Brain metabolism is significantly impaired at blood glucose below 6 mM and brain glucose below 1 mM in patients with severe traumatic brain injury

Abstract

Introduction: The optimal blood glucose target following severe traumatic brain injury (TBI) must be defined. Cerebral microdialysis was used to investigate the influence of arterial blood and brain glucose on cerebral glucose, lactate, pyruvate, glutamate, and calculated indices of downstream metabolism.

Methods: In twenty TBI patients, microdialysis catheters inserted in the edematous frontal lobe were dialyzed at 1 microl/min, collecting samples at 60 minute intervals. Occult metabolic alterations were determined by calculating the lactate- pyruvate (L/P), lactate- glucose (L/Glc), and lactate- glutamate (L/Glu) ratios.

Results: Brain glucose was influenced by arterial blood glucose. Elevated L/P and L/Glc were significantly reduced at brain glucose above 1 mM, reaching lowest values at blood and brain glucose levels between 6-9 mM (P < 0.001). Lowest cerebral glutamate was measured at brain glucose 3-5 mM with a significant increase at brain glucose below 3 mM and above 6 mM. While L/Glu was significantly increased at low brain glucose levels, it was significantly decreased at brain glucose above 5 mM (P < 0.001). Insulin administration increased brain glutamate at low brain glucose, but prevented increase in L/Glu.

Conclusions: Arterial blood glucose levels appear to be optimal at 6-9 mM. While low brain glucose levels below 1 mM are detrimental, elevated brain glucose are to be targeted despite increased brain glutamate at brain glucose >5 mM. Pathogenity of elevated glutamate appears to be relativized by L/Glu and suggests to exclude insulin- induced brain injury.

Figure 1

Changes in brain glucose determined by microdialysis (grey box plots) and calculated brain-to-blood glucose ratio (white box plots) in pre-defined blood glucose clusters, ranging from less than 5 mM to more than 9 mM in 1 mM buckets. At arterial blood glucose levels exceeding 6 mM brain glucose was significantly increased. With increasing arterial blood glucose and brain glucose levels calculated brain-to-blood glucose ratio was significantly decreased, reflecting reduced cerebral uptake. Increases across the pre-defined blood glucose clusters compared with low arterial blood glucose levels were significant (*P < 0.001; analysis of variance on ranks, post hoc Dunn's test).

Figure 2

Changes in calculated brain lactate-to-pyruvate (grey box plots) and lactate-to-glucose (white box plots) ratio determined by microdialysis reflecting influence of blood glucose on downstream cerebral metabolism in pre- defined arterial blood glucose clusters, ranging from less than 5 mM to more than 9 mM in 1 mM buckets. At arterial blood glucose levels exceeding 6 mM brain lactate-to-glucose ratio was significantly decreased. Decreases across the pre-defined blood glucose clusters compared with low arterial blood glucose levels were significant (*P < 0.001; analysis of variance on ranks, post hoc Dunn's test).

Figure 3

Changes in brain glutamate (grey box plots) and calculated lactate-to-glutamate (white box plots) ratio determined by microdialysis reflecting influence of blood glucose on downstream cerebral metabolism in pre-defined arterial blood glucose clusters, ranging from less than 5 mM to more than 9 mM in 1 mM buckets. At arterial blood glucose levels exceeding 6 mM brain glutamate was significantly increased. In parallel, calculated lactate-to-glutamate was significantly decreased. Alterations across the pre-defined blood glucose clusters compared with low arterial blood glucose levels were significant (*P < 0.001; analysis of variance on ranks, post hoc Dunn's test).

Figure 4

Changes in calculated brain lactate-to-pyruvate (grey box plots) and lactate-to-glucose (white box plots) ratio determined by microdialysis reflecting influence of brain glucose on downstream cerebral metabolism in pre-defined brain glucose clusters, ranging from less than 1 mM to more than 9 mM in 1 mM buckets. At brain glucose levels exceeding 1 mM brain lactate-to-pyruvate and lactate-to-glucose were significantly decreased. Changes across the pre-defined brain glucose clusters compared with low brain glucose levels (<1 mM) were significant (*P < 0.001; analysis of variance on ranks, post hoc Dunn's test).

Figure 5

Changes in brain glutamate (grey box plots) and calculated brain lactate-to-glutamate (white box plots) ratio determined by microdialysis reflecting influence of brain glucose on downstream cerebral metabolism in pre-defined brain glucose clusters, ranging from less than 1 mM to more than 9 mM in 1 mM buckets. At brain glucose levels exceeding 5 mM brain glutamate was significantly increased. In parallel lactate-to-glutamate was significantly decreased. Changes across the pre-defined brain glucose clusters compared with low brain glucose levels were significant (*P < 0.001; analysis of variance on ranks, post hoc Dunn's test).

Figure 6

Changes in (a) brain glutamate, (b) calculated brain lactate-to-glutamate ratio, and arterial blood glucose determined by cerebral microdialysis investigating the influence of time points with insulin (grey box plots) and without insulin (white box plots) administration in pre-defined brain glucose clusters. At brain glucose levels ≤ 5 mM brain glutamate was significantly increased under the influence of insulin compared with time points without insulin administration (*P < 0.001, Mann-Whitney test). In parallel lactate-to-glutamate was significantly decreased (*P < 0.001, Mann-Whitney test). At brain glucose levels >5 mM insulin did not influence brain glutamate or lactate-to-glutamate ratio. Whenever insulin was administered, arterial blood glucose was significantly increased compared with time points insulin was not given (*P < 0.001, Mann-Whitney test).