• amiloride;
  • birth asphyxia;
  • hypoxia-ischaemia;
  • neonatal encephalopathy;
  • neuroprotection;
  • NHE blockade


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. Disclosures/Conflicts of interest
  9. References
  10. Supporting Information

Na+/H+ exchanger (NHE) blockade attenuates the detrimental consequences of ischaemia and reperfusion in myocardium and brain in adult and neonatal animal studies. Our aim was to use magnetic resonance spectroscopy (MRS) biomarkers and immunohistochemistry to investigate the cerebral effects of the NHE inhibitor, methyl isobutyl amiloride (MIA) given after severe perinatal asphyxia in the piglet. Eighteen male piglets (aged < 24 h) underwent transient global cerebral hypoxia-ischaemia and were randomized to (i) saline placebo; or (ii) 3 mg/kg intravenous MIA administered 10 min post-insult and 8 hourly thereafter. Serial phosphorus-31 (31P) and proton (1H) MRS data were acquired before, during and up to 48 h after hypoxia-ischaemia and metabolite-ratio time-series Area under the Curve (AUC) calculated. At 48 h, histological and immunohistochemical assessments quantified regional tissue injury. MIA decreased thalamic lactate/N-acetylaspartate and lactate/creatine AUCs (both p < 0.05) compared with placebo. Correlating with improved cerebral energy metabolism, transferase mediated biotinylated d-UTP nick end-labelling (TUNEL) positive cell density was reduced in the MIA group in cerebral cortex, thalamus and white matter (all p < 0.05) and caspase 3 immunoreactive cells were reduced in pyriform cortex and caudate nucleus (both p < 0.05). Microglial activation was reduced in pyriform and midtemporal cortex (both p < 0.05). Treatment with MIA starting 10 min after hypoxia-ischaemia was neuroprotective in this perinatal asphyxia model.

Abbreviations used



phosphorus -31


acute energy depletion


advanced method for accurate, robust and efficient spectral fitting with the use of prior knowledge


area over the curve


area under the curve








exchangeable phosphate pool


fraction of inspired oxygen




ionized calcium-binding adaptor molecule 1






mean arterial blood pressure


magnesium and pH from ATP calculation


methyl isobutyl-amiloride


magnetic resonance spectroscopy


N acetyl aspartate


Na+/H+ exchanger isoform 1


Na+/H+ exchanger


nuceltide triphosphate


partial pressure of carbon dioxide


partial pressure of oxygen






intracellular pH


inorganic phosphate


point resolved spectroscopy


echo time


repeat time


transferase mediated biotinylated d-UTP nick end-labelling


ventromedial forebrain


white matter

In the UK and other high-income countries, moderate-severe encephalopathy following perinatal asphyxia occurs in 1–3/1000 term births (Kurinczuk et al. 2010) and is a major cause of neonatal death and disability. Therapeutic hypothermia is the first treatment of newborns for neonatal encephalopathy with consistent evidence of reduced risk of death and severe disability (Edwards et al. 2010). Further neuroprotective agents are needed either to augment hypothermic neuroprotection in severe cases (Kelen and Robertson 2010) or to use in settings where hypothermia has not been established as standard care.

Disruption of ionic homeostasis is a significant consequence of hypoxia-ischaemia (HI) and contributes to the evolving cascade of neurotoxic death (Masereel et al. 2003). The Na+/H+ exchangers (NHEs) are a family of ion transport proteins that maintain normal intracellular pH (pHi) and cell volume (Masereel et al. 2003; Slepkov et al. 2007). The NHEs conduct the electroneutral exchange of protons (H+) and sodium (Na+) ions across cellular membranes down their concentration gradients. Excessive NHE activation during HI leads to intracellular Na+ overload, which subsequently promotes Ca2+ entry via reversal of the Na+/Ca2+ exchanger. Increased cytosolic Ca2+ then triggers events in the neurotoxic cascade leading to cell death (Cheung et al. 1986). Activation of NHE also leads to the rapid normalization of pHi during reperfusion after HI (Wakabayashi et al. 1992). In several cell types including neuronal (Vornov et al. 1996) and cardiac cells (Bond et al. 1991, 1993), a rapid return of pHi to normal or even alkaline levels has been observed and termed the pH paradox, as rather than causing cells to recover, the return of pHi to physiological or alkaline values causes cells to deteriorate. In babies with neonatal encephalopathy studied with phosphorus (31P) magnetic resonance spectroscopy (MRS) an alkaline brain pHi in the first week after birth was associated with an adverse outcome (Robertson et al. 2002).

NHE isoform 1 (NHE 1) plays a role in cerebral damage after HI. NHE1 inhibition, either pharmacologically or by genetic knock-out, attenuated the detrimental consequences of focal cerebral ischaemia in adult mice (Luo et al. 2005) and in a neonatal focal HI model, the potent NHE inhibitor HOE642 reduced neurodegeneration in the striatum, thalamus and hippocampus and improved functional outcome (Cengiz et al. 2011). Using the NHE1 inhibitor N-methyl-isobutyl-amiloride (MIA), preservation of ATP was seen in a neonatal brain slice model (Robertson et al. 2005) and neuroprotection was observed following focal HI in neonatal mice (Kendall et al. 2006). The time course of pHi recovery after cardiac arrest was altered by amiloride (Ferimer et al. 1995). The possibility that reperfusion injury could be ameliorated if brain pHi was slowly rather than abruptly normalized after HI has important implications for neuroprotection.

The aim of this study was to assess the treatment effect of MIA given 10 min after transient global HI in a piglet model of perinatal asphyxia using clinically relevant biomarkers; proton (1H) MRS for lactate, N acetyl aspartate (NAA) and creatine (Cr) and 31P MRS for Phosphocreatine (PCr),nucleotide triphosphate (NTP) and brain pHi. Lac/NAA (Thayyil et al. 2010) and NTP/exchangeable phosphate pool (ePP) (Azzopardi et al. 1989) are sensitive and specific surrogate outcome markers in neonatal encephalopathy. At the end of the 48 h survival period, cell death quantification was performed on tissue stained for transferase mediated biotinylated d-UTP nick end-labelling (TUNEL+) and cleaved caspase 3. Immunohistochemistry for microglial ionized calcium-binding adaptor molecule 1 (IBA1) was performed to reveal the extent of microglial activation.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. Disclosures/Conflicts of interest
  9. References
  10. Supporting Information

Animal preparation and MIA administration

All animal experiments accorded with UK Home Office Guidelines [Animals (Scientific Procedures) Act 1986]. Eighteen large white male piglets aged < 24 h were anesthetized and surgically prepared as described previously (Lorek et al. 1994). Briefly, piglets were sedated with intramuscular midazolam (0.2 mg/kg) and arterial O2 saturation was monitored (Nonin Medical Inc, Plymouth, Minnesota, USA). Isoflurane anaesthesia [initial concentration 4% (v/v)] via a facemask facilitated tracheostomy and intubation and was maintained (3% during surgery and 2% otherwise) with continuous morphine infusion (0.05 mg/kg/h). Mechanical ventilation settings were adjusted to maintain PaO2 and PaCO2 at 8–13 kPa and 4.5–6.5 kPa respectively, allowing for temperature correction of the arterial blood sample.

