Cortical Spreading Depression¶
A depolarizing front traverses cerebral cortex while spontaneous electrical activity is newly suppressed in its wake at successive sites.
Core Idea¶
Cortical spreading depression (CSD) is a traveling cortical event in which a depolarizing disturbance crosses cerebral tissue and, at successive sites with ongoing activity, the spontaneous electrical activity falls as the disturbance arrives. The fall in activity is the “depression” in the name. The local front and successive suppression distinguish it from a simultaneous shutdown and from a depolarization that passes through cortex already electrically silent.[1][2][3]
The event can be induced experimentally or recorded spontaneously after human brain injury. Those settings differ in trigger, tissue condition, and evidence access, but both can exhibit the same paired traveling relation. Recovery of the background, extension of a lesion, and a particular ion-transmitter account are possible contexts or outcomes, not clauses of the definition. The historical name is sometimes used more broadly for spreading depolarizations; here it retains the narrower activity-depression test.[1][2][4]
Structural Signature¶
Signature: local cortical onset → depolarizing front → ordered arrival across connected cortex → new suppression of ongoing electrical activity at those sites. The trigger and later recovery course vary.
- Cortical substrate. The tissue is cerebral cortex with spontaneous activity available to be depressed. A similar event in retina or brainstem is an analogue, not a cortical instance.[1][2]
- Local onset and front. A disturbance starts in one region and produces a slow-potential change associated with depolarization. The measured front need not have the same initiating cause in every case.[1][2]
- Ordered spread. Changes reach spaced cortical recording sites in sequence. A single-channel decrease or essentially synchronous decrease cannot by itself establish a spreading event.[1][3]
- Activity depression. Previously present ECoG background diminishes as the front passes. A slow-potential shift through cortex whose activity had not recovered is a related spreading depolarization but supplies no new depression of ongoing activity there.[1]
- Variable course. The background may recover or remain silent, and subsequent depolarizations can recur. These observations matter for interpretation but do not redefine the first paired event.[1]
What It Is Not¶
CSD is not every spreading depolarization. Fabricius and colleagues reported recurrent slow-potential changes after ECoG background failed to recover in two monitored patients. Those later fronts did not newly suppress activity already absent. In this entry, the narrow label applies where a front and newly spreading depression are both evidenced; broader literature may use “CSD” differently.[1]
Nor is every depression of cortical activity a spreading depression. Strong and colleagues distinguished spreading from essentially synchronous depressions in acutely injured human cortex. In the Carlson mouse experiment, a lateral-cautery trace had near-simultaneous arrival at two electrodes because the electrodes lay about equally distant from the source. That geometry alone does not prove a nonpropagating event; it simply cannot serve as the particular two-site sequential proof used here.[3][2]
A visual aura or BOLD-imaging progression is not, by itself, a direct recording of both neural depolarization and activity depression. Hadjikhani and colleagues observed an occipital BOLD progression during human visual aura that strongly suggested a CSD-like electrophysiological event, but the imaging is a qualified indirect comparator. Conversely, experimental tissue changes or lesion growth following repeated waves do not make injury a defining outcome of every CSD.[5][6]
Scope of Application¶
The literal scope is an electrophysiological process in cerebral cortex, in experimental cortex and in monitored human cortex with a recordable background. Direct identification needs enough spatial and temporal recording to relate the slow-potential front to suppression at successive sites. In injured tissue, recordings must separate an initial spreading depression from later depolarizations after persistent electrical silence.[1][2]
A mouse familial-hemiplegic-migraine model shows that electrically evoked DC-potential fronts can have different induction thresholds and speeds by genotype. That study did not directly record ECoG suppression in its described CSD methods, so it supports the front and condition-dependent rate, not a separate full positive of this narrower identity. The visual-aura fMRI study supports a plausible clinical association but cannot turn all aura episodes into directly verified CSD events.[4][5]
Clarity¶
“Depression” names reduced ongoing activity, not a low membrane potential or a synonym for damage. In the directly monitored cases, a large negative slow-potential change travels with a decrease in ECoG background. The two signals answer different questions: the slow potential tracks depolarization, and the activity record shows whether an existing background was newly suppressed.[1][2]
The most useful discriminating question is: Was activity present to be depressed at each site when the front arrived? If yes, and the paired changes pass sites in order, the narrow identity is supported. If the background was already silent, describe a spreading depolarization without silently counting another new depression. If the activity drop occurs essentially at once without a traveling front, examine a synchronous mechanism instead.[1][3]
