Cortical spreading depolarizations induced by surgical field blood in a mouse model of neurosurgery

Restricted access


Cortical spreading depolarization (CSD) has been linked to poor clinical outcomes in the setting of traumatic brain injury, malignant stroke, and subarachnoid hemorrhage. There is evidence that electrocautery during neurosurgical procedures can also evoke CSD waves in the brain. It is unknown whether blood contacting the cortical surface during surgical bleeding affects the frequency of spontaneous or surgery-induced CSDs. Using a mouse neurosurgical model, the authors tested the hypothesis that electrocautery can induce CSD waves and that surgical field blood (SFB) is associated with more CSDs. The authors also investigated whether CSD can be reliably observed by monitoring the fluorescence of GCaMP6f expressed in neurons.


CSD waves were monitored by using confocal microscopy to detect fluorescence increases at the cortical surface in mice expressing GCaMP6f in CamKII-positive neurons. The cortical surface was electrocauterized through an adjacent burr hole. SFB was simulated by applying a drop of tail vein blood to the brain through the same burr hole.


CSD waves were readily detected in GCaMP6f-expressing mice. Monitoring GCaMP6f fluorescence provided far better sensitivity and spatial resolution than detecting CSD events by observing changes in the intrinsic optical signal (IOS). Forty-nine percent of the CSD waves identified by GCaMP6f had no corresponding IOS signal. Electrocautery evoked CSD waves. On average, 0.67 ± 0.08 CSD events were generated per electrocautery episode, and multiple CSD waves could be induced in the same mouse by repeated cauterization (average, 7.9 ± 1.3 events; maximum number in 1 animal, 13 events). In the presence of SFB, significantly more spontaneous CSDs were generated (1.35 ± 0.37 vs 0.13 ± 0.16 events per hour, p = 0.002). Ketamine effectively decreased the frequency of spontaneous CSD waves (1.35 ± 0.37 to 0.36 ± 0.15 CSD waves per hour, p = 0.016) and electrocautery-stimulated CSD waves (0.80 ± 0.05 to 0.18 ± 0.08 CSD waves per electrocautery, p = 0.00002).


CSD waves are detected with far greater sensitivity and fidelity by monitoring GCaMP6f signals in neurons than by monitoring IOSs. Electrocautery reliably evokes CSD waves, and the frequency of spontaneous CSD waves is increased when blood is applied to the cortical surface. These experimental conditions recapitulate common scenarios in the neurosurgical operating room. Ketamine, a clinically available pharmaceutical agent, can block stimulated and spontaneous CSDs. More research is required to understand the clinical importance of intraoperative CSD.

ABBREVIATIONS CSD = cortical spreading depolarization; IOS = intrinsic optical signal; IP = intraperitoneal; NMDA = N-methyl-d-aspartate; SAH = subarachnoid hemorrhage; SFB = surgical field blood.

Article Information

Correspondence Anja I. Srienc: Washington University School of Medicine, St. Louis, MO.

INCLUDE WHEN CITING Published online April 5, 2019; DOI: 10.3171/2018.12.JNS181130.

Disclosures The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

© AANS, except where prohibited by US copyright law.



  • View in gallery

    Electrocautery mouse model of neurosurgery. A 2 × 2–mm thinned-skull cranial window was created on the right side of the head, with the medial border of the window approximately 2 mm from the sagittal suture and the posterior border of the window approximately 0 mm from the lambdoid suture. A burr hole was then drilled 1 mm anterior to the cranial window, approximately at the center of the anterior border of the window, with subsequent durotomy. Through this burr hole, tail vein blood could be applied to the cortical surface and/or the cortex could be cauterized with bipolar electrocautery. During experiments, the mouse was fixed to a head bar attached to a treadmill under the microscope objective. Drawing by Pei-Pei Chiang; used with permission.

