- Email: [email protected]
Nontraumatic Convexal Subarachnoid Hemorrhage Concomitant with Acute Ischemic Stroke
Nontraumatic Convexal Subarachnoid Hemorrhage Concomitant with Acute Ischemic Stroke Makoto Nakajima, MD, PhD,* Yuichiro Inatomi, MD, PhD,* Toshiro Yonehara, MD, PhD,* Teruyuki Hirano, MD, PhD,† and Yukio Ando, MD, PhD‡
Background: Nontraumatic convexal subarachnoid hemorrhage (cSAH) rarely occurs subsequent to acute ischemic stroke. The incidence, clinical background characteristics, and outcomes in acute ischemic stroke patients with cSAH were investigated. Methods: Our stroke center database was reviewed to identify patients with acute ischemic stroke/transient ischemic attack (TIA) who demonstrated acute cSAH within 14 days of admission between 2005 and 2011. Background characteristics, clinical course, and outcomes at discharge and 3 months after onset were investigated in these patients. Results: Of 4953 acute stroke/TIA patients, cSAH was observed in 8 (.14%) patients (7 men, mean age 71 years): 7 were detected incidentally, and the other was found immediately after a convulsion. Two patients died during their hospital stay, 1 died after discharge, and 3 were dependent at 3 months. Major artery occlusion or severe stenosis was observed in 5 patients. Two patients subsequently developed subcortical hemorrhage. On gradient echo imaging, lobar cerebral microbleeds were observed in 2 patients, and chronic superficial siderosis was observed in 2 patients. Conclusions: In this retrospective review of cases with ischemic stroke and cSAH, over half of patients had occlusion of major arteries. Cerebral amyloid angiopathy was suggested by magnetic resonance imaging findings and subsequent events in 3 patients. The overall outcome was unfavorable although the causal relationship with cSAH was unclear. Key Words: Carotid stenosis— cerebral amyloid angiopathy—convexal subarachnoid hemorrhage—intracranial stenosis—superficial siderosis. Ó 2014 by National Stroke Association
Introduction Nontraumatic convexal subarachnoid hemorrhage (cSAH) without any causative intracranial aneurysm
From the *Department of Neurology, Stroke Center, Saiseikai Kumamoto Hospital, Kumamoto; †Department of Neurology and Neuromuscular Disorder, Oita University Faculty of Medicine, Yufu; and ‡Department of Neurology, Graduate School of Medical Sciences, Kumamoto University, Kumamoto, Japan. Received November 26, 2013; revision received December 23, 2013; accepted December 25, 2013. Conflicts of interest: None. Source of funding: None. Address correspondence to Makoto Nakajima, MD, PhD, Department of Neurology, Graduate School of Medical Sciences, Kumamoto University, 1-1-1, Honjo, Chuo-ku, Kumamoto, 860-8556, Japan. E-mail: [email protected]. 1052-3057/$ - see front matter Ó 2014 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2013.12.046
has been increasingly recognized as a separate entity with distinct pathophysiological mechanisms.1-9 Various vascular or nonvascular mechanisms have been suggested for cSAH. Some authors reported patients who developed cSAH subsequent to acute ischemic events,2,6-9 but the incidence, clinical characteristics, and outcomes of such patients have not yet been established. Our database of patients with acute ischemic stroke/ transient ischemic attack (TIA) was reviewed to investigate clinical characteristics and courses in patients with cSAH. In addition, possible underlying pathogenesis among these patients was speculated.
Materials and Methods Subjects Our database of patients who were admitted to our stroke center with a diagnosis of acute ischemic
Journal of Stroke and Cerebrovascular Diseases, Vol. -, No. - (---), 2014: pp 1-7
1
M. NAKAJIMA ET AL.
2
Table 1. Summary of patients who developed cortical subarachnoid hemorrhage after ischemic stroke
Case
Age (y), sex
Underlying disease
1
63, M DM, HT
2
72, M CHF, AF, CKD, ICH 72, M AF
3
4
5
Arterial disease Right extracranial ICAO UD (because of motion artifact) N
70, M Chronic hepatitis, Right extracranial DM, HT, DLP ICAS and bilateral VAS 66, M OMI, HT, DLP N
Initial NIHSS score 6
Antithrombotics
Stroke subtype
Site of infarction (arterial territory)
Initial gradient echo image
Argatroban, aspirin, cilostazol Heparin, aspirin
LA
Right MCA
Normal
CE
0
N
TIA
Bilateral PICA, right MCA Right MCA
1
Ozagrel, heparin, cilostazol
LA
BA, right PICA
Old putaminal hematoma 4 lobar CMBs and superficial siderosis Normal
1
Argatroban, aspirin
UD
Left MCA
Normal
21
6
77, M DLP, gout, lung carcinoma
Right extracranial ICAO
0
Argatroban, aspirin
LA
Right MCA
Normal
7
80, F N
Left MCAO
3
N
UD
Left MCA
1 lobar CMB
8
65, M AF, HT, CKD
Right extracranial ICAO
Heparin, warfarin
CE
Right MCA
Superficial siderosis
18
Abbreviations: AF, atrial fibrillation; AMI, acute myocardial infarction; BA, basilar artery; CCA, cerebral amyloid angiopathy; CE, cardioembolic stroke; CHF, congestive heart failure; CKD, chronic kidney disease; CMB, cerebral microbleed; cSAH, cortical subarachnoid hemorrhage; DLP, dyslipidemia; DM, diabetes mellitus; HT, hypertension; ICAO, internal carotid artery occlusion; ICAS, internal carotid artery stenosis; ICH, intracerebral hemorrhage; LA, large artery atherosclerosis; LOS, length of hospital stay; mRS, modified Rankin Scale; MCA, middle cerebral artery; NA, not applicable; NIHSS, National Institutes of Health Stroke Scale; PICA, posterior inferior cerebellar artery; TIA, transient ischemic attack; UD, undetermined; VAS, vertebral artery stenosis. *Standard value is #1.0 mg/mL.
stroke/TIA between 2005 and 2011 was reviewed. All patients underwent both brain computed tomography (CT) and magnetic resonance imaging (MRI; except for those with contraindications) on admission. In addition, almost every patient routinely underwent follow-up CT and/or MRI at least once during the hospital stay to detect recurrence, hemorrhagic complications, or changes in arterial findings. Initial MRI included the following sequences: diffusion-weighted imaging, T2-weighted imaging, fluid-attenuated inversion recovery (FLAIR) imaging, and magnetic resonance angiography (MRA). On follow-up MRI, gradient echo imaging (GRE) was added on the sequences of initial MRI. If cSAH was detected, causative arterial abnormalities were further investigated using CT angiography or reconstruction of MRA. All images were double reviewed separately by the attending physician and a radiologist.
