Clin Neuroradiol DOI 10.1007/s00062-015-0409-8
O r i g i n a l A rt i c l e
Etiology of Corpus Callosum Lesions with Restricted Diffusion C.A. Wilson · M.T. Mullen · B.P. Jackson · K. Ishida · S.R. Messé
Received: 30 November 2014 / Accepted: 6 May 2015 © Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose Infarction of the corpus callosum is rare, and other conditions can cause magnetic resonance imaging (MRI) restricted diffusion in the callosum, leading to diagnostic uncertainty. We sought to characterize the etiology of lesions with diffusion restriction in the corpus callosum. Methods Callosal lesions with restricted diffusion were identified at our institution between January 2000 and December 2010. Radiographic and clinical data were reviewed to determine whether the lesion was vascular and if so, to identify the underlying mechanism. Results A total of 174 cases were reviewed in depth; 47 % were vascular and 53 % were nonvascular. Among vascular cases, atypical mechanisms of stroke (e.g., vasculitis/vasculopathy, hypercoagulable state) were most common (37 %), followed by cardioembolism (28 %). Vascular splenial lesions in particular were likely due to atypical causes of stroke. The most common nonvascular etiologies were trauma (44 %), tumor (22 %), and demyelination (15 %). Vascular lesions were more common in older, non-Caucasian pa-
tients with vascular risk factors. Nonvascular lesions were more likely to be found in association with T2-hyperintense cortical lesions, focal intraparenchymal enhancement, or edema/mass effect on MRI. Conclusions More than half of lesions with diffusion restriction in the corpus callosum are due to a nonvascular cause. Clinical and radiographic characteristics can help distinguish vascular from nonvascular lesions in the corpus callosum. Nonvascular lesions are more likely to be seen in younger patients without vascular risk factors and are more often accompanied by enhancement and edema. Vascular lesions are most commonly due to atypical stroke etiologies, and these patients may require additional diagnostic testing. Keywords Corpus callosum · Stroke · Restricted diffusion · MRI Abbreviations CC Corpus callosum Introduction
C.A. Wilson, MD, PhD () Department of Neurology, University of Florida, 32610 100236, Gainesville, FL, USA e-mail: [email protected] B.P. Jackson, MD, MHA Department of Pathology, Thomas Jefferson University, Philadelphia, PA, USA K. Ishida, MD Department of Neurology, New York University, New York, NY, USA M.T. Mullen, MD · S.R. Messé, MD Department of Neurology, University of Pennsylvania, Philadelphia, PA, USA
Magnetic resonance imaging (MRI) restricted diffusion in the corpus callosum (CC), the white matter tract connecting the two cerebral hemispheres, may lead to diagnostic uncertainty. Although diffusion restriction often reflects a vascular etiology, it can occur with many other disease processes, [1] and the CC may be particularly predisposed to restricted diffusion from causes other than acute infarction [2]. Further, CC infarcts often have accompanying radiographic features that may cause confusion as to the primary disease process. For example, extension across the midline may raise concern for malignancy [3, 4]. Indeed, in one case series, three of five patients with CC infarction underwent a
13
2
neurosurgical biopsy before the correct diagnosis was made [3]. Infarction of the CC is a relatively rare occurrence, as only 3–8 % of infarcts involve the CC [5, 6]. Based on the current literature, there is uncertainty surrounding the characteristics of CC infarcts. Some studies have suggested that the genu and body of the callosum are at greatest risk, [3, 5, 7–9] and others suggested the splenium [6]. The most likely vascular etiology has been variably reported as small-vessel disease, large-vessel atherosclerosis, and cardioembolism [5, 6, 9]. Prior studies are limited by their small size, generally fewer than 10 cases, and only a subset of these studies applied a mechanism-based classification system [5, 9]. In this study, we sought to describe the clinical and radiographic characteristics of restricted diffusion in the CC, to determine the etiology of the lesion, and, in vascular cases, to elucidate the mechanism. We hypothesized that restricted diffusion in the callosum is most likely due to atypical vascular causes, or to nonvascular disease process. Methods This study was a retrospective consecutive cohort of patients with restricted diffusion in the CC. Approval for the study was obtained from the Institutional Review Board at the University of Pennsylvania. The combined inpatient and outpatient radiology database of the Hospital of the University of Pennsylvania was searched using the search terms “diffusion”, “callosum”, and “restricted/restriction” to identify all patients with restricted diffusion of the CC noted on MRI performed between January 2000 and December 2010. We excluded cases with history of neurosurgical procedure or recent endovascular intervention to eliminate cases that may have been precipitated by instrumentation. All MRI scans were performed on one of the multiple 1.5-T scanners in clinical use at the University of Pennsylvania. At minimum, the sequences obtained included T1-, T2-, fluid-attenuated inversion recovery (FLAIR), diffusion-weighted imaging (DWI), and apparent diffusion coefficient. DWI was performed in axial planes as per standard protocol at our institution. Additional sequences, including gadolinium-enhanced T1 and gradient echo (GRE), were obtained if ordered by the clinician caring for the patient. The study was formally read by a neuroradiologist at the time the study was performed, as per standard clinical practice. All MR images and the corresponding radiology reports were reviewed to determine the distribution of restricted diffusion and the presence of other cross-sectional imaging characteristics including cortical or subcortical T2/FLAIR lesions, intraparenchymal hemorrhage, microhemorrhage on GRE sequence, edema/mass effect, and gadolinium
13
C.A. Wilson et al.
