Reduced ADAMTS13 Activity in Delayed Cerebral Ischemia after Aneurysmal Subarachnoid Hemorrhage

Abstract

The pathogenesis of delayed cerebral ischemia (DCI) after aneurysmal subarachnoid hemorrhage (SAH) remains unknown. Besides vasospasm, microthrombosis might have an important function. As in patients with thrombotic thrombocytopenic purpura an A Disintegrin And Metalloprotease with ThromboSpondin repeats-13 (ADAMTS13) deficiency leads to higher concentrations of large von Willebrand factor (vWF) multimers resulting in microthrombosis, our purpose was to compare ADAMTS13 and vWF in patients with and without DCI after aneurysmal SAH. We measured ADAMTS13 activity, vWF antigen, vWF propeptide, and vWF ristocetin cofactor activity in plasma at standard intervals. Thirty-one patients were included. Eleven patients (35%) developed DCI. No differences were observed in baseline characteristics between patients with and without DCI. Patients with DCI had a stronger decrease in ADAMTS13 activity, and a more profound increase in vWF antigen, vWF propeptide, and vWF activity in the first few days after the hemorrhage (P-values for difference in polynomial time trend 0.0001, 0.020, 0.004, and 0.188, respectively). No indication of correlation between vWF antigen and ADAMTS13 was found (r=−0.027, P=0.736). Our results suggest that microthrombosis has a role in the pathogenesis of DCI, as a result of decreased ADAMTS13 activity and endothelium dysfunction.

Keywords

ADAMTS13 delayed cerebral ischemia subarachnoid hemorrhage von Willebrand factor

Introduction

Delayed cerebral ischemia (DCI) is a common complication after aneurysmal subarachnoid hemorrhage (SAH), which occurs in approximately 30% of patients (Roos et al, 2000). Although DCI is common, the pathogenesis of DCI has not yet been elucidated. Clinical signs and symptoms of DCI are often attributed to macrovascular vasospasm shown by angiography. However, DCI can occur in the absence of vasospasm, and, conversely, severe vasospasm can occur in the absence of symptoms of DCI. In an autopsy study, cortical infarcts were found in 78% of cases and represented the most common infarct pattern (Neil-Dwyer et al, 1994; Stoltenburg-Didinger and Schwarz, 1987). These lesions typically occurred in areas covered with subarachnoid blood, suggesting local effects of the clot. Especially for these cortical infarctions, several factors other than macrovascular spasm can be responsible, such as microthrombosis, embolic infarcts, microvascular spasm, and cortical spreading ischemia (Vergouwen et al, 2008; Romano et al, 2002; Dreier et al, 2006).

Von Willebrand factor (vWF) is a glycoprotein that induces platelet adhesion and aggregation at sites of vascular injury or under stress conditions. VWF is secreted as UltraLarge von Willebrand factor (ULvWF) multimers from Weibel–Palade bodies in the vascular endothelium and from α-granules of platelets. Under normal circumstances ULvWF is cleaved by a protease called A Disintegrin And Metalloprotease with ThromboSpondin repeats-13 (ADAMTS13) (Dong et al, 2002). As large vWF multimers are more potent mediators of platelet thrombus formation than small vWF multimers, cleavage of ULvWF by ADAMTS13 results in lower-molecular weight vWF forms with reduced adhesive and aggregation potential. In contrast, an ADAMTS13 deficiency, such as in patients with thrombotic thrombocytopenic purpura, leads to higher concentrations of ULvWF, which results in microthrombosis followed by ischemic complications, such as cerebral infarction (Moake et al, 1982).

As in patients with aneurysmal SAH microthrombosis might have a role in the pathogenesis of DCI, the purpose of this study was to investigate whether a similar pathogenesis of microthrombi, as has been shown in thrombotic thrombocytopenic purpura, is also present in SAH. Therefore, we compared serial measurements of vWF and ADAMTS13 in patients with and without DCI after aneurysmal SAH.

