Brain Swelling and Loss of Gray and White Matter Differentiation in Human Postmortem Cases by Computed Tomography

Abstract

The purpose of this study was to evaluate the brain by postmortem computed tomography (PMCT) versus antemortem computed tomography (AMCT) using brains from the same patients. We studied 36 nontraumatic subjects who underwent AMCT, PMCT, and pathological autopsy in our hospital between April 2009 and December 2013. PMCT was performed within 20 h after death, followed by pathological autopsy including the brain. Autopsy confirmed the absence of intracranial disorders that might be related to the cause of death or might affect measurements in our study. Width of the third ventricle, width of the central sulcus, and attenuation in gray matter (GM) and white matter (WM) from the same area of the basal ganglia, centrum semiovale, and high convexity were statistically compared between AMCT and PMCT. Both the width of the third ventricle and the central sulcus were significantly shorter in PMCT than in AMCT (P < 0.0001). GM attenuation increased after death at the level of the centrum semiovale and high convexity, but the differences were not statistically significant considering the differences in attenuation among the different computed tomography scanners. WM attenuation significantly increased after death at all levels (P<0.0001). The differences were larger than the differences in scanners. GM/WM ratio of attenuation was significantly lower by PMCT than by AMCT at all levels (P<0.0001). PMCT showed an increase in WM attenuation, loss of GM–WM differentiation, and brain swelling, evidenced by a decrease in the size of ventricles and sulci.

Can postmortem computed tomography detect antemortem hypoxic–ischemic encephalopathy?

“…Brain swelling [9,10] and the loss of GM-WM differentiation [8] are often observed on PMCT. However, brain swelling and other common PMCT findings also resemble the findings for HIE.…”

“…Previous studies have revealed that brain PMCT shows an increase in white matter (WM) attenuation, a decrease in gray matter (GM) attenuation, the loss of GM-WM differentiation [8,9], and diffuse brain swelling represented by a decrease in the size of the ventricles and sulci [9,10]. In most cases, these changes are nonspecific and not pathognomonic.…”

Evaluation of post-mortem lateral cerebral ventricle changes using sequential scans during post-mortem computed tomography

“…For practical applications of PMCT, it is important to differentiate disease-oriented pathological change from post-mortem change. However, systematic studies evaluating post-mortem change in PMCT are rarely focusing on longitudinal post-mortem observations[ 21 – 25 ], particularly for post-mortem brain alterations [ 20 , 26 ]. To address this issue, the current study included an assessment of post-mortem change in PMCT, which will be referred to as a “sequential scan project.” This sequential scan paradigm includes successive, periodical CT scans of the same post-mortem subject.…”

“…Using a sequential scan procedure, it is possible to observe the unfolding of post-mortem change inside the whole body, without a destructive inspection. Post-mortem brain alterations including CSF volume changes have been described [ 3 , 20 , 26 , 28 ], but there is still a need for clarification of the extent of these phenomena in comparison to antemortem brain edema as some authors denied brain swelling in the very early PMI [ 20 ]. The LCV volumes have recently been described as weak indicators of ante mortem cerebral edema [ 28 ] but their post mortem alteration had not been exactly deliminated in that retrospective study.…”

“…Another important issue is the Hounsfield units (HU) attenuation change which has been described for gray and white matter of the brain [ 16 , 26 ] but not for the CSF in the lateral ventricles and particularly for its implications regarding PMI estimations. For the current study, the lateral cerebral ventricles (LCVs) were evaluated using this sequential scan method, because they offer the most convenient measurement of HU radiodensity in comparison to other CSF spaces as unwanted inclusion of boundaries with the risk of partial volume artifacts can be minimized.…”