enhancement. In addition, it was noted whether diffusion restriction was isolated to the CC or also present in additional areas of the brain and the region of CC involvement (rostrum/genu, body, or splenium). Clinical information was obtained by reviewing medical records for each case. Specific information gathered included demographic data (age, gender, and race), vascular risk factors (e.g., history of hypertension, hypercholesterolemia, diabetes, tobacco use, coronary artery disease, atrial fibrillation, systolic heart failure, or prior infarct), neurological examination findings, echocardiogram results, vascular study results (MR angiography, computed tomography angiography, ultrasound studies, or conventional angiography), and relevant laboratory values (e.g., sedimentation rate, syphilis testing, lipid panel, cerebrospinal fluid, inflammatory or infectious markers). Using the information gathered, the etiology of the CC restricted diffusion was independently determined by two neurologists (C.A. Wilson and M.T. Mullen). Cases deemed to be vascular were further categorized using a standard and widely implemented causative classification system of ischemic stroke (Stop Stroke Study Trial of Org 10172 in Acute Stroke Treatment [SSS-TOAST]) [10]. Vascular cases were categorized into one of the six following categories: large-artery atherosclerosis, cardioembolism, small-vessel occlusion, other determined etiology (e.g., vasculopathy, vasculitis, a hypercoagulable state, or hematologic disorder), cryptogenic (due to insufficient information or undetermined cause), and multiple determined etiologies. Any disagreement between the two investigators was adjudicated by a third independent neurologist (S.R. Messé). Descriptive statistics were performed for demographic data, cross-sectional imaging characteristics of the lesion, vascular risk factors, etiology of the lesion, and infarct mechanism. Univariate analysis of lesion etiology by demographic data, MRI characteristics, and vascular risk factors was performed by Pearson chi-square or Fisher’s exact test, as appropriate. Significance was established at p
O r i g i n a l A rt i c l e
Etiology of Corpus Callosum Lesions with Restricted Diffusion C.A. Wilson · M.T. Mullen · B.P. Jackson · K. Ishida · S.R. Messé
Received: 30 November 2014 / Accepted: 6 May 2015 © Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose Infarction of the corpus callosum is rare, and other conditions can cause magnetic resonance imaging (MRI) restricted diffusion in the callosum, leading to diagnostic uncertainty. We sought to characterize the etiology of lesions with diffusion restriction in the corpus callosum. Methods Callosal lesions with restricted diffusion were identified at our institution between January 2000 and December 2010. Radiographic and clinical data were reviewed to determine whether the lesion was vascular and if so, to identify the underlying mechanism. Results A total of 174 cases were reviewed in depth; 47 % were vascular and 53 % were nonvascular. Among vascular cases, atypical mechanisms of stroke (e.g., vasculitis/vasculopathy, hypercoagulable state) were most common (37 %), followed by cardioembolism (28 %). Vascular splenial lesions in particular were likely due to atypical causes of stroke. The most common nonvascular etiologies were trauma (44 %), tumor (22 %), and demyelination (15 %). Vascular lesions were more common in older, non-Caucasian pa-
tients with vascular risk factors. Nonvascular lesions were more likely to be found in association with T2-hyperintense cortical lesions, focal intraparenchymal enhancement, or edema/mass effect on MRI. Conclusions More than half of lesions with diffusion restriction in the corpus callosum are due to a nonvascular cause. Clinical and radiographic characteristics can help distinguish vascular from nonvascular lesions in the corpus callosum. Nonvascular lesions are more likely to be seen in younger patients without vascular risk factors and are more often accompanied by enhancement and edema. Vascular lesions are most commonly due to atypical stroke etiologies, and these patients may require additional diagnostic testing. Keywords Corpus callosum · Stroke · Restricted diffusion · MRI Abbreviations CC Corpus callosum Introduction
C.A. Wilson, MD, PhD () Department of Neurology, University of Florida, 32610 100236, Gainesville, FL, USA e-mail: [email protected] B.P. Jackson, MD, MHA Department of Pathology, Thomas Jefferson University, Philadelphia, PA, USA K. Ishida, MD Department of Neurology, New York University, New York, NY, USA M.T. Mullen, MD · S.R. Messé, MD Department of Neurology, University of Pennsylvania, Philadelphia, PA, USA
Magnetic resonance imaging (MRI) restricted diffusion in the corpus callosum (CC), the white matter tract connecting the two cerebral hemispheres, may lead to diagnostic uncertainty. Although diffusion restriction often reflects a vascular etiology, it can occur with many other disease processes, [1] and the CC may be particularly predisposed to restricted diffusion from causes other than acute infarction [2]. Further, CC infarcts often have accompanying radiographic features that may cause confusion as to the primary disease process. For example, extension across the midline may raise concern for malignancy [3, 4]. Indeed, in one case series, three of five patients with CC infarction underwent a
