On behalf of the RSNA Educational Committee, our track chair, Dr. Diane Strollo, and our moderator, Dr. Vikas Agarwal, we welcome you to the Essentials of Neuroimaging course MSES 34. We have four excellent speakers and topics. I'd like to thank the program committee for inviting me to speak. I'm Dr. Jim Chen, and I'm speaking on spinal infections, toxic and metabolic abnormalities. I have no relevant disclosures. Our goals for today are to discuss the imaging, clinical, and laboratory features of spinal infections and some of their mimics. We're going to do this by going from outside the spinal canal and work our way into the spinal canal. On our first patient, what we see on these T2-weighted, fluid-sensitive sequences is edema or bright T2 signal, both within the disc as well as two adjacent vertebral bodies with some material in the ventral epidural space and paraspinal soft tissues. Here on T1-weighted images, we see dark T1 signal. All of this enhances, and we have enhancement in the epidural space without a rim of enhancement. Our axials show the same thing with enhancement of the disc and involvement of the paraspinal soft tissues going into the ventral epidural space without rim enhancement. This case is a classic case of pyogenic discitis osteomyelitis with epidural phlegmon. Our key findings, as we've seen, are fluid signal within the involved disc and adjacent signal abnormality and destruction of the adjacent endplates and vertebral bodies. You may have paraspinal involvement that may develop abscesses, and extension into the epidural space may result in compression of the neural elements, which we need to report. The patient's sed rate and C-reactive protein are usually markedly elevated. In patients who have spinal infection, you may have worsening of that infection in the epidural space to develop rim-enhancing collections, such as in this patient where we see a larger enhancing collection posteriorly, as well as another collection ventrally that is compressing the cauda equina. We also notice that there are paraspinal soft tissue abscesses here as well. This is a patient with discitis and osteomyelitis with epidural and paraspinal abscesses. Primary epidural abscesses tend to be posteriorly, whereas direct extension from discitis osteomyelitis does tend to be more ventral. One thing that can be confusing can be degenerative endplate changes, MODIC type 1, which is oftentimes difficult to distinguish from infection, such as in this patient who has abnormal signal in two adjacent vertebral bodies along the endplates with loss of disc height here. But unlike our infected patient, we don't have T2 bright signal within the disc, even though we do have enhancement of those areas of abnormal signal. Another differentiator is we see a lack of paraspinal soft tissue involvement. If we are fortunate, sometimes you may see disc space gas. Usually that's from vacuum phenomenon and usually not infected, although there are rare times where you may have a gas-forming organism. In these patients, typically, who have MODIC type 1 degenerative changes, have a normal or only slightly elevated sed right or C-reactive protein. Sometimes biopsy is necessary to make differentiation between those two entities, though. If you are fortunate enough to have a patient on the table, and you're trying to figure this out, you can try adding diffusion-weighted images, which can be helpful. In this particular case, Dr. Tannenbaum and his group described the diffusion claw sign, where you can see in between the area of signal abnormality here at the endplates and between the normal marrow is the presence of this bright diffusion signal, or the diffusion claw. When you see this, the patient is usually infection-free the vast majority of the time, whereas the absence of the claw sign, the patient is usually infected. Even if there's just a probable claw sign, most are infection-free. Our next patient shows destruction of the fets that's over here, where we see erosions, and on MRI, we see fluid signal and edema within the facets extending into the neural foramen associated with enhancement and extension into the soft tissues as well. Our axial images show that there is some involvement of the epidural space, both ventrally, laterally, and posteriorly. Our diffusion-weighted images show abnormal diffusion in those areas of signal abnormality. This is a patient who has a septic facet joint with epidural involvement. Typically these patients are quite febrile and usually have unilateral pain. As on this patient, we typically see erosions as well as inflammation centered on the facet and involvement of the adjacent soft tissues. Our next patient is a man with cough and several weeks of back pain. What we see here are two adjacent vertebral bodies with abnormal enhancement and fluid signal, but there's relative sparing of the disc. Ventrally, what we see