Excitotoxicity and Metabolic Changes in Association With Infarct Progression
Background and Purpose—We investigated to what extent excitotoxicity and metabolic changes in the peri-infarct region of patients with malignant hemispheric stroke are associated with delayed infarct progression.
Methods—In 18 patients with malignant hemispheric stroke, 2 microdialysis probes were implanted within the peri-infarct tissue at a distance of 5 and 15 mm to the infarct. Precise probe placement was achieved by intraoperative laser speckle imaging. Glutamate, glucose, pyruvate, and lactate levels were monitored for 5 days after surgery. Delayed infarct progression was determined from serial MRI on the day after surgery and after the monitoring period.
Results—Initial stroke volume ranged from 122 to 479 cm3 with a median of 295 cm3. Nine of 18 patients (50%) had delayed infarct progression (median, 44 cm3; range, 19–93 cm3). In these patients, glucose and individual pyruvate levels were significantly lower when compared with patients without infarct progression, whereas glutamate and the lactate–pyruvate ratio were significantly elevated in patients with infarct progression early after surgery (12–36 hours) at the 15-mm microdialysis probe location. Lactate was elevated but without difference between groups.
Conclusions—Excitotoxic or metabolic impairment was associated with delayed infarct progression and could serve as a treatment target.
Infarct progression is one of the most serious in-hospital complications after stroke. Among several involved factors, excitotoxicity and metabolic compromise are discussed as major contributors to infarct progression.1,2
In patients with large space-occupying infarcts, decompressive hemicraniectomy is often performed to prevent herniation. This surgical exposure of the brain allows probe implantation to monitor tissue at risk for secondary brain damage. Although a number of investigations have focused on biochemical changes within the peri-infarct tissue,3,4 precise probe positioning may have been hampered because probe placement was not performed under direct visual confirmation of the infarct border.
Here, we investigated biochemical alterations in the peri-infarct tissue of patients with malignant hemispheric stroke in relation to delayed infarct progression. Precise probe implantation was achieved through visual confirmation of the infarct border by intraoperative laser speckle imaging.
Methods are available in the online-only Data Supplement.
Forty-seven patients with subtotal or total middle cerebral artery infarction with or without additional infarction of the ipsilateral anterior- or posterior-cerebral artery territory and surgical indication for decompressive hemicraniectomy were screened for eligibility of study participation between May 19, 2009, and April 30, 2011. A total of 18 patients were included after informed consent was obtained.
After bone removal and durotomy, the infarct border was localized by cerebral blood flow measurement using a laser speckle imager (MoorFLPI; Moor Instruments Ltd, Axminster, United Kingdom). A sharp regional drop in cortical perfusion ≤20% of normal was used as a threshold for ischemic tissue (Figure IA in the online-only Data Supplement) and judged as valid after independent confirmation of a macroscopically visible change in cortical tissue appearance at the suspected infarct border. Next, 2 microdialysis catheters (100 kDa; CMA, Stockholm, Sweden) were implanted 5 and 15 mm from the infarct in the peri-infarct tissue. Because of the fact that we were only able to identify the infarct border at the brain surface, microdialysis probes were implanted directly subpially to the cortex.
Glutamate, glucose, pyruvate, and lactate levels were monitored for 5 days after surgery.
Infarct Volume and Outcome Analysis
MRI or computed tomography was performed after surgery and at the end of the monitoring period on days 5 and 6.
With exception of patient numbers 6 and 13, where an MRI scan was not possible for logistic reasons, matched diffusion-weighted imaging and fluid attenuated inversion recovery sequences from the MRI performed postoperatively and on days 5 and 6 were used for determination of the infarct volume and brain swelling with iPlan Cranial software. In patient numbers 6 and 13, infarct volume and swelling were assessed from serial computed tomographic scans. For volumetric analysis, the infarct volume was corrected for hemispheric swelling. The initial neurological deficit was assessed using the Glasgow Coma Scale and Modified National Institutes of Health Stroke Scale. Outcome was evaluated at 6 months using the modified Rankin Scale.
The initial stroke volume ranged from 122 to 479 cm3 with a median volume of 295 cm3. Nine of 18 patients (50%) had delayed infarct progression (median 44 cm3; range, 19–93 cm3; Figure IB in the online-only Data Supplement). A mild perfusion mismatch of 15, 20, and 15 cm3 was found at the early MRI time-point in patient numbers 11, 15, and 17, respectively. All other patients with available MRI data did not show a perfusion mismatch. Demographic, clinical, and postoperative monitoring data are listed in Table I in the online-only Data Supplement.
The mean glutamate concentrations were not significantly different (5 mm: 49±41 versus 52±68 μmol/L in patients with and without infarct progression; 15 mm: 64±62 versus 21±20 μmol/L in patients with and without infarct progression). However, early after surgery (12–36 hours), glutamate levels at the 15-mm distance were significantly elevated in patients where infarct progression occurred (Figure).