An umbilical venous catheter was inserted for infusion of maintenance fluids (10% dextrose, 60 mL/kg/day), morphine (10–40 μg/kg/h) and antibiotics (benzylpenicillin 50 mg/kg and gentamicin 2.5 mg/kg, every 12 h). An umbilical arterial catheter was inserted for continuous heart rate and arterial blood pressure monitoring, and intermittent blood sampling for PaO2, PaCO2, pH, electrolytes, glucose and lactate (Abbot Laboratories, Maidenhead, UK). Bolus infusions of colloid (Gelofusin, B Braun Medical Ltd. Emmenbrucke, Switzerland) and dopamine (5–20 μg/kg/min) maintained mean arterial blood pressure (MABP) > 40 mmHg. All animals received continuous physiological monitoring and intensive life support throughout experimentation. Arterial lines were maintained by infusing 0.9% saline (Baxter, 1 mL/h): heparin sodium was added at a concentration of 1 IU/mL to prevent line blockage.

Both common carotid arteries were surgically isolated at the level of the fourth cervical vertebra and encircled by remotely controlled vascular occluders (OC2A, In Vivo Metric, Healdsburg, CA, USA). After surgery, piglets were positioned prone in a plastic pod; the scalp was firmly immobilized below the MRS surface coil and the pod was inserted into the MRS system. Piglets were maintained normothermic (38.5°C) throughout experimentation using a warmed water mattress and circulating warm air.

Cerebral hypoxia-ischaemia

Whilst in the MRS system, transient HI was induced by occluding both common carotid arteries, using the inflatable vascular occluders, and reducing fractional inspired (Fi) O2 to 12% (v/v). During HI cerebral-energetic changes were observed every 2 min by 31P MRS and the β-NTP peak height was measured automatically. HI was continued until 10 min after the β-NTP peak height had fallen to 30% of baseline: during this 10 min period, FiO2 was adjusted to maintain this β-NTP level. After this 10 min, the occluders were deflated and FiO2 normalized. 31P spectra were acquired for a further hour to monitor recovery. The time integral of the reduction in β-NTP relative to the exchangeable phosphate pool [ePP = Pi + PCr + (2γ + β)−NTP] during HI and the first 60 min of resuscitation quantified acute energy depletion (insult severity, AED) as described previously (Lorek et al. 1994).

Sample size

Our previous studies of therapeutic hypothermia after HI have suggested that the change in Lac/NAA over 48 h varied between normothermic and hypothermic groups by 1.0 U with a standard deviation of 0.75 U (both log scale). Assuming a similar effect of MIA compared with placebo, with 5% significance and 90% power, nine subjects would be required in each group.

Experimental groups

Baseline data were acquired before transient HI following stabilization of the animal in the MRS system. Following resuscitation, piglets were randomized into two groups (each n = 9): (i) saline placebo and normal care; and (ii) MIA 3 mg/kg intravenously commencing 10 min post-insult and 8 hourly thereafter and normal care. Normothermia was maintained throughout.

For intravenous administration, 100 mg MIA dissolved in 1 mL of 350 mmol/L acetic acid, and diluted with saline to give 2 mg/mL MIA. pH was checked to ensure normality. Aliquots of 6 mg in 3 mL MIA were made up and stored at −20°C. Just before use, aliquots were defrosted rapidly in a water bath to avoid crystallization.


Spectra were acquired as previously described (Faulkner et al. 2011) using a 6.5 cm × 5.5 cm elliptical transmit-receive surface-coil tunable for both 1H and 31P nuclei in a 4.7 Tesla Biospec Avance (Bruker Medizintechnik, Karlsruhe, Germany). Spectra were acquired prior to HI (after stabilization of the animal in the MRS system), during HI and 1 h recovery (31P only), and thereafter continuously about every ~5 h. During the whole experiment the piglet remained in the bore of the magnet continuously receiving intensive care support and monitoring. 1H PRESS spectra [repetition time (TR) 5 s, 128 summed transients, echo time (TE) 288 ms] were acquired from two voxel positions (Fig. 1a): deep grey matter centred on both lateral thalami and hypothalami [ventromedial forebrain (vmFB); 14 mm × 14 mm × 7 mm voxel]; and dorsal right subcortical white matter at the centrum semiovale level dorsal subcortical white matter (dscWM); 9 mm × 4 mm × 20 mm voxel). 31P MR spectra were acquired from whole brain using single-pulse acquire with TR 10 s.


Figure 1. Methyl isobutyl amiloride (MIA) reduces the cerebral metabolic changes seen by 1H magnetic resonance spectroscopy (MRS) following hypoxia-ischaemia in the neonatal piglet. (a) Coronal image of piglet brain (midparietal plane) showing the 1H MRS voxels in dorsal subcortical (dsc) white matter (WM) and in ventro medial forebrain (vmFB). 31P MRS used a surface coil therefore most signal originated from superficial brain. (b, c) 1H spectra from the WM voxel of a naive piglet (b) and 48 h after transient, global hypoxia-ischaemia (c). Note the increase in lactate (Lac) at 1.3 ppm and the decrease in N-Acetylaspartate (NAA) at 2 ppm. The 3 ppm creatine (Cr) peak remains largely unchanged. (d, e) 31P spectra from a naive piglet (d) and 33 h after hypoxia-ischaemia (e). Note the increased inorganic phosphate (Pi) and decreased phosphocreatine (PCr), phosphoethanolamine (PET) and γ, α and β adenosine triphosphate (ATP) in (e). (f–m) vmFB Lac/Cr (f, j), Cho/Cr (g, k), NAA/Cr (h, l) and Lac/NAA (i, m) log10 peak-area ratios. Individual plots are shown for each animal randomly assigned to placebo or MIA. MIA reduced the overall Lac/Cr and Lac/NAA increases for up to 48 h. (n) AUC summary [mean + standard error (SEM)] for vmFB 1H MRS (left), WM 1H MRS (center) and 31P MRS whole brain (WB; right). Open columns - placebo, filled columns - MIA-treated. Note the consistent lower AUC trend with MIA reaching significance (*< 0.05, t-test) for vmFB Lac/Cr and Lac/NAA. NTP/ePP and pH (Pi) were borderline significant (= 0.056 and 0.053, respectively).