Materials and methods

We used plasma samples from an exploratory single center, prospective, randomized, double-blind, placebo-controlled trial investigating the effects of simvastatin on endothelial function, coagulation, fibrinolysis, and inflammation in patients with aneurysmal SAH (Vergouwen et al, 2009). This study was registered in the International Standard Randomised Controlled Trial registry (ISRCTN45662651) and approved by the local Institutional Review Board. In this study, use of simvastatin did not influence any parameters of endothelial function (including vWF antigen), coagulation, fibrinolysis, and inflammation, and no effect was observed on vasospasm as detected with transcranial Doppler (TCD) or clinical signs and symptoms of DCI (Vergouwen et al, 2009). Therefore, all patients of the study, except one patient who had no aneurysm on angiography, were included in this study.

Patients

Patients with signs and symptoms of aneurysmal SAH admitted to the Academic Medical Center, Amsterdam, the Netherlands were included if a computed tomography (CT) scan showed an aneurysmal or perimesencephalic bleeding pattern, in combination with the presence of an appropriate aneurysm at angiography, and if written informed consent was obtained. Exclusion criteria were (1) under 18 years of age; (2) if death seemed imminent; (3) patients using aspirin, warfarin, or statins; (4) more than 72 h after SAH; (5) contraindication for simvastatin (active liver disease, liver alanine aminotransferase, or aspartate aminotransferase more than three times the normal upper limit, myopathy); (6) kidney insufficiency; and (7) pregnancy or lactation. All patients received standard care including treatment with nimodipine 360 mg a day orally (60 mg every 4 h). Initiation of hypertension and hypervolemia therapy was at the discretion of the treating neurosurgeon, according to the Academic Medical Center Subarachnoid Hemorrhage Treatment guideline to which all neurologists, neurosurgeons, and intensive care physicians adhere.

At baseline, several baseline characteristics were recorded such as age, sex, Glasgow Coma Scale, presence of focal neurologic deficits, loss of consciousness during ictus, warning leak, blood pressure at admission, smoking, history of hypertension, and location of aneurysm.

DCI was defined as the gradual onset of new focal neurologic impairment and/or a decreased level of consciousness of at least two points as recorded on the Glasgow Coma Scale, either with cerebral infarction on CT scan compatible with clinical presentation or proven at autopsy, or in case no CT scan or autopsy was obtained, suspect for infarction with exclusion of other causes by appropriate laboratory studies (Van Gijn et al, 1994). Events were scored by one investigator (MDIV), and in case of uncertainty discussed with another investigator (YBWEMR).

Ancillary Investigations

The amount of blood on admission CT scan was calculated using the Hijdra score (Hijdra et al, 1990). Blood withdrawals were performed at six standardized moments during hospitalization: 2±1, 4±1, 7±1, 10±1, 14±1, and 17±1 days after SAH, between 8 a.m. and 9.30 a.m. avoiding the effect of diurnal fluctuations in blood parameters. During the first blood withdrawal at day 2, and the last at day 17 after SAH, patients did not use study medication. All blood withdrawals were performed by the same investigator (MDIV). In case of hospital discharge before day 17 after SAH, patients did not return to the hospital for the remaining blood withdrawals. After withdrawal, blood was directly processed and stored in a −80°C freezer, until laboratory analyses were performed. In citrate plasma we measured ADAMTS13 activity (Kokame et al, 2005), vWF antigen (ELISA using antibodies from DAKO, Glostrup, Denmark), vWF propeptide (ELISA from Sanquin, Amsterdam, the Netherlands), and vWF ristocetin cofactor activity (BC von Willebrand, Dade Behring, Marburg, Germany). ADAMTS13 antigen and autoantibodies against ADAMTS13 were determined with Imubind ELISA kits from American Diagnostica, only at day 7 after SAH (Greenwich, CT, USA). A plasma sample was considered positive for autoantibodies if the value exceeded that of a locally prepared pooled normal plasma (pool of more than 200 healthy volunteers). Measurements of prothrombin fragment F1+2 (Dade Behring) were performed by ELISA. Interleukin (IL)-6 and tumor-necrosis factor-α (TNF-α) were measured by cytometric beads array multiplex assay (BD Biosciences, San Jose, CA, USA). The detection level for IL-6 and TNF-α was 2.5 pg/ml.

Statistics and Analyses

To investigate differences in baseline characteristics between patients with and without DCI we used the unpaired t-test, Mann–Whitney test, χ², or Fisher's exact test where appropriate.

