Symmetry in computed tomography of the brain: the pitfalls

Symmetry of computerised tomography of the brain in traumatic brain injury: a quality improvement audit

BACKGROUND

Non-contrast Computerised Tomography (NCCT) of brain is the gold standard investigation for diagnosis and management of Traumatic brain injury (TBI). Asymmetrical CT brain images as a result of improper head positioning in the CT gantry will compromise the diagnostic value. Therefore, this audit aimed to assess the degree of asymmetry in CT brain studies carried out in TBI patients.

METHODS

This audit was carried out at a level one trauma centre and included CT scans of TBI patients with a Glasgow come scale (GCS) score ≤ 13, admitted to the Neurological intensive care unit (NICU). The first cycle involved a period of three months. The data collected included demographic data and variables such as GCS at the time of the scan and whether the patient was intubated or not. The visualisation of bilateral internal auditory meatuses was used as landmark to determine scan symmetry. If the internal auditory meatus on both sides were visible on the same slice of CT scan, it was considered symmetric. The degree of asymmetry was gauged based on the axial slice difference between bilateral meatuses. The data collected was tabulated and presented to Neurosurgery residents and a checklist was formulated which had to be followed while positioning the patient on CT table prior to imaging.

RESULTS

The first cycle of the audit showed that 83.8% of scans were asymmetric and among them 44.1% revealed gross asymmetry affecting interpretation of the scan. Following, implementation of the checklist the percentage of gross asymmetry dropped to 21.86% in the second and to 22.22% in the third audit.

CONCLUSION

The use of checklist prior to CT brain studies showed sustainable improvement in reducing gross asymmetry and in acquisition of symmetrical CT brain images.

The pseudo-SAH sign: an imaging pitfall in postmortem computed tomography

Postmortem computed tomography (PMCT) of the brain has an important role in detection of subarachnoid hemorrhage (SAH), which has a high mortality rate. However, a phenomenon known as "pseudo-SAH," or high-attenuation areas along the cisterns mimicking SAH, may be seen on CT. The aim of this study was to evaluate the diagnostic accuracy of brain PMCT for SAH and to identify the characteristics of pseudo-SAH. Findings on PMCT (sulcal effacement, asymmetry, maximum thickness of SAH signs, presence of acute/subacute intraventricular/intraparenchymal hemorrhage) and clinical history (left ventricular assist device [LVAD] implantation, anticoagulation therapy/coagulation disorder, global ischemia) were compared between subjects with true SAH and those with pseudo-SAH. Twenty eight of 128 enrolled subjects had positive signs of SAH on PMCT, 20 (71.4%) had SAH on autopsy, and 8 (28.6%) did not. The sensitivity, specificity, positive predictive value, and negative predictive value of SAH signs seen on PMCT were 95.2, 94.6, 71.4, and 99.3%, respectively. Asymmetry of SAH signs and acute/subacute intraventricular and intraparenchymal hemorrhage were significantly more common in true SAH cases than in pseudo-SAH cases. The maximum thickness of SAH signs was significantly greater in true SAH cases. A history of LVAD implantation, anticoagulation therapy, and/or a coagulation disorder were more common in true SAH cases but not significantly so. A history of global ischemia was significantly more common in pseudo-SAH cases. If signs of SAH are observed on PMCT, it is important to look for other signs on PMCT and carefully review the clinical history to avoid a diagnostic error.

Head computed tomography interpretation in trauma: a primer

Noncontrast computed tomography (CT) provides important diagnostic information for patients with traumatic brain injury. A systematic approach to image interpretation optimizes detection of pathologic air, fractures, hemorrhagic lesions, brain parenchymal injury, and abnormal cerebrospinal fluid spaces. Bone and brain windows should be reviewed to enhance injury detection. Findings of midline shift and mass effect should be noted as well as findings of increased intracranial pressure such as hydrocephalus and cerebral edema, because these may immediately influence management. Compared with CT, magnetic resonance imaging may provide more sensitive detection of diffuse axonal injury but has no proven improvement in clinical outcomes. This article discusses key CT interpretation skills and reviews important traumatic brain injuries that can be discerned on head CT. It focuses on imaging findings that may deserve immediate surgical intervention. In addition, the article reviews the limits of noncontrast CT and discusses some advanced imaging modalities that may reveal subtle injury patterns not seen with CT scan.