is subligamentous spread of material between the two vertebral bodies. On our axial images, what we see is involvement of the paraspinal soft tissues. This is what we typically see in a patient with tuberculous spine infection. Unlike bacterial organisms, mycobacteria lack elastase, so therefore disc destruction occurs late rather than early and is one of the ways that you can differentiate between tuberculous infection and pyogenic infections. Although pyogenic infections are so common relative to tuberculous infections that you may see sparing of the disc space more commonly in pyogenic infections rather than tuberculous. This is what the patient's chest radiograph looks like. This is a paraplegic patient with an incidentally discovered abnormality on routine chest radiography where we see this abnormal looking vertebral body, which appears partially destroyed with some calcifications. On MRI, what we see is this fluid signal cleft on both T2 and T1 weighted images, as well as this T2 dark material in the paraspinal space that enhances quite intensely, as do many of the adjacent vertebral bodies and some of this epidural and paraspinal spaces as well. Paraxial images show a very similar finding with fluid signal in the disc, enhancement around the disc, and some of the soft tissues with T2 dark material. On CT, what we see is calcification that looks quite disorganized throughout this as well as sclerosis. This is a patient who has neuropathic or charcot spine. This is usually incidentally discovered and the patient usually has a lack of symptoms for the degree of the imaging abnormality and is characterized by disorganization as well as calcification. The sed rate and C-reactive protein can be slightly elevated but are usually normal. And unlike infection, what we see here is extensive emolument, both of the vertebral body as well as both of the posterior elements. For infection, you would typically have to have an incredibly severe infection to have this extent of abnormality. This is a chronic dialysis patient with an incidentally discovered abnormality. Again, here we see abnormality of the disc as well as adjacent vertebral body erosion and sclerosis. We see a fluid cleft in this vertebral body and disc space that extends toward the facet. But notice the relative lack of enhancement and the relative T2 darkness of the adjacent end plates. Our axial images show the same thing, although it looks as though there may be some paraspinal involvement too and some fluid signal extending into this cleft. This is a patient with dialysis-related or amyloid spondyloarthropathy. This typically occurs in patients who have long-term dialysis, although can be seen as early as two to four years, long-term being about 15 years or so. The risk of this varies depending on the type of dialysis, the type of membrane that's being used. High-flux membranes that remove more of the beta-2 microglobulins resulting in amyloid as well as biocompatible membranes, use of those membranes will actually decrease the risk relative to low-flux or bio-incompatible membranes. Again, similar to our Charcot spine, the imaging is out of proportion to the symptoms. Our next patient is a patient who has areas of discoverable sclerosis at multiple levels, but involvement of the sternum as well. On our contrast-enhanced images, what we see is that there's a fair amount of contrast enhancement throughout the vertebral body that is far greater than the edema or signal abnormality within the corners on just our STIR images. This is fairly typical of SAFO, where we see this particular collection of findings. SAFO is a clinical syndrome in which it is characterized by synovitis, acne, palmoplantar pustulosis, hyperostosis, and osteitis. The most common sites are going to be the sternoclastoclavicular joint, and therefore it's prudent to look at the medial sternoclavicular joint for erosions or sclerosis when you're considering this diagnosis. Unfortunately, not all patients present with the skin acne. This next patient is a man with neck pain for two months. That is not due to this old DENS fracture. What we do see is edema in the prevertebral soft tissues, as well as some T2 dark signal just underneath the anterior arch of C1, which we also see on these axial images. On radiography, we see a chunky calcification within the insertion of the longus coli tendon, and this is a case of calcific tendinosis of the longus coli. The key here is that we have prevertebral fluid, but no rim enhancement, differentiating that from a prevertebral abscess. Patients don't usually have fever or elevated white count, and respond well to conservative therapy or NSAIDs. This is a patient who has E. coli bacteremia and back pain. On T1-weighted images, what we see is that there's a smudgy appearance of the CSF, where it's very difficult to distinguish between the conus and cauda equina and