| OCR Text |
Show NEURO-OPHTHALMOLOGY AT LARGE The International Stroke Conference 2008 New Orleans, Louisiana February 20-22, 2008 The 2008 International Stroke Conference was held at the Ernest N. Morial Convention Center in New Orleans, Louisiana, February 20-22, 2008 and had more than 5,000 attendees. The meeting included more than 600 oral and poster presentations. The following papers were of special interest. ACUTE ISCHEMIC STROKE 1. EPITHET Intravenous tissue plasminogen activator (tPA) is currently administered to eligible patients within 3 hours of ischemic stroke onset. If an ischemic penumbra can be documented, the possibility exists for extending the time window to increase the number of patients treated. Stephen M. Davis, MD, University of Melbourne, Australia, reported the results of the Echo-Planar Imaging Thrombolysis Evaluation Trial (EPITHET) with full publication in Lancet Neurol (2008;7:299-309). The study is based on the hypothesis that a perfusion and diffusion MRI mismatch would signify potentially salvageable tissue. Thrombolysis might attenuate infarct growth by increased reperfusion of this area. The study was a double-blind, placebo-controlled, phase 2 study at 15 centers in Australia, New Zealand, Scotland, and Belgium. Patients who were tPA-eligible by National Institute of Neurological Disorders and Stroke criteria presenting 3-6 hours from onset of stroke symptoms underwent diffusion, perfusion, and MRA imaging. After imaging, one half of the patients received tPA and the remainder received placebo. Repeat imaging was performed at days 3-5, and T2MRI was performed after 90 days. The primary outcome was infarct growth between baseline diffusion imaging and the day 90 T2 lesion in mismatch patients using a ratio of geometric means. A total of 101 patients were enrolled in the trial. Data were analyzed from 44 patients in the tPA group and 47 patients in the placebo group. Mismatch was defined as perfusion imaging > diffusion imaging volume by 20% and perfusion-diffusion image volume of at least 10 mL. Using this definition, 37 of 44 tPA patients and 43 of 47 placebo patients had a mismatch. The primary outcome of relative infarct volume growth was 1.24 for the tPA group and 1.78 for the placebo group (P = 0.24). The major secondary outcome was reperfusion, defined as >90% reduction between baseline and day 3 perfusion imaging volumes. This outcome significantly favored the tPA group, with 56% reduction in the tPA group and 26% reduction in the placebo group (P = 0.01; 95% CI: 9%-51%). There was no significant clinical difference in the mismatch patients who received tPA compared with those who received placebo. Four patients, all in the tPA treatment group, had symptomatic hemorrhage. Because of the small number of non-mismatch patients, the authors were unable to compare the mismatch and non-mismatch groups. The trial results support the feasibility of performing MRI in acute stroke patients, although the feasibility of processing the perfusion image acutely for real-time clinical decision-making was not addressed and represents a major limitation of this work. The authors concluded that EPITHET results provide biologic support for extension of the time window for tPA and the need for a larger trial. Although not conclusive, this study provides evidence that further trials examining the use of acute MRI to extend the tPA window are needed. 2. CCHIPS Blood pressure reduction is beneficial for primary and secondary stroke prevention. The management of hypertension in the setting of acute stroke is unclear. John Potter, MD, University of East Anglia, United Kingdom, presented data from the Control of Hypertension and Hypotension Immediately Post-Stroke (CHIPPS) pilot trial. CHHIPS is a multicenter, prospective, randomized, double-blind, placebo-controlled titrated dose phase 2 trial in the United Kingdom. Patients aged older than 18 years without recent antihypertensive treatment who have systolic blood pressure (SBP) greater than 160 mmHg were randomized within 24 hours and, later in the study, up to 36 hours after onset of their suspected stroke symptoms. Patients un-dergoing thrombolysis were excluded. Patients were randomized to oral medications of 50 mg labetalol or 5 mg lisinopril or placebo if they were able to swallow. Patients who were unable to swallow were given sublingual lisinopril, intravenous labetalol or placebo. The blood J Neuro-Ophthalmol, Vol. 28, No. 3, 2008 245 pressure readings were repeated at 4 and 8 hours and, if the target systolic blood pressure of 145-155 mm Hg or a 15 mm Hg fall in blood pressure was not achieved, the antihypertensive medication was repeated. The titrated dose was continued for 14 days. In the dysphagic group, the medications were administered by nasogastric or gastro-stomy tube. A total of 179 patients were randomized (58 patients to the labetalol group, 58 to the lisinopril group, and 63 to the placebo group). Nearly half of each group was com-posed of dysphagic patients. The study included patients with ischemic and hemorrhagic strokes. Patients undergo-ing thrombolysis were excluded. The mean National Insti-tutes of Health Stroke Scale (NIHSS) score was 11. The primary outcome was the proportion of patients dead or dependent (dependence defined as a modified Rankin Scale score >3 at poststroke day 14). There was no difference between the active treatment group and the placebo group regarding the primary outcome of death and dependence at 14 days (odds ratio: 1.08; 95% CI: 0.55-2.03). Lowering blood pressure did not have an adverse effect on neurologic status, defined as increase in NIHSS$4 or death at 72 hours (antihypertensive drug 7% vs placebo 10%, P = 0.56). The major strengths of this study were the use of a clinical end point and the ease