Manages Complexity¶
A depolarization front, activity suppression, and tissue outcome are often bundled into one story. Separating them prevents the event count from absorbing unlike observations. Fabricius and colleagues reported matched spreading slow-potential and ECoG changes in injured patients, then later recurrent slow potentials without ECoG recovery in some. Both observations can matter clinically, but only the first directly contains the newly depressed-background role.[1]
The separation also keeps trigger and outcome from becoming false universal mechanisms. Bipolar cautery is the stimulus in Carlson's mouse model; electrical visual-cortex stimulation is used in Leo's genotype comparison; spontaneous episodes occur in the monitored human injury series. Nakamura's experimental lesion enlargement is important in its focal-ischemia setting, but it is not a claim that every event, including an aura-associated one, enlarges tissue injury.[2][4][1][6]
Abstract Reasoning¶
Start with a time-ordered map of cortical sites, not a diagnostic label. At each site, mark the slow-potential onset and whether spontaneous ECoG activity existed before it and fell after it. Compare arrival times between sites. A positive narrow case has both the front and activity suppression moving through cortex; a stationary shift, simultaneous depression, or later front through pre-silent tissue fails a different part of that test.[1][3][2]
Then separate what the trace establishes from what it suggests. In Carlson's mouse series, 10 of 16 DC shifts were accompanied by ECoG suppression in aggregate. The indexed Figure 2a caption explicitly gives one anterior-cautery event with both changes traveling from Channel 1 to Channel 2. The aggregate does not establish that every one of those ten had individually verified sequential paired spread. The Figure 2b near-simultaneous two-channel arrival reflects electrode geometry and cannot be used as a separate ordered two-site example.[2]
Finally, treat rate, recovery, and lesion effects as measured attributes of particular instances. Leo's mutant mice showed lower induction threshold and faster DC-potential propagation than wild type, but the authors' astrocytic glutamate-clearance explanation remains a hypothesis. Nothing in that experiment shows a universal transmitter mechanism for all cortical spreading depression.[4]
Knowledge Transfer¶
The role test transfers between experimental and clinical cortex: local onset, traveling depolarization, new activity suppression, and site-order evidence. It lets investigators ask the same Classification question of a cautery-evoked mouse trace and a spontaneous human recording without treating their triggers or tissue states as the same.[2][1]
The larger Prime Propagation supplies the source–medium–spread lens. It directs attention to recording-site spacing, cortical condition, velocity, and what counts as an observed path. CSD adds the domain-specific coupled electrophysiological signature. A label such as “wave” alone is weaker here: the live Wave Prime includes properties such as dispersion and superposition that these cortical reports do not establish for this nonlinear front.[1][4]
Examples¶
Mouse cortex under simulated neurosurgical cauterization¶
Carlson, Carter, and Shuttleworth applied bipolar cautery to mouse cortex under simulated intraoperative conditions and recorded DC potential and ECoG. Their report gives 10 of 16 DC shifts with ECoG suppression in aggregate. The indexed Figure 2a description shows a specific anterior-cautery event in which a negative DC shift and reduced ECoG activity reached Channel 1 and then Channel 2, at a reported 3.41 mm/min. Its tissue, trigger, and observed site order give one direct experimental positive. The lateral Figure 2b has near-simultaneous channel arrival because the sites are equidistant from its stimulus; it is not the ordered-pair example.[2]
Mapped back: Cautery is the local onset, living mouse cortex is the connected medium, the slow potential marks the depolarizing front, and the ECoG loss supplies newly depressed ongoing activity at successive sites in the Figure 2a event. Neither the aggregate 10/16 count nor the paper's metabolic findings prove that all mouse shifts show this complete trace or that human neurosurgery has the same event rate.[2]
Spontaneous recordings in acutely injured human cortex¶
Fabricius and colleagues' original abstract reports 73 spontaneous episodes in six of twelve monitored patients. Across recording sites, slow-potential changes accompanied spreading ECoG depression with a median propagation rate of 3.3 mm/min. ECoG recovered spontaneously in four of the six affected patients; in two, recurrent slow potentials later occurred without background recovery for hours. The abstract establishes these reported observations, while full traces and uninspected methods are not claimed here.[1]
Mapped back: Injured human cortex is the medium, a spontaneous event supplies local onset, ordered slow potentials mark propagation, and the simultaneously observed spreading loss of existing ECoG background supplies the depression role. Later fronts in silent tissue are explicitly separated. Human spontaneous occurrence and experimental mouse induction instantiate the same narrow relation without sharing a trigger or a guaranteed recovery outcome.[1]