the CSF. On our T2 and STIR images, the fluid signal that we typically see for CSF is actually fairly dark compared to the fluid in this disc space. And on post-contrast images, we see intense enhancement of all of the CSF surrounding the conus, outlining the conus, making the conus look relatively dark. On axial images, we see the same findings, enhancement and smudginess of our CSF. This is a patient with spinal meningitis with E. coli pus within the CSF. And this is characterized by smudginess of that CSF signal as well as enhancement. From meningitis, if things progress into the cord, what we can see from this case from the literature is rim enhancement with a T2 dark rim, as well as central diffusion restriction. Just like in the brain, this is a cord abscess, which is characterized by rim enhancement with central diffusion restriction with the collection of pus. Something similar appearing is this patient who has a rim enhancing collection or rim enhancing fluid signal lesion within the conus. But when you magnify this, you can see a T2 dark dot centrally within it that faintly enhances. What we see in this particular case then is, again, on the axials, rim enhancement, T2 dark dot centrally. This is the vesicular form of neurosissicosis within the conus. Unlike our abscess, there is no central diffusion restriction, and we have that central scolex. Something similar appearing is this fluid signal lesion within the conus, but unlike our last case, there is no enhancement. In this particular case, this is characteristic of a ventriculous terminalis, also known as the fifth ventricle. Typically it's an incidental discovery, and oftentimes in the central location of the distal spinal cord along where you expect this central canal of the cord to be. Oftentimes it's just focal dilatation, and some people consider it a fourth ventricle. The key here is that there is no nodule. There's no septation or enhancement within this. In conclusion, we've talked about a bunch of spinal infections as well as some of their mimics and some of their key imaging findings. Using some of those clinical features can be helpful to distinguish these. Sometimes biopsy is necessary, so thank you for your attention. I am delighted to introduce our next speaker, Dr. Sapna Rawal. Thank you very much to Dr. Estraldo and Dr. Agrawal for the opportunity to present a brief approach to the imaging of cognitive impairment. Cognitive impairment can be separated into rapidly progressive dementia, which evolves over weeks to months, or chronic dementia, which evolves over months to years. For rapidly progressive dementia, begin by excluding major structural disease such as tumor, stroke, or infection, and then considering diseases that affect key areas like the gray matter, the limbic system, as well as classic patterns of metabolic disease or venous congestion. For chronic dementia, we'll begin by discussing regional atrophy patterns, and then focusing on white matter disease, and finally concluding with a few comments on normal pressure hydrocephalus. So, for rapidly progressive dementia syndromes, the first important gray matter pathology to keep in mind is Kurzweil-Jakob disease, here manifesting in the typical imaging pattern for sporadic form with asymmetric basal ganglia involvement that shows up best on diffusion-weighted imaging and is more subtle on FLIR imaging, as well as with cortical involvement that is, again, subtle on FLIR imaging but more obvious on diffusion-weighted imaging. This manifests with cognitive decline that is rapidly progressive, as well as with myoclonus, ataxia, and pyramidal, as well as extrapyramidal features, and is generally rapidly fatal. Hyperamynemic encephalopathy also preferentially affects the gray matter, and this occurs in patients with fulminant liver failure, characteristically affecting the insula and the cingulate gyrus preferentially, but can also extend to the remainder of the cortex, although relative sparing of the parirulandic and the occipital regions has been described. With regard to the limbic system, we can consider perineoplastic autoimmune encephalitis, shown here with asymmetric involvement of the mesial-temporal regions seen best on FLIR imaging, without significant restricted diffusion or enhancement in a patient with small-cell lung cancer, or alternatively non-perineoplastic causes, such as this patient with Hashimoto's encephalopathy, who developed limbic encephalitis, seen here with bilateral but asymmetric involvement of the mesial-temporal regions, in this case with some associated enhancement, in the context of thyroid dysfunction, with an ultrasound showing an enlarged, multinodular thyroid gland. A few important metabolic diseases that can present with rapid cognitive decline include Wernicke's, with the classic triad of ophthalmoplegia, ataxia, and confusion, and synoval abnormalities, which