The results of this study provide insight into the effects of prone positioning, head rotation and positive pressure breathing on brain perfusion and brain venous drainage. The prone position with sideways rotated head and positive breathing narrowed the ipsilateral internal jugular vein together with a ~10% reduction in MCA Vmean in spite of an elevated MAP. These results suggest that for this commonly applied anesthetic approach both cerebral blood flow and cerebrovenous drainage are optimal only with the head centered.
Transcranial Doppler ultrasound monitors blood peak flow velocity rather than volume flow, and changes in the diameter of the insonated vessel could modulate velocity independently of volume flow. During craniotomy, the diameter of the MCA remains unchanged by even large changes in arterial pressure [20
]. Constancy of the diameter of the MCA as determined with magnetic resonance imaging during changes in carbon dioxide tension and in simulated orthostasis further supports that the MCA is not involved in regulation of cerebral vascular resistance [21
] linking changes in MCA Vmean
to those in cerebral blood flow [22
All Finometer and TCD variables were A/D converted and stored for off-line analysis. The off-line analysis involved detection and manually removal of artifacts with subsequent semiautomated data extraction. This semiautomated process leaves little potential for introduction of bias. Jugular vein areas were obtained off-line by manually tracking the outline of the vein. Potential for bias exist as we consider blinding impractical because the position of the head was revealed by the distortion of the structures in the ultrasonographic picture.
With prone position, we observed a small reduction in cardiac stroke volume and output () in accordance with an earlier report in anesthetized patients likely reflecting reduced venous return [23
]. The approximate doubling of internal jugular vein cross-sectional area when positioned below heart level suggests passive gravitational jugular vein dilation (). However, MCA Vmean
was largely unaffected ().
Head rotation in the prone position increased MAP by ~4
mmHg () possibly reflecting unloading of carotid baroreceptors by altered pressure from the soft tissues with a reflex increase in sympathetic nervous activity [24
]. Head-down rotation engages the otolith organs and vestibular otolith stimulation may increase sympathetic activity during baroreflex unloading [25
]. Positive pressure breathing also enhances resting muscle sympathetic activity and the reduction of urinary output and sodium excretion associated with prolonged positive pressure breathing has been attributed to cardiopulmonary receptor unloading [26
]. A contribution of enhanced sympathetic activity modulating cerebral vascular tone may be considered. The increase in MAP from supine to the prone position presumably by increased sympathetic activity was not accompanied by changes in MCA Vmean
reflecting integrity of cerebrovascular autoregulation mechanisms. In contrast, positive pressure breathing reduced cardiac output together with MCA Vmean
whereas MAP increased. A restricted CO that challenges MAP limits flow to the brain [27
], but under conditions where MAP is not challenged by a restricted CO, the influence of sympathetic stimulation on CBF is not manifested [29
]. Thus, in this study, we consider an altered cerebral venous outflow resistance rather than sympathetic activity to dominate CBF.
With the head-turned constancy of MCA Vmean
was maintained notwithstanding a ~50% reduction in AjugR
. The cerebral venous system is characterized by many collaterals and void of valves directing blood flow [30
], explaining why compression of one jugular vein was of no consequence for cerebral blood flow as long as the central blood volume was maintained. Rotation per se appears to be of little influence for the cerebral blood supply since rotation of the head in the supine position leaves the diameter of the common and internal carotid artery unchanged [31
]. Our subjects had their head supported in a horseshoe headrest avoiding external pressure on the neck, and we consider compromised arterial cerebral inflow unlikely to have caused the reduction in MCA Vmean
. However, carotid compression cannot be excluded in prone-positioned anesthetized patients, for example, by external compression of supporting pillows and/or inappropriate extention/rotation. Only ~40% of normal subjects have a complete Circle of Willis maintaining blood flow to the contralateral side of the brain with unilateral occlusion [32
]. Thus, variations in the Circle of Willis may have influenced cerebral hemodynamics following head rotation in the prone position. Furthermore, especially with incomplete Circle of Willis, the side of head rotation may be of importance for the hemodynamic response, but this was not investigated in our study.
During positive pressure breathing when supine MCA Vmean
was reduced together with SV and CO conforming an earlier report [11
] and reflecting a reduction of the central blood volume imposed by positive pressure breathing [33
]. Under the conditions of this study, absence of changes in PaCO2
with CPAP renders an effect of CO2
on CBF unlikely.
During positive pressure breathing in the head centered prone position MCA Vmean
was maintained despite a further reduction in SV and CO. This observation is compatible with the notion that an increase in cerebral perfusion pressure in the prone position may sustain CBF () [34
reached a nadir of ~10% below baseline in the head-rotated prone position with positive pressure breathing. This reduction is within the range observed under everyday physiological challenges, that is, standing up [36
]. By comparison a ~50% decline is associated with clinical ischemia [22
] and syncope [33
]. As general anesthesia has a favorable supply/demand profile for cerebral oxygen flux—even during induction, where the cardiovascular depression often is largest [37
], it could be argued that the changes induced by prone position are of little clinical relevance to most patients. However, even small reductions in cerebral blood supply could be deleterious in the setting of an already compromised cerebral perfusion, for example, in elderly subjects with vascular disease [12
The applied pressure of 10
O of CPAP is (on average) similar to the prevailing intrathoracic pressure when ventilating healthy patients in the prone position [39
] whereas in the presence of lung disease or marked obesity inflation pressures may surpass this value. The reduction of the central blood volume as an accompanying effect of positive pressure breathing did reduce CO and MCA Vmean
. These new findings are of clinical relevance since they indicate that CBF may be compromised by venous compression in the head-turned prone position when the central blood volume is challenged by positive pressure breathing.
In healthy awake individuals, the prone position with positive pressure breathing reduced MCA Vmean especially when the head was rotated to the side. These results may have implications for the anesthetized and ventilated patient. The hypothesis that in the prone position both CBF and cerebrovenous drainage are optimal only with the head centered should be tested in patients undergoing general anesthesia.