Delayed Cerebral Ischemia and Spreading Depolarization in Absence of Angiographic Vasospasm after Subarachnoid Hemorrhage

Abstract

It has been hypothesized that vasospasm is the prime mechanism of delayed cerebral ischemia (DCI) after aneurysmal subarachnoid hemorrhage (aSAH). Recently, it was found that clusters of spreading depolarizations (SDs) are associated with DCI. Surgical placement of nicardipine prolonged-release implants (NPRIs) was shown to strongly attenuate vasospasm. In the present study, we tested whether SDs and DCI are abolished when vasospasm is reduced or abolished by NPRIs. After aneurysm clipping, 10 NPRIs were placed next to the proximal intracranial vessels. The SDs were recorded using a subdural electrode strip. Proximal vasospasm was assessed by digital subtraction angiography (DSA). 534 SDs were recorded in 10 of 13 patients (77%). Digital subtraction angiography revealed no vasospasm in 8 of 13 patients (62%) and only mild or moderate vasospasm in the remaining. Five patients developed DCI associated with clusters of SD despite the absence of angiographic vasospasm in three of those patients. The number of SDs correlated significantly with the development of DCI. This may explain why reduction of angiographic vasospasm alone has not been sufficient to improve outcome in some clinical studies.

Keywords

microcirculation neurovascular coupling spreading depression subarachnoid hemorrhage vasospasm

Introduction

Delayed cerebral ischemia (DCI) describes a specific subtype of ischemic stroke with delay of ∼1 week after aneurysm rupture in patients suffering from aneurysmal subarachnoid hemorrhage (aSAH). Erythrocyte degradation products are assumed to induce DCI (Macdonald and Weir, 1991). Traditionally, proximal spasm of basal cerebral arteries has been regarded as the prime mechanism of DCI. Consistently, the predictive value of absent angiographic vasospasm for nonappearance of DCI was high in previous studies (Burns et al, 2009a) and more recent studies confirmed an association between anigographic vasospasm and brain perfusion after aSAH (Burns et al, 2009b; Dupont et al, 2009). However, it is controversial whether angiographic vasospasm alone determines delayed brain damage after aSAH. For example, the large dominance of cortical over territorial infarcts in the pathoanatomical literature suggests alternative pathophysiological mechanisms (Neil-Dwyer et al, 1994). Moreover, a positron emission tomography study showed that highly heterogeneous perfusion patterns ranging from hypoperfusion to hyperperfusion can be associated with DCI suggesting that mechanisms contribute to DCI which are more dynamic than angiographic vasospasm (Minhas et al, 2003). The positive predictive value of vasospasm for DCI is low (Rabinstein et al, 2005; Unterberg et al, 2001). Most recently, this controversy gained momentum by the finding that the endothelin A receptor antagonist clazosentan caused a 65% relative risk reduction of angiographic vasospasm whereas the patient outcome did not improve significantly (Macdonald et al, 2008).

There is a long list of complementary mechanisms and factors that could contribute to DCI including microthrombosis (Stein et al, 2006), impaired large and small vessel reactivity (Wijdicks and Cloft, 2009), volume contraction (Abdelmoneim et al, 2009), impaired autoregulation (Jaeger et al, 2007), spreading ischemia (Dreier et al, 1998), chemical factors such as local potassium or hemoglobin released from the clot (Dreier et al, 1998), oxygen-free radicals (Zubkov and Wijdicks, 2009), bilirubin oxidation products (Stead et al, 2009), hyponatremia (Abdelmoneim et al, 2009), and endothelin-1 or nitric oxide (Pluta et al, 2009).

Spreading depolarization (SD) describes a neuronal and glial mass depolarization wave with breakdown of the ion concentrations gradients, extreme shunt of neuronal membrane resistance, loss of electrical activity (spreading depression), and neuronal swelling and distortion of dendritic spines. Thus, SD is a mechanism underlying cytotoxic edema in the gray matter of the brain and contributes presumably to lesion development and progression. Spreading depolarization is the consequence of various etiologies of ischemia as well as a potential cause since it can induce severe microarterial spasm under pathological conditions (spreading ischemia) (Mateen et al, 2009).

Recently, it was observed that clusters of SD occur time locked to the development of DCI in patients with aSAH (Bosche et al, 2010; Dreier et al, 2006, 2009). However, these patients also suffered from significant proximal vasospasm in 72% (23/32) (Bosche et al, 2010; Dreier et al, 2006, 2009). Thus, it remained unclear whether SD is induced by proximal vasospasm in the delayed period or could also be due to other potential etiologies of delayed ischemia.

Here, we present a prospective case series study of 13 patients who received nicardipine prolonged-release implants (NPRI). The NPRIs were shown previously to reduce or abolish angiographic vasospasm (Barth et al, 2007). The a priori hypothesis of the present study was that blockade of angiographic vasospasm by NPRIs would prevent DCI and SDs.