Download figure to PowerPoint

MRS analysis

1H MR spectra were analysed using LC Model (Provencher 1993); representative MR spectra are shown in Fig. 1b–e. For each spectrum the relative signal amplitudes for choline (Cho), total creatine (Cr), N acetyl aspartate (NAA) and lactate (Lac) were measured. The primary outcome measures nominated a priori were Lac/NAA and Lac/Cr. Metabolite peak area under the curve (AUC) ratios were plotted from baseline to 48 h after HI for each voxel and each subject on a logarithmic scale and analysed according to randomized group (Fig. 1f–m).

31P MR spectra were analysed by AMARES (Vanhamme et al. 1997). Signal amplitudes and chemical shifts were measured for PCr, Pi (three independent components fitted) and α-, β-, and γ-NTP; ePP was calculated. Estimations of pHi via Pi (pH (Pi)) and phosphoethanolamine [pH (PEtA)] used previously published chemical-shift titrations (Petroff and Prichard 1985; Corbett et al. 1987). pHi was also estimated using the NTP chemical shifts [pH (ATP)] using ‘magnesium and pH from ATP calculation’ (MAGPAC) (Williams and Smith 1995). Again, metabolite peak AUC ratios were plotted from baseline to 48 h after HI for each voxel and each subject on a logarithmic scale and analysed according to randomized group.

As some piglets died prematurely, we needed to calculate time-series AUC or area over the curve (AOC) consistently, therefore we used a uniform time window of 0 to 49 h post-insult with the value for the first datum extended horizontally to 0 h and that for the last datum to 49 h. The resulting area was divided by 49 h. If MRS started before 0 h, a 0 h datum was interpolated linearly from those directly before and after 0 h.

In the case of pH measures derived from animals in subgroups without secondary brain acidosis shown in Figure S2c, f and i (5/9 of the placebo group, and 7/9 of the MIA-treated group), the data samples were first tested using repeated measures anova followed by a paired t-test for individual time points against the baseline.


Forty-eight hours after HI, piglets were killed with pentobarbital and the brains fixed via cardiac perfusion with cold 4% paraformaldehyde in phosphate-buffered saline. Brains were then dissected out and post-fixed at 4°C in 2% paraformaldehyde for 7 days. Coronal slices (5 mm thick) of the right hemisphere, starting from anterior to the optic chiasma, were embedded in paraffin wax and sectioned to 5 μm thickness and stained for hematoxylin and eosin (H&E). To assess cell death and glial activation, adjacent sections were stained for nuclear DNA fragmentation using histochemistry with transferase mediated biotinylated d-UTP nick end-labelling (TUNEL), and for the appearance of cleaved caspase 3 and IBA1 immunoreactivity (Ito et al. 2001). Per animal, two sections, placed 5 mm apart, were examined with each stain.

For all histo- and immunohistochemical stains, brain sections were dehydrated in xylene (3 × 10 min) and rehydrated in graded ethanol solutions (100–70%), followed by double-distilled water. For TUNEL, sections were pre-treated for 15 min in 3% H2O2 in methanol to remove endogenous peroxidase, followed by a 15 min peptidase pre-digestion with 20 ug/mL Proteinase K (Promega, Southampton, UK) at 65°C, and then incubated at 37°C for 2 h with the TUNEL solution (Roche, Burgess Hill, UK) containing biotinylated dUTP. For immunohistochemistry, sections were processed for antigen retrieval (800 mW microwave irradiation in 0.1 M citrate buffer, 10 min), followed by overnight incubation with primary rabbit antibody against caspase 3 (1 : 500) or IBA1 (1 : 1000), and then 2 h incubation with a biotinylated, secondary goat-anti-rabbit immunoglobulin antibody (1 : 100, Jackson Labs, West Grove, PA, USA). The biotin residues were detected with the avidin-biotinylated horseradish peroxidise complex (ABC, Vector Laboratories, Peterborough, UK) and visualized with diaminobenzidine/H2O2 (Sigma, Gillingham, UK), with CoCl2 and NiCl2 included to intensify TUNEL histochemistry. The sections were counter-stained with haemotoxylin, dehydrated in graded alcohol and xylene and mounted with Depex (VWR, Leighton Buzzard, UK).

For each animal, section and region, TUNEL+ nuclei were counted in three fields (40x magnification, with an area of 0.76 mm2) and the average converted into counts/mm2. The less numerous caspase immunoreactive cells were also counted in three fields (20x magnification, 3.1 mm2). In the parasagittal cortex, midtemporal cortex (insular region) and hippocampus (dentate gyrus), the three fields were from the superficial, middle and deep cortical layers. IBA1+ microglial cell bodies and branch density were determined at 40x magnification using a 0.24 mm × 0.24 mm square grid, placed in three fields for each brain region by counting the number of cell bodies inside the grid [cell body density (CBD)] and the average number of branches (B) crossing the three horizontal (top, middle and bottom) and the three vertical (left, middle, and right) 0.24 mm long gridlines. This determined the microglial ramification index (B2/CBD).

The threshold for statistical significance was p < 0.05. For statistical analysis in individual brain regions, original counts were normalized by log(x + 1) algorithm conversion (Werner et al. 2001) and the differences between the two groups detected using one-way anova, followed by post hoc TUKEY test. Trend analysis was performed across all seven forebrain regions, using one-way anova followed by Tukey. Statistical significance of correlation R2 was assessed by the F-test.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. Disclosures/Conflicts of interest
  9. References
  10. Supporting Information

Physiology data: time changes and between group differences

Physiological measures were determined at baseline and HI end and averaged over the periods 2.5–26 h and 26.5–48 h after HI. Apart from base excess, there were no intergroup differences in baseline physiological and biochemical measures or in insult severity (Table 1). Base excess was increased in the MIA group compared with placebo at baseline prior to MIA administration. At the end of HI and later the only intergroup difference was a consistently lower heart rate in the MIA group (Table 1).