The possible differences over time in ADAMTS13 activity, vWF antigen, vWF propeptide, and vWF cofactor activity, between the groups of patients with and without DCI were investigated with a linear random effects model, using the nlme package in the R statistical program (R Development Core Team, 2008). As in the analysis of ADAMTS13 activity levels first decreased and than increased, and in the analyses of vWF antigen, vWF propeptide, and vWF cofactor activity first increased and than decreased over time, a polynomial trend over time was assumed, which was allowed to differ by treatment group. Furthermore, because ADAMTS13 activity might be influenced by IL-6, TNF-α, and thrombin, the associations of these parameters (for thrombin the related parameter prothrombin fragment 1+2 was used) with ADAMTS13 were investigated by calculation of correlation coefficients, using the measurements at all time points simultaneously. We used Pearson or Spearman correlation coefficients depending on the observed distribution of the markers. In the same way, the associations between vWF antigen and vWF cofactor activity, vWF antigen and vWF propeptide, and vWF antigen and ADAMTS13 activity were investigated. Finally, the associations were investigated between ADAMTS13 activity and ADAMTS13 antigen, and ADAMTS13 activity and autoantibodies, at day 7 after SAH.

Results

After written informed consent was obtained, 31 patients were included in this study. Baseline characteristics are listed in Table 1. In twenty-four patients (77%) the aneurysms were coiled and in seven patients (23%) clipped. Median day of aneurysm treatment after SAH was day 1 (range 0 to 30).

Table 1

Characteristics of patients with and without delayed cerebral ischemia

	Total (n=31)	No DCI (n=20)	DCI (n=11)	P-value
Mean age, year (s.d.)	53 (11)	52 (11)	55 (10)	0.52^a
No. of women (%)	19 (61)	13 (65)	6 (55)	0.71^b
No. of patients with history of hypertension (%)	6 (19)	4 (20)	2 (18)	>0.99^b
No. of patients smoking (%)	11 (36)	6 (30)	5 (46)	0.45^b
Median WFNS-score (range)	2 (1–5)	2 (1–5)	3 (1–5)	0.27^c
Median GCS score on admission (range)	14 (5–15)	14 (5–15)	13 (5–15)	0.32^c
No. of patients with initial GCS score of 15 (%)	9 (29)	6 (30)	3 (27)	>0.99^b
No. of patients with focal neurologic impairment (%)	9 (29)	5 (25)	4 (36)	0.68^b
No. of patients with loss of consciousness during ictus
Yes (%)	17 (55)	10 (50)	7 (64)	0.71^b
No (%)	12 (39)	8 (40)	4 (36)
Unknown (%)	2 (7)	2 (10)	—
Mean blood pressure on admission (s.d.)
Systolic	161 (34)	156 (32)	171 (38)	0.28^a
Diastolic	91 (18)	90 (14)	95 (25)	0.53^a
Median Hijdra-score admission CT-scan (range)	23 (2–37)	23 (2–37)	28 (17–31)	0.06^c
Location of aneurysm
Anterior circulation (%)	27 (87)	17 (85)	10 (91)	>0.99^b
Posterior circulation (%)	4 (13)	3 (15)	1 (9)
Number of patients using simvastatin during hospitalization (%)	15 (48)	9 (45)	6 (55)	0.61^d

DCI, delayed cerebral ischemia; s.d., standard deviation; GCS, Glasgow Coma Scale; WFNS, World Federation of Neurological Surgeons.

Based on Student's t-test.

Based on Fisher's exact test.

Based on Mann–Whitney U test.

Based on χ².

Eleven patients (35%) developed DCI. Signs of DCI started at a median of 6 days (range 4 to 10 days). No differences were observed in baseline characteristics between patients with and without DCI occurrence, although patients with DCI had, as expected, a higher Hijdra score on admission CT scan (P=0.06).

Patients with DCI showed a stronger decrease of ADAMTS13 in the first few days after SAH (P-value for difference in polynomial time trend 0.0001) (Figure 1A). Autoantibodies could be measured in all samples 7 days after SAH, however in only three patients autoantibodies were above the threshold for positivity (1 out of 11 patients with DCI; 2 out of 20 patients without DCI). In patients who developed DCI vWF antigen, vWF propeptide, and vWF activity showed a more profound increase in the first few days; however, for vWF activity this difference was not significant (P-values for difference in polynomial time trend 0.020, 0.004, and 0.188 respectively) (Figures 1B–1D).