Materials and methods

Patient Management

This study was approved by the local research ethics committee of the Universitätsmedizin Mannheim. Informed consent was obtained from the patient or legal representative. Thirteen consecutive patients with major aSAH and the clinical decision for surgical aneurysm repair were recruited and treated according to guidelines of the German Society of Neurosurgery. Clinical presentation was assessed according to the World Federation of Neurological Surgeons (WFNS) scale (Drake, 1988). Hemorrhage on computed tomography (CT) scans was graded according to the Fisher scale (Kistler et al, 1983). Aneurysm surgery was performed early after aSAH (between 4 and 22 hours) and 10 NPRIs containing 4 mg nicardipine were placed adjacent to the proximal cerebral vessels as described previously (Barth et al, 2007). A 6-contact platinum subdural electrocorticography (ECoG) recording strip (Ad-Tech Medical, Racine, WI, USA) was positioned over the ipsilateral frontal cortex via the craniotomy as reported previously (Dreier et al, 2006). A Clark-type probe (Licox CC1-SB; Integra Neuroscience, Andover, UK) was placed next to the ECoG-strip electrode in an oblique manner to assess cortical tissue partial pressure of oxygen (ptiO₂). A thermal diffusion regional cerebral blood flow (TD-rCBF) probe (Hemedex Cerebral Blood Flow Monitoring System; Codman, Raynham, MA, USA) was placed in the neighboring subcortical white matter.

After surgery, the patients were transferred to the intensive care unit. Intracranial pressure (ICP) was monitored via an external ventricular drain. No prophylactic triple-H-therapy (hypertension, hypervolemia, and hemodilution) was performed. Blood gases, electrolytes, and glucose were controlled every 4 hours and a neurologic examination was performed at least twice per day.

Delayed cerebral ischemia (DCI) was defined either by a new focal deficit, a decrease in the level of consciousness (Glasgow coma scale decline >2 points) or by a new infarct proven by CT or magnetic resonance imaging (MRI). Postoperative CTs were performed 1 day after surgery and then once a week or after clinical deterioration. An MRI was performed before discharge from the intensive care unit.

Transcranial Doppler sonography was performed daily. Vasospasm was suspected above a mean flow velocity>150 cm/s in at least one middle cerebral artery (MCA) or by a rise in flow velocity of >50 cm/s compared with the previous transcranial Doppler sonography examination. If vasospasm was suspected by either transcranial Doppler sonography or clinical deterioration, an immediate DSA was performed. If patients showed no signs of vasospasm by transcranial Doppler sonography and no clinical deterioration, then DSA was performed on day 8±1 after aneurysm rupture. Vasospasm was graded as mild (33% reduction in vessel diameter), moderate (>33% to 66% reduction), or severe (>66% reduction) when present in at least one of the following vessel segments: A1, A2, M1, M2, and C1 to C2 (Unterberg et al, 2001). Magnification errors were corrected by comparing extradural segments of the internal carotid artery (C4 to C5). Assessment of angiographic vasospasm was performed by MD who was blinded to the study results. Symptomatic and/or significant DSA-proven vasospasm was treated by triple-H-therapy. Triple-H-therapy aimed at a mean arterial pressure of 100 to 130 mm Hg, a hematocrit around 30% and a central venous pressure around 10 mm Hg. For this purpose, 250 mL hydroxyethyl starch (10%) was administered four times daily and crystalloids were infused at a rate of 120 mL/h. Norepinephrine was administered if fluid therapy alone failed to secure adequate mean arterial pressure (Unterberg et al, 2001).

Data Collection and Analysis

The ECoG activity was recorded as described previously (Dreier et al, 2006). Electrode one served as ground while the remaining electrodes were connected in a sequential bipolar manner to a GT205 amplifier (0.01 to 100 Hz) (ADInstruments, New South Wales, Australia). Data were sampled at 200 Hz and recorded with a Powerlab 8/SP analogue/digital converter and Chart-5 software (ADInstruments). Spreading depolarization was defined by the sequential onset of a propagating, polyphasic slow potential change (Fabricius et al, 2006) in adjacent channels, corresponding to the negative slow voltage variation described by Leão (1947). The accompanying ECoG depression was defined by a rapid reduction of power in the highpass-filtered (lower frequency limit, 0.5 Hz) ECoG amplitude (Fabricius et al, 2006; Strong et al, 2002). The duration of the depression period was determined as the interval between depression onset and onset of activity restoration using mathematical integration of the power of the highpass-filtered ECoG activity (time constant delay, 60 seconds) as described previously (Dreier et al, 2006). The ECoG analysis was performed in a blinded manner to the clinical course of individual patients by JD, MW, and YD.