Table 1. Physiological data
VariablesTimeMIA group mean (SD)Placebo group mean (SD)p value (MIA vs. Placebo)
  1. Measurements were made at baseline and during three time periods: 0.5–2 h post-HI (HI end), 2.5–26 h post-HI and 26.5–48 h post-HI. Each variable was averaged over each time period for each animal. Intergroup comparisons used the unpaired t-test (right column).

  2. a

    < 0.05 against baseline value in the same group (MIA treated or placebo) using paired t-test. ° Insufficient data for statistical testing. There were no differences between groups at baseline apart from a slightly more alkaline blood pH in the MIA group. There were no physiological and serum electrolyte or biochemical differences between groups following HI apart from a lower heart rate in the MIA group.

Post-natal age (h)21.9 (13–26.5)23.4 (21.5–28)0.47
Bodyweight (g)1870 (220)1830 (190)0.74
Duration of HI (min)22.16 (3.58)24.83 (7.17)0.34
Insult severity (10−2 h)11.6 (1.4)12.9 (3.0)0.25
Rectal temperature (°C)Baseline38.2 (0.5)38.4 (0.4)0.32
HI end38.7 (0.7)38.6 (0.7)0.67
2.5–26 h 38.7 (0.5)a38.6 (0.2)0.39
26.5–48 h 38.5 (0.3)38.5 (0.2)0.91
Heart rate (min−1)Baseline177 (27)195 (19)0.13
HI end152 (27)a194 (21)0.002
2.5–26 h 147 (15)a183 (13)<0.001
26.5–48 h 130 (15)a162 (13)a<0.001
MABP (mmHg)Baseline47 (9)45 (6)0.67
HI end45 (6)44 (4)0.60
2.5–26 h46 (5)45 (6)0.57
26.5–48 h48 (7)46 (7)0.44
Isoflurane (%) Baseline2.7 (0.5)2.7 (0.3)0.95
HI end2.1 (0.0)(°)
2.5–26 h 2.4 (0.2)2.4 (0.5)0.77
26.5–48 h 2.3 (0.3)2.1 (0.6)0.30
FiO2 (%)Baseline32 (6)33 (3)0.71
HI end35 (6)52 (0)
2.5–26 h 33 (17)35 (9)0.75
26.5–48 h 35 (4)51 (13)0.07
Blood pHBaseline7.46 (0.08)7.46 (0.08)0.94
HI end7.44 (0.01)7.39 (0.01)(°)
2.5–26 h 7.46 (0.08)7.49 (0.10)0.54
26.5–48 h 7.45 (0.07)7.38 (0.20)0.36
PaCO2 (kPa)Baseline6.0 (1.3)5.5 (1.4)0.48
HI end5.9 (0.4)6.8 (0.0)(°)
2.5–26 h 5.6 (1.2)5.1 (1.4)0.47
26.5–48 h 5.3 (1.3)5.6 (1.2)0.62
PaO2 (kPa)Baseline8.8 (1.6)9.2 (1.7)0.62
HI end10.5 (3.0)13.2 (1.6)(°)
2.5–26 h 13.2 (3.7)a13.4 (3.1)a0.89
26.5–48 h 15.0 (1.7)a13.8 (2.1)a0.26
Base excess (mmol/L)Baseline6.9 (1.9)4.3 (2.5)0.02
HI end5.5 (0.7)5.0 (0.4)(°)
2.5–26 h 4.9 (3.1)4.7 (5.0)0.90
26.5–48 h 2.4 (3.2)2.1 (4.8)0.87
Blood Sodium (mmol/L)Baseline141 (6)139 (3)0.28
HI end141 (11)139 (0)(°)
2.5–26 h136 (4)a135 (3)a0.52
26.5–48 h133 (9)a133 (6)a0.89
Blood Potassium (mmol/L)Baseline4.0 (0.8)4.1 (1.0)0.81
HI end3.8 (1.8)4.6 (0.6)(°)
2.5–26 h 5.1 (1.4)a5.1 (1.0)a0.99
26.5–48 h 5.6 (1.3)a5.0 (1.5)0.39
Blood Lactate (mmol/L)Baseline2.5 (0.9)3.6 (1.4)0.06
HI end2.5 (1.2)2.5 (0.0)(°)
2.5–26 h 1.7 (1.0)a1.9 (0.7)a0.66
26.5–48 h 1.5 (0.5)a1.3 (0.7)a0.55

There were no temporal changes in either group for MABP, pH and PaCO2. Compared with baseline, rectal temperature was higher in the MIA group 2.5 to 26 h after HI (Table 1) but no difference was seen between groups at any time. Compared with baseline, heart rate was reduced at HI end and at 2.5–26 h and 26.5–48 h in the MIA group; without treatment heart rate was lower at 26.5–48 h only. PaO2 increases were observed in both the MIA and placebo groups compared with baseline. In both groups, blood potassium levels were increased, and sodium levels reduced between 2.5 and 48 h after HI (with the exception of the potassium in the placebo group losing any difference compared with baseline at the 26.5–48 h time period). Blood lactate was lower compared with baseline in both groups following HI. There was no difference in base excess compared with baseline or group following HI.

Two placebo piglets died at 24 and 32 h and 1 MIA piglet died at 30 h; all deaths were because of complications secondary to severe HI and progressive cardiovascular decline not responsive to resuscitation. All other subjects survived until 48 h after HI.

1H and 31P MRS

Spectra from all 18 subjects were analysed. Representative 1H and 31P MR spectra of naïve and injured brain are shown in Fig. 1b–e. In the thalamic voxel [ventromedial forebrain (vmFB)] Lac/Cr and Lac/NAA increased after HI, NAA/Cr decreased and Cho/Cr was unchanged (Fig. 1f–m). Although there was considerable variability in the onset, extent and trajectory of metabolic abnormality with each group, several placebo animals (e.g. #19 and #42) displayed very pronounced, early abnormalities: such abnormalities were clearly less conspicuous in the MIA group. AUC analysis (Fig. 1n) revealed lower thalamic Lac/Cr-AUC and Lac/NAA-AUC with MIA (< 0.05, t-test): NAA/Cr-AOC showed no MIA effect (= 0.069).