Figure 1

(A) ADAMTS13 activity boxplots of patients with and without DCI. On x-axis consecutive blood withdrawals. Symptoms of DCI started at a median of 6 days (range 4 to 10 days). (B) vWF antigen level boxplots of patients with and without DCI. On x-axis consecutive blood withdrawals. Symptoms of DCI started at a median of 6 days (range 4 to 10 days). (C) vWF propeptide level boxplots of patients with and without DCI. On x-axis consecutive blood withdrawals. Symptoms of DCI started at a median of 6 days (range 4 to 10 days). (D) vWF cofactor activity boxplots of patients with and without DCI. On x-axis consecutive blood withdrawals. Symptoms of DCI started at a median of 6 days (range 4 to 10 days).

Several correlations were investigated. No correlation was found between ADAMTS13 and IL-6 (Spearman correlation coefficient r=−0.033, P=0.676), or between ADAMTS13 and prothrombin fragment 1+2 (Spearman correlation coefficient r=−0.38, P=0.634). As in all plasma samples TNF-α levels were below the detection level, no correlation between ADAMTS13 and TNF-α was calculated. VWF antigen was strongly correlated to both vWF propeptide (Spearman correlation coefficient r=0.765, P<0.001) and vWF activity (Pearson correlation coefficient r=0.88, P<0.001). No indication of correlation between vWF antigen and ADAMTS13 was found (r=−0.027, P=0.736). A strong correlation was observed between ADAMTS13 activity and ADAMTS13 antigen at day 7 after SAH (r=0.89, P<0.0001). A trend was observed toward a correlation between ADAMTS13 activity and autoantibodies against ADAMTS13 on the samples from day 7 (r=−0.34, P=0.07).

Discussion

The results of our study show that patients with DCI after aneurysmal SAH have a significantly different development of ADAMTS13 and vWF over time compared with patients without DCI. Patients with DCI had a stronger decrease in ADAMTS13. Interestingly, there was no correlation between vWF antigen and ADAMTS13 activity. In addition, the decrease in ADAMTS13 activity could not be explained as a result of suppression by IL-6, TNF-α, or thrombin (investigated using the substitute parameter prothrombin fragment 1+2). The decrease in ADAMTS13 could have been caused by autoantibodies against the protein, because a trend was found toward a correlation between ADAMTS13 activity and autoantibodies. However, because only three samples were above the threshold for positivity of the assay, and positivity was equally distributed in patients with and without DCI, this might actually not be the cause. Furthermore, the results of this study show that patients with DCI had a more profound increase of vWF antigen, vWF propeptide, and vWF activity in the first few days after the hemorrhage, although for vWF activity this difference was not statistically significant. von Willebrand factor antigen was strongly correlated to both vWF propeptide and vWF activity.

ADAMTS13 has never been studied before in patients with aneurysmal SAH. However, some studies investigated ADAMTS13 in ischemic stroke (Bongers et al, 2006; Zhao et al, 2008). In a case–control study of 124 first-ever ischemic stroke patients and 125 age- and sex-matched controls, vWF antigen and vWF activity were associated with the occurrence of acute ischemic stroke (Bongers et al, 2006). This association was not affected by ADAMTS13 activity. As vWF and ADAMTS13 were negatively associated, similar to earlier observations (Mannucci et al, 2004), the question in that study remains whether low levels of ADAMTS13 resulted in higher vWF levels, or whether increased vWF levels resulted in lower ADAMTS13 levels (Bongers et al, 2006). However, in our study investigating DCI after SAH no correlation between vWF and ADAMTS13 was observed. Nevertheless, it cannot be ruled out whether other factors as a result of ischemia decreased ADAMTS13. Recently, a more causal relation between ADAMTS13 and cerebral ischemia was found in an SAH animal model. It was observed that ADAMTS13-deficient mice had approximately 20% larger infarcts after induction of cerebral ischemia (Zhao et al, 2008). Interestingly, infusion of recombinant human ADAMTS13 in ADAMTS13-deficient mice immediately before reperfusion 2 h after occlusion significantly reduced infarct volume with approximately 30% (Zhao et al, 2008).

