High-physiological and supra-physiological 1,2- 13 C 2 glucose focal supplementation to the traumatised human brain

Abstract

How to optimise glucose metabolism in the traumatised human brain remains unclear, including whether injured brain can metabolise additional glucose when supplied. We studied the effect of microdialysis-delivered 1,2-¹³C₂ glucose at 4 and 8 mmol/L on brain extracellular chemistry using bedside ISCUSflex, and the fate of the ¹³C label in the 8 mmol/L group using high-resolution NMR of recovered microdialysates, in 20 patients. Compared with unsupplemented perfusion, 4 mmol/L glucose increased extracellular concentrations of pyruvate (17%, p = 0.04) and lactate (19%, p = 0.01), with a small increase in lactate/pyruvate ratio (5%, p = 0.007). Perfusion with 8 mmol/L glucose did not significantly influence extracellular chemistry measured with ISCUSflex, compared to unsupplemented perfusion. These extracellular chemistry changes appeared influenced by the underlying metabolic states of patients’ traumatised brains, and the presence of relative neuroglycopaenia. Despite abundant ¹³C glucose supplementation, NMR revealed only 16.7% ¹³C enrichment of recovered extracellular lactate; the majority being glycolytic in origin. Furthermore, no ¹³C enrichment of TCA cycle-derived extracellular glutamine was detected. These findings indicate that a large proportion of extracellular lactate does not originate from local glucose metabolism, and taken together with our earlier studies, suggest that extracellular lactate is an important transitional step in the brain’s production of glutamine.

Keywords

Brain metabolism microdialysis NMR traumatic brain injury (human)

Introduction

The leading cause of death and disability in young adults in the developed world is traumatic brain injury (TBI). Initial survivors of severe TBI may subsequently suffer further brain damage through secondary insults including dysfunction of brain energy metabolism, despite the best current clinical management, affecting patient outcome.¹ While various brain metabolic fuels are reported, including lactate,^2
–5 glucose is conventionally regarded as the primary substrate for brain energy metabolism.^6
–8

After TBI, changes in brain glucose metabolism have been demonstrated with typically lower cerebral microdialysate concentrations of glucose (e.g. 1.1 mmol/L)⁹ than non-TBI control patients (1.7–2.0 mmol/L).^10,11 It is unclear whether low microdialysis glucose within TBI cohorts is pathological or adaptive. Furthermore, extremes of both high and low microdialysis glucose are associated with worse outcomes.^8,9 Low microdialysate glucose (neuroglycopenia) may result from reduced glucose transport across the blood brain barrier, or greater cellular uptake of glucose from the extracellular space, or stem from ischemia. The latter is less likely with modern neuro-intensive care protocols targeting partial brain oxygen pressure (PbtO₂) and cerebral perfusion pressure (CPP) thresholds, but the traumatised brain may have higher serum glucose requirements than the uninjured brain. Inadequate cellular uptake may explain high microdialysate glucose associated with unfavourable outcome in a large study of TBI patients.⁹ Studies comparing ‘tight’ and ‘loose’ control of serum glucose reported that tight serum glucose control was associated with more frequent episodes of critically low microdialysis glucose,^12
–14 so more permissive serum glucose control is often advocated.

Cerebral glucose metabolism changes after TBI are thus complex. The injured brain appears more dependent on serum glucose to maintain cerebral glucose, and since both high and low brain glucose levels are statistically associated with unfavourable outcome, an optimal range seems likely. There are currently insufficient data to define this,⁸ and no existing evidence whether supplementing the traumatised brain with additional glucose will support its metabolism and be beneficial,⁸ as while serum glucose and glycaemic control influence brain glucose, the relationship may be altered in injured brain.⁸

We aimed to determine if the traumatised human brain could metabolise supplementary glucose when delivered focally into the brain extracellular fluid. We expected additional glucose metabolism to be expressed as an increase in its metabolites extracellularly: pyruvate and lactate in the recovered microdialysates.¹⁵ 1,2-¹³C₂ glucose was perfused via a cerebral microdialysis catheter at a ‘high-physiological’ concentration (4 mmol/L), and a ‘supra-physiological’ concentration (8 mmol/L) to study any dose response to this focal supplementation, thereby avoiding systemic metabolism and the blood brain barrier. We used ¹³C high-resolution nuclear magnetic resonance (NMR) analysis of the recovered microdialysates, expecting to detect ¹³C lactate and ¹³C glutamine as these glucose metabolites are typically sustained and relatively abundant extracellularly in brain.^3,16,17

Materials and methods

Patients

We recruited patients aged >16 years with a severe/moderate cranial trauma (post-resuscitation Glasgow Coma Scale (GCS) ≤ 12), who additionally required sedation and mechanical ventilation for intracranial hypertension and/or airway protection. Patients were treated using our standard TBI management protocols including endotracheal intubation, ventilation, sedation, neuromuscular blockade and maintenance of serum glucose concentration within the target range 4–10 mmol/L.¹⁸ Informed written assent was obtained from patients’ relatives. The study was conducted in conformation with the spirit and the letter of the Declaration of Helsinki. The National Research Ethics Service, Committee East of England–Cambridge Central (REC Reference No.11/EE/0463) approved the protocol. Data from four of the patients who received high-physiological (4 mmol/L) 1,2-¹³C₂ glucose by microdialysis were from a previous study.¹¹ The others were not in any previous ¹³C studies.

Perfusion fluid and ¹³C-labelled substrate

CNS Perfusion Fluid (M Dialysis AB, Stockholm, Sweden) consisted of NaCl (147 mmol/L), KCl (2.7 mmol/L), CaCl₂ (1.2 mmol/L), and MgCl₂ (0.85 mmol/L) in water. 1,2-¹³C₂ glucose (isotopic enrichment 99%, chemical purity 99%) from Cambridge Isotope Laboratories (Tewksbury, MA, USA) was formulated at 4 or 8 mmol/L in CNS perfusion fluid by the Pharmacy Manufacturing Unit, Ipswich Hospital NHS Trust (Ipswich, UK) who tested the formulations to verify purity, sterility, freedom from endotoxins and absence of pyrogenicity, compliant with current regulations.

Microdialysis technique

CMA 71 microdialysis catheters (membrane 10 mm, cut-off 100 kDa, M Dialysis AB) were directed into normal-appearing brain (white matter); neither into nor adjacent to CT-visible lesions, per the 2014 Consensus Statement guidelines.⁸ The microdialysis catheter was placed either via a craniotomy, or through a triple-lumen cranial access device (Technicam, Newton Abbot, UK) together with an intracranial pressure monitor (Codman, Raynham, MA, USA) and a Licox PbtO₂ sensor (GMS, Kiel-Mielkendorf, Germany) when available. Catheters were perfused at 0.3 µL/min with plain, unsupplemented CNS Perfusion Fluid for at least 24 h. The first hour of microdialysate collected was never used for clinical monitoring (to eliminate any artefacts from insertion trauma and the pump flush sequence). Then, for a period of 24 h, the perfusion fluid was changed to CNS perfusion fluid supplemented with either 4 mmol/L or 8 mmol/L 1,2-¹³C₂ glucose. Afterwards, the perfusion fluid was reverted to plain unsupplemented CNS perfusion fluid. Microdialysate collection vials were changed hourly and analysed on a bedside ISCUSflex analyser (M Dialysis AB) for glucose, lactate, pyruvate, glycerol and glutamate. Microdialysates were frozen at -80°C if storage >24 h was necessary before pooling for subsequent NMR analysis.

High resolution NMR acquisition and analysis

After bedside ISCUSflex analysis, brain microdialysate samples from patients perfused with 8 mmol/L 1,2-¹³C₂ glucose were collected and pooled into 24-hour supplementation periods for each patient. Samples from patients perfused with 4 mmol/L 1,2-¹³C₂ glucose were not analysed with ¹³C NMR due to limited availability of this resource and the existence of our similar previous studies.¹¹ Each patient’s pooled sample (180 µL) was added to 20 µL of deuterium oxide (D₂O) and 50 µL of 24 mmol/L 2,2-dimethyl-2-silapentane-5-sulfonate sodium salt (DSS) internal reference standard (Sigma-Aldrich, Poole, Dorset, UK). NMR analysis was performed using 3 mm NMR tubes (Hilgenberg GmbH, Malsfeld, Germany) in a Bruker Avance III HD 500 MHz spectrometer (Bruker BioSpin GmbH, Karlsruhe, Germany) using a dual ¹H/¹³C cryoprobe (CP DUL500C/H, Bruker BioSpin GmbH) for ¹³C and ¹H spectral acquisition (for further details, see Supplementary Information online). Peaks were identified by reference to our own standards and online NMR databases (BMRB—Biological Magnetic Resonance Bank, University of Wisconsin¹⁹ and HMDB—Human Metabolome Database, Genome Alberta²⁰) Chemical shifts were expressed in ppm (Hz per MHz) referenced to DSS at zero ppm. Peak areas for DSS and lactate signals were integrated using TopSpin.

To quantify the lactate signals in the ¹³C NMR spectra, calibration was performed with a series of known concentrations of standard lactate,¹¹ in CNS perfusion fluid and added to NMR tubes at the same volumes as for microdialysates, with the same fixed concentrations and volumes of D₂O and DSS as an internal standard as used for the microdialysates samples and run under identical NMR conditions. To quantify the microdialysate spectra, peak areas relative to the DSS internal standard were used with reference to calibration curves derived from the lactate standards, showing a linear relationship between peak areas (ratio to DSS internal standard) and concentrations.