surround the third ventricle, involving mammillary bodies, as well as the periaqueductal gray matter, here seen best on post-gatilineum imaging. We can also consider MELAS in a patient that presents with transient or fluctuating cortical FLIR signal abnormalities, such as in this patient who developed right occipital cortical abnormality initially, which subsequently resolved, and then returned with left-sided occipital abnormality. Or we can also consider Wilson's disease in a patient with symmetric signal abnormality at the basal ganglia, here more pronounced at the level of the putamen bilaterally, involvement of bilateral thalami, as well as involvement of the midbrain, resulting in the characteristic face of the genic panda sign, reflected by dark signal within the red nuclei bilaterally, on a background of high signal within the midbrain tegmentum. We also see increased susceptibility effect within affected areas, reflecting increased mineralization. Finally, remember to think about venous congestion as a potential cause of rapid cognitive decline, as seen in this patient with deep white matter calcifications on CT, prominent transmedullary veins seen on T2 and post-gatilineum imaging, arterialization of veins on susceptibility-weighted imaging, and a DSA demonstrating a left tentorial duralative fistula supplied predominantly by the left occipital artery with reflux into cerebellar veins. Moving on to chronic dementia syndromes, the first item to consider is regional atrophy, which can be broken down into medial temporal or parietal atrophy, as seen in Alzheimer's dementia, or anterior lateral temporal or frontal atrophy, which is characteristically seen in the context of frontotemporal dementia. That can be further broken down into the major subtypes of behavioral variant FTD and primary progressive aphasia that have specific atrophy patterns. This is a classic imaging example of Alzheimer's dementia, which is the most common cause of dementia in the elderly, where we see preferential mesial temporal atrophy in both coronal and axial plane, which is out of proportion to the degree of atrophy seen elsewhere in the brain. Contrast that with the behavioral variant of frontotemporal dementia, which presents with personality changes and disinhibition, where we see preferential anteromedial frontal atrophy with a very pronounced anterior to posterior gradient. We also see flattening of the insula. And in this case, we also see atrophy of the anterior and lateral margins of the temporal lobe, which again is not preferentially medial, as we would expect to see in Alzheimer's disease. The non-fluent variant of primary progressive aphasia presents with expressive language deficits, with preferential left perisylvian and left frontal opercular atrophy, with marked asymmetric widening of the left sylvian fissure. The semantic variant of primary progressive aphasia presents with a loss of semantic knowledge of words, and affects the anterior temporal lobes, often asymmetrically, with slightly different clinical manifestations, depending on whether the left or right side is predominantly affected. Here we see that the right side is predominantly affected in this patient, and this patient presented with more of a behavioral syndrome, whereas if the left side is predominantly affected, those patients would present with more of a language syndrome. Moving on to white matter disease, vascular dementia is the second most common cause of dementia, and progresses in a stepwise manner with each new insult. This patient manifested with extensive subcortical disease, with advanced leukoencephalopathy affecting the deep white matter of bilateral cerebral hemispheres, as well as the pons, as well as multiple old subcortical lacunar infarcts, predominantly within the basal ganglia. But causes can broadly include predominant large artery infarctions, multiple small artery infarctions, as well as advanced microangiopathic disease. Related disorders include cataclysm, which is an autosomal dominant inherited disorder related to mutations in the Notch 3 gene, manifesting with extensive leuko-aeriosis, multiple old subcortical infarcts, with characteristic involvement of the anterior temporal regions, as well as the external capsules. And also cerebral amyloid angiopathy, which manifests generally in patients over 60, with advanced white matter changes, microbleeds, which are predominantly located at the cortical subcortical junction within the posterior parieto-occipital region, and potentially macrobleeds in an acute setting. While white matter processes often present with chronic dementia, a few notable exceptions are inflammatory cerebral amyloid angiopathy, which tends to affect slightly younger patients and manifests with acute cognitive deterioration that may be responsive to steroid therapy. The imaging presentation is that of regional vasogenic