) should have been 24 (4) mm Hg but actually was 50 (14) mm Hg. In 73 patients, the effect of an average of two to eight vials of bicarbonate increased HCO(3) from 10.5 (3.1) mEq/L to 16.8 (4.0) mEq/L. In addition, PaCO(2) increased from 44 (9) mm Hg to 51 (11) mm Hg and end-tidal CO(2) stayed relatively constant (26 [6] to 25 [5]). This resulted in a increase in P(a-ET)CO(2) from 17 (9) mm Hg to 24 (13) mm Hg, affecting survival. In the final values after resuscitation, the P(a-ET)CO(2) in the 75 patients who survived was 10 (6) mm Hg, while the 103 patients who died in the operating room or within 48 hours of surgery had a P(a-ET)CO(2) of 23 (10) mm Hg (p < 0.001).
Conclusion: In severely acidotic, critically injured patients, reducing the PaCO(2) to less than 40 mm Hg and decreasing the P(a-ET)CO(2) to 10 (6) mm Hg should be attempted, using as little HCO(3) therapy as possible. Bicarbonate should be given only if severe acidosis persists despite resuscitation and if PaCO(2) levels near those which are appropriate can be obtained.
Level of evidence: Therapeutic study, level IV.