Human visual aura — qualified indirect comparator¶
In three human subjects, Hadjikhani and colleagues observed a contiguous occipital BOLD progression during visual aura, reported at 3.5 ± 1.1 mm/min. Their interpretation strongly suggested a CSD-like event, but BOLD imaging did not directly record the required electrical front and spreading ECoG suppression. This is relevant to the hypothesis, not a third directly verified positive example.[5]
Structural Tensions¶
Depolarization detection versus narrow classification. In injured cortex, a slow-potential trace can continue to register traveling depolarizations after the ECoG background has become silent. Counting every front under one “depression” label retains a broad depolarization tally but loses the distinction between a new depression and activity already absent. Restricting the term to paired fronts and activity loss preserves the narrower identity, while requiring separate reporting of later fronts that may still matter. Diagnostic: Was ongoing activity present at each site to be newly suppressed? This is a measurement and classification tension, not a claim that one clinical monitoring policy is universally preferable.[1]
Structural–Framed Character¶
Cortical site order, electrical potential, and loss of background activity can be described without an institution's goals or local policy terms. The process is therefore more structural than framed: it is a physiological relation, whether seen in an experimental animal or in injured human tissue. Judgments about when to monitor a patient, whether a particular recording qualifies, and what harm the event predicts are evidence and clinical questions around the process rather than part of its definition.[2][1]
The paired front-and-depression relation is recognized in the cautery-evoked mouse trace and the spontaneous human recordings without importing the mouse trigger, the human injury course, or a clinical monitoring policy from one setting to the other. Conversely, importing the CSD label into a retinal event or a BOLD-only aura observation would bypass the cortical, direct-electrophysiological test. “Cortical” and “electrocorticographic activity” are therefore identity conditions, not arbitrary labels. Its character: structural with a biological substrate bound. The relation is evidence-tested and largely neutral, but remains within cortical electrophysiology rather than an unrestricted formal relation.[2][1][5]
Structural Core vs. Domain Accent¶
The portable core is a local state change traveling through a connected medium with an order and rate of arrival. That is what Propagation contributes. The domain-specific accent is decisive rather than decorative: cerebral-cortical tissue, a depolarizing slow-potential front, and newly suppressed spontaneous ECoG activity at successive sites. Remove those electrophysiological roles and one has generic propagation, not CSD.[1][2]
Mouse and human cases show that the accent is stable within the domain even when initiation and condition differ. The FHM2 genotype changes measured threshold and speed; injury changes the recovery context; neither change erases the paired-front criterion. The edge to Propagation abstracts the travel relation while the entry retains the observable cortical differentia.[4][1]
Instantiates / Related Primes¶
This entry is a kind of Propagation.
Strict parent — Propagation (Propagation). Each full CSD case starts locally and spreads a cortical state through tissue at a site-observable rate. The parent holds across many media and signals, while CSD adds specific cortical depolarization and activity loss. The subtype relation is child-to-parent subsumption/kind_of: if spread is removed, the event no longer satisfies this identity. The cited recordings show paths over electrode spans; they do not measure an event's complete terminal reach in every case.[1][2][4]
Related, not parent — Wave (Wave). These reports use “wave” descriptively for a front, but the live Wave identity asks for a broader physical signature including dispersion or superposition. Sequential cortical spread alone does not prove those properties. Related measurement — Electrocorticography (Electrocorticography). ECoG helps identify depression; the physiological process is not a recording method.
Relationships to Other Abstractions¶
Current abstraction Cortical Spreading Depression Domain-specific
Parents (1) — more general patterns this builds on
-
Cortical Spreading Depression is a kind of Propagation Prime
A cortical depolarization and activity-depression state spreads systematically from local onset through excitable tissue at a measurable, condition-dependent rate.Every admitted cortical spreading depression has a local cortical onset, connected excitable medium, and sequential travel of a depolarization together with newly depressed spontaneous activity across observed tissue. Human and mouse recordings measure site order and speed over electrode spans; a mouse genotype experiment shows that cortical condition can change propagation rate. These instantiate Propagation's source–medium–spread relation without claiming that the complete terminal reach or stopping point was measured in every case. Remove spatial travel and one has synchronous depression or local depolarization rather than this event. Propagation also occurs without cortical activity depression in many other media, so this is a strict specialized kind of the parent, not a synonym.