edema with clustered microbleeds and may show leptomeningeal or parenchymal enhancement. Another potential exception includes viral infections such as PML, which tend to affect immunocompromised patients and manifests with subacute neurological and cognitive decline. The imaging presentation includes multifocal but asymmetric subcortical flare hyperintense lesions, typically involving the U-fibers, without significant mass effect, restricted diffusion or enhancement. A few final thoughts on normal pressure hydrocephalus. Since we are often asked to exclude or confirm this diagnosis on imaging, and so it's important to remember that there are no consensus clinical or imaging criteria to define normal pressure hydrocephalus, and in particular to define those patients who will respond well to shunt therapy. This despite the fact that there are some proposed objective features, including a high Evans index, which is defined as the ventricular diameter of the frontal horns divided by the largest cranial diameter, as well as an acute colossal angle. A few additional findings that are thought to be characteristic of neural pressure hydrocephalus are prominence of the ventricular system and the sylvian fissures out of proportion to the size of the sulci, particularly at the cranial vertex, as well as the presence of an aqueductal flow void that's thought to represent hyperdynamic flow. However, again, it's important to remember that these are not necessarily diagnostic of the disease and that there may also be other coexisting neurodegenerative disease in these patients, and so although we can help to support a strong clinical suspicion of NPH, we shouldn't feel pressured to come down hard in the diagnosis based on imaging alone. So in summary, we reviewed a quick approach to imaging cognitive impairment, dividing this into chronic versus rapidly progressive disease, beginning by ruling out major structural abnormalities, and then focusing in on key regions or imaging patterns that can help make the diagnosis. Thank you very much for your attention, and I'm delighted to introduce our next speaker, Dr. Suyash Mohan, who will be speaking on the value of imaging in the found down patient. Thank you, Dr. Raval. In this talk, we will discuss the value of imaging in a found down patient. No relevant disclosures. So we will discuss the differential diagnosis for a found down patient, recognize conditions, and imaging findings that will acutely change patient management, highlighting the value of imaging. Now, whenever we are dealing with a found down patient, we are usually in a hurry and possibly under pressure to start our imaging interpretation, but before we discuss imaging, we should keep in mind that our role is primarily twofold. The first thing we have to tell, is there something which is actionable? Is there something that can acutely kill the patient? Keep in mind, triple H, herniation, hydrocephalus, hemorrhage, or if there is something that is potentially reversible. So tip number one, we should try and get as much information as we can. Some questions to ask is what exactly happened? When did it happen? Were there any pre-existing conditions? Any information regarding patient's immune status? Is the patient febrile, etc. Okay, so this was a 44-year-old who was found down in the bathroom with worst headache of life. Non-contrast HETs showed extensive subarachnoid hemorrhage in the basilar systems in the right more than left Silvian fissures. CT and geography shows a right MCA bifurcation aneurysm. So this was a case of ruptured aneurysm leading to subarachnoid hemorrhage, which remains the most common cause of non-traumatic subarachnoid hemorrhage. And trauma is by far the most common cause of subarachnoid hemorrhage from all causes. Now, looking at the pattern of subarachnoid hemorrhage, we can perhaps suggest the underlying etiology. If you have extensive hemorrhage in the perimes encephalic systems, in the basilar systems, in the Silvian fissures, we should think about aneurysmal pattern. If you have subarachnoid hemorrhage along the cerebral convexities, we should think about trauma. And this image over here is classic for a perimes encephalic pattern of subarachnoid hemorrhage. Now, when you have a patient who presents with worst headache of life, it is not always subarachnoid hemorrhage. We should remember these six entities that can also present with worst headache of life. Now, this case is a 59-year-old who was found collapsed in a crowded street. Notice there are ill-defined areas of low density on CT, extensive abnormal findings in bilateral deep brain nuclei. Notice there is involvement of thalamide. Notice there is involvement of the cerebellum. Extensive areas of reduced diffusion on these ADC maps. And this was a case of hypoxemic ischemic encephalopathy. This patient had cardiac arrest. So usually, history is