Fractional enrichment (%) is defined as 100×[¹³C]/([¹³C]+[¹²C]) where square brackets indicate concentrations of the relevant species. [¹³C] was determined from the calibrated ¹³C NMR spectra (above). [¹²C] was determined from the ¹H spectra, measuring the peak areas of lactate C3 methyl protons that give an ¹H NMR doublet at 1.32 ppm (from protons attached to ¹²C), relative to the peak area of DSS internal standard.

¹³C natural abundance is 1.1% of all carbon atoms, and ¹³C results for the lactate ¹³C singlet signals were expressed after subtracting this natural background. ¹³C doublet signals were not background-subtracted because the probability of two ¹³C atoms occurring next to each other naturally is 0.01% (=1.1% × 1.1%).

The pentose phosphate pathway (PPP)-derived 3-¹³C lactate ratio to glycolysis-derived 2,3-¹³C₂ lactate was expressed in the form 1:N where N is the concentration of lactate derived from the ¹³C doublet at 22.8, divided by the concentration of lactate represented by the ¹³C singlet at 22.8 ppm (after first subtracting the singlet’s ¹³C natural abundance background).

Statistical analysis

Statistical data analysis utilised R (version 3.5.1, www.r-project.org). Demographics of patients who received 4 mmol/L and 8 mmol/L glucose were compared using Mann-Whitney U test. For ISCUSflex results, changes between baseline and glucose supplementation periods were excluded for 2 h after changeover from pre-supplementation to glucose supplementation, and changeover from glucose supplementation to post-supplementation, to allow for ‘run-in’ and ‘washout’.¹⁷ Each patient’s pre- and post-supplementation data were combined as a baseline period for that individual, thereby allowing for underlying trends in patient physiology. Heterogeneity (frequent in TBI cohorts) was accounted for by a linear mixed effects model (‘lmer’ in R package lme4²¹) that allows for clustering of data, and has a different random effect for each patient. Also, Wilcoxon signed rank test was used to corroborate statistical results. Only those patients with microdialysates from both pre- and post-supplementation baseline periods plus a supplementation data period were included in the ISCUSflex data analysis. The Shapiro-Wilk normality test was also performed. Our detailed data analysis of ISCUSflex results focussed mainly on microdialysis glucose, lactate, pyruvate and lactate/pyruvate ratio (LPR), as our primary question was whether the traumatised brain could metabolise additional glucose, and interpretation of microdialysate glycerol is ambiguous (as it can be a glucose metabolite or a breakdown product of lipid),⁸ and several patients had incomplete glutamate microdialysis data.

Results

Patients and monitoring

Twenty TBI patients (14 M, 6 F) were recruited; mean age 39 y (median 33 y, range 16 y–65 y), and post-resuscitation mean GCS 6. Injury causes were mostly road traffic collisions and falls. The commonest surgical diagnoses were cerebral contusions and acute subdural haematomas. For detailed demography, see Supplementary Table 1. There were no complications from insertion of microdialysis catheters, nor related to the perfusion of study substrate; specifically, no cerebral haematomas or contusions were noted on CT imaging after catheter insertion. Throughout the study period, adequate CPP was maintained in all patients (>60 mmHg) and no patients suffered from intractable raised ICP (>25 mmHg). PbtO₂ data was available for 7/20 patients (mean PbtO₂ 27 mmHg). No patients suffered from cerebral ischaemia (PbtO₂<15 mmHg).

Patient baseline characteristics

There were no significant differences (Mann-Whitney U, p > 0.1) between the two groups of patients – those who received 4 mmol/L or 8 mmol/L glucose – for baseline GCS, Extended Glasgow Outcome Scale (GOS-E),²² time interval between injury and period of supplementation (mean 2.9 days), nor serum glucose or serum lactate during the 1,2-¹³C₂ glucose supplementation period. However, there was a statistically significant difference (p < 0.0001, lmer) between the two patient groups for microdialysate glucose, lactate and pyruvate at baseline. For ’baseline’ definition, see Statistical Analysis above. Baseline glucose differences between the two groups of patients are likely due to inherent heterogeneity between TBI patients, and our cohort being small. Microdialysate baseline glucose was 32% higher in the 8 mmol/L group versus the 4 mmol/L group, while lactate and pyruvate were 21% and 13% lower respectively, although the baseline lactate/pyruvate ratio (LPR) was only 4% lower in the 8 mmol/L than in the 4 mmol/L group (Table 1, Supplementary Table 2).

Table 1.

ISCUSflex results for microdialysates at baseline and during microdialysis supplementation with 1,2-¹³C₂ glucose.

ISCUSflex analyte	Supp. conc. (mmol/L)	Baseline Mean (sd)	Supp. Mean (sd)	% change mean	lmer p	Baseline Median (IQR)	Supp. Median (IQR)	% change median	Wilcoxon p
Glucose (mmol/L)	4	1.22 (0.70)	4.07 (0.80)	↑233%	0.0001	1.05 (0.79)	3.77 (0.96)	↑259%	0.001
Glucose (mmol/L)	8	1.61 (1.38)	7.56 (0.91)	↑309%	0.0001	0.92 (1.14)	7.47 (1.13)	↑708%	0.008
Lactate (mmol/L)	4	3.26 (1.36)	3.89 (1.92)	↑19%	0.0001	3.07 (1.28)	3.71 (1.91)	↑21%	0.01
Lactate (mmol/L)	8	2.56 (1.25)	2.50 (0.93)	↓2%	0.17	2.18 (0.52)	2.16 (0.30)	↓1%	0.8
Pyruvate (µmol/L)	4	113 (41)	132 (52)	↑17%	0.0001	106 (35)	122 (72)	↑15%	0.04
Pyruvate (µmol/L)	8	98 (37)	95 (31)	↓2%	0.12	78 (33)	86 (40)	↑10%	0.64
LPR	4	27.8 (9.2)	29.3 (9.8)	↑5%	0.002	23.6 (12.0)	26.4 (12.2)	↑12%	0.007
LPR	8	26.7 (6.7)	27.1 (6.1)	↑1.5%	0.17	26.0 (3.4)	27.6 (4.8)	↑6%	0.33
Glycerol (µmol/L)	4	109 (76)	118 (85)	↑8%	0.0001	110 (79)	88 (82)	↓20%	0.2
Glycerol (µmol/L)	8	93 (37)	107 (55)	↑15%	0.0001	85 (36)	101 (36)	↑19%	0.2
Glutamate(µmol/L)	4	3.24 (3.37)	8.06 (8.11)	↑149%	0.3	2.35 (2.35)	6.75 (4.27)	↑187%	0.9
Glutamate(µmol/L)	8	15.30 (34.10)	12.10 (29.50)	↓21%	0.0001	3.89 (3.68)	1.90 (1.06)	↓51%	0.016

Group means (with sd) and medians (with IQR) from periods of baseline (pre- and post- supplementation) and during supplementation with 1,2-¹³C₂ glucose (Supp.). Total number (n) of patients whose data were included in this table was 18. One of these 18 patients had two catheters, so the results here in Table 1 are from 19 catheters. For further details, see Supplemental Table 1 (patient demographics). Significance was determined using R by linear mixed effects model (lmer) of patient individual hourly data and Wilcoxon signed rank of patient averaged data. Test for normality of metabolite data was performed in R using Shapiro-Wilk test. Patient hourly data were not normally distributed. Although the majority of patient-averaged data was found to be normally distributed, Wilcoxon signed rank test was used due to the small number of observations and the inclusion of analytes with non-normally distributed data. Up- and down- arrows denote % change increases and decreases respectively. sd; standard deviation; IQR; interquartile range.

Evidence from ¹³C labelling patterns

The appearance of doubly-labelled lactate (2,3-¹³C₂ lactate) in the recovered microdialysates showed unequivocally that the brain metabolised by glycolysis the exogenous 1,2-¹³C₂ glucose delivered via the microdialysis catheter. The 2,3-¹³C₂ lactate ‘signature’ of characteristic doublets (Figure 1) in the ¹³C NMR spectrum is evidence that the ¹³C-¹³C bond stays intact, because the probability of two ¹³C atoms occurring next to each other naturally without exogenous double-labelling is 0.01%. Thus, the 8 mmol/L 1,2-¹³C₂ glucose delivered directly to the brain’s extracellular space by microdialysis catheters was taken up by brain cells, metabolised by glycolysis as the main pathway into lactate, and this ¹³C-labelled lactate was exported into the brain extracellular space, where it was recovered by microdialysis catheters for analysis by ISCUSflex (Figure 2) and ¹³C NMR (Figure 3). This concurs with our previous microdialysis studies using 1,2-¹³C₂ glucose at 2 mmol/L and 4 mmol/L concentrations.^3,11 Also as previously, a lower enrichment of singly-labelled lactate (3-¹³C lactate) showed the PPP to be a lesser contributor to lactate production than glycolysis (2,3-¹³C₂ lactate) from the exogenously-supplied 1,2-¹³C₂ glucose.¹¹

Figure 1.