Hierarchy path (1) — routes to 1 parentless root
- Cortical Spreading Depression → Propagation
Neighborhood in Abstraction Space¶
Cortical Spreading Depression sits in a sparse region of the domain-specific corpus (97th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Unclustered & Miscellaneous (2551 abstractions)
Nearest neighbors
- Hyperpolarization (biology) — 0.78
- Place Field — 0.76
- Flux Limiter — 0.76
- Spike-Timing-Dependent Plasticity — 0.76
- Retinotopy — 0.76
Computed from structural-signature embeddings · 2026-10-08
Not to Be Confused With¶
- Spreading depolarization: the broader front can be detected even in electrically silent cortex; it is not always a newly spreading depression of ongoing activity.[1]
- Synchronous cortical depression: an activity decrease without sequential spread across sites, contrasted in the original injured-human report.[3]
- Migraine aura: a clinical experience with indirect imaging evidence compatible with CSD, not itself a direct paired cortical recording.[5]
- Electrocorticography: a measurement method, not the measured event.
- Inevitable tissue injury: a possible experimental consequence in particular conditions, not a defining outcome.[6]
References¶
[1] Fabricius M, Fuhr S, Bhatia R, et al., Cortical spreading depression and peri-infarct depolarization in acutely injured human cerebral cortex, Brain 129 (2006), 778–790, doi:10.1093/brain/awh716. Original paper abstract, Methods/Results/Conclusions; full text and electrode figures were not independently inspected. Supports the observed human spreading ECoG/slow-potential relation, episode count, speed, and variable recovery within the sampled patients. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s ↩t ↩u ↩v ↩w ↩x ↩y
[2] Carlson AP, Carter RE, Shuttleworth CW, Vascular, electrophysiological, and metabolic consequences of cortical spreading depression in a mouse model of simulated neurosurgical conditions, Neurological Research 34 (2012), 223–231, doi:10.1179/1743132811Y.0000000077. Original abstract Results and indexed author-manuscript Results/Figure 2a–b caption; direct PMC page open was blocked. Supports the aggregate 10/16 ECoG-suppression observation and the individually described sequential paired Figure 2a event, not ten individually verified sequential paired traces or human-surgery incidence. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q
[3] Strong AJ, Fabricius M, Boutelle MG, et al., Spreading and synchronous depressions of cortical activity in acutely injured human brain, Stroke 33 (2002), 2738–2743, doi:10.1161/01.STR.0000043073.69602.09. Original abstract Methods/Results; full text was not independently inspected. Supports the spreading-versus-synchronous activity-depression contrast. registry ↩a ↩b ↩c ↩d ↩e ↩f
[4] Leo L, Gherardini L, Barone V, et al., Increased Susceptibility to Cortical Spreading Depression in the Mouse Model of Familial Hemiplegic Migraine Type 2, PLOS Genetics 7 (2011), e1002129, Fig. 4 and CSD Methods/Discussion. Full original article. Supports electrically evoked DC-potential propagation and genotype threshold/speed effects; astrocytic glutamate clearance is a hypothesis. The described methods do not directly record ECoG suppression. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g
[5] Hadjikhani, N., Sanchez del Rio, M., Wu, O., et al., Mechanisms of Migraine Aura Revealed by Functional MRI in Human Visual Cortex, PNAS 98 (2001), 4687–4692, Abstract and Results/Figs. 1–3. Full original article. Supports an occipital BOLD progression during visual aura and a qualified CSD-like interpretation, not direct electrophysiological proof. registry ↩a ↩b ↩c ↩d ↩e
[6] Nakamura H, Strong AJ, Dohmen C, et al., Spreading depolarizations cycle around and enlarge focal ischaemic brain lesions, Brain 133 (2010), 1994–2006, Abstract and experimental Results. Full original article. Supports lesion-cycling and stepwise growth in its experimental focal-ischemia context, not inevitable injury in every CSD event. registry ↩a ↩b ↩c