An ICP monitor is considered the gold standard in neurocritical care. However, recent studies have suggested that cerebral ischemia can occur despite normal ICP and cerebral perfusion pressure (CPP) values (Adams et al, 2007). To mirror the hemodynamic situation more precisely, many neurointensivists have adopted flow assessment (Barth et al, 2010) or surrogate flow assessment by continuous ptiO₂ measurement in patients with aSAH (Barth et al, 2010; Jaeger et al, 2007). The ptiO₂ is affected by both oxygen delivery and demand. Apart from the CBF, this balance is influenced by the arterial oxygen content and the cerebral metabolism with the arterial oxygen pressure as the main driving force for oxygen diffusion into the tissue. Based on the previous studies, a lower threshold value of 10 to 15 mm Hg has been suggested for prevention of ischemia (Kiening et al, 1996). The ptiO₂ and TD-rCBF were assessed with a Licox system (Integra NeuroSciences, Hamphire, UK) and the Hemedex Cerebral Blood Flow Monitoring System (Hemedex, Raynham, MA, USA), respectively. The ptiO₂ and TD-rCBF were recorded simultaneously with mean arterial pressure and ICP at 0.3 Hz using a custom-made analogue/digital converter (MIDAS, University of Mannheim, Germany). The ptiO₂ and TD-rCBF changes were considered significant if a rapid (120 seconds) ptiO₂ (±2 mm Hg) or TD-rCBF deviation (±2 mL/100 g/min) from baseline occurred in a temporal range of 6 minutes around SD occurrence not accompanied by a similar change in CPP (Bosche et al, 2010).

Statistical Analysis

Data are presented as mean ±95% confidence interval. One patient (no. 7) with ischemia related to the surgical procedure was excluded from the analysis of comparing patients with and without DCI. Statistical tests are given in the text. P<0.05 was considered statistically significant.

Results

Incidence of Angiographic Vasospasm and Delayed Cerebral Ischemia

Digital subtraction angiography revealed no vasospasm in 8 of 13 patients (62%) and only mild or moderate vasospasm in the remaining. The detailed description is given in Table 1.

Table 1

Summary of demographic, clinical, and monitoring data

No.	Age (years), sex	WFNS grade/Fisher grade	Additional ICH, volume (mL)	Location of aneurysm	Degree of DSA proven vasospasm/Day of DSA	Clinical DCI	Delayed CT/MRI-proven infarct	Duration of ECoG monitoring (hours)	Total number of SDs	Total duration of ECoG depression period (minutes)	Number of decreases, biphasic responses and increases of ptiO ₂	eGOS
1	71, f	4/3	N	MCA right	None/7	Y	Y	172	42	383	7/3/2	3
2	69, m	5/3	Y/(80)	MCA right	None/8	N	Y	228	31	294	2/4/0	1
3	43, f	5/2	N	pericalA	None/7	Sedated^a	Y	139	116	1,361	6/3/9	3
4	49, m	4/3	Y/(4)	ACoA	None/9	Sedated	Y	136	31	366	16/0/0	3
5	47, f	5/3	N	ACoA	Moderate/7	Y	N	166	45	247	3/0/0	5
6	35, f	2/3	N	MCA right	Mild/7	Y	N	225	142	656	5/91/21	7
7	52, f	2/3	N	MCA right	Mild/9	Sedated	N^b	307	50	878	2/8/0	2
8	62, f	5/3	N	ICA right	Moderate/6	N	N	334	51	557	0/3/2	4
9	63, m	2/3	N	ACoA	None/8	N	N	218	22	141	0/0/0	7
10	47, f	1/3	N	ICA right+PCoA left	Moderate/8	N	N	214	4	17	0/0/0	3
11	54, m	3/3	Y/(95)	MCA left	None/8	Sedated	N	158	0	0	0/0/0	5
12	67, f	2/3	Y/(5)	ICA right	None/7	N	N	166	0	0	0/0/0	7
13	50, f	2/2	N	ACoA	None/9	N	N	160	0	0	0/0/0	5

WFNS, World Federation of Neurological Surgeons scale; ICH, intracerebral hemorrhage; DSA, digital subtraction angiography; DCI, delayed cerebral ischemia; ECoG, electrocorticography; SD, spreading depolarization; ptiO₂, tissue partial pressure of oxygen; eGOS, extended Glasgow outcome score; f, female; m, male; MCA, middle cerebral artery; pericalA, pericallosal artery; ACoA, anterior communicating artery; ICA, internal carotid artery; PCoA, posterior communicating artery; CT, computed tomography; MRI, magnetic resonance imaging.

Due to small subdural hematoma with brain swelling and mild midline shift during the first 4 days.

Postoperative ischemia right frontal region.

A total number of six patients developed DCI. Four of those patients developed delayed CT- or MRI-proven infarcts. In all four patients, proximal vasospasm was excluded as etiology of DCI. However, one patient (no. 3) had a possible hemodynamic coimpairment due to mild brain swelling and a small subdural hematoma with mild midline shift. The ICP was successfully controlled in this patient by sedation and cerebrospinal fluid drainage.

Of the two patients who developed transient clinical DCI without infarcts, one patient showed moderate vasospasm in the right anterior cerebral artery and posterior cerebral artery and mild vasospasm in both MCAs and the left anterior cerebral artery. The other patient exhibited mild vasospasm in the A2 and M2 segments bilaterally.

One patient (no. 7) suffered from a postoperative partial MCA infarct in the right frontal territory. Since ischemia is associated with a high incidence of SDs (Zubkov et al, 2008) and this infarct developed early after intervention and was not attributed to the delayed time period after aSAH we have excluded this patient from further analysis.