Example ¹³C NMR spectra: Panel a CNS perfusion fluid supplemented with 8 mmol/L 1,2-¹³C₂ glucose, prior to infusion. The large α and β peaks for ¹³C at glucose position C1 and C2 can be seen dominating the spectrum, with small peaks representing natural abundance of ¹³C at the unenriched positions within the molecule. Panel b pooled microdialysate from a patient during their 24 h of 8 mmol/L 1,2-¹³C₂ glucose supplementation. Lactate can be seen at 22.8 ppm (C3) and 71.2 ppm (C2). The C3 peak is expanded, showing in greater detail the ¹³C labelled doublet and the small ¹³C singlet inside the doublet. The doublet represents glycolytic metabolism of supplemented 1,2-¹³C₂ glucose, and the singlet a combination of pentose phosphate pathway-metabolised 1,2-¹³C₂ glucose supplement and the natural abundance of ¹³C label in endogenous lactate. This singlet within a doublet was visible in all patients’ microdialysates analysed with high resolution NMR. Panel c pooled microdialysate from a patient during a period of ‘baseline’ perfusion with unsupplemented perfusion fluid. Small singlet peaks can be seen corresponding to lactate and glucose, representing the natural abundance background of ¹³C in these molecules. All spectra were analysed with TopSpin (Bruker GmbH), referenced to internal standard DSS. Small peaks for ¹³C-labelled fructose are visible in Panel a and Panel b, representing a small proportion of ¹³C-labelled glucose that has undergone isomerisation which occurs spontaneously in solution. Glc: Glucose; Fruc: fructose; DSS: 4,4-dimethyl-4-silapentane-1-sulfonate sodium salt internal reference standard; Lac: lactate.

Figure 2.

ISCUSflex bedside analyser measurements showing the effect of 1,2-¹³C₂ glucose supplementation via microdialysis on brain extracellular chemistry. Each pair of data-points indicates mean levels at baseline and during 1,2-¹³C₂ glucose perfusion, respectively, for that patient. Supplementation was for ≈ 24 h. Baseline denotes ≈ 24 h of perfusion with plain (unsupplemented) CNS-perfusion fluid; the baseline value represents combined data pre- and post-supplementation (to account for any underlying trends). Upper 4 Panel a (circles) denote results from patients supplemented with 1,2-¹³C₂ glucose at 4 mmol/L, middle 4 Panel b (triangles) denote results from patients supplemented with 1,2-¹³C₂ glucose at 8 mmol/L, lower 4 Panel c (circles and triangles) denote results from subset of these patients from both groups with low baseline cerebral glucose (<0.8 mmol/L), black crossbars on graphs denote averages (means) of individual patient means. P-values calculated using linear mixed effects model in R. LPR: lactate/pyruvate ratio. Note that in Panel a the LPR data points for two of the patients who received 4 mM 1,2-¹³C₂ glucose almost exactly coincide (see Supplemental Table 2 for patients GL-17 and GL-19).

Figure 3.

Bar chart of mean percentage ¹³C enrichment in lactate carbon position 3 after supplementation with 8 mmol/L 1,2-¹³C₂ glucose, and correlation between glycolysis and PPP derived lactate. Panel a: Error bars denote standard deviation. ¹H and ¹³C high-resolution NMR analysed with TopSpin. ‘Total 3-¹³C Lactate enrichment at C3’ denotes total fractional (percentage) ¹³C enrichment at the C3 position of lactate (the methyl group), which represents glycolytic and pentose phosphate pathway (PPP) metabolism of supplemented glucose. ‘2,3-¹³C₂ Lactate’ denotes percent of all lactate that is doubly-labelled (2,3-¹³C₂ lactate doublet), representing glycolytic-derived lactate. ‘3-¹³C Lactate (endog. subtracted)’ denotes percent of all lactate that is singly labelled (3-¹³C lactate singlet after subtraction of the natural abundance background ¹³C) produced via the PPP. There was a strong linear correlation between glycolysis-derived lactate (2,3-¹³C₂ lactate) and pentose phosphate pathway (PPP)-derived lactate (3-¹³C lactate, excluding natural abundance) with a y-intercept of close to zero in these patients who received 8 mmol/L 1,2-¹³C₂ glucose, suggesting that these two metabolic pathways are regulated in unison in TBI. Relationships expressed as fractional (percentage) ¹³C enrichment (Panel b) and ¹³C concentration (mmol/L) (Panel c). Each data-point represents 24 x 1 h pooled microdialysis vials from each individual patient. Numerical values are in Table 2. Panel b: 3-¹³C lactate fractional (percentage) enrichment (y-axis), plotted versus 2,3-¹³C₂ lactate fractional (percentage) enrichment (x-axis). Linear relationship, y = 0.1195x + 0.1959, Pearson’s r² = 0.7299. Panel c: 3-¹³C lactate concentration in mmol/L (y-axis), plotted versus 2,3-¹³C₂ lactate concentration in mmol/L (x-axis). Linear relationship, y = 0.148x - 0.0052, Pearson’s r² = 0.9437. Thus, approximately 1 molecule of 3-¹³C lactate is produced by the PPP for every 6.8 molecules of 2,3-¹³C lactate produced by glycolysis, when 1,2-¹³C₂ glucose is metabolised.

Glucose supplementation at 4 mmol/L: brain biochemistry compared to baseline

Eleven patients, who received microdialysis perfusion supplemented with 4 mmol/L 1,2-¹³C₂ glucose, possessed complete baseline (both pre- and post-supplementation) ISCUSflex datasets (Supplementary Table 1). Mean baseline concentrations were glucose 1.22 mmol/L, lactate 3.26 mmol/L and pyruvate 113 µmol/L. Glucose supplementation increased microdialysate glucose by 234%, produced a mean 19% increase (versus baseline) in extracellular lactate (p < 0.0001, lmer) and 17% increase in extracellular pyruvate (p < 0.0001, lmer). A mean 5% increase in LPR (p = 0.002, lmer), was unlikely to be biologically or clinically significant (Figure 2). Glycerol rose by a mean 8% (p < 0.0001, lmer), but there was no significant change in extracellular glutamate.

Glucose supplementation at 8 mmol/L: brain biochemistry compared to baseline

Twelve patients received microdialysis with CNS perfusion fluid supplemented with 8 mmol/L 1,2-¹³C₂ glucose. One of these 12 patients also had a second microdialysis catheter simultaneously supplemented with 4 mmol/L 1,2-¹³C₂ glucose. Eight of these 12 patients had complete ISCUSflex datasets comprising pre-supplementation, during-supplementation, and post-supplementation. Mean baseline concentrations were glucose 1.61 mmol/L, lactate 2.56 mmol/L, and pyruvate 98 µmol/L. Compared to baseline perfusion with plain (unsupplemented) CNS perfusion fluid, glucose supplementation increased microdialysate glucose by 370%, but did not result in a statistically significant change (lmer) in extracellular lactate concentration (average 2% decrease, p = 0.17) or pyruvate (average 2% decrease, p = 0.12). Similarly, there was no significant change in LPR (average 1.5% increase, p = 0.17, lmer) (Figure 2). There was a statistically significant 15% increase in glycerol (p < 0.0001, lmer) and 21% fall in glutamate (p < 0.0001, lmer).

Glucose supplementation in the neuroglycopenic brain

Seven patients had a low baseline extracellular brain glucose, defined as <0.8 mmol/L.^8,23 Analysis of microdialysis (ISCUSflex) data from neuroglycopenic patients – 3 patients supplemented with 8 mmol/L 1,2-¹³C₂ glucose and 4 patients supplemented with 4 mmol/L 1,2-¹³C₂ glucose – revealed a similar pattern as supplementation in the 4 mmol/L 1,2-¹³C₂ glucose group as a whole: a mean 15% increase in lactate (p < 0.001), 15% increase in pyruvate (p < 0.001), and only 4% increase in LPR (p = 0.003) (Figure 2).

NMR analysis of microdialysates from 8 mmol/L 1,2-¹³C₂ glucose perfusion

Lactate is an important glucose metabolite found at mmol/L concentrations in cerebral extracellular fluid in health and TBI.¹¹ The mean percentage of lactate labelled with ¹³C at the C3 position in recovered microdialysates was 17.6% (s.d. 5.4%). This comprised 14.7% (s.d. 4.9%) doubly-labelled (2,3-¹³C₂ lactate, 0.39 mmol/L) derived from glycolysis, and 2.0% (s.d. 0.7%) singly-labelled (3-¹³C lactate, 0.03 mmol/L, 1.1% ¹³C natural abundance subtracted) derived from PPP metabolism (Figure 3, Table 2 and Supplementary Figure 1). For further explanation of PPP biochemistry, see Carpenter et al.⁷ and Jalloh et al.¹¹

Table 2.

NMR results.