WM abnormalities after HI resembled those in the thalamus with increased Lac/Cr and Lac/NAA, decreased NAA/Cr and unchanged Cho/Cr (Figure S1a–h). However, AUC and AOC results were similar for MIA and placebo (Fig. 1n, middle).

31P MRS showed a gradual increase in Pi/EPP (Figure S1i and l), and decreases in PCr/EPP (Figure S1j and m), NTP/EPP (Figure S1k and n) and pHi(Pi) (Figure S2a and b) in the severely affected animals: the MIA NTP/EPP-AOC and pHi(Pi)-AOC decreases did not reach significance versus saline (= 0.056 and 0.053 respectively).

If the severely affected piglets that died prematurely are removed from the MRS analysis, significance was lost for any protective effect of MIA. However, we designed the study to replicate as much as possible the successful clinical hypothermia neuroprotection trials in babies. In these studies an adverse outcome is defined as those babies who die and those with an abnormal neurodevelopmental outcome – i.e. the deaths and severe neurodevelopmental outcome are grouped together (Edwards et al. 2010).

31P MRS brain pHi

As MIA inhibition of the Na+/H+ transporters responsible for proton efflux could influence pHi, we conducted a detailed analysis using separate titration algorithms based on the Pi, PETA and ATP chemical shifts (Petroff and Prichard 1985; Corbett et al. 1987; Williams and Smith 1995). 4/9 placebo animals (#14,#19,#25,and#42) demonstrated clear reductions in pH(Pi) to ≤ 6.92 within 40 h of HI (Figure S2a). With MIA only 2/9 (#13 and #22) had such a decrease (Figure S2b) and only later than 40 h after HI (< 0.05 in χ2 test). pHi(PEtA) changes were similar to those for pHi(Pi) (Figure S2d and e) whereas pHi(ATP) showed no acidosis in either group (Figure S2g and h): the latter could be because of decreased NTP signal reducing sensitivity (Fig. 1e). Direct comparison of pHi(Pi) (Figure S2a and b) with PCr/ePP and NTP/ePP (Figure S1j, k, m and n) for animals with acidic pHi(Pi) showed that the first time points with pH(Pi) ≤ 6.92 were associated with PCr/ePP ~55% and NTP/EPP ~60% of their baselines.

Nevertheless, because large piglet subgroups (5/9 placebo; 7/9 MIA) did not demonstrate acidosis after HI, we next assessed whether MIA modulated pHi in the non-acidic subgroup. The non-acidotic placebo subgroup revealed mild but significant pHi(Pi) and pHi(PEtA) increases versus baseline of ~0.05–0.09 2–30 h following HI, (Figure S2c and f; < 0.05 for both, paired two-sided t-test); pHi(ATP) showed no such increase (Figure S2i). The baseline comparison of the non-acidotic MIA subgroup (Figure S2c, f and i) revealed no pHi(PETA) or pHi(ATP) changes. However, pHi(Pi) was increased at 24–48 h.


One saline-group brain (#25) was excluded from TUNEL analysis because of inadequate fixation associated with non-specific staining; all brains underwent caspase 3 and IBA1 analysis. Representative histology and immunohistochemistry results for pyriform cortex are shown in Fig. 2a–o.


Figure 2. (a–o) Methyl isobutyl amiloride (MIA) treatment decreases histological damage. Haematoxylin-eosin (H&E) histology (a–c), transferase mediated biotinylated d-UTP nick end-labelling (TUNEL) (d–f) and immunohistochemistry for caspase 3 (g–i) and ionized calcium-binding adaptor molecule 1 (IBA1) (j–o) in naïve animals (a, d, g, j, m), and 48 h after hypoxia-ischaemia for placebo (b, e, h, k, n), and for post-insult MIA (c, f, i, l, o). Compared with naive (left column), placebo animals lost deeper layer neurons on H&E (b), and had a TUNEL+ cell preponderance (e), occasional caspase 3 immunoreactive neurons (h, arrows), and a redistribution of microglial IBA1 immunoreactivity from a very uniform, diffuse pattern (j) to a retraction from deep subcortical areas and large columnar clusters around layer 4 (k, arrows). At high magnification, the placebo group (n) also shows loss of microglial branching and microglial transformation into rounded macrophages (arrows). The cortical changes are to a large extent reversed by MIA, but the hippocampal dentate gyrus (hip) appears unaffected. Bar scales – a–i and m–o: 100 um, j–l: 1 mm.

Download figure to PowerPoint

For comparison, naïve tissue is shown in the left column of Fig. 2; compared with naïve (Fig. 2a), H&E staining of the placebo animals 48 h following HI demonstrated pronounced neuronal injury in cortical (Fig. 2b) and deep grey matter regions (not shown), with neuropil vacuolation, eosinophilia, frequent satellitosis in the superficial neuronal layer of the cerebral cortex, and loss of deeper neurons. These changes were reduced by MIA (Fig. 2c). To evaluate cell death, we next assessed nuclear DNA fragmentation using TUNEL and the appearance of neurons immunoreactive for cleaved caspase 3 antigen. There were almost no TUNEL+ cells in naïve brain (Fig. 2d) but many reactive nuclei, particularly in superficial cortex, in the placebo group (Fig. 2e) and substantially less with MIA treatment (Fig. 2f). There were less cleaved caspase 3 immunoreactive cells than TUNEL+ cells but the former were readily detected in the superficial neuronal cortical layer of the placebo group (Fig. 2h).

Compared with placebo, the number of TUNEL+ cells per 40x eyefield was less with MIA in dorsal parietal cerebral cortex (dCTX), midtemporal cortex (mCTX), ventral pyriform cortex (vCTX), periventricular subcortical white matter (pvWM) and thalamus (Fig. 3a). In MIA animals activated caspase 3 positive cells were reduced in the ventral pyriform cortex and the caudate nucleus (CDT) (Fig. 3b). The absolute activated caspase 3 positive cell counts were 5–20% of those for TUNEL+: this concurs with previous studies comparing both stains simultaneously on the same tissue (Hristova et al. 2010; Faulkner et al. 2011).