The results of this study add to an accumulating body of evidence that DCI cannot be fully explained by macrovascular vasospasm. The results of this study suggest that, besides vasospasm-related hypoperfusion, microthrombosis has a more important function in the pathogenesis of DCI than is generally accepted (Vergouwen et al, 2008). Indeed, in autopsy studies microthrombi have been observed consisting of fibrin and activated platelets (Suzuki et al, 1983; Suzuki et al, 1990). Patients who died from DCI had significantly more microthrombi in clinically ischemic regions and in areas showing cerebral infarction on computed tomographic scan, when compared with patients who died from rebleeding or acute hydrocephalus (Suzuki et al, 1990). Furthermore, in up to 70% of SAH patients microembolic signals can be observed with transcranial Doppler ultrasonography, with a trend toward an increased incidence of microemboli in patients with symptomatic vasospasm (Romano et al, 2002). Interestingly, in the described study the incidence of microembolic signals was lower in vessels with radiologic vasospasm compared with those without vasospasm, although not significantly different, which suggests that microemboli are probably not a direct result of macrovascular vasospasm.

As not all patients with DCI have macrovascular vasospasm, and conversely, not all patients with vasospasm have DCI, cerebral microthrombosis might be a conditio sine qua non. In other words, it could be that patients who have angiographic vasospasm only develop symptoms of DCI when at the same time the coagulation cascade is activated or the fibrinolytic system impaired. Patients without macrovascular vasospasm can also have symptoms of DCI as a result of microthrombosis. As several studies found a strong association between angiographic vasospasm and DCI, vasospasm might be a confounder. This is probably the explanation why treatments aiming to prevent macrovascular vasospasm, for example endothelin receptor antagonists or prophylactic balloon angioplasty, had no beneficial effects, despite the fact that vasospasm is associated with DCI (Shaw et al, 2000; Zwienenberg-Lee et al, 2008). Recently, a large double-blind, placebo-controlled phase II study investigating the effect of the endothelin receptor antagonist clazosentan in patients with aneurysmal SAH for the first time showed a 65% relative risk reduction (95% confidence interval 47% to 78%) of angiographically demonstrated vasospasm in the group of patients treated with the highest dose of clazosentan (15 mg/h), however without an effect on the clinically more important predefined combined morbidity and mortality endpoint (Macdonald et al, 2008). It has been suggested that the pathogenesis of DCI should be reconsidered (Macdonald et al, 2007; Pluta et al, 2009).

As vWF is a parameter of endothelium function and an important contributor of thrombus formation, the results of this study suggest that endothelium dysfunction and the hemostatic system are involved in the development of DCI. The mechanism by which ADAMTS13 decreases in patients with DCI remains to be elucidated. A plausible explanation might be that an increased use of ADAMTS13 results from higher vWF antigen levels, however in our study no correlation was found between ADAMTS13 activity and vWF antigen. Earlier it has been shown that thrombin and plasmin potentiate the proteolytic inactivation of ADAMTS13, and therefore another explanation might be that ADAMTS13 activity is suppressed by thrombin and plasmin present in the subarachnoid space after the hemorrhage (Crawley et al, 2005). However, in our study no correlation was found between ADAMTS13 activity and prothrombin fragment 1+2, and therefore it is unlikely that the decrease in ADAMTS13 activity in DCI is caused by thrombin. ADAMTS13 activity might also be suppressed by IL-6, because IL-6 inhibits the cleavage of ULvWF by ADAMTS13 under flow conditions (Bernardo et al, 2004). In patients with aneurysmal SAH increased levels of IL-6 are associated with DCI (Hendryk et al, 2004). However, in our study no correlation was found between ADAMTS13 activity and IL-6. Finally, decreased ADAMTS13 activity might be caused by locally present neutralizing antibodies against ADAMTS13. However as discussed above, although we found a trend toward a correlation between ADAMTS13 activity and autoantibodies, the evidence from this study that autoantibodies are the cause of the decrease in ADAMTS13 activity is weak.