  • View in gallery

    Detecting CSD with GCaMP6f fluorescence is superior to detecting CSD with IOS. Image frames are displayed at 4 different time points from a single trial in which GCaMP6 and IOS were monitored simultaneously while a CSD event was elicited with electrocautery at time = 0 seconds (s). The CSD wave in the Raw IOS series could not be detected. However, when the raw image stack was processed using a custom MATLAB routine to highlight changes in signal brightness, the CSD became evident (Processed IOS). Meanwhile, the CSD wavefront during the same trial could be clearly observed in the raw GCaMP6f fluorescence signal (Raw Ca2+) and was further enhanced after image processing (Processed Ca2+).

  • View in gallery

    Most electrocautery-evoked CSD waves occur within 82 seconds of electrocautery. Histograms representing the distribution of the number of CSD events after electrocautery in the No SFB group (A), the SFB group (B), and the SFB group that was treated with ketamine (SFB+KET) (C). Electrocautery occurred at time 0. Most electrocautery-evoked CSD waves occurred within 82 seconds of electrocautery. The widened distribution of CSD events in the SFB group is thought to represent the observation of increased spontaneous CSD waves in that group.

  • View in gallery

    Frequency, latency, and velocity of CSD in the mouse electrocautery model. A: Electrocautery without SFB elicited CSD in the mouse cortex (No SFB group) at an average rate of 0.67 ± 0.08 waves per stimulus (stim). In the presence of SFB (SFB group), 0.80 ± 0.05 CSD waves occurred per stimulus (p = 0.186). However, in SFB mice treated with ketamine (SFB+KET group), the rate of electrocautery-induced waves was decreased (0.18 ± 0.08) compared to the rate in the animals in the No SFB and SFB groups (p = 0.0005 and p = 0.00002, respectively). B: In cortexes previously electrocauterized, the average rate of spontaneous CSD waves in the SFB group was 1.35 ± 0.37 events per hour. This rate was significantly higher than the rate in the No SFB group (0.13 ± 0.16, p = 0.002). Mice in the SFB+KET group showed reduced spontaneous CSD waves in SFB trials compared to SFB group mice that did not receive ketamine (0.36 ± 0.15 events per hour, p = 0.016). For cortexes that did not receive electrocautery, no waves were generated in trials in control mice that underwent the thinned-skull window procedure without a burr hole (Without burr hole) (0.00 ± 0.00). This result was not significantly different from that for mice with no electrocautery in the Burr hole group (0.45 ± 0.25 events per hour, p = 0.11) or in the Burr hole+SFB group (0.38 ± 0.26 events per hour, p = 0.25). C: In the mice with no electrocautery in the Burr hole+SFB group, the occurrence of spontaneous CSD waves did not vary as a function of exposure time to SFB. CSD events occurred randomly over 180-minute trials. D: For CSD wave latency, no statistically significant difference was observed from the time of electrocautery between the No SFB (19.9 ± 3.4 seconds), SFB (23.7 ± 3.5 seconds), and SFB+KET (26.5 ± 5.0) groups (p = 0.21). E: For wave velocity, when CSD waves were elicited by electrocautery (Stim Waves), the average CSD velocity in the No SFB group was 3.46 ± 0.16 mm/min, while in the SFB group it was 3.51 ± 0.19 mm/min (p = 0.99). The average CSD velocity in the SFB+KET group was 1.87 ± 0.13 mm/min, significantly less than that for the SFB group, which was not treated with ketamine (p = 0.0002). When CSD events occurred spontaneously (Spon Waves), there was no difference in CSD velocity between the No SFB (3.2 ± 0.27 mm/min), SFB (3.47 ± 0.27 mm/min), Burr hole (3.81 ± 0.32 mm/min), and Burr hole+SFB groups (3.87 ± 0.34 mm/min); however, ketamine significantly reduced velocity in the SFB+KET group (1.7 ± 0.21 mm/min, p = 0.0009). For all groups, n = 8 animals, except the control Burr hole and Without burr hole mice (all of which underwent the thinned-skull window procedure), for which n = 6 animals. Cautery = electrocautery. *Statistically significant at an alpha of 0.05.





All Time Past Year Past 30 Days
Abstract Views 335 335 59
Full Text Views 68 68 10
PDF Downloads 51 51 8
EPUB Downloads 0 0 0


Google Scholar