Figure 2. The relationship among intracranial pressure (ICP), central venous pressure (CVP), and mean arterial blood pressure (MAP) in a lightly anesthetized, mechanically ventilated dog. The horizontal lines approximate the mean baseline values for each variable. Before initiating a noxious stimulus (A), baseline ICP, CVP, and MAP were stable, yet reflected minor fluctuations induced by mechanical ventilation. During delivery of the 1-min noxious stimulus (B), CVP demonstrated positive and negative deflections that correlated with the dog's attempts to inspire and expire. However, the mean CVP response during this period did not vary meaningfully from baseline. In contrast, mean ICP increased by more than twofold. Early in the stimulus period, MAP decreased at a time when ICP was dramatically increased. Later, MAP increased to--and eventually exceeded--baseline values, such that, cerebral perfusion pressure (defined as MAP-ICP) was not diminished thereafter. After completion of the stimulus (C), the dog continued to move. Mean ICP and MAP values remained elevated at a time in which mean CVP differed little from baseline. As the dog ceased moving (D), both ICP and CVP values returned to baseline. When the same stimulus was initiated after neuromuscular block with pancuronium (data not shown), neither ICP, CVP, nor MAP meaningfully changed from baseline. (Note: the large artifacts in the MAP tracing are the result of sampling blood from--and flushing--the arterial catheter at the 2- and 4- min measurement intervals.)