	Glycolytic 2,3- ¹³C₂ lactate (calculated from C3 doublet)		PPP 3- ¹³C lactate^a (calculated from C3 singlet)		Total ¹³C-lactate^a C3 enrichment (calculated from C3 singlet + C3 doublet)		Percentage of total ¹³C at C3 of lactate that is derived from PPP	Ratio of PPP 3-¹³C lactate to glycolytic 2,3-¹³C₂ lactate
ID	Conc. mmol/L	FE %	Conc. mmol/L	FE %	Conc. mmol/L	FE %	%	Ratio
GL-01	0.15	8	0.015	0.8	0.18	8.9	9.1	1:10.0
GL-02	0.27	19	0.040	2.9	0.32	21.9	13	1:6.8
GL-05	0.61	13	0.081	1.7	0.73	14.7	12	1:7.5
GL-07	0.83	13	0.130	2.1	1.00	15.1	14	1:6.4
GL-08	0.24	14	0.020	1.2	0.27	15.2	8	1:12.0
GL-09	0.19	10	0.022	1.2	0.23	11.2	10	1:8.6
GL-11	0.22	13	0.031	1.8	0.26	14.8	12	1:7.1
GL-12	0.22	12	0.039	2.2	0.28	14.2	15	1:5.6
GL-13	0.38	18	0.053	2.5	0.46	20.5	12	1:7.2
GL-14	0.6	15	0.084	2.1	0.72	16.9	12	1:7.1
GL-15	0.63	27	0.073	3.1	0.72	30.1	10	1:8.6
GL-16	0.39	15	0.049	1.9	0.46	16.9	11	1:8.0
Mean	0.39	15	0.053	2.0	0.47	16.7	12	1:7.9

Results of high-resolution NMR analysis of microdialysate (each patient’s NMR sample was a pool of 24 × 1 h vials) during supplementation with 8 mmol/L 1,2-¹³C₂ glucose. Total number (n) of patients whose data were included in this table was 12. FE: fractional enrichment – see Methods for details of calculation.

^aThe 3-¹³C lactate fractional enrichment results are presented after background subtraction to remove the contribution from natural abundance background ¹³C that is 1.1% of all carbon atoms. 2,3-¹³C₂ lactate results are not background-subtracted as the probability of two natural ¹³C atoms occurring next to each other by chance is only 0.01% (=1.1% × 1.1%).

NMR spectroscopy showed no evidence for TCA cycle metabolism of 1,2-¹³C₂ glucose supplemented at 8 mmol/L. Notably, doubly-labelled ¹³C glutamine was undetectable in microdialysates from all 12 patients with NMR analysis; only two of these showed singly-labelled glutamine but not above background ¹³C natural abundance.

Discussion

Here we have shown that the traumatised human brain can metabolise additional glucose to a limited degree when delivered directly into the brain extracellular fluid via a microdialysis catheter. Glucose supplementation may be a useful strategy in TBI patients in certain circumstances, discussed below.

Significance of ¹³C-labelling in metabolites

The substrate 8 mmol/L 1,2-¹³C₂ glucose was metabolised to 2,3-¹³C₂ lactate, detected by NMR in the recovered microdialysates, providing clear unambiguous evidence of glycolysis, with a smaller amount of 3-¹³C lactate indicating metabolism via the PPP as a minor route for lactate production.

Despite this unambiguous evidence of glycolytic (and PPP) metabolism of 1,2-¹³C₂ glucose, there was no detectable ¹³C NMR evidence for TCA cycle metabolism – notably no doubly-labelled glutamine was identified, while singly-labelled glutamine was only at natural-abundance background level, similar to previous studies employing lower concentrations of 1,2-¹³C₂ glucose.^3,11

The lack of change in ISCUSflex-measured extracellular metabolites (whereby labelled and unlabelled molecules are not differentiated) during perfusion with 8 mmol/L 1,2-¹³C₂ glucose suggests several explanations, below.

Limited effect on glycolytic activity – changes in extracellular lactate and pyruvate measured by bedside ISCUSflex

Predictably, 1,2-¹³C₂ glucose supplementation caused a marked, significant rise in microdialysate levels of glucose measured on the ISCUSflex bedside analyser. Supplementation with 4 mmol/L glucose produced significant rises in lactate (19%) and pyruvate (17%), suggesting increased glycolytic activity and possibly PPP too. However, supplementation with 8 mmol/L glucose did not increase lactate and pyruvate concentrations further.

Probably these different responses are at least partly due to patient differences. Baseline glucose was 32% lower in the 4 mmol/L supplementation group (1.22 mmol/L) than the 8 mmol/L supplementation group (1.61 mmol/L); while lactate and pyruvate were 14% and 21% higher, respectively. Supplementation of additional glucose appeared less effective at increasing glycolysis in patients with an ‘already-high’ baseline glucose, in contrast to patients with lower baseline glucose. This is evident in subgroup analysis of neuroglycopaenic patients, where there were modest, statistically significant increases in lactate, pyruvate and LPR when supplemented with 1,2-¹³C₂ glucose via microdialysis, regardless of supplementation concentration. This ‘neuroglycopaenic’ subgroup nevertheless showed a similar pattern of metabolite labelling to the study cohort overall, wherein ¹³C labelling in lactate indicated glycolysis and PPP, but no detectable ¹³C labelling in glutamine.

Additionally, the relatively high baseline lactate and pyruvate in patients who received 4 mmol/L glucose supplementation may indicate tissue able to metabolise glucose relatively freely – and thus can metabolise additional glucose when supplemented. Contrastingly, the relatively lower baseline lactate and pyruvate in those patients who subsequently received 8 mmol/L glucose may represent underlying relative impairment of cerebral glycolysis (and PPP), hindering upregulation of metabolism when given excess glucose.

The biochemical behaviour pattern of TBI patients who received 8 mmol/L glucose concurs with earlier findings by Jalloh et al. in their non-TBI “normal brain” group, where 4 mmol/L glucose elicited a proportionate rise in ISCUSflex-measured glucose but no significant changes in lactate or pyruvate, and a minor (statistically significant) increase in LPR.¹¹ Their statistical analysis utilised pooled data rather than individual biochemistry.¹¹ Although statistically non-significant (possibly because of low number of patients), median levels of for lactate and pyruvate increased with glucose supplementation in non-TBI “normal brain”.¹¹ Conversely, in TBI patients, lactate and pyruvate did not increase with glucose supplementation.¹¹ The ability of TBI brain to metabolise additional glucose via glycolysis and the PPP might depend on its baseline metabolic ‘health’ that may be poorer in TBI patients.

The self-limiting nature of glycolysis may also be relevant. Cells take up glucose via specific transporters^24,25 then phosphorylate it with hexokinase, locking it inside the cell in the first step of glycolysis. Normally the rest of glycolysis would follow, leading to pyruvate (and then lactate and/or to acetyl-CoA and TCA cycle). However, if there is too much glucose the hexokinase is product-inhibited by glucose-6-phosphate and the cell stops taking up more glucose, thereby limiting glycolysis.²⁶ This may be relevant to the present study, particularly with high (8 mmol/L) glucose supplementation. The potentially deleterious effect of severe hyperglycaemia in the brain after TBI was reported^24,25 in studies considering elevated serum glucose rather than increased brain interstitial glucose. Such adverse effects might manifest at least partly through effects on brain vasculature, less relevant to this study. Nevertheless, a comprehensive study of graded hyperglycaemia in humans and rabbits found that more extreme hyperglycaemia may cause neuronal injury and suppress astrocyte activation in the frontal cortex, whereas moderate hyperglycaemia may not.²⁷ Furthermore, a positive statistical relationship between microdialysate glucose concentration and mortality was reported in TBI patients, alongside a negative relationship between pyruvate concentration and mortality. This concurs with the idea that abnormally high levels of brain extracellular glucose may exist because injured brain cells are poorly able to metabolise it.

A caveat to our present study’s technique is that glucose given directly into the brain extracellular fluid might differ in its subsequent processing compared to glucose delivered via the circulation (e.g. intravenous glucose, discussed below).

Extracellular ¹³C lactate – dilution, glycolysis, and the pentose phosphate pathway

Supplementing the brain extracellular space with 8 mmol/L 1,2-¹³C₂ glucose yielded no evidence of TCA cycle metabolism (the spin-out product ¹³C-labelled glutamine was undetected) and there were no significant changes in total labelled plus unlabelled concentrations of lactate and pyruvate. However, the ¹³C-labelled glucose was clearly metabolised by glycolysis and the PPP, evidenced by recovered ¹³C-lactate with the diagnostic labelling patterns in microdialysates. This implies that the metabolism of the exogenous ¹³C labelled glucose was seemingly in place of that of endogenous unlabelled glucose, discussed below. The ¹³C NMR results appear to reveal the ceiling of cerebral glycolytic metabolism of this cohort.

When glycolysis breaks down one mole of 1,2-¹³C₂ glucose, it produces 1 mole of 2,3-¹³C₂ lactate and 1 mole of unlabelled lactate (via their respective precursor pyruvates), thereby self-diluting the labelled lactate. Thus, the theoretical maximum level of enrichment of lactate at the region of interest (ROI) addressed by the microdialysis catheter assuming exposure to 1,2-¹³C₂ glucose at 99% enrichment is 49%. However, when baseline extracellular unlabelled lactate is considered, this may be expected to fall to around 44% (assuming an unlabelled baseline glucose concentration of 1 mmol/L). The mean enrichment level observed in the C3 position of lactate (combined glycolytic 2,3-¹³C₂ lactate and PPP 3-¹³C lactate) in this study was under half this, at 16.7%. The highest individual level of ¹³C enrichment observed in lactate was 30%, which is over three-fifths of the theoretical maximum level of enrichment of 49% if the ROI addressed by the microdialysis catheter was exposed solely to 1,2-¹³C₂ glucose at 99% enrichment. This relative paucity of lactate ¹³C enrichment was unexpected given the high concentration of glucose perfused, with potential explanations below.

