Uncoupling of cerebral blood flow and brain oxygen saturation under non-pulsatile flow conditions

Abstract

Introduction Non-pulsatile blood flow has been associated with altered microcirculatory perfusion. Soon after loss of pulsatility a marked heterogenous flow profile in the very small vessels ensues that is coupled with impaired oxygen extraction [1]. We investigated if changes in cerebral blood flow during non-pulsatile perfusion would affect brain tissue oxygenation assessed by near-infrared spectroscopy. Methods After approval by the local ethics committee thirty-two patients were included in this trial. They were studied during general anaesthesia, and stable hemodynamic conditions. Cerebral blood flow velocity in the middle cerebral artery was measured during non-pulsatile flow in the course of cardiac surgery on total extracorporeal bypass and again postoperatively on the intensive care unit during pulsatile flow. We modified cerebral blood flow by changing arterial partial pressures of carbon dioxide between 30, 40 and 50 mmHg. This was conducted either by a change of sweep gas flow at the heart lung machine or by changing respiratory rate at the respirator. In addition to flow measurements patients' regional brain oxygenation was determined. Relative changes in blood flow velocity were plotted against relative changes in regional brain saturation for both flow conditions and compared using Pearson's correlation. A p-value Results Absolute cerebral blood flow velocity was slightly higher during non-pulsatile flow when serum hemoglobin was diluted. Relative changes in cerebral blood flow velocity and regional brain oxygenation were tightly coupled independent of the flow condition. R-values for non-pulsatile and pulsatile blood flow were 0.724 and 0.796, respectively. However, during non-pulsatile flow the slope of the regression line was less steep as compared with pulsatile flow (P Discussion Despite greater cerebral perfusion during hypercapnia and non-pulsatile flow there was no equivalent rise in cerebral oxygenation. Assuming similar cerebral oxygen delivery this finding could be explained by impaired brain oxygen extraction in the high flow range while partial pressure of carbon dioxide was elevated. In contrast, during hypocapnia, when cerebral blood flow was decreased to a greater extent as during pulsatile flow, the concomitant desaturation was less pronounced. Both findings are in line with shunting on the microcirculatory level, whereby hyperperfused vessels coexist with vessels lacking adequate flow. These findings substantiate that loss of pulsatility seems to negatively affect nutritive tissue perfusion. Raised neurological morbidity in patients supported with non-pulsatile extracorporeal cardiac assist devices could potentially be attributable to the observed perfusion deficit [2].