The results of this study imply that antithrombotic or anticoagulant drugs might be beneficial in the prevention of DCI. However, earlier attempts to influence procoagulant activity in SAH patients were not successful or inconclusive. A randomized controlled trial investigating the effect of aspirin was stopped prematurely, after an interim analysis showed that the probability of a beneficial effect was negligible (Van den Bergh et al, 2006). Explanations provided were that a dose of 100 mg once daily might not be appropriate, or that aspirin might not be the right antiplatelet drug for the prevention of DCI. Another explanation could be that the effect of aspirin is too selective by only inhibiting platelet function. A Cochrane analysis investigating the effect of antiplatelet therapy showed a strong trend toward a decreased incidence of DCI and poor outcome, but because the results were not statistically significant and antiplatelet therapy was associated with an increased risk of intracranial hemorrhagic complications, the conclusion was that antiplatelet use could not be recommended in SAH patients (Dorhout Mees et al, 2007). Earlier, in a single-blind controlled trial no effect of dipyridamole on the incidence of postoperative ischemic deficits was observed (Shaw et al, 1985). Two randomized controlled trials investigating the effect of enoxaparin showed contradictory results, and therefore it remains unknown whether enoxaparin is effective (Wurm et al, 2004; Siironen et al, 2003). In conclusion, earlier studies showed that investigated drugs influencing the coagulation pathway were inconclusive, and therefore large, well-powered, randomized controlled trials are needed to investigate the effects of these drugs on the incidence of DCI.

A possible limitation of our study is that only systemic levels of biomarkers were assessed instead of local levels from the cerebral circulation. Therefore, the results might not be a good representation of pathophysiological processes in the cerebral circulation. However, our results may also be interpreted as diluted values of locally increased levels, and therefore an underestimation of the local pathophysiological processes. Another limitation is that our study was not designed to measure ADAMTS13 and vWF levels at the days of clinically manifest DCI. Therefore, many blood samples were not taken on the day of first signs of DCI, and our data only indicate that ADAMTS13 decreases and vWF antigen level increases in the time period that DCI develops. Furthermore, this study was a post hoc study from an exploratory randomized placebo-controlled study in which the biologic effects of simvastatin in SAH patients were investigated. Although in the original study no effect of simvastatin on endothelial function was observed, statins have effects on endothelium function, such as eNOS, in animals. Finally, the use of CT instead of magnetic resonance imaging limits the study, because the sensitivity of magnetic resonance imaging is higher for smaller lesions.

As in this study no explanation was found for the decrease in ADAMTS13 activity in patients with DCI, and the role of ADAMTS13 and vWF in the development of DCI should be further investigated, we conclude that the results of this study have an important translational aspect, from bedside to bench. Little is known about the factors that increase coagulation activity in the pathogenesis of DCI. The role of Weibel–Palade bodies in the vascular endothelium needs further investigation. Weibel–Palade bodies are storage sites in both the microvascular and macrovascular endothelium, which can rapidly respond to changes in its microenvironment by the release of various substances. Well-known stimulators of Weibel–Palade bodies are (among others) thrombin, complement factors, and epinephrine. In patients with SAH, Weibel–Palade bodies are probably activated by thrombin, present in the subarachnoid space as a direct result of the hemorrhage, inflammatory reactions, or from the stress response. After stimulation, Weibel–Palade bodies release substances such as vWF, endothelin, tissue plasminogen activator, P-selectin, and IL-8. Recent studies indicate that substance release from Weibel–Palade bodies can be selective (Babich et al, 2008). Therefore, in patients with DCI only procoagulatory substances might be released, and not substances such as tissue plasminogen activator, which is profibrinolytic. The fact that substance release from Weibel–Palade bodies can be selective might also explain why macrovascular vasospasm does not always result in clinical signs and symptoms of DCI, and vice versa, not all patients with DCI have vasospasm. For instance, in patients with macrovascular vasospasm without symptoms of DCI, endothelin might be released from Weibel–Palade bodies in the macrovascular endothelium, resulting in vasospasm, without release of procoagulatory substances. In patients with symptoms of DCI without macrovascular vasospasm procoagulatory substances are released without release of endothelin from the macrovascular endothelium. In the latter group of patients it could also be that endothelin is released from the microvascular endothelium, resulting in microvascular vasospasm. In patients with both macrovascular vasospasm and symptoms of DCI both endothelin and procoagulatory substances are released.