Microdialysis delivery of 8 mmol/L 1,2-¹³C₂ glucose was used to create focal supra-physiological cerebral concentrations of glucose in the brain extracellular fluid. However, this might not entirely reproduce the process of glucose delivery from the circulation. The brain has an extensive vascular supply, and astrocyte end-feet are firmly attached to the endothelial cells of the blood capillaries constituting the tight junctions of the blood-brain barrier (Figure 4). Glucose uptake from blood into brain is controlled via specific transporters.^24,25 Glucose transfer from blood in the capillary into the astrocytes directly adjoining the endothelial cells might conceivably be more effective at feeding glucose into the astrocytes than direct uptake from the extracellular fluid (Figure 4), perhaps explaining the discrepancy between observed and theoretical ¹³C lactate enrichment.

Figure 4.

Proposed schematics of intercellular metabolic pathways including astrocyte-neuron lactate shuttle hypothesis and explanations for low ¹³C lactate enrichment and absent ¹³C glutamine enrichment – indicating the importance of extracellular lactate in glutamine production. Panel a: Simplified single-cell metabolism of glucose. Glucose from the vasculature is metabolised to form pyruvate by glycolysis or the pentose phosphate pathway. Pyruvate may then either be converted to lactate and exported extracellularly, or incorporated into the mitochondrial TCA cycle, leading to ATP synthesis (oxidative phosphorylation). Glutamate is also produced as a spin-off from alpha-ketoglutarate for release into the synaptic cleft as a neurotransmitter, before being taken up and converted to glutamine. Panel b: Simplified schematic of metabolic trafficking in the brain according to the astrocyte-neuron astrocyte shuttle hypothesis.³⁴ Glucose from the vasculature is metabolised by astrocytes to lactate, then exported to local neurons via the extracellular fluid. Lactate is incorporated into the neuronal mitochondrial TCA cycle, leading to ATP synthesis and glutamate production for release into the synaptic cleft as a neurotransmitter. Leftover released glutamate is taken up by adjacent astrocytes, converted to glutamine and delivered back to neurons via the extracellular fluid. Extracellular lactate and glutamine are important extracellular mediators in this model. Panel c: Potential explanations for relatively low ¹³C lactate enrichment (16.7%; 0.45 mmol/L) despite high concentration ¹³C glucose supplementation via microdialysis: (1) Glucose may more readily be absorbed by glia and neurons from capillaries via its natural route, rather than through the extracellular space. (2) Exchange of ¹³C lactate with unlabelled lactate from the systemic circulation. (3) Local diffusion of ¹³C lactate away from the region surrounding the catheter. (4) Insufficient rate of delivery of ¹³C glucose by catheter compared to the cellular metabolic rate of glucose. Panel d: Microdialysis delivered ¹³C labelled lactate yields ¹³C labelled glutamine. Delivery of modest concentrations ¹³C lactate (4 and 8 mmol/L) by microdialysis catheter results in ¹³C labelling of glutamine.^3,16 Panel e: Extracellular lactate is an important step in the brain’s production of glutamine. Delivery of ¹³C glucose by microdialysis catheter does not result in ¹³C labelling of glutamine, but does produce low percent enrichment ¹³C lactate labelling (9-30%). This suggests that additional dilution of the ¹³C label occurs at the point of extracellular lactate, so pathway ‘x’ that shuttles lactate between cells (coloured green) is more likely to represent the dominant pathway of glutamine production in the traumatised human brain than pathway ‘y’, coloured red, that does not involve extracellular lactate. Pathway ‘y’ would have expected to produce ¹³C labelled glutamine when ¹³C glucose is perfused at high concentrations, but not when ¹³C lactate is perfused. Glc, glucose; Lac, lactate; Pyr, pyruvate; Glt, glutamate; Gln, glutamine.

Other likely sources of dilution include active or passive exchange with endogenous lactate from the systemic circulation, and passive diffusion with adjacent brain. Indeed, arteriovenous difference measurements have shown that the brain undergoes periods of net uptake and net export of lactate.⁷

Glycolysis is the primary metabolic path for glucose metabolism in the brain, generating 2 moles of ATP per mole of glucose, whereas the PPP is an alternative pathway (a complex detour around some of the steps of glycolysis) that neither consumes nor produces ATP, but sacrifices some of the brain’s glucose for the sake of producing intermediates for protecting and repairing the brain^7,11 (Supplementary Figure 1). This was confirmed in the 7.9:1 mean ratio of 2,3-¹³C₂ lactate to 3-¹³C lactate in this study (Table 2). There was a close positive linear correlation between 2,3-¹³C₂ lactate and 3-¹³C lactate (Pearson’s r² = 0.73 for fractional enrichments and r² = 0.94 for concentrations) (Figure 3(b) and (c) respectively). Thus, the minor PPP lactate production mirrored the major glycolytic lactate production. The PPP may be increased after TBI, evidenced by rodent studies of moderate-severe TBI,^28,29 but a microdialysis study in human non-TBI controls and TBI patients suggested a more varied response in the ratio of PPP to glycolytic metabolism, with the tendency for more patients to have a reduction in their relative PPP metabolism.¹¹

TCA cycle activity – absence of recovered ¹³C glutamine

No evidence of supplemented 1,2-¹³C₂ glucose metabolism via the TCA cycle was found using ¹³C NMR analysis of microdialysis samples as ¹³C glutamine was only found at its natural-abundance background level. Previous microdialysis studies in TBI and non-TBI surgical patients in which ¹³C glucose was delivered at lower concentrations to the human brain similarly have failed to produce ¹³C glutamine enrichment.^3,11 Conversely, clear evidence of TCA cycle ¹³C labelling in glutamine was found when the three other substrates 2,3-¹³C₂ succinate, 2-¹³C acetate, and 3-¹³C lactate were perfused by microdialysis (Figure 4).^3,16,17

Literature from magnetic resonance studies shows that intravenous supplementation of ¹³C-labelled glucose can be metabolised by the brain in-vivo via the TCA cycle, evidenced by detection of ¹³C-label in glutamine and glutamate. Those studies address whole brain tissue, dominated by the intracellular compartment, thereby differing from microdialysis that samples extracellular molecules. Studies in healthy human volunteers and animals receiving intravenous ¹³C-glucose have shown ample evidence of TCA cycle activity by cerebral ¹³C MRS in-vivo,^30,31 however this has not yet been performed in TBI patients. In animals, intravenous ¹³C-glucose both in TBI-models and controls showed TCA cycle evidence, by ex-vivo NMR analysis of ¹³C in brain tissue extracts, showing ¹³C labelling in glutamate and/or glutamine.^28,29

Contrary to evidence from brain tissue, evidence of TCA cycle production of ¹³C glutamine from microdialysis delivery of ¹³C glucose seems consistently lacking, previously ascribed to ¹³C label dilution by endogenous material given the lower glucose concentrations perfused, as many biosynthetic steps and intermediates are involved in the pathway between glucose and glutamate-glutamine.^3,11 TCA cycle ¹³C labelling of glutamine by microdialysis delivery of other substates – 2,3-¹³C₂ succinate,¹⁷ 2-¹³C acetate,³ and 3-¹³C lactate^3,16 – is explicable because these three substrates are closer (in the number of main biosynthetic steps) to the TCA cycle than glucose – succinate is itself a TCA cycle intermediate, acetate is one step from the TCA cycle, and lactate is three steps from the TCA cycle. Microdialysis delivery of 3-¹³C lactate (8 mmol/L) with 99% fractional enrichment resulted in ¹³C labelling in the product glutamine, showing NMR singlets for C4, C3 and C2 with maximal fractional enrichments of 15.8%, 11.9% and 7.5% respectively.¹⁶ Contrastingly, in the present study, microdialysis administration of 1,2-¹³C₂ glucose (8 mmol/L,) with a 99% fractional enrichment only resulted in 2,3-¹³C₂ lactate and 3-¹³C lactate with maximal fractional enrichments of 31% and 3.6% respectively (average total ¹³C lactate enrichment 16.7%; 0.45 mmol/L), and no detectable ¹³C-labelling in glutamine. Thus, despite 1,2-¹³C₂ glucose administration at a supra-physiologically high concentration and high ¹³C fractional enrichment, insufficient enrichment was attained at the metabolic stage of lactate, and consequently undetectable ¹³C-labelling in glutamine.

Another possible explanation of lack of ¹³C labelling in glutamine is mitochondrial dysfunction. In an earlier study of 3-¹³C lactate (8 mmol/L) given by microdialysis, most of the patients showed TCA cycle metabolism evident from ¹³C labelling in glutamine, but a few were ‘non-responders’ who gave no ¹³C-glutamine production, attributed to mitochondrial dysfunction.¹⁶ However, it is unlikely that all of the patients tested with 1,2-¹³C₂ glucose in the present and previous study¹¹ would have been non-responders because of mitochondrial dysfunction, particularly as many of them had baseline LPR < 25.

