An Analysis of Spatial Clustering of Stroke Types, In-hospital Mortality, and Reported Risk Factors in Alberta, Canada, Using Geographic Information Systems.

BACKGROUND: Despite advances in the quality and delivery of stroke care, regional disparities in stroke incidence and outcome persist. Spatial analysis using geographic information systems (GIS) can assist in identifying high-risk populations and regional differences in efficacy of stroke care. The aim of this study was to identify and locate geographic clusters of high or low rates of stroke, risk factors, and in-hospital mortality across a provincial health care network in Alberta, Canada. METHODS: This study employed a spatial epidemiological approach using population-based hospital administrative data. Getis-Ord Gi* and Spatial Scan statistics were used to identify and locate statistically significant "hot" and "cold" spots of stroke occurrence by type, risk factors, and in-hospital mortality. RESULTS: Marked regional variations were found. East central Alberta was a significant hot spot for ischemic stroke (relative risk [RR] 1.43, p<0.001), transient ischemic attack (RR 2.25, p<0.05), and in-hospital mortality (RR 1.50, p<0.05). Hot spots of intracerebral hemorrhage (RR 1.80, p<0.05) and subarachnoid hemorrhage (RR 1.64, p<0.05) were identified in a major urban centre. Unexpectedly, stroke risk factor hot spots (RR 2.58, p<0.001) were not spatially associated (did not overlap) with hot spots of ischemic stroke, transient ischemic attack, or in-hospital mortality. CONCLUSIONS: Integration of health care administrative data sets with geographic information systems contributes valuable information by identifying the existence and location of regional disparities in the spatial distribution of stroke occurrence and outcomes. Findings from this study raise important questions regarding why regional differences exist and how disparities might be mitigated.

The ICV sign as a marker of increased cerebral blood transit time.

OBJECTIVE/BACKGROUND: We describe the internal cerebral vein (ICV) sign, which is a hypo-opacification of the ICV on computed tomogram angiography (CTA) as a new marker of increased cerebral blood transit-time in ipsilateral internal carotid artery occlusions (ICAO). METHODS: A retrospective analysis of 153 patients with acute unilateral M1 middle cerebral artery (MCA) occlusions ± ICAOs was performed. The degree of contrast opacification of the ICV on the ipsilesional side was compared to that of the unaffected side. RESULTS: Of 153 patients in our study, 135 had M1 MCA occlusions ± intra-cranial ICAO (M1±iICAO) and 18 had isolated extracranial ICAO (eICAO). In the patients with proximal M1±iICAO, 57/65 (87.1%) showed the ICV sign. Of the 8 patients without the ICV sign in this group, 6 had prominent lenticulostriate arteries arising from the non-occluded M1 segment, 1 had a recurrent artery of Huebner, and 1 had filling of distal ICA/M1 segment through prominent Circle of Willis collaterals. For the 70 patients with isolated distal M1±iICAO, 7/70 (10%) showed the ICV sign, with all 7 showing occluded lenticulostriate arteries. Of the patients with eICAO, 8/18 showed the ICV sign, all 8 with the ICV sign had poor Circle of Willis collaterals. CONCLUSIONS: The ICV sign correlates well with presence of proximal M1±iICAO in patients with either occluded lenticulostriate arteries or poor Circle of Willis collaterals. In patients with eICAO, the sign correlates with reduced Circle of Willis collaterals and may be a marker of increased ipsilateral cerebral blood transit time.

Tissue Window in Stroke Thrombolysis study (TWIST): a safety study.

BACKGROUND: Stroke thrombolysis is limited by the "last-seen well" principle, which defines stroke onset time. A significant minority of stroke patients (~15%) awake with their symptoms and are by definition ineligible for thrombolysis because they were "last-seen well" at the time they went to bed implying an interval that is most often greater than three hours. METHODS: A single-centre prospective, safety study was designed to thrombolyse 20 subjects with stroke-on-awakening. Patients were eligible for inclusion if they were last seen well less than 12 hours previously, specifically including those who awoke from sleep with their stroke deficits. They had a baseline computed tomogram (CT) scan with an ASPECTS score greater than 5, no evidence of well-evolved infarction and a CT angiogram / Trans-cranial Doppler ultrasound study demonstrating an intracranial arterial occlusion. Patients fulfilled all other standard criteria for stroke thrombolysis. The primary outcome was safety defined by symptomatic ICH or death. RESULTS: Among 89 screened patients, 20 were treated with thrombolysis. Two patients (10%) died due to massive carotid territory stroke and two patients (10%) died of stroke complications. Two patients (10%) showed asymptomatic intracerebral hemorrhage (ICH) (petechial hemorrhage) and none symptomatic ICH. Reasons for exclusion were: (a) ASPECTS ≤ 5 (29); (b) well-evolved infarcts on CT (19); (c) historical mRS > 2 (17); (d) no demonstrable arterial occlusion or were too mild to warrant treatment (10). CONCLUSIONS: Patients who awake with their deficits can be safely treated with thrombolysis based upon a tissue window defined by NCCT and CTA/TCD.

Anterior temporal artery sign in CT angiography predicts reduced fatal brain edema and mortality in acute M1 middle cerebral artery occlusions.

BACKGROUND: Mortality in acute ischemic middle cerebral artery (MCA) stroke ranges from 5% to 45%. We identify a vascular imaging sign, presence of "prominent anterior temporal artery" on computed tomography (CT) angiography (CTA) and investigate whether it predicts mortality in acute M1-MCA occlusions. METHODS: One hundred and two patients with acute M1-MCA occlusions from 2003-to 2007 were included in the study. A prominent anterior temporal artery arising from proximal M1 MCA was identified by two readers blinded to clinical outcome. Primary clinical outcome was survival (modified Rankin Scale [mRS] 0-5) at 3 months. RESULTS: An anterior temporal artery arising from M1 MCA was present in 20/102 (20%). Eighteen of 20 (90%) patients with this sign survived at 3 months (mRS 0-5) when compared to 66/82 (80.4%) patients without the sign (odds ratio 2.2 CI(95) .5-10.4). The sign has a sensitivity of 21% (CI(95) .13-.25) but specificity of 89% (CI(95) .64-.98) in predicting survival at 3 months. Positive predictive value was 90% with likelihood ratio of 1.9 (CI(95) .9-7.6). CONCLUSION: Presence of prominent anterior temporal artery in M1-MCA occlusions on CTA identifies a group of patients with reduced case fatality. The mechanism is likely related to a reduced chance of malignant cerebral edema.