Although it would be interesting to further investigate the role of Weibel–Palade bodies in the pathogenesis of DCI, it is currently not an easy scenario, because it is not possible to study Weibel–Palade bodies in vivo in humans. Presently, Weibel–Palade body function can only be studied indirectly by measuring vWF in blood. Furthermore, although SAH animal models can be used to investigate whether the animal develops spastic arteries, it is much more difficult to evaluate whether the animal develops signs and symptoms of DCI. A better approach to investigate Weibel–Palade bodies in animals would be to compare Weibel–Palade bodies in animals with and without cerebral infarctions at postmortem examination. Furthermore, ADAMTS13 should be further studied in animal models. For example, local levels of ADAMTS13 activity and autoantibodies in the cerebral circulation should be investigated. A more causal relation between ADAMTS13 and cerebral infarction after SAH can be studied by comparing ischemic stroke volume in ADAMTS13-deficient and -sufficient animals. Finally, it would be interesting to investigate the effect of infusion of recombinant human ADAMTS13 in SAH animal models.

We conclude that patients with DCI after aneurysmal SAH show a stronger decrease in ADAMTS13 activity and a more profound increase of both vWF antigen and vWF propeptide levels within the first days after the hemorrhage compared with patients without DCI. Although the results of this study do not prove a causal relationship between decreased ADAMTS13 activity and the pathogenesis of DCI, our results suggest that microthrombosis has a role in the pathogenesis of DCI.

Footnotes

The authors declare no conflict of interest.

References

Babich

Meli

Knipe

Dempster

Skehel

Hannah

Carter

(2008) Selective release of molecules from Weibel-Palade bodies during a lingering kiss. Blood 111:5282–90

Bernardo

Ball

Nolasco

Moake

Dong

(2004) Effects of inflammatory cytokines on the release and cleavage of the endothelial cell-derived ultralarge von Willebrand factor multimers under flow. Blood 104:100–6

Bongers

de Maat

van Goor

Bhagwanbali

van Vliet

Gómez García

Dippel

Leebeek

(2006) High von Willebrand factor levels increase the risk of first ischemic stroke: influence of ADAMTS13, inflammation, and genetic variability. Stroke 37:2672–7

Crawley

Lam

Rance

Mollica

O0Donnell

Lane

(2005) Proteolytic inactivation of ADAMTS13 by thrombin and plasmin. Blood 105:1085–93

Dong

Moake

Nolasco

Bernardo

Arceneaux

Shrimpton

Schade

McIntire

Fujikawa

López

(2002) ADAMTS-13 rapidly cleaves newly secreted ultralarge von Willebrand factor multimers on the endothelial surface under flowing conditions. Blood 100:4033–9

Dorhout Mees

van den Bergh

Algra

Rinkel

(2007) Antiplatelet therapy for aneurysmal subarachnoid haemorrhage. Cochrane Database Syst Rev 4:CD006184

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

Hendryk

Jarzab

Josko

(2004) Increase of the IL-1 beta and IL-6 levels in CSF in patients with vasospasm followinganeurysmal SAH. Neuro Endocrinol Lett 25:141–7

Hijdra

Brouwers

Vermeulen

van Gijn

(1990) Grading the amount of blood on computed tomograms after subarachnoid hemorrhage. Stroke 21:1156–61

10.

Kokame

Nobe

Kokubo

Okayama

Miyata

(2005) FRETS-VWF73, a first fluorogenic substrate for ADAMTS13 assay. Br J Haematol 129:93–100

11.

Macdonald

Pluta

Zhang

(2007) Cerebral vasospasm after subarachnoid hemorrhage: the emerging revolution. Nat Clin Pract Neurol 3:256–63

12.

Macdonald RL, Kassell NF, Mayer S, Ruefenacht D, Schmiedek P, Weidauer S, Frey A, Roux S, Pasqualin A, on behalf of the CONSCIOUS-1 Investigators (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

13.

Mannucci

Capoferri

Canciani

(2004) Plasma levels of von Willebrand factor regulate ADAMTS-13, its major cleaving protease. Br J Haematol 126:213–8

14.

Moake

Rudy

Troll

Weinstein

Colannino

Azocar

Seder

Hong

Deykin

(1982) Unusually large plasma factor VIII:von Willebrand factor multimers in chronic relapsing thrombotic thrombocytopenic purpura. N Engl J Med 307:1432–5

15.

Neil-Dwyer

Lang

Doshi

Gerber

Smith

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

16.

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

17.

R Development Core Team (2008) R: a language and environment for statistical computing. Vienna, Austria: ISBN 3-900051-07-0 URL http://www.R-project.org

18.

