Background: Bicarbonate therapy for metabolic acidosis can cause tissue hypercapnia. We tested the hypothesis that CO2 retention is initially confined to blood and that tissue hypercapnia is consequent to blood hypercapnia.
Design: prospective study.
Setting: general Intensive Care Unit in an University Hospital.
Patients: ten patients, aged 34-84, undergoing fixed mechanical ventilation for stroke, respiratory failure, polytrauma, or coma and presenting fairly stabilized circulatory condition were included in the study providing that definite exclusion criteria were absent. Sodium bicarbonate, 0.08, 0.17, and 0.25 mEq/kg, was infused in 5 minutes during stages A, B, and C. During each stage, before and after bicarbonate infusion, heart rate, arterial pressure, central venous pressure, and PetCO2 were registered and arterial and central venous haemogasanalysis were performed. CO2 output was measured during and after bicarbonate infusion and total CO2 output over basal values was calculated.
Results: Heart rate, arterial pressure, central venous pressure did not change in any stage, while PetCO2 increased in all. Following bicarbonate infusion, pH increased more in venous than in arterial blood and was significantly affected by the amount of bicarbonate infused. In arterial blood both PCO2 and total CO2 content (TCO2) increased. In venous blood PCO2 did not change while TCO2 increased so that arterial-venous rl CO2 difference was unaffected. CO2 output changed in no patient but two during stage A, and increased by 0.54 +/- 0.16 mL/kg during stage B and by 1.18 +/- 0.26 mL/kg during stage C so that 14.5 +/- 4.3 and 21.2 +/- 4.7% of given bicarbonate was released as CO2 through the lungs. CO2 output increased after a latency that was significantly shorter during stage C (145 +/- 51 sec) than during stage B (226 +/- 53 sec).
Conclusions: Since arterial-venous TCO2 difference was unaffected and haemodynamic parameters were stable, the conclusion is drawn that CO2 and bicarbonate did not diffuse out of vascular compartment during this study. However CO2 and bicarbonate diffusion out of blood would have occurred if bicarbonate infusion had continued, because CO2 excess was released through the lungs after a latency and extended after the infusion. Consequently, tissue hypercapnia during bicarbonate infusion could probably be prevented by improving CO2 release through the lungs, i.e. by hyperventilating the patient.