Diagnosis of cSAH Of a total of 4953 patients during the study period, patients with a diagnosis of acute cSAH, which was de-
tected on CT (and/or MRI) and rapidly resolved on the follow-up imaging, were selected. Patients in whom cSAH was detected 14 days or more after admission were excluded because the aim was to assess cSAH in ischemic stroke patients during their acute phase. Focal linear hypointense signal changes in the superficial layers of the cerebral cortex on GRE (superficial siderosis),4,10 which were not detected by CT or FLAIR imaging but detected only by GRE were regarded as old cSAH. Patients with only superficial siderosis were excluded from this study. In addition, patients with a preceding episode of head trauma, intracranial aneurysm, thrombolytic therapy, or intracranial dissection were also excluded.
Data Collection The following clinical information of patients with cSAH was collected: age and sex, underlying disease, site of stenosis greater than 50% of the corresponding artery for the index stroke, National Institutes of Health Stroke Scale score on admission, acute antithrombotic
ACUTE STROKE AND CONVEXAL SUBARACHNOID HEMORRHAGE
mRS score
Days for cSAH detection D-dimer* after ischemic stroke (mg/mL) .4
1 (CT) 9 (MRI)
11.0
2 (CT) 3 (MRI)
1.1
.3
UD: on admission (CT 1 MRI) 13 (CT 1 MRI)
.2
3 (MRI)
1.1
3 (CT 1 MRI)
1.1
UD: on admission (CT 1 MRI) 1 (MRI) 5 (CT)
.7
3
Site of cSAH (lobe) Right frontal
Trigger for cSAH detection
Complications
Neurologic deterioration Right parietal N Lerish syndrome, AMI Left frontal N Subcortical hemorrhage (10 M) Right parietal N Recurrence of ischemic stroke, pneumonia Left N Subcortical frontoparietal hemorrhage (day 15) Headache and Frequent TIA Right frontoparietal convulsion Left frontal N Neurological deterioration Right parietal
N
N
therapy when cSAH was detected, stroke subtype according to the Trial of Org 10172 Acute Stroke Treatment classification, site of the index ischemic stroke, initial GRE finding, D-dimer level during the hospital stay, time between the onset of the index stroke and detection of cSAH, trigger for detection of cSAH, other complications during the hospital stay, length of hospital stay, modified Rankin Scale scores at discharge and at 3 months, and diagnosis of cerebral amyloid angiopathy (CAA) based on the Boston criteria.11
Clinical Workup for Diagnosis The site of arterial stenosis was divided into the extracranial internal carotid artery (ICA), the intracranial ICA, or the other main cerebral arteries based on the results of carotid ultrasonography and MRA. Because the exact onset time of cSAH that was detected incidentally could not be determined, the medical record in each case was carefully reviewed to search for new clinical events or symptom changes possibly related to the cSAH between the last brain imaging examination without cSAH and the first evidence of cSAH (such as sudden headache, consciousness disturbance, or other neurologic deficits). A standard blood examination including prothrombin time, activated partial prothrombin time, blood cell count, fibrinogen, and D-dimer was performed in all patients. In patients who were suspected to have abnormal coagulation–fibrinolysis activity, fibrin degradation products
N
At LOS discharge
At 3 months
Diagnosis of CAA
23
4
UD
NA
11
6
6
NA
14
0
40
6
6
NA
6
0
5
Possible CAA
17
2
6
NA
20
4
3
Possible CAA
22
4
4
NA
0 / 5 (10 M) Probable CAA
(FDPs) or antithrombin III was also assessed. Both transthoracic echocardiography and 24-hour Holter electrocardiogram monitoring were performed to detect embolic sources in patients with no history of atrial fibrillation.
Results Of 4953 patients, 16 patients showed cSAH. Eight patients were excluded because of superficial siderosis without acute cSAH (n 5 1), head trauma (n 5 1), ruptured aneurysm (n 5 1), precedent thrombolytic therapy (n 5 2), or a diagnosis of intracranial arterial dissection (n 5 3). Therefore, 8 patients (.14%; 7 men; aged 71 6 6 years) demonstrated cSAH during the treatment of acute ischemic stroke/TIA (Table 1, Fig 1): 7 were detected incidentally and 1 was detected immediately after a convulsion (case 6). cSAH was confirmed on CT, FLAIR, and GRE in all but 1 patient (without CT; case 5). Two patients who had superficial siderosis (cases 3 and 8) on the initial GRE were included in the 8 patients because they also demonstrated acute cSAH during their hospital stay. Two patients (cases 3 and 7) demonstrated both cSAH and acute ischemic lesions on admission. Their overall outcome was poor: 2 patients died during their hospital stay because of ventricular fibrillation (case 2) and severe aspiration pneumonia (case 4). Two patients (cases 3 and 5) discharged without any neurologic deficit developed subcortical hemorrhage thereafter. In detail, the former
4
M. NAKAJIMA ET AL.
Figure 1. Brain images of 8 acute ischemic stroke patients with convexal subarachnoid hemorrhage (cSAH; upper: diffusion-weighted images on admission; middle, fluid-attenuated inversion recovery images; lower, computed tomography except for case 6 of gradient echo imaging). cSAH (arrowheads) was not observed in all patients on admission. All patients except for cases 3 and 4 demonstrate focal cSAH in the same vascular territory as the index ischemic lesions.
(case 3) started warfarin at discharge because of atrial fibrillation and developed hemorrhage in the left parietal lobe (ie, distinct from the site of cSAH) after 10 months. The latter (case 5) continued aspirin, which he had been prescribed because of coronary heart disease, and developed hemorrhage in the left frontoparietal lobe on day 15. The hematoma included the site of cSAH. In both of them, GRE before hemorrhage did not demonstrate microbleeds on the bleeding site, and no abnormal findings were observed on cerebral angiography performed after hemorrhage. One patient (case 6) died after discharge because of bleeding from lung cancer, and at least 3 patients were dependent at 3 months.