13
2
neurosurgical biopsy before the correct diagnosis was made [3]. Infarction of the CC is a relatively rare occurrence, as only 3–8 % of infarcts involve the CC [5, 6]. Based on the current literature, there is uncertainty surrounding the characteristics of CC infarcts. Some studies have suggested that the genu and body of the callosum are at greatest risk, [3, 5, 7–9] and others suggested the splenium [6]. The most likely vascular etiology has been variably reported as small-vessel disease, large-vessel atherosclerosis, and cardioembolism [5, 6, 9]. Prior studies are limited by their small size, generally fewer than 10 cases, and only a subset of these studies applied a mechanism-based classification system [5, 9]. In this study, we sought to describe the clinical and radiographic characteristics of restricted diffusion in the CC, to determine the etiology of the lesion, and, in vascular cases, to elucidate the mechanism. We hypothesized that restricted diffusion in the callosum is most likely due to atypical vascular causes, or to nonvascular disease process. Methods This study was a retrospective consecutive cohort of patients with restricted diffusion in the CC. Approval for the study was obtained from the Institutional Review Board at the University of Pennsylvania. The combined inpatient and outpatient radiology database of the Hospital of the University of Pennsylvania was searched using the search terms “diffusion”, “callosum”, and “restricted/restriction” to identify all patients with restricted diffusion of the CC noted on MRI performed between January 2000 and December 2010. We excluded cases with history of neurosurgical procedure or recent endovascular intervention to eliminate cases that may have been precipitated by instrumentation. All MRI scans were performed on one of the multiple 1.5-T scanners in clinical use at the University of Pennsylvania. At minimum, the sequences obtained included T1-, T2-, fluid-attenuated inversion recovery (FLAIR), diffusion-weighted imaging (DWI), and apparent diffusion coefficient. DWI was performed in axial planes as per standard protocol at our institution. Additional sequences, including gadolinium-enhanced T1 and gradient echo (GRE), were obtained if ordered by the clinician caring for the patient. The study was formally read by a neuroradiologist at the time the study was performed, as per standard clinical practice. All MR images and the corresponding radiology reports were reviewed to determine the distribution of restricted diffusion and the presence of other cross-sectional imaging characteristics including cortical or subcortical T2/FLAIR lesions, intraparenchymal hemorrhage, microhemorrhage on GRE sequence, edema/mass effect, and gadolinium
13
C.A. Wilson et al.
enhancement. In addition, it was noted whether diffusion restriction was isolated to the CC or also present in additional areas of the brain and the region of CC involvement (rostrum/genu, body, or splenium). Clinical information was obtained by reviewing medical records for each case. Specific information gathered included demographic data (age, gender, and race), vascular risk factors (e.g., history of hypertension, hypercholesterolemia, diabetes, tobacco use, coronary artery disease, atrial fibrillation, systolic heart failure, or prior infarct), neurological examination findings, echocardiogram results, vascular study results (MR angiography, computed tomography angiography, ultrasound studies, or conventional angiography), and relevant laboratory values (e.g., sedimentation rate, syphilis testing, lipid panel, cerebrospinal fluid, inflammatory or infectious markers). Using the information gathered, the etiology of the CC restricted diffusion was independently determined by two neurologists (C.A. Wilson and M.T. Mullen). Cases deemed to be vascular were further categorized using a standard and widely implemented causative classification system of ischemic stroke (Stop Stroke Study Trial of Org 10172 in Acute Stroke Treatment [SSS-TOAST]) [10]. Vascular cases were categorized into one of the six following categories: large-artery atherosclerosis, cardioembolism, small-vessel occlusion, other determined etiology (e.g., vasculopathy, vasculitis, a hypercoagulable state, or hematologic disorder), cryptogenic (due to insufficient information or undetermined cause), and multiple determined etiologies. Any disagreement between the two investigators was adjudicated by a third independent neurologist (S.R. Messé). Descriptive statistics were performed for demographic data, cross-sectional imaging characteristics of the lesion, vascular risk factors, etiology of the lesion, and infarct mechanism. Univariate analysis of lesion etiology by demographic data, MRI characteristics, and vascular risk factors was performed by Pearson chi-square or Fisher’s exact test, as appropriate. Significance was established at p