Incidence of Spreading Depolarizations

During a total recording time of 2,625 hours, 534 SDs were assessed in 10 of 13 patients (77%). The mean depression period per SD was 10.4±1.5 minutes, resulting in a total depression period of 81.6 hours in all patients. The SD propagation velocity was 2.5±0.1 mm/min assuming an ideal linear propagation along the electrode strip. Figure 1 and Table 1 summarize the demographic, clinical, and monitoring data.

Figure 1

Temporal correlation of neurologic and clinical scores, spreading depolarization (SD) and coregistered responses of the tissue partial pressure of oxygen (ptiO₂). The figure summarizes 10 of 13 patients in whom SDs occurred. The length of the black horizontal lines marks the individual monitoring period in each patient. The length of the black vertical lines illustrates the duration of the depression period. Negative black vertical lines denote the occurrence of a slow potential change before recovery of the electrocorticography (ECoG) activity. Red vertical lines illustrate the coregistered increases and decreases of the ptiO₂ in response to SD. The gray-shaded area indicates periods with sedation. Arrows indicate the occurrence of delayed neurologic deficits, defined by focal deficits or a decrease of the level of consciousness by at least two Glasgow coma scale points (DOC=decrease of consciousness). In a similar manner, new CT- or MRI-proven infarcts are given (MCA, middle cerebral artery; ACA, anterior cerebral artery; PCA, posterior cerebral artery). Patient no. 7 suffered from an early postoperative frontal MCA ischemia, which is marked by an asterisk. Patient no. 6 developed a long lasting cluster of SDs without recovery of the ECoG activity (striped area). CT, computed tomography; MRI, magnetic resonance imaging; SAH, subarachnoid hemorrhage.

Patients with clinical or CT/MRI-proven DCI had a significantly higher number of SDs and significantly longer periods of depressed ECoG activity per day recording time than patients without DCI (P<0.01; Mann—Whitney rank-sum tests; Figure 2). The mean number of SDs per day recording time was 8.7±12.2, 10.8±54.9, and 1.1±1.7 in patients with infarcts, patients with only clinical DCI, and patients without DCI, respectively. The mean time of depressed ECoG activity per day recording time was 96.0±149.1, 52.8±217.7, and 9.6±16.9 minutes, respectively. Apart from patient no. 4, no infarct occurred within the recording area and all infarcts were relatively small (6, 3, 18, and 7 cm³ in patient nos. 1, 2, 3, and 4, respectively).

Figure 2

Delayed cerebral ischemia (DCI) is possibly associated with the number of spreading depolarizations (SDs) and total duration of electrocorticography (ECoG) depression periods.

Delayed Cerebral Ischemia and Clusters of Spreading Depolarizations in the Absence of Proximal Vasospasm—Illustrative Case 1

This 71-year-old female patient suffered from major aSAH (WFNS grade 4/Fisher grade 3) due to a ruptured right-sided MCA aneurysm (Figure 3B). After surgery, the patient was transferred to the intensive care unit and had an uneventful course until day 7 when several SDs with depression periods of up to 18.5 minutes occurred. On day 7, a cluster of SDs coincided with left-sided hemiparesis and decrease in consciousness necessitating reintubation. Proximal vasospasm was excluded as possible etiology of DCI by DSA (Figure 3D). On day 13, MRI revealed a new laminar cortical necrosis in the right MCA (Figures 3E and 3F) and to a lesser degree in the anterior cerebral artery territory.

Figure 3

Electrocorticography (ECoG) of a 71-year-old female (patient no. 1) suffering from aneurysmal subarachnoid hemorrhage (aSAH) (World Federation of Neurological Surgeons (WFNS) grade 4/Fisher grade 3). The patient showed several spreading depolarizations (SDs) (arrows) during the first 7 days after aneurysm rupture (compare Figure 1 for the temporal course). On day 7, the patient developed a left-sided hemiparesis and loss of consciousness which necessitated reintubation. Digital subtraction angiography (DSA) (D) excluded proximal vasospasm as a source for the delayed neurologic deficit. Magnetic resonance imaging with diffusion weighted imaging (DWI, E) and fluid-attenuated inversion recovery (Flair) pulse sequences (F) on day 13 revealed a laminar cortical necrosis in the right middle cerebral artery territory and anterior cerebral artery territory (not shown). The ECoG in panel A shows bipolar recordings between electrodes 2 and 3 (channel A), 3 and 4 (channel B), 4 and 5 (channel C), and 5 and 6 (channel D). Recordings from a coimplanted Clark-type oxygen sensor showed monophasic increases (not shown), biphasic changes (not shown), or monophasic decreases (lower trace) of tissue partial pressure of oxygen (ptiO₂) in close temporal relationship to SDs. (B) The initial computed tomography (CT) with major aSAH is shown. The CT scan performed on day 6 (C) shows artifacts of the electrode strip implanted over the right frontal cortex. No signs of ischemic stroke are visualized at this time point.