Patients with PbtO₂ monitoring showed mean 27 mmHg, with none <15 mmHg, thus not hypoxic, and all 20 patients in the present study were treated with the same protocol-driven therapy aimed at adequate cerebral perfusion. Thus, the absence of TCA cycle ¹³C labelling in glutamine is unlikely to be due to oxygen deficiency. Hyperoxia may suppress cellular glycolysis;³² its effects are variable on cerebral microdialysis lactate and LPR, and may increase cerebral oxygen extraction in acute severe TBI patients.³³ However, none of the patients in our study demonstrated cerebral hyperoxia (highest PbtO₂ 38 mmHg; patient GL-13), nor were they exposed to extreme arterial hyperoxaemia (PaO₂ ≤17 kPa; except patient GL-05 whose mean PaO₂ was 19 kPa).

Extracellular lactate as a key step in brain metabolism

The lack of ¹³C glutamine detected in this study suggests that extracellular lactate is an important intermediate step in brain glucose metabolism (Figure 4). Since we achieved an average 16.7% enrichment of ¹³C lactate by glycolysis of ¹³C glucose, similar enrichment would be expected in equivalent downstream metabolites within a single cell ‘independently’ performing all steps of glycolysis and the TCA cycle. However, if the ¹³C label was further diluted by its export as ¹³C lactate to the extracellular environment where it mixed with unlabelled lactate, before a population of cells metabolised it to glutamine, the final ¹³C glutamine enrichment may be insufficient to detect. When 3-¹³C lactate (99% enrichment) is supplemented directly at 4 or 8 mmol/L concentration, ¹³C glutamine is detected,^3,16 whereas when 3-¹³C lactate is delivered indirectly (at average 0.45 mmol/L concentration, 16.7% enrichment) in this study by supplementation with 1,2-¹³C₂ glucose at 8 mmol/L ¹³C, glutamine is not detected.

Debate exists whether astrocytic glycolysis-derived lactate is the preferred energy substrate for neurons as in the astrocyte-neuron lactate shuttle hypothesis (ANLSH – see below), or whether both astrocytes and neurons independently metabolise glucose per their needs.^34
–36 Supporting the ‘independent model’ (vs. ANLSH) is a kinetic metabolic modelling study in rats.³⁵

In the ANLSH (Figure 4),^2,34,36 metabolic trafficking between astrocytes and neurons can be summarised as follows.³ Glucose from the vasculature is taken up by astrocytes that metabolise it by glycolysis to form lactate that is exported out of the astrocyte, then taken up by neurons that metabolise it via the TCA cycle. A portion of the TCA cycle intermediate alpha-ketoglutarate is converted into the spin-off product glutamate released from the pre-synaptic neuron into the synaptic cleft where it can stimulate glutamate receptors on the post-synaptic neuron. Extracellular glutamate is taken up by astrocytes, converted into glutamine³⁷ and exported into the extracellular fluid, then taken up by neurons and converted back into glutamate.³⁸ Outside of the synaptic cleft, extracellular glutamate concentrations are typically low (1–20 µmol/L, rising to 100–200 µmol/L in ischaemia); glutamine concentrations are typically 400–1000 µmol/L.^39,40

Both glucose metabolic models – ‘independent’ and ANLSH – feature glutamate-glutamine cycling between neurons and astrocytes.³⁵ Our study’s lack of ¹³C glutamine when microdialysis-delivered 1,2-¹³C₂ glucose is the substrate is compatible with the ANLSH – implying that lactate is the favoured substrate for the neuronal TCA cycle.⁴¹ Therefore, when 1,2-¹³C₂ glucose is the substrate (even when supra-physiological and highly enriched) the ensuing 2,3-¹³C₂ lactate becomes too diluted with unlabelled lactate. Moreover, the astrocytic glycolysis-derived lactate is not only self-diluted with the lactate from the unlabelled half of the 1,2-¹³C₂ glucose molecule (C4, C5 and C6), but also exits into the extracellular pool, where it can mix with endogenous lactate, and is then taken up by neurons for TCA cycle processing resulting in glutamine. Furthermore, astrocytes/glia can themselves synthesise glutamate de novo via the TCA cycle alpha-ketoglutarate, and ensuing glutamate converted to glutamine by the same astrocytes/glia.^42,43

In future, self-dilution could be overcome by using uniformly ¹³C labelled glucose. Glycolysis then yields two molecules of 1,2,3-¹³C₃ pyruvate, then forms 1,2,3-¹³C₃ lactate, with no self-dilution. This would facilitate detecting TCA cycle products, albeit with more complex ¹³C NMR spectral splitting patterns thus smaller peaks.

Clinical relevance of glucose delivery and alternative fuels in TBI

The ISCUSflex and ¹³C results of this and previous¹¹ studies together suggest that glucose supplementation may have limited utility in human TBI. Evidence for glycolysis (and minor PPP) rather than TCA cycle were found, but no shift towards high-ATP-yield oxidative TCA cycle metabolism (no decrease in LPR and no detectable ¹³C glutamine). Furthermore, glycolysis, which produces 2 moles of ATP per mole of glucose metabolised, only increased in patients with either low baseline glucose, or evidence of functioning metabolism producing higher concentrations of extracellular lactate and pyruvate at baseline.

A caveat is that glucose was delivered directly into the brain extracellular fluid, which may be less efficient than by its ‘natural’ route from the systemic circulation. Blood glucose concentration and glycaemic control normally influence brain glucose, though this relationship may be weaker in injured brain.^{12,14,44
–49} In the present study there was a non-significant positive trend (Pearson r² = 0.17, p = 0.08) between blood glucose and brain microdialysate (ISCUSflex) glucose concentrations during the microdialysis supplementation period, but no significant relationships between blood glucose concentrations and brain microdialysate (ISCUSflex) concentrations of lactate, pyruvate or LPR. Elsewhere, blood glucose levels reportedly influenced brain metabolism in TBI patients, evidenced by arterial-jugular venous difference measurements⁵⁰ and cerebral microdialysis.^51,52 In these, blood glucose levels 6–9 mmol and cerebral microdialysate glucose 1–5 mmol were associated with minimizing the corresponding microdialysate LPR and glutamate concentration. High microdialysate glucose >5 mmol was associated with high glutamate, while low microdialysate glucose <1 mmol was associated with high LPR, in a small study.⁵¹ Moreover, data from 619 TBI patients also showed high microdialysate LPR associated with low microdialysate glucose.⁵³ In the present small study, there was no significant relationship between microdialysate LPR and microdialysate glucose.

Delivering additional glucose to the whole brain using intravenous infusions to achieve iatrogenic hyperglycaemia may have risk. A large study of general critical care patients (mainly non-TBI) revealed worse outcome in those with loosely-controlled serum glucose,⁵⁴ but unconfirmed in a later, similar-sized study.⁵⁵ A more recent meta-analysis in TBI patients found that tight glucose control was associated with better neurological outcome than loose.⁵⁶ Conversely, other TBI studies suggested loose glucose control was better than tight.^12,14 An observational study of 86 TBI patients⁵⁷ with blood glucose targeted between 6-8 mmol/L (using an insulin sliding scale) achieving a mean 6.6 (SD 1.1) mmol/L concluded that increased blood glucose may impair cerebrovascular reactivity, suggesting mechanistic linkage between increased blood glucose and poorer outcome post-TBI. Defining the optimal ranges of both cerebral and plasma glucose is beyond the scope of the present study and requires well-designed prospective studies. However, the present study does indicate that adding glucose focally to the extracellular space does not cause a major change in LPR: administration of 4 mM or 8 mM glucose via the microdialysis catheter resulted in mean LPR increases of 5% and 1.5% respectively.

Discussion grows on whether substrates other than glucose have therapeutic potential in TBI. Debate exists⁵⁸ on the suitability of administering exogenous lactate for TBI therapy,^59
–61 although it is clearly metabolised via the TCA cycle,¹⁶ and its metabolism appears unsuppressed after TBI.⁴ However, the administration time post-injury may be important.⁵ Evidence increases that succinate may support metabolism of the acutely traumatised human brain.^17,62,63 Whether such alternative substrates have therapeutic roles need more extensive comparative studies against glucose.

Conclusions

Using a reductionist approach, directly delivering glucose into the extracellular space of the traumatised human brain, we have shown limited ability of the brain to metabolise additional glucose in only some patients. This appears to be principally determined by a patient’s underlying disturbance in their injured brain’s ability to metabolise glucose, or a saturation of their existing metabolic pathways, as its utilisation may be greater in cases of relative neuroglycopaenia.

As expected, we found that supplemented ¹³C glucose yielded extracellular ¹³C lactate derived from both glycolysis and the PPP. However, ¹³C lactate did not attain sufficient enrichment to allow subsequent detection of TCA cycle spin-out products, thus no ¹³C glutamine was detected – despite it being found in studies that directly supplemented ¹³C lactate or ¹³C acetate.^3,16 Our findings suggest dilution of the ¹³C label with extracellular endogenous metabolites occurs after glycolysis, and during glycolysis where self-dilution with unlabelled carbons from the breakup of 1,2-¹³C₂ glucose forming a 1:1 mixture of labelled and unlabelled pyruvate occurs. The metabolism of glucose and its products appears likely to be shared between multiple cells in-vivo, rather than occurring to completion in a single ‘stand-alone’ cell, and extracellular lactate is an important step for such appearance of extracellular glutamine in the traumatised brain. Further studies in TBI patients with intravenous ¹³C-labelled glucose and monitoring by in-vivo MRS and cerebral microdialysis are warranted, to address respectively the intracellular and extracellular compartments.