Occlusion or stenosis of a major artery ipsilateral to the cSAH was detected in 5 patients. Of these, hemodynamic compromise in the affected area was seen in 2 patients: broad hypoperfusion in the right anterior circulation on single-photon emission CT (case 1; Fig 2) and severe delayed collateral flow of the right anterior circulation on cerebral angiography (case 6; Fig 3). Three patients (cases 1, 4, and 6) had severe atherosclerotic changes in the extracranial ICA and/or the vertebral arteries but no proximal embolic sources. Case 7 had neither embolic sources nor severe atherosclerotic changes in the ICA. On the other hand, 1 patient with atrial fibrillation (case 8) demonstrated extracranial ICA occlusion on admission.
ACUTE STROKE AND CONVEXAL SUBARACHNOID HEMORRHAGE
5
Figure 2. Single-photon emission computed tomography using 99m Tc-ECD without acetazolamide in case 1. The cerebral blood flow to the right cerebral hemisphere is extensively decreased, even outside the infarcted region (arrowheads).
Follow-up CT detected cSAH on day 5, and MRA demonstrated recanalization of the right ICA on day 7. In 1 patient (case 5), blood pressure on admission was as high as 187/89 mm Hg and then it gradually decreased. The other patients experienced neither hypertension more than 180 mm Hg nor abrupt elevation of blood pressure between admission and detection of cSAH. Platelet counts during the hospital stay were within normal limits in all subjects. The serum D-dimer level was elevated substantially in 1 patient (case 2); the serum FDP level (13.7 mg/dL) and fibrinogen (477 mg/dL) were also elevated, whereas antithrombin III was within the normal limit (77%). The serum D-dimer level was elevated marginally in 3 other patients, one of which with lung carcinoma (case 6) showed elevated serum fibrinogen (508 mg/dL), normal FDP level (4.0 mg/dL), and antithrombin III activity (88%).
vascular anomaly, infectious disease, venous thrombosis, various vasculopathies, coagulation disorder, or brain tumor.2,9 In an observational study of 24 cSAH cases, Beitzke et al8 found 5 cases with concurrent ischemic lesions (including 3 asymptomatic cases), as in the present study. However, they did not mention the mechanism of ischemia. The etiology of cSAH in each patient in the present case series is now discussed. In cases 3, 5, and 7, CAA was suspected as the most plausible cause of cSAH. They fulfilled the criteria of ‘‘probable CAA’’ or ‘‘possible CAA.’’4,11 CAA is one of
Discussion Overall, 8 (.14%) of 4953 patients showed cSAH during the acute treatment of ischemic stroke. This was a small case series; however, we would emphasize the poor outcomes of the patients. Although the prevalence of cSAH is reported to be 5%-7.5% in spontaneous SAH patients,3,12 little has been known regarding the incidence of cSAH concomitant with acute ischemic stroke. On the other hand, many different etiologies of cSAH have been proposed: CAA, arterial occlusive disease,
Figure 3. Intra-arterial subtraction angiography of the right vertebral artery in case 6. The territory of the right anterior circulation is perfused by a compensated bypass from the right vertebral artery, muscle branches (arrowheads), external carotid artery (retrograde flow), internal carotid artery, and also by leptomeningeal anastomosis.
M. NAKAJIMA ET AL.
6 3-5,8
the most common pathogeneses of cSAH. On the other hand, CAA can cause brain infarction because of the deposition of amyloid in the wall of meningeal and parenchymal arteries or arterioles, which presumably induces a disturbance of vascular reactivity or focal circulation.13 Some authors speculate that small ischemia and hemorrhage in CAA patients may share common pathophysiologic steps.14 Therefore, if the 3 patients had CAA, the prior ischemic event might also have been associated with CAA. Five patients (cases 1, 4, 6, 7, and 8) had occlusion or severe stenosis of major arteries. Of these, the first 3 patients (cases 1, 4, and 6) were suspected to have chronic occlusive arterial disease from the sonographic findings and the absence of emboligenic diseases. The main mechanism of cSAH because of chronic arterial occlusive diseases is speculated to be the rupture of dilated fragile compensatory pial vessels.2,6,7 Imaging findings that suggested hemodynamic compromise in cases 1 and 6 seemed to be consistent with this hypothesis. Importantly, most previous cases of cSAH associated with carotid artery occlusion/severe stenosis demonstrated acute ischemic lesions in the territory of the affected artery.2,6,7 Simultaneous occurrence of an ischemic event and cSAH might indicate that hemodynamic insufficiency because of arterial occlusion/stenosis reached a critical point. One possible scenario is that, for example, after an acute ischemic stroke because of occlusive disease via an embolic and/or hemodynamic mechanism, subsequent dynamic changes of intracranial perfusion pressure might trigger cSAH. The other 2 (cases 7 and 8) cases were rather complicated. The first 1 (case 7) demonstrated intracranial arterial occlusion, but there were few findings that supported an embolic or atherothrombotic mechanism. Regarding the mechanism of cSAH, as mentioned earlier, CAA would be a possible cause. The latter one (case 8) had atrial fibrillation and demonstrated recanalization of the occluded artery, which was diagnosed as cardioembolic stroke. It would be rather difficult to conclude that the cSAH resulted from the same mechanism as in those who had chronic occlusive diseases. This patient did not fulfill the CAA criteria11 but did match possible CAA in the modified criteria of Linn et al,4 based on the chronic superficial siderosis. Another possibility is reperfusion injury, although the exact temporal relationship between reperfusion of the ICA and cSAH could not be determined. Malignancy was observed in 1 patient (case 6). Coagulation–fibrinolysis activity is frequently unstable in patients with malignant diseases. It may cause either ischemic stroke (embolism) or hemorrhagic stroke.15 In a recent study of 208 intracranial hemorrhages in patients with cancer, one of the common causes of intracranial hemorrhage was coagulopathy (67%).16 In the present patient (case 6), an abnormal hemostatic status might