Spreading Depolarization-Coupled Cerebral Blood Flow and Tissue Partial Pressure of Oxygen Responses

In 8/13 patients, 187 SDs (35%) were associated with monophasic decreases (n=41), biphasic responses (112), or monophasic increases (34) of ptiO₂. The median amplitude and duration of the ptiO₂ decrease in response to SD were −3.5±0.6 mm Hg and 3.8±0.3 minutes, respectively. The median amplitude and duration of the ptiO₂ increase in response to SD were 5.8±0.5 mm Hg and 6.4±0.7 minutes, respectively. The summarized data of the ptiO₂ responses are given in Table 1. A corresponding TD-rCBF response in the subcortical white matter was only noted during 6 out of 534 SDs in the current data set.

Cerebral Perfusion Pressure Seems To Determine the Tissue Partial Pressure of Oxygen Responses—Illustrative Case 2

Patient no. 6 (35-year-old female) presented with frequent SDs during the entire monitoring period (Figure 1). In all, 5 of the 117 SDs were associated with monophasic decrease, 91 with biphasic, and 21 with monophasic increase of the ptiO₂ response.

To investigate the ptiO₂ responses without the confounding factor of proximal vasospasm, we performed a detailed ptiO₂-response analysis during the first 4 days after aneurysm rupture. The analysis revealed a CPP-dependent increase/decrease of the ptiO₂ response (Figure 4). The higher the baseline CPP the larger was the resulting increase of the ptiO₂ response (R²=0.43, P<0.001, linear regression). Conversely, the lower the baseline CPP the larger was the resulting decrease of the ptiO₂ response (R²=0.54, P<0.001, linear regression).

Figure 4

Coregistered responses of tissue partial pressure of oxygen (ptiO₂) in temporal relationship to spreading depolarizations (SDs) (A) observed during the first 4 days in patient no. 6. Monophasic increases, decreases, or biphasic responses were detected. A lower baseline level of cerebral perfusion pressure (CPP) was associated with less pronounced increase and more pronounced decrease of the ptiO₂ response. Whereas panel B depicts the negative correlation between baseline CPP and the ptiO₂ decrease in response to SD (∗∗P<0.001, linear regression correlation), panel C shows the positive correlation between baseline CPP and the ptiO₂ increase in response to SD (∗∗P<0.001, linear regression correlation).

Discussion

Delayed cerebral ischemia is considered to be one of the major in-hospital predictors for poor clinical outcome (Haley et al, 1997; Lanzino and Kassell, 1999; Lanzino et al, 1999; Rabinstein et al, 2005). Neurosurgeons, neurologists, and neurointensivists have often attributed DCI to proximal vasospasm. Consequently, a great effort was undertaken in basic and clinical research aiming at the prevention of angiographic vasospasm. On this way, arguments were found in support of this hypothesis as well as against it (Dreier et al, 2002; Macdonald et al, 2008; Minhas et al, 2003; Rabinstein et al, 2005). In the present prospective, diagnostic study, we found that DCI and delayed ischemic stroke can also occur in the absence of angiographic vasospasm, consistent with some previous reports (Brouwers et al, 1992; Weidauer et al, 2008). Moreover, we provide evidence that clusters of SDs can occur in absence of angiographic vasospasm (Dreier et al, 2006, 2009). Furthermore, we found preliminary evidence that SD-induced changes of ptiO₂ showed significant correlation with CPP similar to animal experiments (Sukhotinsky et al, 2010).

Delayed Cerebral Ischemia in Absence of Angiographic Vasospasm

In a patient population like ours, which is characterized mostly by Fisher grade 3 and high WFNS grades, the incidence rate of vasospasm is typically very high (between 70% and 80%) (Barth et al, 2007). In such a population, also the incidence of DCI is high compared with that in patients with lower WFNS and Fisher grades. In total, DCI occurs in 33% to 38% of all patients with aSAH and led to CT-proven delayed infarcts in 10% to 13% in the large tirilazad mesylate trials (Haley et al, 1997; Lanzino and Kassell, 1999; Lanzino et al, 1999). A surprising finding in our study was that there remained a relatively high rate of DCI although, similarly to a previous study, angiographic vasospasm was significantly reduced or even abolished by the treatment with NPRIs (Barth et al, 2007). In the present study, the incidence rate of delayed stroke was 31% whereas it was 14% in the NPRI arm of the study by Barth et al (2007). In their control group, the incidence rate of CT-proven delayed infarcts had been 47% similar to the incidence rate reported by Rabinstein et al (2005) (39%). This seems relatively high compared with other studies (Dreier et al, 2002; Weidauer et al, 2008) but presumably reflects improved CT technology rather than a real increase in lesion occurrence over the last decade. Our present findings do not contradict an association between angiographic vasospasm and DCI in untreated patients. Rather, our findings put into question that proximal vasospasm is the single cause of DCI. Possibly, breakdown products of erythrocytes lead to both angiographic vasospasm and other etiologies of DCI as listed in the introduction (Rabinstein and Wijdicks, 2009). This hypothesis is also supported by the different infarct patterns reported previously (Rabinstein et al, 2005; Sukhotinsky et al, 2010) and alternative etiologies could explain why blockade of angiographic vasospasm alone was not sufficient to improve patient outcome in a previous study with an endothelin A receptor antagonist (Macdonald et al, 2008).