self-evident in these cases. On CT, you can see the reversal sign or the white cerebellum sign, as I'm highlighting in this particular example. Now, in contrast, notice that the imaging findings are very similar in this case, who was also found down but had history of diabetes. Notice there is a sparing of the thalamide. And further down, I'm not showing over here, there was a sparing of the brainstem and cerebellum, but there was involvement of the splenium of the corpus callosum in a classic distribution. And this was a case of hypoglycemia. Now, this classic white matter involvement, the splenial involvement, is also known as boomerang sign. The value of imaging in these cases is that this can be rapidly reversible if we administer glucose in time. And we recognize these imaging findings. The key differences between hypoxia and hypoglycemia is in hypoglycemia, there is a sparing of the thalamide, there is a sparing of the cerebellum, and generally, the white matter is relatively spared in the earlier part of the disease. Another example, insulin overdose. Notice there is, again, sparing of the thalamide, involvement of gray matter everywhere else. Again, the teaching point here is that if you have involvement of bilateral basal ganglia with diffusion restriction and cortical abnormalities, we should think about hypoxia, hypoglycemia, seizures, encephalitis, CJD, and hyperammonemia. Another example, this was an alcoholic who was presented with acute onset of impaired consciousness. Notice there is abnormal signal involving mammillary bodies, abnormal signal involving the periaqueductal gray, and the basal ganglia bilaterally. And this was a case of Wernicke's encephalopathy. Now, location becomes classic for Wernicke's encephalopathy with involvement of the portion of the thalamus, which is next to the third ventricle, which is called paraventricular thalamic involvement. There is involvement of mammillary bodies that show enhancement, and if they enhance, it becomes pathognomonic for Wernicke's encephalopathy. Again, notice involvement of periaqueductal gray and reduced diffusion in the perirolandic areas. Another example of Wernicke's encephalopathy. Now, again, this is a condition where rapid identification and administration of theamine can save somebody's life. It is a neurological emergency, and if untreated, it can cause death. Another example, this was an unknown male, and it highlights the problems, the challenges that our clinical colleagues face in the trauma bay and the emergency department. Nobody knows anything about this patient. Nobody knows what exactly happened. There is no exam. And when they looked at him, they could see that there were abrasions along bilateral shins, and they thought this was a case of trauma. So it became, we got a whole body CAT scan, and HCT showed extensive areas of hemorrhage surrounding edema in both hemispheres, and somebody thought that we are dealing with sagittal sinus thrombosis, and morphinography was performed, and noticed that the superior sagittal sinus is intact, completely fated. Instead, we saw this normal anatomic variant. It's an unusual variant where the straight sinus kind of takes an inferior course and drains into the right transfer sinus. Now, notice there was abnormal signal in the cerebellum, which would also be difficult to explain based on sagittal sinus thrombosis. This was indeed a case of severe breast, severe breast. There was extensive swelling. Patient had to undergo decompressive hemicraniactomy, but we could not save him, highlighting that this is not always reversible. Patient passed away on day number eight. Now, last scenario, this was a patient who was found down with froth at the angle of mouth. EEG showed status epilepticus, arising from the right hemisphere. Now, on imaging, we see tiny formed areas of increased T2N flare signal in the right hemisphere, increased signal in the posterior and the medial part, the pulmonary hypothalamus, with areas of reduced diffusion. This was transient reversible edema from status epilepticus. Notice that all of these changes completely reversed on the next follow-up MRI in one week. Transient reversible edema, this is usually a combination of cytotoxic and vasogenic anema, also known as seizure-related changes, commonly involving the hippocampi, as I'm showing you in this case. And this was, again, a recent case, a COVID-19 positive patient. A large vessel occlusion involving the right M1, underwent a decompressive hemicraniactomy, subsequent hemorrhage, and then was found in status epilepticus with involvement of bilateral deep gray nuclei. So I hope I was able to give you some highlights of how imaging can be very valuable in a found down patient. Again, keep in mind, our role is to tell our clinical colleagues if there is something actionable, something that can acutely kill the patient, something that is potentially