have had an effect on cSAH. Another possible cause of cSAH in patients with malignancy is intracranial microaneurysm based on nonbacterial thromboendocarditis. However, transthoracic echocardiography did not detect evidence of vegetations in this patient. One patient (case 2), who had many underlying diseases and demonstrated severely elevated D-dimer level, developed systemic embolic complications. Although abnormal coagulation–fibrinolysis activity or arterial occlusive diseases might have been involved in the occurrence of cSAH and the clinical course, these hypotheses could not be validated. Finally, influence of antithrombotic agents on the occurrence of cSAH other than in cases 3 and 7 should be considered. Even patients with unruptured intracranial aneurysms do not usually develop subarachnoid hemorrhage during treatment of acute ischemic stroke.17 However, some patients whose intracranial arteries are vulnerable because of the aforementioned specific pathogeneses such as CAA or arterial stenotic disease might develop cSAH because of intensive antithrombotic therapy. Because this was a retrospective study, it is difficult to clarify the relationship between antithrombotics and cSAH. Ischemic lesions and cSAH are observed in the different arterial territory in 2 patients (cases 3 and 4). Given that either had a potential embolic source (AF in case 3 and multiple arterial stenosis in case 4), it is possible that the ischemic event and cSAH occurred coincidentally. Likewise, we must take into account the same possibility also in the other 6 cases. This study had some limitations. First, some patients with asymptomatic cSAH might have been missed, although most patients underwent a follow-up CT or MRI during their hospital stay. Second, because this was a retrospective study using clinical records, the relationships between cSAH and symptoms, results of examinations, or comorbidities could not be completely proven. Besides, we might have possibly overlooked faint cSAH or clinical signs related to certain pathogenesis in some patients; the incidence of cSAH among ischemic stroke should be interpreted with cautious. To overcome these issues, a prospective case–control study with a definite protocol for examination, imaging, and collecting clinical signs would be required. Third, because cerebral angiography or autopsy was not performed in most of the patients, small aneurysms locating at the branches or focal venous thrombosis may have been dismissed. In summary, 8 cases with cSAH observed in the acute phase of ischemic stroke were reported. Clinical courses vary among the cases and most patients died or were dependent; however, cSAH did not directly affect these unfavorable outcomes. Although some comorbidities including occlusive arterial disease and CAA were suspected to be associated with cSAH, the relationships could not be confirmed. Particular MRI sequences such
ACUTE STROKE AND CONVEXAL SUBARACHNOID HEMORRHAGE
as GRE or susceptibility-weighted imaging may facilitate detection of asymptomatic cSAH.9 Careful observation of images and clinical symptoms would be required for clinicians to detect and collect these patients. Acknowledgment: We express our sincere gratitude to Dr Hiroyuki Kawano (Saiseikai Kumamoto Hospital) for the assistance in collecting the patient data and for his professional comments. We are also deeply grateful to Dr Toru Nishi and staff members of the Department of Neurosurgery, Saiseikai Kumamoto Hospital for their ungrudging clinical supports and professional advices.
References 1. Ducros A, Fiedler U, Porcher R, et al. Hemorrhagic manifestations of reversible cerebral vasoconstriction syndrome: frequency, features, and risk factors. Stroke 2010;41:2505-2511. 2. Cuvinciuc V, Viguier A, Calviere L, et al. Isolated acute nontraumatic cortical subarachnoid hemorrhage. AJNR Am J Neuroradiol 2010;31:1355-1362. 3. Kumar S, Goddeau RP Jr, Selim MH, et al. Atraumatic convexal subarachnoid hemorrhage: clinical presentation, imaging patterns, and etiologies. Neurology 2010;74:893-899. 4. Linn J, Halpin A, Demaerel P, et al. Prevalence of superficial siderosis in patients with cerebral amyloid angiopathy. Neurology 2010;74:1346-1350. 5. Raposo N, Viguier A, Cuvinciuc V, et al. Cortical subarachnoid haemorrhage in the elderly: a recurrent event probably related to cerebral amyloid angiopathy. Eur J Neurol 2011;18:597-603. 6. Geraldes R, Santos C, Canhao P. Atraumatic localized convexity subarachnoid hemorrhage associated with acute carotid artery occlusion. Eur J Neurol 2011; 18:e28-e29.
7
7. Chandra RV, Leslie-Mazwi TM, Oh D, et al. Extracranial internal carotid artery stenosis as a cause of cortical subarachnoid hemorrhage. AJNR Am J Neuroradiol 2011; 32:E51-E52. author reply E53. 8. Beitzke M, Gattringer T, Enzinger C, et al. Clinical presentation, etiology, and long-term prognosis in patients with nontraumatic convexal subarachnoid hemorrhage. Stroke 2011;42:3055-3060. 9. Renou P, Tourdias T, Fleury O, et al. Atraumatic nonaneurysmal sulcal subarachnoid hemorrhages: a diagnostic workup based on a case series. Cerebrovasc Dis 2012;34:147-152. 10. Linn J, Herms J, Dichgans M, et al. Subarachnoid hemosiderosis and superficial cortical hemosiderosis in cerebral amyloid angiopathy. AJNR Am J Neuroradiol 2008; 29:184-186. 11. Knudsen KA, Rosand J, Karluk D, et al. Clinical diagnosis of cerebral amyloid angiopathy: validation of the Boston criteria. Neurology 2001;56:537-539. 12. van Gijn J, Rinkel GJ. Subarachnoid haemorrhage: diagnosis, causes and management. Brain 2001;124: 249-278. 13. Cadavid D, Mena H, Koeller K, et al. Cerebral beta amyloid angiopathy is a risk factor for cerebral ischemic infarction. A case control study in human brain biopsies. J Neuropathol Exp Neurol 2000;59:768-773. 14. Kimberly WT, Gilson A, Rost NS, et al. Silent ischemic infarcts are associated with hemorrhage burden in cerebral amyloid angiopathy. Neurology 2009;72: 1230-1235. 15. Rogers LR. Cerebrovascular complications in patients with cancer. Semin Neurol 2010;30:311-319. 16. Navi BB, Reichman JS, Berlin D, et al. Intracerebral and subarachnoid hemorrhage in patients with cancer. Neurology 2010;74:494-501. 17. Oh YS, Lee SJ, Shon YM, et al. Incidental unruptured intracranial aneurysms in patients with acute ischemic stroke. Cerebrovasc Dis 2008;26:650-653.