Delayed Cerebral Ischemia, Angiographic Vasospasm, and Spreading Depolarization

In the present study, no patient developed severe angiographic vasospasm in presence of NPRIs although most patients had suffered from Fisher grade 3 hemorrhages. A considerable number of CT/MRI-proven delayed infarcts (4/13; 31%) and clinical DCI (3/13; 23%), respectively, were detected although angiographic vasospasm was absent in 8 of 13 patients (62%) and only mild or moderate in 2 (15%) and 3 (23%) patients, respectively. However, a high number of 534 SDs were recorded in 10 of 13 patients (77%). Both the number of SDs and the total duration of ECoG depression correlated significantly with the incidence of DCI. In all patients with CT/MRI-proven infarcts, proximal vasospasm was excluded by DSA. In a previous report on aSAH patients without NPRI treatment, the incidence of SDs was 72% (Dreier et al, 2006). The similar incidence in the present study suggests that SDs are triggered by another mechanism than proximal vasospasm. However, in contrast to previous reports (Dreier et al, 2006, 2009), the patients in our study developed only relatively small infarcts (ranging from ∼3 to 18 cm³). This could be due to a nicardipine-induced effect on calcium-dependent potassium channels but more likely, this is explained as follows: In presence of breakdown products of erythrocytes, SD is one of the most potent inducers of acute arterial spasm currently known in animals (Mateen et al, 2009). This acute vasospasm causes a spreading ischemia in the cortex whereas under physiological conditions vasodilatation and hyperemia are coupled to SD (Dreier et al, 1998, 2000). Similar spreading ischemia was measured recently in response to SD in patients suffering from aSAH using subdural opto-electrode strips for ECoG and laser-Doppler flowmetry (Dreier et al, 2009): A preexisting perfusion deficit due to proximal vasospasm could largely augment the severity, duration, and spatial extension of spreading ischemia and thus, it could expand the lesions. Consistently, experimental evidence suggests that spreading ischemia is augmented and prolonged by a decline in CPP (Dreier et al, 2000) while an artificial CPP increase by Triple-H-therapy can convert the spreading ischemic to an almost normal spreading hyperemic response to SD (Sukhotinsky et al, 2010). These experimental studies are also in line with the present finding that the ptiO₂ decreases in response to SD were more pronounced when CPP was decreased (Figure 4).

The ptiO₂ findings are consistent with previous reports (Bosche et al, 2010; Dreier et al, 2009). However, in contrast to previous measurements of cortical rCBF in a small sample volume of ∼1 mm³ using laser-Doppler flowmetry (Dreier et al, 2009), subcortical rCBF recordings in a sample volume of ∼0.5 cm³ using thermal diffusion flowmetry did not reveal any significant perfusion changes in response to the large majority of SDs in the present study. This observation could be related to the fact that propagation of SD is restricted to the cortical compartment. Thus, the hemodynamic responses to SD could be restricted to the cortex as well. However, we cannot exclude that more pronounced responses of rCBF to SD would have been detected in the subcortical white matter when proximal vasospasm would have been present and may have prevented compensation for SD-induced perfusion deficits by collateral blood supply.

The present study has limitations. This concerns the small patient number and the lack of a control group. To exclude the possible bias in the clinical assessment to detect DCI in the absence of vasospasm, a large randomized trial of NPRIs is needed which should be flanked by multimodal neuromonitoring technology. Moreover, using this technology it will have to be investigated whether the observed SD-induced decreases in ptiO₂ are in fact of biological significance for the tissue and patient outcome.

In summary, we found that (1) there is a high incidence of SDs in patients with aSAH despite the treatment of proximal vasospasm, (2) DCI may occur in the absence of proximal vasospasm with a possible link to SDs, and we found preliminary evidence that (3) the baseline level of CPP shows an effect on the SD-induced ptiO₂ responses in the human brain in a similar manner to that in animals (Sukhotinsky et al, 2010). Furthermore, robust clinical trials are necessary to confirm the possible association between SD and DCI.

Angiographic vasospasm remains a target for therapy but potential alternative mechanisms of DCI should be considered as well in the future drug development. Possibly, combination therapies contain the answer to this clinical problem. However, clinical testing of combination therapies will be challenging and will have to be well prepared from bench to bedside.

Footnotes

The authors declare no conflict of interest.

References

Abdelmoneim

Wijdicks

Lee

Daugherty

Bernier

Pellikka

Mulvagh

(2009) Real-time myocardial perfusion contrast echocardiography and regional wall motion abnormalities after aneurysmal subarachnoid hemorrhage. Clinical article. J Neurosurg 111:1023–8.

Adams

Jr Del Zoppo

Alberts

Bhatt

Brass

Furlan

Grubb

Higashida

Jauch

Kidwell

Lyden

Morgenstern

Qureshi

Rosenwasser

Scott

Wijdicks

(2007) Guidelines for the early management of adults with ischemic stroke: a guideline from the American Heart Association/American Stroke Association Stroke Council, Clinical Cardiology Council, Cardiovascular Radiology and Intervention Council, and the Atherosclerotic Peripheral Vascular Disease and Quality of Care Outcomes in Research Interdisciplinary Working Groups: The American Academy of Neurology affirms the value of this guideline as an educational tool for neurologists. Circulation 115:478–534.