Supplemental Material

sj-pdf-1-jcb-10.1177_0271678X231173584 - Supplemental material for High-physiological and supra-physiological 1,2-¹³C₂ glucose focal supplementation to the traumatised human brain

Supplemental material, sj-pdf-1-jcb-10.1177_0271678X231173584 for High-physiological and supra-physiological 1,2-¹³C₂ glucose focal supplementation to the traumatised human brain by Matthew G Stovell, Duncan J Howe, Eric P Thelin, Ibrahim Jalloh, Adel Helmy, Mathew R Guilfoyle, Peter Grice, Andrew Mason, Susan Giorgi-Coll, Clare N Gallagher, Michael P Murphy, David K Menon, T Adrian Carpenter, Peter J Hutchinson and Keri LH Carpenter in Journal of Cerebral Blood Flow & Metabolism

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Medical Research Council (Grant Nos. G0600986 ID79068 and G1002277 ID98489) and National Institute for Heath and Care Research Biomedical Research Centre, Cambridge (Neuroscience Theme; Brain Injury and Repair Theme). Authors’ support: MGS – National Institute for Health and Care Research Biomedical Research Centre, Cambridge; Cambridge; EPT – Strategic Research Area for Neuroscience (StratNeuro, Karolinska Institutet), The Erling-Persson Family Foundation, Region Stockholm (Clinical Research Appointment) and the Swedish Brain Foundation (#FO2019-0006); KLHC – National Institute for Health and Care Research (NIHR) Biomedical Research Centre, Cambridge (Neuroscience Theme; Brain Injury and Repair Theme), by an NIHR i4i Challenge Award II-C5-0715-20005, an NIHR i4i Product Development Award NIHR200986, and the NIHR Brain Injury MedTech Co-operative; CG – the Canadian Institute of Health Research; AH – Medical Research Council/Royal College of Surgeons of England Clinical Research Training Fellowship (Grant no. G0802251), Royal College of Surgeons of England Pump Priming Grant; DKM – National Institute for Health and Care Research Senior Investigator Awards; PJH – the NIHR (Senior Investigator Award, Cambridge BRC; Medtech Co-operative on Acute Brain Injury) and the Royal College of Surgeons of England. MPM – Medical Research Council UK (MC_UU_00028/4) and a Wellcome Trust Investigator award (220257/Z/20/Z). The views expressed in this paper are those of the Authors and are not necessarily those of the NIHR or any of the other funding bodies.

Acknowledgements

We thank Prof. John D. Pickard for his advice and support setting up the study, the patients for participating, and their relatives for their assent.

Declaration of conflicting interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: PJH is a Director of Technicam, the company that manufactures the triple lumen cranial access device used in the study. PJH has no competing non-financial interests with this study. The other authors have no competing financial interests.

Authors’ contributions

The study concept was by PJH, DKM, KLHC, AH, TAC, MPM and CNG. The study design was by AH, IJ and TAC. MGS, DJH, IJ, PG, AM and SGC performed the experimental work. MGS, EPT, MRG, AH, IJ, MPM, DKM, TAC, KLHC and PJH assessed and analysed the data. MGS, EPT, KLHC, DKM and PJH wrote the manuscript. All authors reviewed and edited the manuscript.

Supplemental material

Supplemental material for this article is available online.

ORCID iDs

Matthew G Stovell

Adel Helmy

Andrew Mason

Clare N Gallagher

References

Maas

AIR

Menon

Adelson

, et al. Traumatic brain injury: integrated approaches to improve prevention, clinical care, and research. Lancet Neurol 2017; 16: 987–1048.

Bouzier-Sore

A-K

Voisin

Canioni

, et al. Lactate is a preferential oxidative energy substrate over glucose for neurons in culture. J Cereb Blood Flow Metab 2003; 23: 1298–1306.

Gallagher

Carpenter

KLH

Grice

, et al. The human brain utilizes lactate via the tricarboxylic acid cycle: a 13C-labelled microdialysis and high-resolution nuclear magnetic resonance study. Brain 2009; 132: 2839–2849.

Glenn

Martin

Horning

, et al. Lactate: brain fuel in human traumatic brain injury: a comparison with normal healthy control subjects. J Neurotrauma 2015; 32: 820–832.

Greco

Vespa

Prins

ML.

Alternative substrate metabolism depends on cerebral metabolic state following traumatic brain injury. Exp Neurol 2020; 329: 113289.

Prins

Greco

Alexander

, et al. The pathophysiology of traumatic brain injury at a glance. Dis Model Mech 2013; 6: 1307–1315.

Carpenter

KLH

Jalloh

Gallagher

, et al. 13C-labelled microdialysis studies of cerebral metabolism in TBI patients. Eur J Pharm Sci 2014; 57: 87–97.

Hutchinson

Jalloh

Helmy

, et al. Consensus statement from the 2014 international microdialysis forum. Intensive Care Med 2015; 41: 1517–1528.

Timofeev

Carpenter

KLH

Nortje

, et al. Cerebral extracellular chemistry and outcome following traumatic brain injury: a microdialysis study of 223 patients. Brain 2011; 134: 484–494.

10.

Reinstrup

Ståhl

Mellergård

, et al. Intracerebral microdialysis in clinical practice: baseline values for chemical markers during wakefulness, anesthesia, and neurosurgery. Neurosurgery 2000; 47: 701–710. discussion 709–710.

11.

Jalloh

Carpenter

KLH

Grice

, et al. Glycolysis and the pentose phosphate pathway after human traumatic brain injury: microdialysis studies using 1,2-(13)C2 glucose. J Cereb Blood Flow Metab 2015; 35: 111–120.

12.

Vespa

McArthur

Stein

, et al. Tight glycemic control increases metabolic distress in traumatic brain injury: a randomized controlled within-subjects trial. Crit Care Med 2012; 40: 1923–1929.

13.

Plummer

Notkina

Timofeev

, et al. Cerebral metabolic effects of strict versus conventional glycaemic targets following severe traumatic brain injury. Crit Care 2018; 22: 16.

14.

Oddo

Schmidt

Carrera

, et al. Impact of tight glycemic control on cerebral glucose metabolism after severe brain injury: a microdialysis study. Crit Care Med 2008; 36: 3233–3238.

15.

Hutchinson

O'Connell

Seal

, et al. A combined microdialysis and FDG-PET study of glucose metabolism in head injury. Acta Neurochir (Wien) 2009; 151: 51–61; discussion 61.

16.

Jalloh

Helmy

Howe

, et al. A comparison of oxidative lactate metabolism in traumatically injured brain and control brain. J Neurotrauma 2018; 35: 2025–2035.

17.

Jalloh

Helmy

Howe

, et al. Focally perfused succinate potentiates brain metabolism in head injury patients. J Cereb Blood Flow Metab 2017; 37: 2626–2638.

18.

Menon

Ercole

Critical care management of traumatic brain injury. Handb Clin Neurol 2017; 140: 239–274.

19.

Ulrich

Akutsu

Doreleijers

, et al. BioMagResBank. Nucleic Acids Res 2008; 36: 402–408.

20.

Wishart

Tzur

Knox

, et al. HMDB: the human metabolome database. Nucleic Acids Res 2007; 35: 521–526.

21.

Bates

Mächler

Bolker

, et al. Fitting linear Mixed-Effects models using lme4. J Stat Soft 2015; 67. DOI: 10.18637/jss.v067.i01.

22.

Teasdale

Pettigrew

Wilson

, et al. Analyzing outcome of treatment of severe head injury: a review and update on advancing the use of the Glasgow Outcome Scale. J Neurotrauma 1998; 15: 587–597.

23.

Stein

McArthur

Etchepare

, et al. Early cerebral metabolic crisis after TBI influences outcome despite adequate hemodynamic resuscitation. Neurocrit Care 2012; 17: 49–57.

24.

Tomlinson

Gardiner

NJ.

Glucose neurotoxicity. Nat Rev Neurosci 2008; 9: 36–45.

25.

Jauch-Chara

Oltmanns

KM.

Glycemic control after brain injury: boon and bane for the brain. Neuroscience 2014; 283: 202–209.

26.

Berg

Jeremy

Tymoczko

, et al. Biochemistry. 5th ed. New York: W.H. Freeman, 2002.

27.

Sonneville

den Hertog

Güiza

, et al. Impact of hyperglycemia on neuropathological alterations during critical illness. J Clin Endocrinol Metab 2012; 97: 2113–2123.

28.

Bartnik

Sutton

Fukushima

, et al. Upregulation of pentose phosphate pathway and preservation of tricarboxylic acid cycle flux after experimental brain injury. J Neurotrauma 2005; 22: 1052–1065.

29.

Bartnik

Lee

Hovda

, et al. The fate of glucose during the period of decreased metabolism after fluid percussion injury: a 13C NMR study. J Neurotrauma 2007; 24: 1079–1092.

30.

de Graaf

Mason

Patel

, et al. In vivo1H-[13C]-NMR spectroscopy of cerebral metabolism. NMR Biomed 2003; 16: 339–357.