Figure 3. (a–c) Effects of methyl isobutyl amiloride (MIA) on log regional density (Mean ± SEM, n = 9 animals per group) of transferase mediated biotinylated d-UTP nick end-labelling (TUNEL)+ (a) and caspase 3+ cells (b) and microglial ramification index (c) in thalamus (THLM), caudate nucleus (CAUDT), putamen (PTMN), periventricular white matter (pvWM), internal capsule (InCa), dorsal cortex – parietal parasagittal gyrus (dCTX), midtemporal (insula-region) cortex (mCTX) and hippocampal dentate gyrus (HIP). Microglial ramification index = (branch density)2/cell density. Overall, MIA consistently decreased the numbers of TUNEL and caspase-3+ cells and increased microglial ramification across different forebrain regions. *< 0.05, **< 0.025 (t-test).

Download figure to PowerPoint

At 2x magnification, cortical microglia had a very diffuse pattern in naïve brain (Fig. 2j) but after HI without MIA there were large columnar clusters in cortical layers 2–4 (Fig. 2k, arrows) as well as layer-wide microglia redistribution in the hippocampal dentate gyrus. MIA greatly reduced the cortical columnar-cluster pattern (Fig. 2l); but microglia redistribution remained in the dentate gyrus.

As injury-associated microglial activation and phagocytosis of neural debris are associated with rapid microglial branch loss (compare Fig. 2m and n), we next assessed the microglial ramification index. This revealed an overall trend towards ramification preservation by MIA and in mCTX and vCTX ramification preservation was significant (Fig. 3c) confirming the impression of visual examination (Fig. 2m–o). There was approximately three to five times more microglial ramification in the mid-temporal and ventral (pyriform) cortex with MIA (< 0.05 and < 0.025 respectively). If the severely affected animals dying prematurely from the effects of severe hypoxia-ischaemia were removed from the analysis, we still observed a protective effect of MIA even though significance was seen in fewer regions – two regions compared with five for TUNEL and one compared with two for microglial ramification (Figure S3a and c). However, more areas showed a difference for activated caspase 3 staining - four regions with the severely affected piglets removed compared with two regions when including all piglets (Figure S3b). This may be because of apoptotic cell death being the predominant form of cell death in this subset of piglets and MIA having a greater effect.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. Disclosures/Conflicts of interest
  9. References
  10. Supporting Information

This study in newborn piglets demonstrates that 3 mg/kg MIA administered intravenously 10 min after the end of transient global HI and repeated every 8 h improved cerebral energy metabolism in the 48 h after HI using 1H MRS biomarkers (thalamic Lac/NAA and Lac/Cr). Correlating with improved cerebral energy metabolism, TUNEL positive nuclei were reduced in the MIA group compared with placebo in the dorsal, mid and ventral cortical areas, periventricular subcortical white matter and thalamus and there was reduced activated caspase 3 in the ventral cortex and caudate nucleus. MIA treatment was also associated with reduced pro-inflammatory microglial activation in the mid and ventral cortical areas, based on preserved microglial ramification in these areas. The addition of 3 mg/kg MIA at 10 min after HI and repeated every 8 h did not alter the serum electrolytes, mean arterial blood pressure, need for inotropes or glucose and lactate levels after treatment. Although the heart rate was significantly lower in the MIA group, this difference started prior to MIA administration; the continuing difference after treatment may reflect better cardiovascular stability in MIA treated animals.

Improved brain energy metabolism with MIA

We monitored brain metabolism using MRS biomarkers that are known to change following HI in the piglet (Lorek et al. 1994; Penrice et al. 1997), serve as linking biomarkers in infants with neonatal encephalopathy (Robertson et al. 1999; Cheong et al. 2006), and are currently used as a surrogate outcome measure in clinical neuroprotection trials for neonatal encephalopathy (Azzopardi 2011). Specifically, high levels of Lac/NAA on thalamic MRS in neonates at 48 h are predictive of poorer 12–18 month neurodevelopmental outcome (Robertson et al. 1999; Cheong et al. 2006; Thayyil et al. 2010). In our perinatal asphyxia piglet model, Lac/NAA levels and TUNEL histopathology are correlated (Faulkner et al. 2011). As shown in Fig. 1, MIA reduced Lac/NAA (Fig. 1i and m) and Lac/Cr (Fig. 1f and j) increases in the forebrain/thalamic voxel. The MRS AUC summary data for each group and metabolite peak area ratio are shown in Fig. 1n. The MIA changes were most marked in the thalamic voxel. No significant differences in brain energetics were seen in the white matter voxel with MIA nor in the whole brain 31P MRS voxel for Pi/ePP, PCr/ePP and NTP/ePP (although NTP/ePP approached significance (= 0.056)). Previous studies of therapeutic hypothermia in this model have shown improved levels of NTP/ePP and PCr/ePP with cooling compared with normal temperature management (Thoresen et al. 1995); the lack of an effect with MIA may relate to a severe HI insult in this study. Our model was designed to cause severe hypoxic-ischaemic cerebral injury by reducing brain NTP to 30% of baseline and maintaining that decrement for 10 min rather than stabilizing NTP at 40% of baseline as previously done (Faulkner et al. 2011). Previous clinical hypothermia studies showed that cooling is less beneficial in severely damaged neonatal patients (Edwards et al. 2010), although still has a protective effect (Tagin et al. 2012).

Overstimulation of NHE is triggered by acidosis during HI and/or by protein phosphorylation mediated by ERK-p90 ribosomal S6 kinase (p90RSK) in ischaemic neurons (Luo et al. 2007). NHE responds to intracellular acidosis and normalizes pHi; the concomitant intracellular Na+ accumulation reverses the Na+/Ca2+ exchanger resulting in Ca2+ overload that contributes to the cascade of ischaemic cell death (Cheung et al. 1986). Previous studies have shown that NHE activity contributes to ionic imbalances after ischaemia and reperfusion in the heart (Avkiran and Marber 2002) and the brain (Luo et al. 2005) and that pharmacological or partial knock out inhibition of NHE-1 during early reperfusion reduces ischaemic damage as measured by changes in apparent diffusion coefficient and T2 on MRI in a rodent model (Ferrazzano et al. 2011). In our study, NHE-1 inhibition with MIA improved cerebral energy metabolism (thalamic Lac/NAA and Lac/Cr) in the early reperfusion period 48 h after HI; this may be due reduced intracellular Ca2+ accumulation and hence a reduction in energy utilization by ATP-dependent Ca2+ transporters attempting to reduce Ca2+ accumulation and Ca2+-mediated uncoupling of oxidative phosphorylation in those animals treated with MIA.

