Comparison of ¹⁵oxygen positron emission tomography and near-infrared spectroscopy for measurement of cerebral physiology

Introduction

The gold standard technique for imaging cerebral blood flow (CBF) and metabolism is ¹⁵oxygen positron emission tomography (¹⁵O PET). Continuous near-infrared spectroscopy (NIRS) has been used to assess adequacy of cerebral oxygenation following stroke, traumatic brain injury and subarachnoid haemorrhage [1], and measurements have been compared with jugular oxygen saturation. In this study we compared NIRS with ¹⁵O PET within healthy volunteers.

Methods

Fifteen healthy subjects were recruited (12 male, average age 38 years); PET precluded females of reproductive age. Steady-state ¹⁵O PET with arterial sampling was performed to measure CBF, cerebral metabolic rate of oxygen (CMRO₂), oxygen extraction ratio (OEF) and cerebral blood volume (CBV) [2]. Simultaneously, NIRS data were collected using a Hamamatsu NIRO 200 with sensors on either side of the forehead. NIRS OEF was calculated from tissue oxygen saturation, SaO₂ and an assumed arterial/venous blood volume ratio of 30/70 [3].

Results

The frontal region ¹⁵O PET CBF, CMRO₂, OEF and CBV were mean (SD) 44.9 (10) ml/100 ml/minute, 158.7 (24.7) μmol/100 ml/minute, 45.8 (7.3)%, and 2.8 (0.8) ml respectively, and there was no relationship between NIRS and ¹⁵O PET (Figure 1).

Figure 1

Linear correlation between NIRS and PET OEF.

Conclusion

We found no relationship between NIRS and baseline physiology as determined by ¹⁵O PET. Further studies should assess the dynamic response of NIRS to a measured change in physiology in patients. Further confines of NIRS include its limited and focal brain coverage.