Cerebrospinal fluid sampling technique and Astrup pH and PCO2 values

Abstract

The pH and PCO2 values measured by the Astrup technique were compared in cerebrospinal fluid (CSF) obtained using two different sampling techniques: 1) a direct or in vivo technique and 2) the widely accepted syringe sampling technique. In 65 pairs of measurements in 9 dogs it was found that the pH was always overestimated and the PCO2 always underestimated in the syringe sample when compared to the in vivo sample. The equations describing the relationships are as follows: 1) pH (syringe = 0.995 pH (in vivo) + 0.084 and 2) PCO2 (syringe) = 0.873 PCO2 (in vivo) + 0.2. The amount by which the syringe sample underestimated the true PCO2 value increased with the absolute PCO2 value, consistent with the possibility of there being a diffusional loss of CO2 during the transfer of CSF from the syringe to the pH electrode (PCO2 (in vivo)- PCO2 (syringe) = 2.4, 4.9, 7.5, and 10.0 mmHg at in vivo PCO2's of 20, 40, 60, and 80 mmHg). This study indicates that the technique used for sampling CSF is crucial to the expected accuracy of the results and that the number of transfers of CSF during the sampling and measurement procedures should be minimized in order to obtain reliable results.