Background: Nontraumatic convexal subarachnoid hemorrhage (cSAH) rarely occurs subsequent to acute ischemic stroke. The incidence, clinical background characteristics, and outcomes in acute ischemic stroke patients with cSAH were investigated. Methods: Our stroke center database was reviewed to identify patients with acute ischemic stroke/transient ischemic attack (TIA) who demonstrated acute cSAH within 14 days of admission between 2005 and 2011. Background characteristics, clinical course, and outcomes at discharge and 3 months after onset were investigated in these patients. Results: Of 4953 acute stroke/TIA patients, cSAH was observed in 8 (.14%) patients (7 men, mean age 71 years): 7 were detected incidentally, and the other was found immediately after a convulsion. Two patients died during their hospital stay, 1 died after discharge, and 3 were dependent at 3 months. Major artery occlusion or severe stenosis was observed in 5 patients. Two patients subsequently developed subcortical hemorrhage. On gradient echo imaging, lobar cerebral microbleeds were observed in 2 patients, and chronic superficial siderosis was observed in 2 patients. Conclusions: In this retrospective review of cases with ischemic stroke and cSAH, over half of patients had occlusion of major arteries. Cerebral amyloid angiopathy was suggested by magnetic resonance imaging findings and subsequent events in 3 patients. The overall outcome was unfavorable although the causal relationship with cSAH was unclear. Key Words: Carotid stenosis— cerebral amyloid angiopathy—convexal subarachnoid hemorrhage—intracranial stenosis—superficial siderosis. Ó 2014 by National Stroke Association
Introduction Nontraumatic convexal subarachnoid hemorrhage (cSAH) without any causative intracranial aneurysm
From the *Department of Neurology, Stroke Center, Saiseikai Kumamoto Hospital, Kumamoto; †Department of Neurology and Neuromuscular Disorder, Oita University Faculty of Medicine, Yufu; and ‡Department of Neurology, Graduate School of Medical Sciences, Kumamoto University, Kumamoto, Japan. Received November 26, 2013; revision received December 23, 2013; accepted December 25, 2013. Conflicts of interest: None. Source of funding: None. Address correspondence to Makoto Nakajima, MD, PhD, Department of Neurology, Graduate School of Medical Sciences, Kumamoto University, 1-1-1, Honjo, Chuo-ku, Kumamoto, 860-8556, Japan. E-mail: [email protected]. 1052-3057/$ - see front matter Ó 2014 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2013.12.046
has been increasingly recognized as a separate entity with distinct pathophysiological mechanisms.1-9 Various vascular or nonvascular mechanisms have been suggested for cSAH. Some authors reported patients who developed cSAH subsequent to acute ischemic events,2,6-9 but the incidence, clinical characteristics, and outcomes of such patients have not yet been established. Our database of patients with acute ischemic stroke/ transient ischemic attack (TIA) was reviewed to investigate clinical characteristics and courses in patients with cSAH. In addition, possible underlying pathogenesis among these patients was speculated.
Materials and Methods Subjects Our database of patients who were admitted to our stroke center with a diagnosis of acute ischemic
Journal of Stroke and Cerebrovascular Diseases, Vol. -, No. - (---), 2014: pp 1-7
1
M. NAKAJIMA ET AL.
2
Table 1. Summary of patients who developed cortical subarachnoid hemorrhage after ischemic stroke
Case
Age (y), sex
Underlying disease
1
63, M DM, HT
2
72, M CHF, AF, CKD, ICH 72, M AF
3
4
5
Arterial disease Right extracranial ICAO UD (because of motion artifact) N
70, M Chronic hepatitis, Right extracranial DM, HT, DLP ICAS and bilateral VAS 66, M OMI, HT, DLP N
Initial NIHSS score 6
Antithrombotics
Stroke subtype
Site of infarction (arterial territory)
Initial gradient echo image
Argatroban, aspirin, cilostazol Heparin, aspirin
LA
Right MCA
Normal
CE
0
N
TIA
Bilateral PICA, right MCA Right MCA
1
Ozagrel, heparin, cilostazol
LA
BA, right PICA
Old putaminal hematoma 4 lobar CMBs and superficial siderosis Normal
1
Argatroban, aspirin
UD
Left MCA
Normal
21
6
77, M DLP, gout, lung carcinoma
Right extracranial ICAO
0
Argatroban, aspirin
LA
Right MCA
Normal
7
80, F N
Left MCAO
3
N
UD
Left MCA
1 lobar CMB
8
65, M AF, HT, CKD
Right extracranial ICAO
Heparin, warfarin
CE
Right MCA
Superficial siderosis
18
Abbreviations: AF, atrial fibrillation; AMI, acute myocardial infarction; BA, basilar artery; CCA, cerebral amyloid angiopathy; CE, cardioembolic stroke; CHF, congestive heart failure; CKD, chronic kidney disease; CMB, cerebral microbleed; cSAH, cortical subarachnoid hemorrhage; DLP, dyslipidemia; DM, diabetes mellitus; HT, hypertension; ICAO, internal carotid artery occlusion; ICAS, internal carotid artery stenosis; ICH, intracerebral hemorrhage; LA, large artery atherosclerosis; LOS, length of hospital stay; mRS, modified Rankin Scale; MCA, middle cerebral artery; NA, not applicable; NIHSS, National Institutes of Health Stroke Scale; PICA, posterior inferior cerebellar artery; TIA, transient ischemic attack; UD, undetermined; VAS, vertebral artery stenosis. *Standard value is #1.0 mg/mL.
stroke/TIA between 2005 and 2011 was reviewed. All patients underwent both brain computed tomography (CT) and magnetic resonance imaging (MRI; except for those with contraindications) on admission. In addition, almost every patient routinely underwent follow-up CT and/or MRI at least once during the hospital stay to detect recurrence, hemorrhagic complications, or changes in arterial findings. Initial MRI included the following sequences: diffusion-weighted imaging, T2-weighted imaging, fluid-attenuated inversion recovery (FLAIR) imaging, and magnetic resonance angiography (MRA). On follow-up MRI, gradient echo imaging (GRE) was added on the sequences of initial MRI. If cSAH was detected, causative arterial abnormalities were further investigated using CT angiography or reconstruction of MRA. All images were double reviewed separately by the attending physician and a radiologist.