Romano

Forteza

Concha

Koch

Heros

Morcos

Babikian

(2002) Detection of microemboli by transcranial Doppler ultrasonography in aneurysmal subarachnoid hemorrhage. Neurosurgery 50:1026–30

19.

Roos

de Haan

Beenen

Groen

Albrecht

Vermeulen

(2000) Complications and outcome in patients with aneurysmal subarachnoid haemorrhage: a prospective hospital based cohort study in The Netherlands. J Neurol Neurosurg Psychiatry 68:337–41

20.

Shaw

Foy

Conway

Pickard

Maloney

Spillane

Chadwick

(1985) Dipyridamole and postoperative ischemic deficits in aneurysmal subarachnoid hemorrhage. J Neurosurg 63:699–703

21.

Shaw

Vermeulen

Murray

Pickard

Bell

Teasdale

(2000) Efficacy and safety of the endothelin, receptor antagonist TAK-044 in treating subarachnoid hemorrhage: a report by the Steering Committee on behalf of the UK/Netherlands/Eire TAK-044 Subarachnoid Haemorrhage Study Group. J Neurosurg 93:992–7

22.

Siironen

Juvela

Varis

Porras

Poussa

Ilveskero

Hernesniemi

Lassila

(2003) No effect of enoxaparin on outcome of aneurysmal subarachnoid hemorrhage: a randomized, double-blind, placebo-controlled clinical trial. J Neurosurg 99:953–9

23.

Stoltenburg-Didinger

Schwarz

(1987) Brain lesions secondary to subarachnoid hemorrhage due to ruptured aneurysms. Stroke and microcirculation. New York, NY: Raven Press, 471–80

24.

Suzuki

Iwabuchi

Kamata

(1983) Role of multiple cerebral microthrombosis in symptomatic cerebral vasospasm: with a case report. Neurosurgery 13:199–203

25.

Suzuki

Kimura

Souma

Ohkima

Shimizu

Iwabuchi

(1990) Cerebral microthrombosis in symptomatic cerebral vasospasm - a quantitative histological study in autopsy cases. Neurol Med Chir (Tokyo) 30:309–16

26.

van den Bergh WM, Algra A, Dorhout Mees SM, van Kooten F, Dirven CM, van Gijn J, Vermeulen M, Rinkel GJ, MASH Study Group (2006) Randomized controlled trial of acetylsalicylic acid in aneurysmal subarachnoid hemorrhage: the MASH Study. Stroke 37:2326–30

27.

Van Gijn

Bromberg

Lindsay

Hasan

Vermeulen

(1994) Definition of initial grading, specific events, and overall outcome in patients with aneurysmal subarachnoid hemorrhage. A survey. Stroke 25:1623–7

28.

Vergouwen

Vermeulen

Coert

Stroes

Roos

(2008) Microthrombosis after aneurysmal subarachnoid hemorrhage—an additional explanation for delayed cerebral ischemia. J Cereb Blood Flow Metab 28:1761–70

29.

Vergouwen

Meijers

Geskus

Coert

Horn

Stroes

van der Poll

Vermeulen

Roos

(2009) Biological effects of simvastatin in patients with aneurysmal subarachnoid hemorrhage—a double-blind, placebo-controlled randomized trial. J Cereb Blood Flow Metab; advance online publication 20 May, 2009 [e-pub ahead of print]

30.

Wurm

Tomancok

Nussbaumer

Adelwohrer

Holl

(2004) Reduction of ischemic sequelae following spontaneous subarachnoid hemorrhage: a double-blind, randomized comparison of enoxaparin versus placebo. Clin Neurol Neurosurg 106:97–103

31.

Zhao

B-Q

Chauhan

Patten

Dockal

Scheiflinger

Wagner

(2008) VWF-cleaving protease ADAMTS13 reduces brain injury following ischemic stroke in mice: essential role for VWF in stroke 50th ASH Annual Meeting and Exposition (abstract 259), San Francisco, USA

32.

Zwienenberg-Lee

Hartman

Rudisill

Madden

Smith

Eskridge

Newell

Verweij

Bullock

Baker

Coplin

Mericle

Dai

Rocke

Muizelaar

(2008) Effect of prophylactic transluminal balloon angioplasty on cerebral vasospasm and outcome in patients with Fisher grade III subarachnoid hemorrhage. Results of a phase II multicenter, randomized, clinical trial. Stroke 39:1759–65