Reduced cell death and inflammatory markers with MIA

Using three techniques to assess cell death and tissue damage following HI (TUNEL histochemistry, and immunohistochemistry for cleaved caspase 3 and microglial IBA1), our results demonstrate MIA reduced injury particularly in the cortex and deep grey matter. Quantification of nuclei with fragmented DNA showed substantial post-insult death in forebrain regions sensitive to HI; the appearance of neurons immunoreactive for cleaved caspase 3, and with typical pyknotic or fragmented nuclei suggests that this was at least partly associated with the classical apoptotic signalling pathway. TUNEL was more sensitive than MRS in detecting pvWM and cortical changes with MIA as the MRS dsc WM voxel did not show significant differences between groups. Although MRS and TUNEL changes were seen in the thalamus, the number of caspase 3 immunoreactive positive cells in the thalamus was not significantly different, however reduced caspase 3 immunoreactive cells were seen in the vCTX and caudate in the MIA group compared with placebo. Staining for caspase 3 immunoreactive cells was much less than TUNEL positive staining; the fact that many cells with similar nuclear morphology were caspase 3 negative, and that their cell density was just 5–20% of that of TUNEL+ cells, could be because of very transient presence of cleaved caspase 3 (Hristova et al. 2010), explaining the current results. One limitation of this study is that some neurons may die after HI through a non-apoptotic process and without the appearance of DNA fragmentation on TUNEL staining (Sun et al. 2009). However, because of technical constraints and model termination 48 h after HI we were unable to independently reconfirm fewer surviving neurons in the untreated group, and increased survival with MIA.

The reduced neuronal cell death seen in the MIA group with TUNEL and caspase 3 immunoreactivity is likely to be because of a reduction in cytosolic Na+ and Ca2+ accumulation and mitochondrial Ca2+ overload, (Teshima et al. 2003). The many deleterious effects of increased cytosolic Ca2+ (Cheung et al. 1986) on different enzymes, including proteases, lipases and calcium dependent nitric oxide synthase isoforms (neuronal and endothelial) are reduced with NHE-1 inhibition. It is likely also that prevention of the alkaline pHi overshoot by MIA inhibition of NHE activation following HI is neuroprotective (Currin et al. 1991; Vornov et al. 1996), although this amelioration or delay in the alkaline pHi shift was only seen in those piglets not suffering secondary energy failure and only in the brain pHi measured using the chemical shift of phosphoethanolamine and inorganic phosphate, but not for that derived from ATP (please see 'Discussion' on brain pHi).

Activation of cerebral microglia comprises a set of highly conserved cellular responses to brain injury (Hristova et al. 2010). Metabolic activation, migration and loss of ramification associated with phagocytosis are rapid events that can occur within hours following injury; microglia proliferation is slower, peaking at 2–4 days following injury. In this study, HI without MIA redistributed the normally very diffuse microglial immunoreactivity (Fig. 2j), with reduced cellular density in deeper cortex and subcortical WM and the appearance of clusters in superficial cortical layers 2 and 4 (Fig. 2k). At high magnification, it was apparent that microglia lost most of their branches, with many transforming into completely rounded macrophages (Fig. 2n). Microglial distribution and ramification was preserved with MIA in predominantly cortical areas (Fig. 2l and o) in keeping the differences seen on TUNEL. However, differences in the microglial ramification were not seen in other brain regions where protection was seen with MIA on TUNEL and MRS. NHE-1 plays an important role in proinflammatory microglial activation in cultured microglia (Liu et al. 2010) and in vivo following focal cerebral HI (Shi et al. 2011). Elevated NHE-1 activity is required for H+ extrusion during respiratory burst activity of microglia, because NADPH oxidase activity is sensitive to pHi (Henderson et al. 1988). NHE-1 prevents intracellular acidosis and allows NADPH oxidase function. In an adult mouse model of focal HI, the NHE-1 protein was highly expressed in activated microglia and NHE-1 inhibition with HOE 642 or transgenic knockdown of NHE-1 significantly attenuated the proinflammatory activation of microglial in the peri-infarct area and reduced proinflammatory cytokine formation (Shi et al. 2011).

Protection with MIA was still observed when piglets dying before the end of the experiment were excluded from the analysis, particularly with the activated caspase 3 staining (Figure S3a–c).

Strengths and limitations of the study

We designed the study to replicate the successful clinical hypothermia neuroprotection trials in babies (Edwards et al. 2010). In these studies an adverse outcome is defined as those babies who die and those with an abnormal neurodevelopmental outcome – i.e. the deaths and severe neurodevelopmental outcome are grouped together. Thus, in this study we did not exclude the piglets that were lost prematurely because of progressive cerebrovascular/cardiovascular decline from the MRS analysis. A possible limitation to the study is that in these studies, piglets were anaesthetized throughout and treatment effects may differ in un-anaesthetized subjects. Furthermore, cooling for 24 h is shorter than the current clinical protocol of 72 h (Azzopardi et al. 2009) but the neuroprotective efficacy of this regimen has previously been validated in this piglet model (O'Brien et al. 2006). The piglet model of HI used in this study has strong preclinical value, as it allows hypothermia to be applied as it would in a clinical setting, and an intensive care level that is on par with a neonatal intensive care unit. This includes supported ventilation, maintenance of cardiac output, blood volume and electrolytes and cardiac resuscitation (if required). These studies have been designed in such a way that they are proof of principle, i.e. constructed to maximize the demonstrable therapeutic effect. Thus, in this study piglets were subjected to a severe HI insult and intervention with MIA was started at 10 min of completing HI; this would be difficult to perform in the clinic. Further studies are needed to assess the therapeutic window of MIA.

MIA and brain pHi

We assessed brain pHi using three different algorithms based on the chemical shift of inorganic phosphate (Pi), phosphoethanolamine (PEtA) and adenosine triphosphate (ATP). The first two pHi traces (Pi and PEtA) revealed a sudden drop in brain pHi in animals with secondary energy failure, when the NTP/ePP levels decreased to ∼60% (Figure S1k and n), and PCr/ePP to ~55% (Figure S1j and m) of basal values. The untreated animal subgroup that did not show secondary energy failure also revealed a mild but significant increase in pH(Pi) and pH(PEtA) of approx. 0.05–0.09 points at 2–30 h following HI insult (Figure S2c and f); this effect was delayed [pHi(Pi)] or abolished [pHi(PEtA)] in the non-SEF subgroup treated with MIA. Unexpectedly, at 24–48 h, the pHi(Pi) was more alkaline in the non-SEF MIA subgroup compared with placebo. Interestingly, none of these effects was observed with pHi(ATP). It is possible that this differential response may be partly to do with the health status of cells preferentially detected by one of the three algorithms. Injured or dead cells could show depletion of high-energy phosphates, particularly of ATP, and a corresponding increase in inorganic phosphate – only present in low levels in healthy cells. Thus, using Pi would preferentially target the pHi in injured and dead cells. However, this would not explain why the alkaline shift in non-SEF animals was only detected with Pi and PEtA but not with ATP, or why PEtA also works in mapping the acidic secondary energy failure, as PEtA levels also decrease following HI insult (Fig. 1d and e). At the moment, these technical issues would suggest the preferential use of Pi and PETA in detecting brain pHi changes following HI insult with MRS.

