Clinical Chemistry, Vol 27, 1676-1681, Copyright © 1981 by American Association for Clinical Chemistry

Estimation, prevention, and quality control of carbon dioxide loss during aerobic sample processing

ZL Bandi

We find that 2 to 6 mmol of carbon dioxide per liter (mean: 4.1 mmol/L) is lost during routine laboratory processing of patients' serum samples after centrifugation. Additional CO2 may be lost if evacuated blood- collection tubes are not filled completely during phlebotomy. More than 2 mmol of CO2 per liter is lost from samples stored in tightly stoppered tubes for 120 min if the tubes are less than half full. In extreme cases, 8 mmol of CO2 per liter may be lost from samples exposed to room air in open cups of automated micro-sample instruments. Clinically significant CO2 loss (greater than 2 mmol/L) before analysis is not detected by many laboratories because the generally accepted quality-control programs monitor only the very last step of the analytical process. A valid CO2 quality-control program should include samples with high as well as the generally used low pCO2 values. Alkalinization of serum and plasma samples with tert-butylamine prevents CO2 loss. Optimum tert-butylamine concentration, pH, and pCO2 were about 14 to 16 mmol/L, 9.0 to 9.3, and 0.4 to 1.5 mmHg (about 50 to 200 Pa).