Barth

Capelle

Weidauer

Weiss

Munch

Thome

Luecke

Schmiedek

Kasuya

Vajkoczy

(2007) Effect of nicardipine prolonged-release implants on cerebral vasospasm and clinical outcome after severe aneurysmal subarachnoid hemorrhage: a prospective, randomized, double-blind phase IIa study. Stroke 38:330–6.

Barth

Woitzik

Weiss

Muench

Diepers

Schmiedek

Kasuya

Vajkoczy

(2010) Correlation of clinical outcome with pressure-, oxygen-, and flow-related indices of cerebrovascular reactivity in patients following aneurysmal SAH. Neurocrit Care 12:234–43.

Bosche

Graf

Ernestus

Dohmen

Reithmeier

Brinker

Strong

Dreier

Woitzik

(2010) Recurrent spreading depolarizations after subarachnoid hemorrhage decreases oxygen availability in human cerebral cortex. Ann Neurol 67:607–17.

Brouwers

Wijdicks

Van Gijn

(1992) Infarction after aneurysm rupture does not depend on distribution or clearance rate of blood. Stroke 23:374–9.

Burns

Rabinstein

Cloft

Lanzino

Daniels

Wijdicks

(2009a) Minimally conscious state after ruptured giant basilar aneurysm. Arch Neurol 66:786–8.

Burns

Schiefer

Wijdicks

(2009b) Large and small: a telltale sign of acute pontomesencephalic injury. Neurology 72:1707.

Drake

(1988) Report of World Federation of Neurological Surgeons Committee on a Universal Subarachnoid Hemorrhage Grading scale. J Neurosurg 68:985–6.

10.

Dreier

Ebert

Priller

Megow

Lindauer

Klee

Reuter

Imai

Einhaupl

Victorov

Dirnagl

(2000) Products of hemolysis in the subarachnoid space inducing spreading ischemia in the cortex and focal necrosis in rats: a model for delayed ischemic neurological deficits after subarachnoid hemorrhage? J Neurosurg 93:658–66.

11.

Dreier

Korner

Ebert

Gorner

Rubin

Back

Lindauer

Wolf

Villringer

Einhaupl

Lauritzen

Dirnagl

(1998) Nitric oxide scavenging by hemoglobin or nitric oxide synthase inhibition by N-nitro-L-arginine induces cortical spreading ischemia when K+ is increased in the subarachnoid space. J Cereb Blood Flow Metab 18:978–90.

12.

Dreier

Major

Manning

Woitzik

Drenckhahn

Steinbrink

Tolias

Oliveira-Ferreira

Fabricius

Hartings

Vajkoczy

Lauritzen

Dirnagl

Bohner

Strong

(2009) Cortical spreading ischaemia is a novel process involved in ischaemic damage in patients with aneurysmal subarachnoid haemorrhage. Brain 132:1866–81.

13.

Dreier

Windmuller

Petzold

Lindauer

Einhaupl

Dirnagl

(2002) Ischemia triggered by red blood cell products in the subarachnoid space is inhibited by nimodipine administration or moderate volume expansion/hemodilution in rats. Neurosurgery 51:1457–65.

14.

Dreier

Woitzik

Fabricius

Bhatia

Major

Drenckhahn

Lehmann

Sarrafzadeh

Willumsen

Hartings

Sakowitz

Seemann

Thieme

Lauritzen

Strong

(2006) Delayed ischaemic neurological deficits after subarachnoid haemorrhage are associated with clusters of spreading depolarizations. Brain 129:3224–37.

15.

Dupont

Wijdicks

Manno

Lanzino

Brown

Jr Rabinstein

(2009) Timing of computed tomography and prediction of vasospasm after aneurysmal subarachnoid hemorrhage. Neurocrit Care 11:71–5.

16.

Fabricius

Fuhr

Bhatia

Boutelle

Hashemi

Strong

Lauritzen

(2006) Cortical spreading depression and peri-infarct depolarization in acutely injured human cerebral cortex. Brain 129:778–90.

17.

Haley

Jr Kassell

Apperson-Hansen

Maile

Alves

(1997) A randomized, double-blind, vehicle-controlled trial of tirilazad mesylate in patients with aneurysmal subarachnoid hemorrhage: a cooperative study in North America. J Neurosurg 86:467–74.

18.

Jaeger

Schuhmann

Soehle

Nagel

Meixensberger

(2007) Continuous monitoring of cerebrovascular autoregulation after subarachnoid hemorrhage by brain tissue oxygen pressure reactivity and its relation to delayed cerebral infarction. Stroke 38:981–6.

19.

Kiening

Unterberg

Bardt

Schneider

Lanksch

(1996) Monitoring of cerebral oxygenation in patients with severe head injuries: brain tissue PO2 versus jugular vein oxygen saturation. J Neurosurg 85:751–7.

