The relationship between brain tissue oxygen partial pressure thresholds and microdialysis markers of ischaemia in traumatic brain injury

Abstract

While cerebral ischaemia contributes significantly to secondary brain injury following trauma, there are no data on critical thresholds of brain tissue pO2 (PbO2) for metabolic compromise. The aim of this study was to identify ischaemic thresholds for changes in brain tissue lactate concentrations and lactate/pyruvate ratio. Eleven patients with severe traumatic brain injury, with brain tissue oxygen tension (NeurotrendTM) and microdialysis monitoring in place, were studied within the first 4 days following injury, over 4 h epochs during which physiology was carefully monitored and stabilised as part of a protocol involving transfusion of red cells. All patients were managed according to the Addenbrooke's Hospital Neuro-critical Care Unit head injury algorithm. Data for brain PbO2, and microdialysate glucose, lactate and pyruvate were measured over 20-minute intervals. Relationships between PbO2 and brain lactate level as well as lactate/pyruvate ratio were examined, and the significance of relationships assessed using non-parametric statistical methods. The eleven patients had a mean age of 34 years (range: 17 – 64), a median post-resuscitation GCS of 9 (range: 4 – 13) and 55% were female. Seven patients had the monitors placed in structurally normal-appearing tissue on CT scan, while in four patients the monitors were in abnormal tissue. Pooled mean ( standard deviation) PbO2 was 2.15 1.3 kPa, lactate 2.77 1.58 mmol/l, pyruvate 105 52 mol/l, and L/P ratio 26.6 5.4. There were no significant differences between normal and abnormal tissue, but PbO2 was less than 1 kPa only in abnormal tissue. Fitting with a Lowess function appeared to show an inflection point for increases in lactate at PbO2 below 1.5 kPa (Fig 1;). While the relationship between PbO2 and L/P ratio was also statistically significant (Spearman-Rank : - 0.358, p < 0.001), clear PbO2 threshold levels were less well defined. Brain tissue oxygen displays a threshold at about 1.5 kPa below which brain lactate increases in abnormal tissue. The absence of clear increases in L/P ratio below this threshold raises the possibility that brain lactate elevation in these areas may not entirely be due to tissue hypoxia. As PbO2 in normal tissue did not drop below 1 kPa, our data may be less representative of physiology in these areas. Further investigation is required in patients with low PbO2.