of the treatment protocol. Blood pressure reduction did not alter death or disability at 2 weeks, and it was not associated with deterioration of neurologic status at 72 hours. The authors concluded that a large trial is needed to confirm these results. INTRACEREBRAL HEMORRHAGE 1. Subanalysis of FAST Trial During the first 3 hours after intracerebral hemor-rhage (ICH), hematoma expansion is common and associated with a worse outcome. In the phase 2b trial of the Factor Seven for Acute Hemorrhagic Stroke Treatment (FAST) trial, a dose-related reduction in mean percentage change in ICH volume at 24 hours was shown in patients treated with recombinant activated factor VII (rFVIIa). The phase 3 FAST trial randomized acute intracerebral hemorrhage patients to receive 80 mg/kg or 20 mg/kg of rFVIIa or placebo. The primary outcome was death or severe disability, defined as a 90-day modified Rankin Scale score of 5 or 6. There was no significant difference in severe disability or mortality between the placebo and rFVIIa-treated groups. However, there was a significant decrease in mean percent change in hematoma growth measured at baseline and at 24 hours in the patients who were treated with 80 mg/kg of rFVIIa (11%) compared with the placebo patients (26%, P = 0.0004). Stephan Mayer, MD, Columbia University, New York, presented the results of a subgroup analysis of patients treated with 80 mg/kg rFVIIa compared with placebo. The data were examined continuously and then combined into various clinically feasible groups to identify an optimal subgroup. The data from this group were then evaluated in the phase 2 trial. The optimized target population was found to be patients aged younger than 70 years who were treated within 150 minutes and who had less than 5 mL of intraventricular hemorrhage (IVH) volume and less than 60 mL of ICH volume. The odds ratio of death or disability for all patients treated with 80 mg/kg rFVIIa in the phase 3 trial was 1.39 compared with 0.28 for the optimized group, a result that was not significant The authors proposed a clinical trial with rFVIIa for ICH in this optimized patient population. It was disappointing that the phase 3 FAST trial did not demonstrate clinical efficacy after the promising results of the phase 2b trial. Whether rFVIIa will be a clinically useful treatment for acute ICH remains doubtful. 2. INTERACT Elevated blood pressure after ICH is common, and the optimal management strategy has not been established in a large clinical trial. Such a strategy likely would need to strike a balance between presumed reduction in infarct growth from persistent hypertension and maintenance of cerebral perfusion to prevent extension of brain injury. Craig Anderson, MD, University of Sydney, Australia, presented the results of the first phase of the Intensive Blood Pressure Reduction in Acute Cerebral Hemorrhage Trial (INTERACT). This vanguard (pilot) phase was intended to gather feasibility and recruitment capacity data. It also examined hematoma expansion, an important surrogate end point. Patients were eligible if they had a CT-confirmed ICH and elevated SBP between 150 and 220 mm Hg. Subjects were randomized to intensive treatment (target SBP of 140 mmHg within 6 hours) or current standard treatment (American Stroke Association guideline target SBP of 180 mmHg.) As this was an international trial, the investigators at each site could choose different blood pressure-lowering agents. The trial hypothesis was that the intensity of blood pressure lowering and not the choice of a specific agent would drive the clinical effect. The primary outcome measure was proportional hematoma volume growth at 24 hours using digital analysis of CT images. The safety of the treatment strategy was assessed by measuring clinical outcome at 90 days. Centers from Australia, China, and Korea enrolled 404 patients in this pilot trial between November 2005 and April 2007. About 95% of the total enrolled patients were from China. Baseline characteristics of the treatment and control groups were similar. The SBP was 13.3 mm Hg lower in the treatment group at 1 hour and 10.8 mm Hg lower during hours 1-24 (P < 0.0001 for comparison 246 q 2008 Lippincott Williams & Wilkins J Neuro-Ophthalmol, Vol. 28, No. 3, 2008 Skolarus and Meurer between treatment and control groups at both time periods). The primary outcome, mean proportional hematoma growth, was 13.7% in the treatment group versus 36.3% in the control group with an absolute difference of 22.6% (95% CI: 0.6%-44.6%). The total hematoma volume in the control group was 12.7 mL at baseline and 15.4 mL at 24 hours. In the treatment group, the hematoma volume was 14.2 mL at baseline and 15.2 mL at 24 hours. No significant differences between the treatment and control groups were observed for clinical or safety outcomes. Case fatality was 10% in the treatment group and 12% in the control group. The authors concluded that the larger phase of this trial powered to determine efficacy for a clinical outcome should go forward. The safety of this strategy appears sound. It is somewhat disappointing that a larger clinical effect on patient outcome was not observed, but this pilot study was not designed to detect this. Of additional concern is the generalizability of the results, given that most enrolled patients were from China and that hematomas were relatively small in volume. The small volumes observed were concordant with the authors' observation that the mild clinical status of the patients in this trial was an important limitation, and probably tended to decrease the ability to detect clinical outcome differences. The larger hematoma volumes at baseline in the treatment group probably also contributed to this. The recruitment strategy appeared to be effective, but enrollment of a less