Diagnosis of cSAH Of a total of 4953 patients during the study period, patients with a diagnosis of acute cSAH, which was de-
tected on CT (and/or MRI) and rapidly resolved on the follow-up imaging, were selected. Patients in whom cSAH was detected 14 days or more after admission were excluded because the aim was to assess cSAH in ischemic stroke patients during their acute phase. Focal linear hypointense signal changes in the superficial layers of the cerebral cortex on GRE (superficial siderosis),4,10 which were not detected by CT or FLAIR imaging but detected only by GRE were regarded as old cSAH. Patients with only superficial siderosis were excluded from this study. In addition, patients with a preceding episode of head trauma, intracranial aneurysm, thrombolytic therapy, or intracranial dissection were also excluded.
Data Collection The following clinical information of patients with cSAH was collected: age and sex, underlying disease, site of stenosis greater than 50% of the corresponding artery for the index stroke, National Institutes of Health Stroke Scale score on admission, acute antithrombotic
ACUTE STROKE AND CONVEXAL SUBARACHNOID HEMORRHAGE
mRS score
Days for cSAH detection D-dimer* after ischemic stroke (mg/mL) .4
1 (CT) 9 (MRI)
11.0
2 (CT) 3 (MRI)
1.1
.3
UD: on admission (CT 1 MRI) 13 (CT 1 MRI)
.2
3 (MRI)
1.1
3 (CT 1 MRI)
1.1
UD: on admission (CT 1 MRI) 1 (MRI) 5 (CT)
.7
3
Site of cSAH (lobe) Right frontal
Trigger for cSAH detection
Complications
Neurologic deterioration Right parietal N Lerish syndrome, AMI Left frontal N Subcortical hemorrhage (10 M) Right parietal N Recurrence of ischemic stroke, pneumonia Left N Subcortical frontoparietal hemorrhage (day 15) Headache and Frequent TIA Right frontoparietal convulsion Left frontal N Neurological deterioration Right parietal
N
N
therapy when cSAH was detected, stroke subtype according to the Trial of Org 10172 Acute Stroke Treatment classification, site of the index ischemic stroke, initial GRE finding, D-dimer level during the hospital stay, time between the onset of the index stroke and detection of cSAH, trigger for detection of cSAH, other complications during the hospital stay, length of hospital stay, modified Rankin Scale scores at discharge and at 3 months, and diagnosis of cerebral amyloid angiopathy (CAA) based on the Boston criteria.11
Clinical Workup for Diagnosis The site of arterial stenosis was divided into the extracranial internal carotid artery (ICA), the intracranial ICA, or the other main cerebral arteries based on the results of carotid ultrasonography and MRA. Because the exact onset time of cSAH that was detected incidentally could not be determined, the medical record in each case was carefully reviewed to search for new clinical events or symptom changes possibly related to the cSAH between the last brain imaging examination without cSAH and the first evidence of cSAH (such as sudden headache, consciousness disturbance, or other neurologic deficits). A standard blood examination including prothrombin time, activated partial prothrombin time, blood cell count, fibrinogen, and D-dimer was performed in all patients. In patients who were suspected to have abnormal coagulation–fibrinolysis activity, fibrin degradation products
N
At LOS discharge
At 3 months
Diagnosis of CAA
23
4
UD
NA
11
6
6
NA
14
0
40
6
6
NA
6
0
5
Possible CAA
17
2
6
NA
20
4
3
Possible CAA
22
4
4
NA
0 / 5 (10 M) Probable CAA
(FDPs) or antithrombin III was also assessed. Both transthoracic echocardiography and 24-hour Holter electrocardiogram monitoring were performed to detect embolic sources in patients with no history of atrial fibrillation.
Results Of 4953 patients, 16 patients showed cSAH. Eight patients were excluded because of superficial siderosis without acute cSAH (n 5 1), head trauma (n 5 1), ruptured aneurysm (n 5 1), precedent thrombolytic therapy (n 5 2), or a diagnosis of intracranial arterial dissection (n 5 3). Therefore, 8 patients (.14%; 7 men; aged 71 6 6 years) demonstrated cSAH during the treatment of acute ischemic stroke/TIA (Table 1, Fig 1): 7 were detected incidentally and 1 was detected immediately after a convulsion (case 6). cSAH was confirmed on CT, FLAIR, and GRE in all but 1 patient (without CT; case 5). Two patients who had superficial siderosis (cases 3 and 8) on the initial GRE were included in the 8 patients because they also demonstrated acute cSAH during their hospital stay. Two patients (cases 3 and 7) demonstrated both cSAH and acute ischemic lesions on admission. Their overall outcome was poor: 2 patients died during their hospital stay because of ventricular fibrillation (case 2) and severe aspiration pneumonia (case 4). Two patients (cases 3 and 5) discharged without any neurologic deficit developed subcortical hemorrhage thereafter. In detail, the former
4
M. NAKAJIMA ET AL.
Figure 1. Brain images of 8 acute ischemic stroke patients with convexal subarachnoid hemorrhage (cSAH; upper: diffusion-weighted images on admission; middle, fluid-attenuated inversion recovery images; lower, computed tomography except for case 6 of gradient echo imaging). cSAH (arrowheads) was not observed in all patients on admission. All patients except for cases 3 and 4 demonstrate focal cSAH in the same vascular territory as the index ischemic lesions.
(case 3) started warfarin at discharge because of atrial fibrillation and developed hemorrhage in the left parietal lobe (ie, distinct from the site of cSAH) after 10 months. The latter (case 5) continued aspirin, which he had been prescribed because of coronary heart disease, and developed hemorrhage in the left frontoparietal lobe on day 15. The hematoma included the site of cSAH. In both of them, GRE before hemorrhage did not demonstrate microbleeds on the bleeding site, and no abnormal findings were observed on cerebral angiography performed after hemorrhage. One patient (case 6) died after discharge because of bleeding from lung cancer, and at least 3 patients were dependent at 3 months.
