Mainstream vs. sidestream capnometry for prediction of arterial carbon dioxide tension during supine craniotomy

We compared the performance of mainstream capnometry as a measure of arterial carbon dioxide tension (P (a) co(2) ) with sidestream recordings in adult neurosurgical patients undergoing supine craniotomy. Two hundred and forty patients were randomly assigned so that the end-tidal carbon dioxide tension (P (E) co(2) ) was measured using either a mainstream or sidestream infrared capnometer. All patients received propofol anaesthesia and ventilation was adjusted according to clinical requirement. Arterial blood gas analyses were performed after induction, prior to dural incision, during surgery and before wound closure. Simultaneous haemodynamic and ventilatory parameters were also recorded. For 1007 paired measurements of P (E) co(2) and P (a) co(2) (mainstream, n = 503; sidestream, n = 504), the mean (SD) mainstream arterial to end-tidal carbon dioxide tension difference, 0.64 (0.16) kPa, was smaller than the corresponding sidestream values, 0.99 (0.40) kPa (p < 0.001). The limits of agreement for the mainstream analyser, 0.32-0.96 kPa, were also narrower than the sidestream recordings, 0.19-1.79 kPa (p < 0.001). In both capnometers, the arterial to end-tidal difference in carbon dioxide tension did not change with time. However, there was greater within-patient variation in the sidestream group. Our study showed that mainstream P (E) co(2) provided a more accurate estimation of P (a) co(2) than sidestream measurement.