heterogeneous population for the main phase of this trial will be crucial. STROKE EPIDEMIOLOGY 1. Daytime Sleepiness A growing body of evidence suggests that sleep plays a crucial role in cardiovascular risk. In stroke, this relationship has been examined with respect to disordered sleep, but prospective population-based studies have been limited. Bernadette Boden-Albala, MPH, DrPH, Columbia University, presented findings from the Northern Manhat-tan Study (NOMAS), a population-based prospective cohort study. The goal of the investigators was to explore the relationship between daytime sleepiness and the risk of stroke and other cardiovascular events. Daytime sleepiness was chosen as it was believed to be a good surrogate marker for disordered sleep, an important risk factor for cardiac disease, hypertension, and stroke. Subjects in the study had assessment of the Epworth Sleepiness Scale (ESS) or the number of days in a week spent snoring and amount of choking during sleep. The patients were then followed for the occurrence of vascular events; analysis was performed by Cox proportional hazard ratio estimation. Nearly 2,000 subjects were followed for a mean of 2.3 years. There were 156 deaths, 123 vascular events, and 40 strokes. The cohort had a mean age of 73 years and was 64% female and approximately 60% Hispanic (remainder split between whites and blacks); 72% were hypertensive; diabetes and coronary artery disease were each seen in approximately 20%. Of the subjects in the cohort, 44% reported "no dozing" (ESS = 0), 47% reported "some dozing" (ESS of 1-9), and 9% reported "significant dozing" (ESS of 9 or above). After adjustment for stroke risk factors, the hazard ratio for "some dozing" (with "no dozing" as reference class) was 2.6 (95% CI: 1.1-6) and for "significant dozing" was 4.5 (95% CI: 1.5-13). The investigators concluded that further studies of the relation-ship between stroke and sleep are needed. Although preliminary, this work has probably established daytime sleepiness as a novel risk factor for stroke. These results are impressive in that a simple and easy-to-administer scale (the ESS) was used to define the risk factor. The high hazard ratios reported, along with the dose-response relationship, are additional strengths of this work. The cohort, as evidenced by a high burden of stroke risk factors at baseline, is not reflective of the overall population. Extension of this work to a wider population of subjects at risk for stroke will be necessary to draw broader conclusions. 2. Stroke Risk Factors Are Predictive of Cognitive Decline The relationship between ischemic stroke, even subclinical events, and cognitive functioning is well known. The more subtle, yet very important, relationship between cardiovascular risk factors and cognitive decline has not been explored previously in a large population-based study. George Howard, DrPH, University of Alabama at Birming-ham, presented the results of a study from the REasons for Geographic and Racial Differences in Stroke (REGARDS) study, a large population-based cohort study examining differences in stroke risk between white and African- American subjects. The REGARDS study also examined the contribution to disease burden of the "Stroke Belt," a geographic area of the United States with a significantly higher stroke mortality. The investigators sought to examine whether stroke risk factors predict cognitive decline. Participants in the REGARDS study are recruited community dwellers who undergo baseline interviews, physical examinations, laboratory tests, and electrocardio-grams. There is central monitoring for detection of possible stroke events and cognitive decline. Cognitive function is assessed via a validated, six-item screening test derived from the Mini-Mental Status Examination (MMSE). The stroke risk factors are the components of the Framingham Stroke Risk Function: SBP, medications, diabetes, smok-ing, heart disease, atrial fibrillation, and left ventricular hypertrophy. Demographic, educational, and mood-related 247 Neuro-Ophthalmology At Large J Neuro-Ophthalmol, Vol. 28, No. 3, 2008 (depression) factors were considered as possible effect modifiers. The cohort contained approximately 30,000 subjects. Subjects who had a validated stroke or transient ischemic attack were excluded (at the point of time when the event occurred). The mean stroke risk in the cohort was about 1% per year. The estimates for annual mean change in the cognitive score were reported graphically at the conference. The baseline group (with 0% risk of stroke over 10 years according to Framingham score) had an annual decline of 0.02. The group with the highest risk (31.5% 10-year risk of stroke) had an annual cognitive decline of 0.11. The groups with average risk (10.5% 10-year risk of stroke) and elevated risk (21% 10-year risk of stroke) had annual declines of 0.05 and 0.08, respectively. This relationship was highly statistically significant (P < 0.0001). The investigators have shown that the Framingham Stroke Risk Function is strongly associated with annual rate of cognitive decline. One important limitation of this study is that the clinical importance of the measurement scale for cognitive decline is unclear, especially because it was derived from the MMSE, which has limitations of its own. Even so, the results provide strong evidence that the burden of stroke risk factors probably plays an important role in cognitive decline, even in the absence of clinically apparent stroke. Lesli E. Skolarus, MD Department of Neurology Washington University St. Louis, Missouri William J. Meurer, MD Department of Neurology University of Michigan Ann Arbor, Michigan wmeurer@umich.edu 248 q 2008 Lippincott Williams & Wilkins J Neuro-Ophthalmol, Vol. 28, No. 3, 2008 Skolarus and Meurer |