20.

Kistler

Crowell

Davis

Heros

Ojemann

Zervas

Fisher

(1983) The relation of cerebral vasospasm to the extent and location of subarachnoid blood visualized by CT scan: a prospective study. Neurology 33:424–36.

21.

Lanzino

Kassell

(1999) Double-blind, randomized, vehicle-controlled study of high-dose tirilazad mesylate in women with aneurysmal subarachnoid hemorrhage. Part II. A cooperative study in North America. J Neurosurg 90:1018–24.

22.

Lanzino

Kassell

Dorsch

Pasqualin

Brandt

Schmiedek

Truskowski

Alves

(1999) Doubleblind, randomized, vehicle-controlled study of high-dose tirilazad mesylate in women with aneurysmal subarachnoid hemorrhage. Part I. A cooperative study in Europe, Australia, New Zealand, and South Africa. J Neurosurg 90:1011–7.

23.

Leão

(1947) Further observations on the spreading depression of activity in the cerebral cortex. J Neurophysiol 10:409–14.

24.

Macdonald

Kassell

Mayer

Ruefenacht

Schmiedek

Weidauer

Frey

Roux

Pasqualin

(2008) Clazosentan to overcome neurological ischemia and infarction occurring after subarachnoid hemorrhage (CONSCIOUS-1): randomized, double-blind, placebo-controlled phase 2 dose-finding trial. Stroke 39:3015–21.

25.

Macdonald

Weir

(1991) A review of hemoglobin and the pathogenesis of cerebral vasospasm. Stroke 22:971–82.

26.

Mateen

Nasser

Spencer

Freeman

Shuaib

Demaerschalk

Wijdicks

(2009) Outcomes of intravenous tissue plasminogen activator for acute ischemic stroke in patients aged 90 years or older. Mayo Clin Proc 84:334–8.

27.

Minhas

Menon

Smielewski

Czosnyka

Kirkpatrick

Clark

Pickard

(2003) Positron emission tomographic cerebral perfusion disturbances and transcranial Doppler findings among patients with neurological deterioration after subarachnoid hemorrhage. Neurosurgery 52:1017–22.

28.

Neil-Dwyer

Lang

Doshi

Gerber

Smith

(1994) Delayed cerebral ischaemia: the pathological substrate. Acta Neurochir (Wien) 131:137–45.

29.

Pluta

Hansen-Schwartz

Dreier

Vajkoczy

Macdonald

Nishizawa

Kasuya

Wellman

Keller

Zauner

Dorsch

Clark

Ono

Kiris

Leroux

Zhang

(2009) Cerebral vasospasm following subarachnoid hemorrhage: time for a new world of thought. Neurol Res 31:151–8.

30.

Rabinstein

Weigand

Atkinson

Wijdicks

(2005) Patterns of cerebral infarction in aneurysmal subarachnoid hemorrhage. Stroke 36:992–7.

31.

Rabinstein

Wijdicks

(2009) Fatal brain swelling due to superior vena cava syndrome. Neurocrit Care 10:91–2.

32.

Stead

Wijdicks

Bhagra

Kashyap

Bellolio

Nash

Enduri

Schears

William

(2009) Validation of a new coma scale, the FOUR score, in the emergency department. Neurocrit Care 10:50–4.

33.

Stein

Browne

Chen

Smith

Graham

(2006) Thromboembolism and delayed cerebral ischemia after subarachnoid hemorrhage: an autopsy study. Neurosurgery 59:781–7.

34.

Strong

Fabricius

Boutelle

Hibbins

Hopwood

Jones

Parkin

Lauritzen

(2002) Spreading and synchronous depressions of cortical activity in acutely injured human brain. Stroke 33:2738–43.

35.

Sukhotinsky

Yaseen

Sakadzic

Ruvinskaya

Sims

Boas

Moskowitz

Ayata

(2010) Perfusion pressure-dependent recovery of cortical spreading depression is independent of tissue oxygenation over a wide physiologic range. J Cereb Blood Flow Metab 30:1168–77.

36.

Unterberg

Sakowitz

Sarrafzadeh

Benndorf

Lanksch

(2001) Role of bedside microdialysis in the diagnosis of cerebral vasospasm following aneurysmal subarachnoid hemorrhage. J Neurosurg 94:740–9.

37.

Weidauer

Vatter

Beck

Raabe

Lanfermann

Seifert

Zanella

(2008) Focal laminar cortical infarcts following aneurysmal subarachnoid haemorrhage. Neuroradiology 50:1–8.

38.

Wijdicks

Cloft

(2009) Is prediction of outcome with magnetic resonance imaging in postresuscitation coma achievable and accurate? Ann Neurol 65:364–6.

39.

Zubkov

Rabinstein

Manno

Wijdicks

(2008) Prolonged refractory status epilepticus related to thrombotic thrombocytopenic purpura. Neurocrit Care 9:361–5.

40.

Zubkov

Wijdicks

(2009) Deep venous thrombosis prophylaxis in cerebral hemorrhage. Rev Neurol Dis 6:21–5.