The reliability of continuous fibreoptic oximetry in cerebral venous blood and its correlation with intracranial and cerebral perfusion pressures (pressure-volume curve) were examined in an experimental porcine study.
Methods: The pressure in the infratentorial compartment of 13 domestic pigs (18-24 kg) was gradually increased by inflating a Fogarty balloon catheter placed on the surface of the right cerebellar hemisphere and below the tentorium. Single volumes of 0.4 ml saline were injected into the inflatable balloon at 1-min intervals up to a total volume of 7 ml. Intracranial pressure (ICP), arterial blood pressure, cerebral perfusion pressure (CPP), and cerebral venous saturation measured continuously by fibreoptic oximetry and intermittent blood-gas analyses (SjO2 superior sagittal sinus) were monitored during balloon inflation. All data were down-loaded onto a PC and evaluated off-line by a commercial statistical software package.
Results: Over the whole pressure-volume curve, two phases of SjO2 behaviour were registered by continuous fibreoptic oximetry (Oximetrix 3, Abbott) (Fig. 1). CPP ranges of less than 50% reduction from the initial value showed a linear correlation (rmittl. = 0.712, P < 0.01) between both parameters (CPP-SjO2). In CPP ranges below a crucial point of about 50 mmHg no such correlation was found (rmittl = 0.176, P < 0.5). In contrast, in 3 pigs a very good correlation was found between CPP and SjO2 over the whole pressure-volume curve as measured by blood-gas analyses of samples from the cerebrovenous catheter (rmittl. = 0.84, P < 0.05).
Discussion: We conclude that in physiological CPP ranges down to 50 mmHg, SjO2 measurement is a reliable method of detecting oxygen desaturation in cerebrovenous blood. Below that CPP value, the fibreoptic catheter showed repeated false-high oxygen saturation values. The accuracy of SjO2 measurement seems to depend on sufficient cerebral blood flow (CBF): with decreasing CBF the amount of cerebral venous outflow is diminished. We believe this is why we could not find a correlation in low CPP ranges with the oximetry catheter. This flow-dependency is a new aspect of fibreoptic cerebrovenous oximetry. The authors recommend that rising SjO2 values after desaturation events be confirmed by blood-gas analyses.