How To Read A Head CT
A noncontrast CT of the head has become the dominant modality for an initial assessment of intracranial pathology, especially in the setting of trauma. Many approaches to reading a head CT exist, with the method described below simply being one example. As they say, there are many ways to skin a cat. Of course, the real question in my mind is: why is anyone skinning a cat in the first place? Anyway, um, back to head CTs: no matter how you read a study, it is important that you:
- Cover all the relevant areas of the study
- Go through your system in a consistent manner
Sounds simple enough, but it is very very important to set up a good system and then go through it each time. On some studies, a major finding may jump out at you and distract from other areas, which may contain even more significant findings. This phenomenon has been termed search satisfaction, which is where you note a finding and therefore stop searching for more. One way to avoid this is to try to make at least two findings on every study, no matter how trivial, just to force yourself to cover the entire study.
Hopefully you are not satisfied with this post... yet. Getting back to head CTs, there are some unique features of the brain that assist in one's reading of this type of study. The main feature is symmetry: the two hemispheres of the cerebrum and cerebellum should generally look the same. Unlike the abdomen or even the chest, symmetry in the head allows for easy comparison across the midline whenever a possible abnormality is questioned. Another helpful feature is that the cranium is a confined space, so if there are masses (or abnormalities with mass effect), neighboring structures will typically be compressed rather than merely pushed out of the way. Practically this means that even small lesions can have a significant impact on the appearance of the brain.
A Systematic Approach
Keeping those general principles in mind, here is one general 10 step approach to assessing a head CT.
References:
Hopefully you are not satisfied with this post... yet. Getting back to head CTs, there are some unique features of the brain that assist in one's reading of this type of study. The main feature is symmetry: the two hemispheres of the cerebrum and cerebellum should generally look the same. Unlike the abdomen or even the chest, symmetry in the head allows for easy comparison across the midline whenever a possible abnormality is questioned. Another helpful feature is that the cranium is a confined space, so if there are masses (or abnormalities with mass effect), neighboring structures will typically be compressed rather than merely pushed out of the way. Practically this means that even small lesions can have a significant impact on the appearance of the brain.
A Systematic Approach
Keeping those general principles in mind, here is one general 10 step approach to assessing a head CT.
- Posterior fossa
Check to see if the cerebellum is symmetric. If the cerebellar fissures are prominent, this may represent volume loss. If the tonsils are low-lying (visible below the foramen magnum), this may represent a normal variation or an Arnold-Chiari malformation. - Ventricles
Ensure that the four ventricles are in their expected positions and are not enlarged. The lateral ventricles should be symmetric. - Midline structures
The midline structures should be... midline. Specifically, the septum pellucidum should be exactly in between the lateral ventricles. To check deviation, draw a line from the apex of the inner table of the calvarium anterior to posterior. At the level of the globus pallidus, check for deviation of the septum from the line. - Extra-axial spaces
Look for any collections between the inner table of the calvarium and the brain parenchyma. Blood, CSF fluid, air, and pus can potentially collect here. For bleeds, biconvex collections represent epidural bleeds; lenticular bleeds represent subdural bleeds and can cross suture lines. Subarachnoid bleeds will line the sulci. The base of the skull is a good place to look for bleeds after trauma due to the sharp edges there. - Brain parenchyma
Check for symmetry, gray-white differentiation, areas of hyper- or hypo-attenuation, and any architectural distortion. - Orbits
Check for symmetry, propotosis, lens position and thickness, the optic nerve, artery, and vein. Use soft tissue windows. Make sure no masses are seen behind the globe. This is also a good time to check the lamina papyracea to rule out any fractures. - Paranasal Sinuses
Look for mucosal thickening or polyp formation. In trauma, fluid and blood can collect in the sinuses. - Mastoid Air Cells
Check to make sure that fluid is not accumulating in the air cells, especially in the setting of trauma. - Bones
Fractures can be hard to spot in the skull. The key to identifying an acute fracture is to look for overlying soft tissue swelling, lack of sclerotic margins, and lack of symmetry. Vascular channels are plentiful in the skull and easily confused for a fracture. However, they will have sclerotic margins. Sutures should be symmetric. - Soft Tissues
Look for soft tissue swelling or masses. Outside of the setting of trauma, the soft tissues should generally be unremarkable.
While this is not a fully comprehensive approach to reading a head CT, it provides a framework for assessing the critical structures. By being systematic and thorough each time, it is highly unlikely that you will miss a significant finding. Review your neuroanatomy thoroughly to help you localize a lesion once you do spot an abnormality. For pediatric patients, the head CT may be preceded by a neonatal head ultrasound, which we will cover in a future post.
Normal Head CT Source: Wikipedia |
References:
- The Brant and Helms Solution: Fundamentals of Diagnostic Radiology, Third Edition
- How To Read A Head CT - Henry Z. Wang, University of Rochester
- Introduction to Head CT - UVA
Post a Comment for "How To Read A Head CT"