reversible. And if there is, our job is to pick up the phone, not the dictaphone. Thank you for your attention. So I'm now delighted to introduce our final speaker for this Essentials of Neuroimaging course, Dr. Courtney Tomlinson from Nashville, Tennessee, and she will talk to us about sinonasal anatomy. Hello. I'm Courtney Tomlinson, and today we're going to discuss sinonasal anatomy. Our four objectives are to discuss the function, anatomy, and anatomical variance, a little bit about the pathology, and surgical implications in the sinonasal cavity. So why do we even have sinuses? Well, they help lighten the skull. They filter particulate. They serve as a humidifier for air passing down to the lungs. We know it's important in olfaction, as we've seen in COVID-19. And they serve as a protective crumple zone for the midface. We'll discuss the nasal cavity, of course the sinuses, some of those variants, and three drainage patterns. So let's begin by taking a look at this 3D model. Here, we're looking at the anterior nasal cavity. We can see the cerviform plate and the ethmoid air cells in pink, the infraturbinates in blue, inside the maxillary sinus there in yellow, and looking superiorly toward the cerviform plate, the septum, and the middle turbinates. Axial CT of the nasal cavity demonstrates the nasal ala, or the alar cartilages, and the nasal vestibule. Moving a little more superiorly, as you can see in the right coronal image, we have the nasal vault. The piriform aperture forms the bony inlet of the nasal cavity, the nasal septum, posteriorly midline, and the nasal coena in the back. Piriform aperture stenosis occurs when this piriform aperture is less than 11 millimeters and can be associated with the central mega incisor. Coenal atresia can occur in the posterior coena and can be either bony, as shown here, or membranous. A dacrycystosele occurs when there is obstruction of the nasolacrimal duct, as shown, and this can occur at either the inferior meatus or more approximately. Coronal CT demonstrates the superior turbinates, as well as the meati, the middle turbinates, and middle meatus, which is lateral, and the inferior turbinate and inferior meatus. Antero-coenal polyps occur through a widened maxillary antrum, extending into the middle meatus and posteriorly toward the nasopharynx, as shown in these three cases. So here we have our normal middle turbinates, and here we can see an expanded pneumatized right middle turbinate called a contrabulosa with left nasal septal deviation and ipsilateral inferior turbinate hypertrophy. This is a classic trifecta. Looking at the septum, we see the bony portion, and the superior part is called the perpendicular platelet ethmoid, while the inferior is the vomer. The cartilaginous septum would be anterior to this. Focusing on the central aspect of the ethmoid cavity and the anterior skull base, we can see this perpendicular plate here extending up as the crista galei to the intracranial cavity, as well as the olfactory grooves on either side with the olfactory foramina that allows small olfactory nerve roots to come into the ethmoid cavity. The olfactory recesses are medial to the turbinates, and the fovea ethmoidalis is along the lateral aspect of the ethmoid roof. So anytime we see a paucity of sinus disease, but soft tissue filling the olfactory cleft or recess, we should consider aesthesio-neuroblastoma, as shown in this case. And if I may impart one piece of wisdom on you, it is that not all lidopacifies is snot. Very eloquent. The anterior skull base can have a dehiscence in it, congenital or post-traumatic, or iatrogenic. And as shown here, there is an anterior skull base encephalocele. So not all lidopacifies is snot. Here we see a bony dehiscence of the lamina prapracia on the right, and orbital fat, extraconal fat herniating into that defect. Another thing we should take care to look for is asymmetry within the cribriform plate and the fovea ethmoidalis. So as the depth of the olfactory faucet increases, so does the risk of iatrogenic injury during surgery. Moving on to the sinuses, we'll begin with the maxillary sinuses paired on either side of the nasal cavity. The maxillary ostium is found here, and that is located posterior to the nasolacrimal duct. This level also gives us a good view at the pterygopalatine fosso, or retromaxillary fat, leading into the sphenopalatine foramen here. We have anterior and posterior ethmoid air cells, which are divided by the basal lamella. The sphenoid sinus is posteriorly, and both the posterior ethmoid and the sphenoid sinus drain through the sphenoethmoidal recesses. A couple anatomic pneumatization variants to mention, the Haller cell. This is an infraorbital ethmoid air cell that can cause narrowing at the osteoimmutal unit, and we'll talk about that drainage pathway soon. An auger nasi cell is the anterior most ethmoid air cell, and