In term infants with neonatal encephalopathy, the extent of the post-HI brain alkalosis and injury severity are associated (Robertson et al. 2002). Alkalosis in the early phase following HI has also been seen in vitro (Luo et al. 2005), in experimental rodent models (Mabe et al. 1983) and in adult stroke (Welch et al. 1990), and may vary with species and type of HI injury. In human brains, pHi has been seen to rise by up to 0.60 pH units in severely affected infants with neonatal encephalopathy (Robertson et al. 2002). In this study, based on pH(Pi) and pH(PEtA) early measurements, pHi increased by approx. 0.10. In neonatal rats, a combination of acute hypoxia and hypercapnia was associated with a moderately alkaline pH shift (+0.40 pH units); recovery from the asphyxic conditions was associated with a large seizure burden, which paralleled the alkalosis (Helmy et al. 2011). Interestingly, seizures were blocked by pre-application of MIA (Helmy et al. 2011). In this study, brain alkalosis was restricted to just animals without secondary energy failure; this alkalosis was delayed [pHi(Pi)] or abolished [pHi(PEtA)] in the subgroup treated with MIA. Furthermore, there was a late alkalosis [pHi(Pi)] in the MIA treated group. Because the alkalosis was mild, and because animals with secondary energy failure did not show alkalosis but were affected by MIA, the neuroprotective effects of MIA observed in this study must involve mainly pHi independent mechanisms.

First described in 1967 as a potassium sparing diuretic, amiloride acts directly on the distal renal tubule of the nephron to inhibit sodium-potassium ion exchange (Teiwes and Toto 2007). Higher amiloride doses inhibit the NHE and the Na+/Ca2+ exchanger. Several synthesized amiloride analogues increase specificity towards one of these ion channels (Slepkov et al. 2007). We chose MIA as it targets NHE1 inhibition and crosses the blood brain barrier in adult male rats (approx 2 mg/kg) (Ferimer et al. 1995). Furthermore, neuroprotection following hypoxia-ischaemia was observed with a similar MIA dose in a neonatal rodent model (Kendall et al. 2006). The amiloride dose used clinically in neonates (0.4 mg/kg/24 h) is around one-twentieth of MIA used in our study (3 mg/kg/8 h); nevertheless we saw no difference in any biochemical or physiological marker between placebo and MIA treated groups. Although MIA is a 2nd generation high affinity amiloride analogue, the 3rd generation NHE inhibitors such as Cariporide (HOE642) are more potent inhibitors of NHE1. In the study by Shi et al., 0.5 mg/kg HOE642 reduced proinflammatory microglial activation following ischaemia in mice (Shi et al. 2011); the same dose of 0.5 mg/kg HOE 642 24 and 48 h after HI was neuroprotective in neonatal mice and showed long-term functional improvement (Cengiz et al. 2011).

NHE inhibition and hypothermia

Therapeutic hypothermia is now standard care for infants in the developed world with moderate to severe neonatal encephalopathy (Higgins et al. 2011). It is unknown, if NHE inhibition combined with therapeutic hypothermia might augment hypothermic neuroprotection after perinatal asphyxia. Some in vitro studies suggest that moderate hypothermia (reducing temperature from 37 to 20°C) itself increases NHE activity (Zhou and Willis 1989; Plesnila et al. 2000) whilst others suggest that hypothermia produces a partial inhibition of NHE activity (Hoshino and Avkiran 2001). If the inhibition of NHE is partial with hypothermia or indeed if hypothermia increases NHE activity, the combined effects of an NHE inhibitor with hypothermia could significantly augment hypothermic neuroprotection especially as they act on separate pathways. Importantly, studies suggest that the potency of NHE blockade under hypothermic conditions is not influenced by a change in temperature (Hoshino and Avkiran 2001) and NHE blockade was beneficial in a perfused heart model even after hypothermic ischaemia (Yamauchi et al. 1997). Pre-clinical large animal studies of the potential augmentation of mild therapeutic hypothermia combined with NHE blockade are high priority.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. Disclosures/Conflicts of interest
  9. References
  10. Supporting Information

This study is the first large animal neonatal HI model demonstrating neuroprotection from NHE-1 inhibition with MIA given 10 min after the end of HI. We report here that MIA improved cerebral energy metabolism in the thalamic area in the 48 h after HI on MRS biomarkers and reduced cell death and microglial activation in some brain areas. The use of clinically relevant MRS biomarkers of injury severity in a large animal model of perinatal HI is an important step towards translating future NHE inhibitors into clinical studies. The effect of NHE blockade on brain pHi depended on whether severe secondary energy failure occurred; in those subjects mildly affected, early brain alkalosis was ameliorated with MIA following HI. Additional studies of NHE blockade with hypothermia are warranted to determine additive effects.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. Disclosures/Conflicts of interest
  9. References
  10. Supporting Information

The study was undertaken at UCH/UCL who received a proportion of funding from the UK Department of Health's NIHR Biomedical Research Centres funding scheme. This project was funded by the Medical Research Council, Sport Aiding Research in Kids (SPARKS) and the Great British Sasakawa Foundation.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. Disclosures/Conflicts of interest
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. Disclosures/Conflicts of interest
  9. References
  10. Supporting Information

Figure S1. Dorsal subcortical WM 1H (A,H) and whole-brain 31P (I-N) MRS.

Figure S2. Effects of MIA on brain pH(Pi) (a–c), pH(PETA) (d–f) and pH(ATP) (g–i).

Figure S3. (a–c) Effects of MIA on log regional density excluding those piglets that died prematurely from the effects of severe hypoxia-ischemia.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.