Occlusion or stenosis of a major artery ipsilateral to the cSAH was detected in 5 patients. Of these, hemodynamic compromise in the affected area was seen in 2 patients: broad hypoperfusion in the right anterior circulation on single-photon emission CT (case 1; Fig 2) and severe delayed collateral flow of the right anterior circulation on cerebral angiography (case 6; Fig 3). Three patients (cases 1, 4, and 6) had severe atherosclerotic changes in the extracranial ICA and/or the vertebral arteries but no proximal embolic sources. Case 7 had neither embolic sources nor severe atherosclerotic changes in the ICA. On the other hand, 1 patient with atrial fibrillation (case 8) demonstrated extracranial ICA occlusion on admission.
ACUTE STROKE AND CONVEXAL SUBARACHNOID HEMORRHAGE
5
Figure 2. Single-photon emission computed tomography using 99m Tc-ECD without acetazolamide in case 1. The cerebral blood flow to the right cerebral hemisphere is extensively decreased, even outside the infarcted region (arrowheads).
Follow-up CT detected cSAH on day 5, and MRA demonstrated recanalization of the right ICA on day 7. In 1 patient (case 5), blood pressure on admission was as high as 187/89 mm Hg and then it gradually decreased. The other patients experienced neither hypertension more than 180 mm Hg nor abrupt elevation of blood pressure between admission and detection of cSAH. Platelet counts during the hospital stay were within normal limits in all subjects. The serum D-dimer level was elevated substantially in 1 patient (case 2); the serum FDP level (13.7 mg/dL) and fibrinogen (477 mg/dL) were also elevated, whereas antithrombin III was within the normal limit (77%). The serum D-dimer level was elevated marginally in 3 other patients, one of which with lung carcinoma (case 6) showed elevated serum fibrinogen (508 mg/dL), normal FDP level (4.0 mg/dL), and antithrombin III activity (88%).
vascular anomaly, infectious disease, venous thrombosis, various vasculopathies, coagulation disorder, or brain tumor.2,9 In an observational study of 24 cSAH cases, Beitzke et al8 found 5 cases with concurrent ischemic lesions (including 3 asymptomatic cases), as in the present study. However, they did not mention the mechanism of ischemia. The etiology of cSAH in each patient in the present case series is now discussed. In cases 3, 5, and 7, CAA was suspected as the most plausible cause of cSAH. They fulfilled the criteria of ‘‘probable CAA’’ or ‘‘possible CAA.’’4,11 CAA is one of
Discussion Overall, 8 (.14%) of 4953 patients showed cSAH during the acute treatment of ischemic stroke. This was a small case series; however, we would emphasize the poor outcomes of the patients. Although the prevalence of cSAH is reported to be 5%-7.5% in spontaneous SAH patients,3,12 little has been known regarding the incidence of cSAH concomitant with acute ischemic stroke. On the other hand, many different etiologies of cSAH have been proposed: CAA, arterial occlusive disease,
Figure 3. Intra-arterial subtraction angiography of the right vertebral artery in case 6. The territory of the right anterior circulation is perfused by a compensated bypass from the right vertebral artery, muscle branches (arrowheads), external carotid artery (retrograde flow), internal carotid artery, and also by leptomeningeal anastomosis.
M. NAKAJIMA ET AL.
6 3-5,8
the most common pathogeneses of cSAH. On the other hand, CAA can cause brain infarction because of the deposition of amyloid in the wall of meningeal and parenchymal arteries or arterioles, which presumably induces a disturbance of vascular reactivity or focal circulation.13 Some authors speculate that small ischemia and hemorrhage in CAA patients may share common pathophysiologic steps.14 Therefore, if the 3 patients had CAA, the prior ischemic event might also have been associated with CAA. Five patients (cases 1, 4, 6, 7, and 8) had occlusion or severe stenosis of major arteries. Of these, the first 3 patients (cases 1, 4, and 6) were suspected to have chronic occlusive arterial disease from the sonographic findings and the absence of emboligenic diseases. The main mechanism of cSAH because of chronic arterial occlusive diseases is speculated to be the rupture of dilated fragile compensatory pial vessels.2,6,7 Imaging findings that suggested hemodynamic compromise in cases 1 and 6 seemed to be consistent with this hypothesis. Importantly, most previous cases of cSAH associated with carotid artery occlusion/severe stenosis demonstrated acute ischemic lesions in the territory of the affected artery.2,6,7 Simultaneous occurrence of an ischemic event and cSAH might indicate that hemodynamic insufficiency because of arterial occlusion/stenosis reached a critical point. One possible scenario is that, for example, after an acute ischemic stroke because of occlusive disease via an embolic and/or hemodynamic mechanism, subsequent dynamic changes of intracranial perfusion pressure might trigger cSAH. The other 2 (cases 7 and 8) cases were rather complicated. The first 1 (case 7) demonstrated intracranial arterial occlusion, but there were few findings that supported an embolic or atherothrombotic mechanism. Regarding the mechanism of cSAH, as mentioned earlier, CAA would be a possible cause. The latter one (case 8) had atrial fibrillation and demonstrated recanalization of the occluded artery, which was diagnosed as cardioembolic stroke. It would be rather difficult to conclude that the cSAH resulted from the same mechanism as in those who had chronic occlusive diseases. This patient did not fulfill the CAA criteria11 but did match possible CAA in the modified criteria of Linn et al,4 based on the chronic superficial siderosis. Another possibility is reperfusion injury, although the exact temporal relationship between reperfusion of the ICA and cSAH could not be determined. Malignancy was observed in 1 patient (case 6). Coagulation–fibrinolysis activity is frequently unstable in patients with malignant diseases. It may cause either ischemic stroke (embolism) or hemorrhagic stroke.15 In a recent study of 208 intracranial hemorrhages in patients with cancer, one of the common causes of intracranial hemorrhage was coagulopathy (67%).16 In the present patient (case 6), an abnormal hemostatic status might