when expanded, can narrow the frontal recess, which is located behind it. Here's the crista galei again, and sometimes it can be incidentally pneumatized, but rather an interesting look. Heading back to the sphenoid sinus, here we see an anode cell, which is an intrasphenoid ethmoid air cell characterized by the septa, horizontal septa, with the anode cells on top. And just a variation of that, here we see bilateral anode cells, and pneumatization of the anteroclinoid processes, with a pedicled optic nerve traversing through this aerated space, although this is not dehiscent as it's covered by bone. Sticking with the sphenoid sinus, we can see the pterygoid recess, extending laterally in the sphenoid sinus, and this can be variably pneumatized, which nicely shapes the foramen rotundum above and videon canal below for us. And here's what that pterygoid recess looks like on an axial. The pneumatization pattern of the sphenoid can be highly variable, with conchal, precellar, and cellar or postcellar recesses. The sphenoid can be highly variable, with conchal, precellar, and cellar or postcellar components. And here, the cellar configuration, it's good to understand, increases the risk of iatrogenic injury, given this very thin dorsal clivus. One important surgical consideration is the pedicle of the anteroethmoid arteries. So here, if the ethmoid artery is inadvertently ligated, its muscular wall will recoil into the orbit and cause an expanding hematoma that may need to be surgically decompressed. The intersphenoid sinus septum, another aberrant attachment, can be when it attaches to the carotid canal instead of the midline, and this can put the carotid at risk during sphenoid surgery. And lastly, we have the frontal sinuses. Shown here, we have the interfrontal septum, as well as an isolated intrafrontal air cell. What's interesting is that these can be highly variably pneumatized, and so they're like a box of chocolates. You never know what you're going to get. And there is a classification system for this called the Kuhn classification system. It's something to just keep in mind, and it's not a very common classification. Every time you look at a sinus CT, there are surgical implications. So briefly, let's review three drainage patterns that can be important for obstruction. The first is the osteomeatal unit, which drains the maxillary and anterior ethmoid air cells through the infundibulum into the hiatus semilunaris as it wraps around this uncinnate process to reach the middle meatus. It's important to note whether the uncinnate process is adherent to the lamina paprasia. Here we see an osteomeatal obstruction pattern caused by a small osteoma in this semilunar hiatus. The sphenoethmoidal recess, when obstructed or opacified such as here, can result in posterior ethmoid opacification and plus-minus sphenoid opacification. And finally, the frontal recess, when it becomes opacified, of course, obstructs the frontal sinuses. So why do we care about the sinuses and sinonasal cavity? Well, we want to help the surgeon so you can help the patient, so you can know that you've done your best. But what can go wrong with them? There's actually quite a lot, many things, pathologies, we weren't able to discuss today. But here, I'd like to show you this case. So remember that not everything opacified is snot. And here we can see dehiscence, multifocal dehiscence of the cribriform plate. In this case, MRI shows an asthesia neuroblastoma with its classic peritumorosis. So keep in mind to look around the sinuses for clues. Okay, here, you may be looking at the contrabulosa or perhaps the septal spur. But what we really should pay note to for the surgeon is this high risk of skull base fracture associated with this short vertical lamella seen with contrabulosa. So remember your soft tissue windows, these can be key as in the setting of acute invasive fungal sinusitis where you have premaxillary and retromaxillary fat stranding, as well as non-enhancing tissue in the left maxillary sinus. The scalp and calvarium as well as the brain are all corner shots on your sinus CTs. And this was the result of complicated ethmoid sinusitis, as is this case of a subperiosteal abscess that you can see here. So remember your surgical checklist, these things that we've talked about, and you'll be doing a good service for your patient. In summary, what we've discussed today regarding sinonasal anatomy is some of the imaging considerations, variant anatomy and pathologies, the drainage pathways, as well as surgical implications. I'd like to acknowledge these superstars for helping me put this talk together. Here are a few references for your reading, and feel free to reach out via email or on Twitter. Thank you very much.

neuroimaging

spinal infections

dementia diagnosis

emergency imaging

sinonasal anatomy

Dr. Jim Chen

Dr. Sapna Rawal

Dr. Suyash Mohan