31.

Rothman

de Feyter

de Graaf

, et al. 13C MRS studies of neuroenergetics and neurotransmitter cycling in humans. NMR Biomed 2011; 24: 943–957.

32.

Das

KC.

Hyperoxia decreases glycolytic capacity, glycolytic reserve and oxidative phosphorylation in MLE-12 cells and inhibits complex I and II function, but not complex IV in isolated mouse lung mitochondria. PLoS One 2013; 8: e73358.

33.

Nortje

Coles

Timofeev

, et al. Effect of hyperoxia on regional oxygenation and metabolism after severe traumatic brain injury: preliminary findings. Crit Care Med 2008; 36: 273–281.

34.

Pellerin

Magistretti

PJ.

Glutamate uptake into astrocytes stimulates aerobic glycolysis: a mechanism coupling neuronal activity to glucose utilization. Proc Natl Acad Sci U S A 1994; 91: 10625–10629.

35.

Patel

Lai

JCK

Chowdhury

GMI

, et al. Direct evidence for activity-dependent glucose phosphorylation in neurons with implications for the astrocyte-to-neuron lactate shuttle. Proc Natl Acad Sci U S A 2014; 111: 5385–5390.

36.

Pellerin

Magistretti

PJ.

Sweet sixteen for ANLS. J Cereb Blood Flow Metab 2012; 32: 1152–1166.

37.

Norenberg

Martinez-Hernandez

Fine structural localization of glutamine synthetase in astrocytes of rat brain. Brain Res 1979; 161: 303–310.

38.

Aoki

Kaneko

Starr

, et al. Identification of mitochondrial and non-mitochondrial glutaminase within select neurons and glia of rat forebrain by electron microscopic immunocytochemistry. J Neurosci Res 1991; 28: 531–548.

39.

Hutchinson

Gupta

Fryer

, et al. Correlation between cerebral blood flow, substrate delivery, and metabolism in head injury: a combined microdialysis and triple oxygen positron emission tomography study. J Cereb Blood Flow Metab 2002; 22: 735–745.

40.

Samuelsson

Hillered

Zetterling

, et al. Cerebral glutamine and glutamate levels in relation to compromised energy metabolism: a microdialysis study in subarachnoid hemorrhage patients. J Cereb Blood Flow Metab 2007; 27: 1309–1317.

41.

Schurr

Lactate: the ultimate cerebral oxidative energy substrate?

J Cereb Blood Flow Metab 2006; 26: 142–152.

42.

Pardo

Contreras

Satrústegui

De novo synthesis of glial glutamate and glutamine in young mice requires aspartate provided by the neuronal mitochondrial aspartate-glutamate carrier ARALAR/AGC1. Front Endocrinol (Lausanne) 2013; 4: 149.

43.

Rose

Verkhratsky

Parpura

Astrocyte glutamine synthetase: pivotal in health and disease. Biochem Soc Trans 2013; 41: 1518–1524.

44.

Schlenk

Graetz

Nagel

, et al. Insulin-related decrease in cerebral glucose despite normoglycemia in aneurysmal subarachnoid hemorrhage. Crit Care 2008; 12: R9.

45.

Schmidt

Claassen

S-B

, et al. Nutritional support and brain tissue glucose metabolism in poor-grade SAH: a retrospective observational study. Crit Care 2012; 16: R15.

46.

Vespa

Boonyaputthikul

McArthur

, et al. Intensive insulin therapy reduces microdialysis glucose values without altering glucose utilization or improving the lactate/pyruvate ratio after traumatic brain injury. Crit Care Med 2006; 34: 850–856.

47.

Zetterling

Hillered

Enblad

, et al. Relation between brain interstitial and systemic glucose concentrations after subarachnoid hemorrhage. J Neurosurg 2011; 115: 66–74.

48.

Magnoni

Tedesco

Carbonara

, et al. Relationship between systemic glucose and cerebral glucose is preserved in patients with severe traumatic brain injury, but glucose delivery to the brain may become limited when oxidative metabolism is impaired. Crit Care Med 2012; 40: 1785–1791.

49.

Rostami

Bellander

B-M.

Monitoring of glucose in brain, adipose tissue, and peripheral blood in patients with traumatic brain injury: a microdialysis study. J Diabetes Sci Technol 2011; 5: 596–604.

50.

Holbein

Béchir

Ludwig

, et al. Differential influence of arterial blood glucose on cerebral metabolism following severe traumatic brain injury. Crit Care 2009; 13: R13.

51.

Meierhans

Béchir

Ludwig

, et al. Brain metabolism is significantly impaired at blood glucose below 6 mM and brain glucose below 1 mM in patients with severe traumatic brain injury. Crit Care 2010; 14: R13.

52.

Diaz-Parejo

Ståhl

, et al. Cerebral energy metabolism during transient hyperglycemia in patients with severe brain trauma. Intensive Care Med 2003; 29: 544–550.

53.

Guilfoyle

Helmy

Donnelly

, et al. Characterising the dynamics of cerebral metabolic dysfunction following traumatic brain injury: a microdialysis study in 619 patients. PLoS One 2021; 16: e0260291.

54.

van den Berghe

Wouters

Weekers

, et al. Intensive insulin therapy in critically ill patients. N Engl J Med 2001; 345: 1359–1367.

55.

NICE-SUGAR Study Investigators; Finfer

Chittock

SY-S

, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med 2009; 360: 1283–1297.

56.

Hermanides

Plummer

Finnis

, et al. Glycaemic control targets after traumatic brain injury: a systematic review and meta-analysis. Crit Care 2018; 22: 11.

57.

Donnelly

Czosnyka

Sudhan

, et al. Increased blood glucose is related to disturbed cerebrovascular pressure reactivity after traumatic brain injury. Neurocrit Care 2015; 22: 20–25.

58.

Dienel

Rothman

Nordström

C-H.

Microdialysate concentration changes do not provide sufficient information to evaluate metabolic effects of lactate supplementation in brain-injured patients. J Cereb Blood Flow Metab 2016; 36: 1844–1864.

59.

Bouzat

Sala

Suys

, et al. Cerebral metabolic effects of exogenous lactate supplementation on the injured human brain. Intensive Care Med 2014; 40: 412–421.

60.

Ichai

Armando

Orban

J-C

, et al. Sodium lactate versus mannitol in the treatment of intracranial hypertensive episodes in severe traumatic brain-injured patients. Intensive Care Med 2009; 35: 471–479.

61.

Ichai

Payen

J-FJ-F

Orban

J-CJ-C

, et al. Half-molar sodium lactate infusion to prevent intracranial hypertensive episodes in severe traumatic brain injured patients: a randomized controlled trial. Intensive Care Med 2013; 39: 1413–1422.

62.

Stovell

Mada

Helmy

, et al. The effect of succinate on brain NADH/NAD+ redox state and high energy phosphate metabolism in acute traumatic brain injury. Sci Rep 2018; 8: 11140.

63.

Khellaf

Garcia

Tajsic

, et al. Focally administered succinate improves cerebral metabolism in traumatic brain injury patients with mitochondrial dysfunction. J Cereb Blood Flow Metab 2022; 42: 39–55. 0271678X2110421.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.88 MB

High-physiological and supra-physiological 1,2- 13 C 2 glucose focal supplementation to the traumatised human brain

Abstract

Keywords

Introduction

Materials and methods

Patients

Perfusion fluid and 13C-labelled substrate

Microdialysis technique

High resolution NMR acquisition and analysis

Statistical analysis

Results

Patients and monitoring

Patient baseline characteristics

Evidence from 13C labelling patterns

Glucose supplementation at 4 mmol/L: brain biochemistry compared to baseline

Glucose supplementation at 8 mmol/L: brain biochemistry compared to baseline

Glucose supplementation in the neuroglycopenic brain

NMR analysis of microdialysates from 8 mmol/L 1,2-13C2 glucose perfusion

Discussion

Significance of 13C-labelling in metabolites

Limited effect on glycolytic activity – changes in extracellular lactate and pyruvate measured by bedside ISCUSflex

Extracellular 13C lactate – dilution, glycolysis, and the pentose phosphate pathway

TCA cycle activity – absence of recovered 13C glutamine

Extracellular lactate as a key step in brain metabolism

Clinical relevance of glucose delivery and alternative fuels in TBI

Conclusions

Supplemental Material

sj-pdf-1-jcb-10.1177_0271678X231173584 - Supplemental material for High-physiological and supra-physiological 1,2-13C2 glucose focal supplementation to the traumatised human brain

Footnotes

Funding

Acknowledgements

Declaration of conflicting interests

Authors’ contributions

Supplemental material

ORCID iDs

References

Supplementary Material

Perfusion fluid and ¹³C-labelled substrate

Evidence from ¹³C labelling patterns

NMR analysis of microdialysates from 8 mmol/L 1,2-¹³C₂ glucose perfusion

Significance of ¹³C-labelling in metabolites

Extracellular ¹³C lactate – dilution, glycolysis, and the pentose phosphate pathway

TCA cycle activity – absence of recovered ¹³C glutamine

sj-pdf-1-jcb-10.1177_0271678X231173584 - Supplemental material for High-physiological and supra-physiological 1,2-¹³C₂ glucose focal supplementation to the traumatised human brain