Wi-Sun Ryu1, Dawid Schellingerhout1, Keun-Sik Hong1, Sang-Wuk Jeong1, Min Uk Jang1, Man-Seok Park1, Kang-Ho Choi1, Joon-Tae Kim1, Beom Joon Kim1, Jun Lee1, Jae-Kwan Cha1, Dae-Hyun Kim1, Hyun-Wook Nah1, Soo Joo Lee1, Jae Guk Kim1, Yong-Jin Cho1, Byung-Chul Lee1, Kyung-Ho Yu1, Mi Sun Oh1, Jong-Moo Park1, Kyusik Kang1, Kyung Bok Lee1, Tai Hwan Park1, Sang-Soon Park1, Juneyoung Lee1, Hee-Joon Bae1, Dong-Eog Kim2. 1. From the Stroke Center and Korean Brain MRI Data Center (W.-S.R., S.-W.J., D.-E.K.), Dongguk University Ilsan Hospital, Goyang; Departments of Radiology and Cancer Systems Imaging (D.S.), University of Texas M.D. Anderson Cancer Center, Houston; Department of Neurology (K.-S.H., Y.-J.C.), Ilsan Paik Hospital, Inje University, Goyang; Department of Neurology (M.U.J.), Hallym University Dongtan Sacred Heart Hospital, Hwaseong; Department of Neurology (M.-S.P., K.-H.C., J.-T.K.), Chonnam National University Medical School, Chonnam National University Hospital, Gwangju; Department of Neurology (B.J.K., H.-J.B.), Seoul National University Bundang Hospital, Seongnam; Department of Neurology (Jun Lee), Yeungnam University Hospital, Daegu; Department of Neurology (J.-K.C., D.-H.K., H.-W.N.), Dong-A University Hospital, Busan; Department of Neurology (S.J.L., J.G.K.), Eulji University Hospital, Daejeon; Department of Neurology (B.-C.L., K.-H.Y., M.S.O.), Hallym University Sacred Heart Hospital, Anyang; Department of Neurology (J.M.-P., K.K.), Nowon Eulji Medical Center, Seoul; Department of Neurology (K.B.L.), Soonchunhyang University Hospital, Seoul; Department of Neurology (T.H.P., S.-S.P.), Seoul Medical Center; and Department of Biostatistics (Juneyoung Lee), Korea University College of Medicine, Seoul. 2. From the Stroke Center and Korean Brain MRI Data Center (W.-S.R., S.-W.J., D.-E.K.), Dongguk University Ilsan Hospital, Goyang; Departments of Radiology and Cancer Systems Imaging (D.S.), University of Texas M.D. Anderson Cancer Center, Houston; Department of Neurology (K.-S.H., Y.-J.C.), Ilsan Paik Hospital, Inje University, Goyang; Department of Neurology (M.U.J.), Hallym University Dongtan Sacred Heart Hospital, Hwaseong; Department of Neurology (M.-S.P., K.-H.C., J.-T.K.), Chonnam National University Medical School, Chonnam National University Hospital, Gwangju; Department of Neurology (B.J.K., H.-J.B.), Seoul National University Bundang Hospital, Seongnam; Department of Neurology (Jun Lee), Yeungnam University Hospital, Daegu; Department of Neurology (J.-K.C., D.-H.K., H.-W.N.), Dong-A University Hospital, Busan; Department of Neurology (S.J.L., J.G.K.), Eulji University Hospital, Daejeon; Department of Neurology (B.-C.L., K.-H.Y., M.S.O.), Hallym University Sacred Heart Hospital, Anyang; Department of Neurology (J.M.-P., K.K.), Nowon Eulji Medical Center, Seoul; Department of Neurology (K.B.L.), Soonchunhyang University Hospital, Seoul; Department of Neurology (T.H.P., S.-S.P.), Seoul Medical Center; and Department of Biostatistics (Juneyoung Lee), Korea University College of Medicine, Seoul. kdongeog@duih.org.
Abstract
OBJECTIVE: To define the role and risks associated with white matter hyperintensity (WMH) load in a stroke population with respect to recurrent stroke and mortality after ischemic stroke. METHODS: A total of 7,101 patients at a network of university hospitals presenting with ischemic strokes were followed up for 1 year. Multivariable Cox proportional hazards model and competing risk analysis were used to examine the independent association between quartiles of WMH load and stroke recurrence and mortality at 1 year. RESULTS: Overall recurrent stroke risk at 1 year was 6.7%/y, divided between 5.6%/y for recurrent ischemic and 0.5%/y for recurrent hemorrhagic strokes. There was a stronger association between WMH volume and recurrent hemorrhagic stroke by quartile (hazard ratio [HR] 7.32, 14.12, and 33.52, respectively) than for ischemic recurrence (HR 1.03, 1.37, and 1.61, respectively), but the absolute incidence of ischemic recurrence by quartile was higher (3.8%/y, 4.5%/y, 6.3%/y, and 8.2%/y by quartiles) vs hemorrhagic recurrence (0.1%/y, 0.4%/y, 0.6%/y, and 1.3%/y). All-cause mortality (10.5%) showed a marked association with WMH volume (HR 1.06, 1.46, and 1.60), but this was attributable to nonvascular rather than vascular causes. CONCLUSIONS: There is an association between WMH volume load and stroke recurrence, and this association is stronger for hemorrhagic than for ischemic stroke, although the absolute risk of ischemic recurrence remains higher. These data should be helpful to practitioners seeking to find the optimal preventive/treatment regimen for poststroke patients and to individualize risk-benefit ratios.
OBJECTIVE: To define the role and risks associated with white matter hyperintensity (WMH) load in a stroke population with respect to recurrent stroke and mortality after ischemic stroke. METHODS: A total of 7,101 patients at a network of university hospitals presenting with ischemic strokes were followed up for 1 year. Multivariable Cox proportional hazards model and competing risk analysis were used to examine the independent association between quartiles of WMH load and stroke recurrence and mortality at 1 year. RESULTS: Overall recurrent stroke risk at 1 year was 6.7%/y, divided between 5.6%/y for recurrent ischemic and 0.5%/y for recurrent hemorrhagic strokes. There was a stronger association between WMH volume and recurrent hemorrhagic stroke by quartile (hazard ratio [HR] 7.32, 14.12, and 33.52, respectively) than for ischemic recurrence (HR 1.03, 1.37, and 1.61, respectively), but the absolute incidence of ischemic recurrence by quartile was higher (3.8%/y, 4.5%/y, 6.3%/y, and 8.2%/y by quartiles) vs hemorrhagic recurrence (0.1%/y, 0.4%/y, 0.6%/y, and 1.3%/y). All-cause mortality (10.5%) showed a marked association with WMH volume (HR 1.06, 1.46, and 1.60), but this was attributable to nonvascular rather than vascular causes. CONCLUSIONS: There is an association between WMH volume load and stroke recurrence, and this association is stronger for hemorrhagic than for ischemic stroke, although the absolute risk of ischemic recurrence remains higher. These data should be helpful to practitioners seeking to find the optimal preventive/treatment regimen for poststroke patients and to individualize risk-benefit ratios.