Brain oxygen, CO2, pH, and temperature monitoring: evaluation in the feline brain

Abstract

Currently, no ideal method exists for monitoring the injured brain. Recently, a single, compact, fiberoptic sensor has become available for measuring oxygen, CO2, pH and temperature in blood. We have adapted this instrument for continuous use in brain tissue to measure oxygen tension, carbon dioxide tension (pCO2), pH, and temperature. To evaluate this new technique, we produced hypercapnia, hypocapnia, intracranial pressure increase, and hypoxemia in seven normal cats. In an additional six animals, sensors were placed within a zone of focal brain ischemia induced by occluding the left middle cerebral artery. The sensor readings were compared with cerebral blood flow measurements, intracranial pressure, and brain histological findings. An in vitro experiment was also performed using human blood to test the accuracy of the sensor over a wide range of pCO2 and oxygen tension values. After careful precalibration and rigid cranium fixation, stable measurements could be obtained throughout the 6- to 8-hour experiments. In normal animals, brain oxygen was 42 +/- 9 mm Hg, brain CO2 was 59 +/- 14 mm Hg, brain pH was 7.0 +/- 0.2, and brain temperature was 36.7 +/- 0.7 degrees C. Hypocapnia and hypoxemia produced a significant decline in tissue oxygen (< or = 30 +/- 3 mm Hg; P < 0.001), whereas hypercapnia caused by hypoventilation and intracranial pressure increase produced a significant increase in tissue CO2 (> or = 74 +/- 4 mm Hg; P < 0.001). Focal ischemia produced a rapid 42% decline in brain oxygen (25 +/- 7 mm Hg) and a 25% increase in tissue pCO2 (71 +/- 23 mm Hg). Brain oxygen further decreased to 19 +/- 6 mm Hg toward the end of the experiment, 4 hours later. After middle cerebral artery occlusion, the regional cerebral blood flow decreased to 10 +/- 5 ml per 100 g per minute, within the 1st hour, from a baseline value of 65 +/- 15 ml per 100 g per minute. It then gradually increased to 15 +/- 5 ml per 100 g per minute by the end of the 4-hour experiment. Brain pH was closely and inversely related to brain CO2. The brain temperature in the focally ischemic tissue decreased from 36.7 +/- 0.7 to 35.5 +/- 1.6 degrees C by the end of the experiment. The in vitro experiment demonstrated good linear correlation between the sensor readings and the blood gas analysis. Continuous monitoring of oxygen, CO2, pH, and temperature in damaged or at-risk brain tissue using a single sensor is now feasible and will, thus, allow improved continuous monitoring of neurosurgical patients who are at risk of significant secondary brain damage.