Infarction of 'non-core-non-penumbral' tissue after stroke: multivariate modelling of clinical impact.

TY - JOUR

T1 - Infarction of 'non-core-non-penumbral' tissue after stroke: multivariate modelling of clinical impact.

AU - Alawneh, Josef A

AU - Jones, Peter Simon

AU - Mikkelsen, Irene Klærke

AU - Cho, Tae-Hee

AU - Siemonsen, Susanne

AU - Mouridsen, Kim

AU - Ribe, Lars

AU - Morris, Rhiannon S

AU - Hjort, Niels

AU - Antoun, Nagui

AU - Gillard, Jonathan H

AU - Fiehler, Jens

AU - Nighoghossian, Norbert

AU - Warburton, Elizabeth A

AU - Ostergaard, Leif

AU - Baron, Jean-Claude

PY - 2011/6/1

Y1 - 2011/6/1

N2 - There is considerable intersubject variability in early neurological course after anterior circulation stroke, yet the pathophysiology underlying this variability is not fully understood. Here, we hypothesize that, although not predicted by current pathophysiological models, infarction of 'non-core-non-penumbral' (i.e. clinically silent) brain tissue may nevertheless occur, and negatively influence clinical course over and above the established positive impact of penumbral salvage. In order to test this hypothesis, non-core-non-penumbral tissue was identified in two independent prospectively recruited cohorts, using computed tomography perfusion, and magnetic resonance perfusion- and diffusion-weighted imaging, respectively. Follow-up structural magnetic resonance imaging was obtained about 1 month later in all patients to map the final infarct. The volumes of both the acutely silent but eventually infarcted tissue, and the eventually non-infarcted penumbra, were determined by performing voxel-wise analysis of the acute and follow-up image sets, using previously validated perfusion thresholds. Early neurological course was expressed as change in National Institutes of Health Stroke Scale scores between the acute and 1-month assessments, relative to the acute score. The relationship between the acutely silent but eventually infarcted tissue volume and early neurological course was tested using a multivariate regression model that included the volume of non-infarcted penumbra. Thirty-four and 58 patients were recruited in the computed tomography perfusion and magnetic resonance perfusion cohorts, respectively (mean onset-to-imaging time: 136 and 156 min; 27 and 42 patients received intravenous thrombolysis, respectively). Infarction of acutely silent tissue was identified in most patients in both cohorts. Although its volume (median 0.2 and 2 ml, respectively) was much smaller than that of salvaged penumbra (59.3 and 93 ml, respectively), it was substantial in ?10% of patients. As expected, salvaged penumbra strongly positively influenced early neurological course. Even after correcting for the latter effect in the multivariate model, infarction of acutely silent tissue independently negatively influenced early neurological course in both cohorts (P=0.018 and 0.031, respectively). This is the first systematic study to document infarction of acutely silent tissue after anterior circulation stroke, and to show that it affects a sizeable fraction of patients and has the predicted negative impact on clinical course. These findings were replicated in two independent cohorts, regardless of the perfusion imaging modality used. Preventing infarction of the tissue not initially at risk should have direct clinical benefit.

AB - There is considerable intersubject variability in early neurological course after anterior circulation stroke, yet the pathophysiology underlying this variability is not fully understood. Here, we hypothesize that, although not predicted by current pathophysiological models, infarction of 'non-core-non-penumbral' (i.e. clinically silent) brain tissue may nevertheless occur, and negatively influence clinical course over and above the established positive impact of penumbral salvage. In order to test this hypothesis, non-core-non-penumbral tissue was identified in two independent prospectively recruited cohorts, using computed tomography perfusion, and magnetic resonance perfusion- and diffusion-weighted imaging, respectively. Follow-up structural magnetic resonance imaging was obtained about 1 month later in all patients to map the final infarct. The volumes of both the acutely silent but eventually infarcted tissue, and the eventually non-infarcted penumbra, were determined by performing voxel-wise analysis of the acute and follow-up image sets, using previously validated perfusion thresholds. Early neurological course was expressed as change in National Institutes of Health Stroke Scale scores between the acute and 1-month assessments, relative to the acute score. The relationship between the acutely silent but eventually infarcted tissue volume and early neurological course was tested using a multivariate regression model that included the volume of non-infarcted penumbra. Thirty-four and 58 patients were recruited in the computed tomography perfusion and magnetic resonance perfusion cohorts, respectively (mean onset-to-imaging time: 136 and 156 min; 27 and 42 patients received intravenous thrombolysis, respectively). Infarction of acutely silent tissue was identified in most patients in both cohorts. Although its volume (median 0.2 and 2 ml, respectively) was much smaller than that of salvaged penumbra (59.3 and 93 ml, respectively), it was substantial in ?10% of patients. As expected, salvaged penumbra strongly positively influenced early neurological course. Even after correcting for the latter effect in the multivariate model, infarction of acutely silent tissue independently negatively influenced early neurological course in both cohorts (P=0.018 and 0.031, respectively). This is the first systematic study to document infarction of acutely silent tissue after anterior circulation stroke, and to show that it affects a sizeable fraction of patients and has the predicted negative impact on clinical course. These findings were replicated in two independent cohorts, regardless of the perfusion imaging modality used. Preventing infarction of the tissue not initially at risk should have direct clinical benefit.

KW - Humans

KW - Male

KW - Aged

KW - Female

KW - Middle Aged

KW - Severity of Illness Index

KW - Cohort Studies

KW - Predictive Value of Tests

KW - Time Factors

KW - Magnetic Resonance Imaging

KW - Image Processing, Computer-Assisted

KW - Analysis of Variance

KW - Models, Statistical

KW - Diffusion Magnetic Resonance Imaging

KW - Brain Infarction/etiology/pathology/radionuclide imaging

KW - Brain Mapping

KW - Perfusion Imaging

KW - Stroke/complications

KW - Humans

KW - Male

KW - Aged

KW - Female

KW - Middle Aged

KW - Severity of Illness Index

KW - Cohort Studies

KW - Predictive Value of Tests

KW - Time Factors

KW - Magnetic Resonance Imaging

KW - Image Processing, Computer-Assisted

KW - Analysis of Variance

KW - Models, Statistical

KW - Diffusion Magnetic Resonance Imaging

KW - Brain Infarction/etiology/pathology/radionuclide imaging

KW - Brain Mapping

KW - Perfusion Imaging

KW - Stroke/complications

U2 - 10.1093/brain/awr100

DO - 10.1093/brain/awr100

M3 - SCORING: Journal article

C2 - 21616971

VL - 134

SP - 1765

EP - 1776

JO - BRAIN

JF - BRAIN

SN - 0006-8950

IS - Pt 6

M1 - Pt 6

ER -