have had an effect on cSAH. Another possible cause of cSAH in patients with malignancy is intracranial microaneurysm based on nonbacterial thromboendocarditis. However, transthoracic echocardiography did not detect evidence of vegetations in this patient. One patient (case 2), who had many underlying diseases and demonstrated severely elevated D-dimer level, developed systemic embolic complications. Although abnormal coagulation–fibrinolysis activity or arterial occlusive diseases might have been involved in the occurrence of cSAH and the clinical course, these hypotheses could not be validated. Finally, influence of antithrombotic agents on the occurrence of cSAH other than in cases 3 and 7 should be considered. Even patients with unruptured intracranial aneurysms do not usually develop subarachnoid hemorrhage during treatment of acute ischemic stroke.17 However, some patients whose intracranial arteries are vulnerable because of the aforementioned specific pathogeneses such as CAA or arterial stenotic disease might develop cSAH because of intensive antithrombotic therapy. Because this was a retrospective study, it is difficult to clarify the relationship between antithrombotics and cSAH. Ischemic lesions and cSAH are observed in the different arterial territory in 2 patients (cases 3 and 4). Given that either had a potential embolic source (AF in case 3 and multiple arterial stenosis in case 4), it is possible that the ischemic event and cSAH occurred coincidentally. Likewise, we must take into account the same possibility also in the other 6 cases. This study had some limitations. First, some patients with asymptomatic cSAH might have been missed, although most patients underwent a follow-up CT or MRI during their hospital stay. Second, because this was a retrospective study using clinical records, the relationships between cSAH and symptoms, results of examinations, or comorbidities could not be completely proven. Besides, we might have possibly overlooked faint cSAH or clinical signs related to certain pathogenesis in some patients; the incidence of cSAH among ischemic stroke should be interpreted with cautious. To overcome these issues, a prospective case–control study with a definite protocol for examination, imaging, and collecting clinical signs would be required. Third, because cerebral angiography or autopsy was not performed in most of the patients, small aneurysms locating at the branches or focal venous thrombosis may have been dismissed. In summary, 8 cases with cSAH observed in the acute phase of ischemic stroke were reported. Clinical courses vary among the cases and most patients died or were dependent; however, cSAH did not directly affect these unfavorable outcomes. Although some comorbidities including occlusive arterial disease and CAA were suspected to be associated with cSAH, the relationships could not be confirmed. Particular MRI sequences such
ACUTE STROKE AND CONVEXAL SUBARACHNOID HEMORRHAGE
as GRE or susceptibility-weighted imaging may facilitate detection of asymptomatic cSAH.9 Careful observation of images and clinical symptoms would be required for clinicians to detect and collect these patients. Acknowledgment: We express our sincere gratitude to Dr Hiroyuki Kawano (Saiseikai Kumamoto Hospital) for the assistance in collecting the patient data and for his professional comments. We are also deeply grateful to Dr Toru Nishi and staff members of the Department of Neurosurgery, Saiseikai Kumamoto Hospital for their ungrudging clinical supports and professional advices.
References 1. Ducros A, Fiedler U, Porcher R, et al. Hemorrhagic manifestations of reversible cerebral vasoconstriction syndrome: frequency, features, and risk factors. Stroke 2010;41:2505-2511. 2. Cuvinciuc V, Viguier A, Calviere L, et al. Isolated acute nontraumatic cortical subarachnoid hemorrhage. AJNR Am J Neuroradiol 2010;31:1355-1362. 3. Kumar S, Goddeau RP Jr, Selim MH, et al. Atraumatic convexal subarachnoid hemorrhage: clinical presentation, imaging patterns, and etiologies. Neurology 2010;74:893-899. 4. Linn J, Halpin A, Demaerel P, et al. Prevalence of superficial siderosis in patients with cerebral amyloid angiopathy. Neurology 2010;74:1346-1350. 5. Raposo N, Viguier A, Cuvinciuc V, et al. Cortical subarachnoid haemorrhage in the elderly: a recurrent event probably related to cerebral amyloid angiopathy. Eur J Neurol 2011;18:597-603. 6. Geraldes R, Santos C, Canhao P. Atraumatic localized convexity subarachnoid hemorrhage associated with acute carotid artery occlusion. Eur J Neurol 2011; 18:e28-e29.
7
7. Chandra RV, Leslie-Mazwi TM, Oh D, et al. Extracranial internal carotid artery stenosis as a cause of cortical subarachnoid hemorrhage. AJNR Am J Neuroradiol 2011; 32:E51-E52. author reply E53. 8. Beitzke M, Gattringer T, Enzinger C, et al. Clinical presentation, etiology, and long-term prognosis in patients with nontraumatic convexal subarachnoid hemorrhage. Stroke 2011;42:3055-3060. 9. Renou P, Tourdias T, Fleury O, et al. Atraumatic nonaneurysmal sulcal subarachnoid hemorrhages: a diagnostic workup based on a case series. Cerebrovasc Dis 2012;34:147-152. 10. Linn J, Herms J, Dichgans M, et al. Subarachnoid hemosiderosis and superficial cortical hemosiderosis in cerebral amyloid angiopathy. AJNR Am J Neuroradiol 2008; 29:184-186. 11. Knudsen KA, Rosand J, Karluk D, et al. Clinical diagnosis of cerebral amyloid angiopathy: validation of the Boston criteria. Neurology 2001;56:537-539. 12. van Gijn J, Rinkel GJ. Subarachnoid haemorrhage: diagnosis, causes and management. Brain 2001;124: 249-278. 13. Cadavid D, Mena H, Koeller K, et al. Cerebral beta amyloid angiopathy is a risk factor for cerebral ischemic infarction. A case control study in human brain biopsies. J Neuropathol Exp Neurol 2000;59:768-773. 14. Kimberly WT, Gilson A, Rost NS, et al. Silent ischemic infarcts are associated with hemorrhage burden in cerebral amyloid angiopathy. Neurology 2009;72: 1230-1235. 15. Rogers LR. Cerebrovascular complications in patients with cancer. Semin Neurol 2010;30:311-319. 16. Navi BB, Reichman JS, Berlin D, et al. Intracerebral and subarachnoid hemorrhage in patients with cancer. Neurology 2010;74:494-501. 17. Oh YS, Lee SJ, Shon YM, et al. Incidental unruptured intracranial aneurysms in patients with acute ischemic stroke. Cerebrovasc Dis 2008;26:650-653.