Glucose levels did not differ 5 mm from the infarct (1.8±0.9 versus 1.6±0.5 mmol/L in patients with and without infarct progression), whereas significantly lower glucose levels were found at the 15-mm distance in patients with infarct progression (1.2±0.4 versus 1.9±0.4 mmol/L in patients with and without infarct progression; P<0.05; Figure).
Lactate was significantly elevated when compared with normal range. However, lactate did not differ between patient groups (5 mm: 6.8±3.8 versus 6.7±2.3 mmol/L in patients with and without infarct progression; 15 mm: 5.9±1.2 versus 5.9±2.6 mmol/L in patients with and without infarct progression; Figure).
The mean pyruvate levels did not differ (5 mm: 130±40 versus 168±65 μmol/L in patients with and without infarct progression; 15 mm: 113±63 versus 164±46 μmol/L in patients with and without infarct progression). However, in patients with infarct progression, individual pyruvate levels were significantly lower (5 mm: 108 and 120 hours after surgery; 15 mm: 12 and 108 hours after surgery; Figure).
Similarly, the mean lactate–pyruvate ratio did not differ (5 mm: 63±23 versus 53±33 in patients with and without infarct progression; 15 mm: 106±97 versus 47±31 in patients with and without infarct progression), but individual ratios during the early observation period (15 mm: 12–36 hours) were significantly higher in patients who showed delayed infarct progression (Figure).
Lesion expansion is closely linked to a border zone of malperfusion, the penumbra, surrounding the already infarcted tissue.1 However, at delayed time points, the typical hemodynamic penumbra rarely exists, and lesion expansion may predominantly rely on other mechanisms, such as excitotoxicity, metabolic changes, lactacidosis, apoptosis, inflammation, and spreading depolarizations.1,5 Against this background, we exclusively studied delayed periods of infarct maturation (beyond 24 hours after stroke onset) where no significant hemodynamic perfusion–diffusion mismatch was measurable in our patients. Interestingly, half of our patients with malignant hemispheric stroke nevertheless had significant delayed infarct progression, which was associated with altered levels of glutamate, glucose, and pyruvate in the immediate peri-infarct region.
To investigate whether there is a gradient from the infarct border to the periphery, we implanted 2 microdialysis probes at different distances to the infarct border. Interestingly, all measured biochemical markers were significantly altered at both distances. When distinguishing between patients with and without infarct progression, however, differences in the levels of glutamate, glucose, and the lactate–pyruvate ratio were predominantly found further away (15 mm) from the infarct. Therefore, our data suggest that altered biochemical changes can be detected in a rather widespread area surrounding the infarct. The association between these alterations and delayed infarct progression might be restricted to a rather circumscribed area.
In all patients, extracellular glutamate levels were initially above the ischemia threshold and with ongoing time, glutamate levels steadily decreased but reached normal ranges only in patients without infarct progression (Figure). In patients with infarct progression, glutamate was significantly higher at a distance of 15 mm during the early observation period. This may reflect progressive tissue damage or could also be part of the underlying pathological mechanism for infarct progression itself.
Extracellular Metabolic Biochemistry
Glucose and pyruvate were significantly lower in patients with delayed infarct progression (Figure) and naturally, this indicates a more intact metabolism in patients without infarct progression. However, recent studies have also suggested neuroprotection through pyruvate,6 which needs to be addressed in future studies. The widespread increase in lactate was in line with previous experimental findings.7 The significantly higher lactate–pyruvate ratio during the early monitoring period in patients with infarct progression (Figure) underlines the higher metabolic compromise in this group.
Delayed infarct progression in patients with malignant hemispheric stroke is associated with a disarrangement of biochemical markers within the peri-infarct region. Whether the observed excitotoxic or metabolic impairment rather reflects progressive tissue damage or possibly represents a suitable target for novel treatment strategies should be a main focus of future studies.
We thank S. Seidlitz, J. Kopetzki, C. Altendorf, and N. Gase for research assistance and G. Bohner and J. Fiebach for help with imaging processing.
Sources of Funding
The study was funded by the Deutsche Forschungsgemeinschaft (DFG-WO 1704/1–1, DFG DR 323/5-1); Bundesministerium für Bildung und Forschung (Center for Stroke Research Berlin, 01 EO 0801) and Kompetenznetz Schlaganfall.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.113.004475/-/DC1.
- Received December 17, 2013.
- Revision received January 20, 2014.
- Accepted January 23, 2014.
- © 2014 American Heart Association, Inc.
- Dohmen C,
- Bosche B,
- Graf R,
- Reithmeier T,
- Ernestus RI,
- Brinker G,
- et al
- Ryou MG,
- Liu R,
- Ren M,
- Sun J,
- Mallet RT,
- Yang SH