Top 5 Missed Head & Neck Diagnoses

Introduction

I'm from New York, and Dave is king
and Dave does these top 10 lists, which are
way more amusing than this talk will be.
But that is the general format.
I'm gonna count down the top five.
I'm just gonna, as a disclaimer, state that
I don't know for sure that these are the top five most
frequently missed head and neck diagnoses.
These are just my personal list of things that
I've seen frequently,
and most of them are things that I actually, all
of them are things that I missed myself.
So therefore, they must be very important to discuss.
That's pretty much the basis for this.

Number 5: Mylohyoid Defect with Herniation of Salivary Tissue

Okay, we will start with this mouthful of
diagnosis mylohyoid defect with herniation of salivary tissue.
This has another name, which is, in my opinion, even more
of a mouthful, accessory salivary tissue in the
mylohyoid boutonniere.
And this is essentially a pretty much normal finding,
but it is a radiographic pitfall.
And from what I understand,
an occasional clinical pitfall as well.
And all it is is a defect in the mylohyoid muscle
and defects are present in 75% of normal patients.
But there is associated herniation of
the sublingual glands.
And some writers describe it
as accessory salivary tissue.
I don't believe that. I think that it's just the
sublingual glands through these mylohyoid defects.
So through the floor of mouth
and into the submandibular triangle.
And the key to not falling victim
to this pitfall is to notice that the density
and the signal characteristics on MR
of these so-called masses is identical
to normal salivary tissue.

Just to review the anatomy, the mylohyoid forms the bulk
of the floor mouth, and it connects the mandible
to the hyoid bone.
If you look at a sagittal cross-section, you can see
that the sublingual gland normally will sit on top of it.
The submandibular gland kind of sits behind it
and wraps around, the submandibular duct, of course,
is through the oral cavity towards its punctum in the anterior floor mouth.
And this anatomy is actually really nicely depicted on imaging.
Here it is, just a graphic in the coronal plane showing
you the normal location of the sublingual gland
above the mylohyoid muscle, the submandibular gland below,
but the very thin sling of the mylohyoid muscle
that you can see very nicely even on this
non-contrast coronal CT image, again, the thin sling
of the mylohyoid muscle, forming the bulk of the floor
of mouth with the anterior belly of the digastric in the midline contributing to the floor
of mouth in the midline.

So what do these look like?
What they look like are extra masses in
the submandibular triangle.
So there's not much that lives in the
submandibular triangle.
I think it's an anatomically very easy area.
You have your submandibular glands, you've got some lymph nodes, a lot of fat, the occasional nerve and vessel here and there.
But in this patient, in addition
to his normal submandibular glands, it's like he has an additional set of two submandibular glands.
They look virtually identical.
This is a non-contrast CT scan,
but I think even on this image, you can see
that the density is virtually identical to that
of the normal submandibular glands.
The arrows are pointing
to the little defect in the mylohyoid muscle.
And when this occurs, it very often kind
of swings out laterally,
so it makes it a little bit more obvious to see same patient, the coronal scan.
You can actually see the mylohyoid muscle.
You can see the defect in the muscle,
and you can see the sublingual gland herniating through
that defect into the submandibular triangle.

This was sent to me by my former training director who doesn't do all that much head
and neck radiology and is now in Florida reading very lucratively from open magnets all over the country.
And he sent me this, I don't even know
what these sequences are exactly.
This is a T1, this is a T2,
and this is some kind of gradient sequence.
The scans are really ugly, and he wanted to know what these big mass was.
But even on these really ugly scans, you can see
that the mass has signal characteristics identical
to the normal submandibular gland on all
of these pulse sequences, whatever they are.
And that's the clue that this is just a very large appearance of a sublingual gland that's herniated
through a defect in the mylohyoid muscle into the
submandibular triangle of the neck.
These can be very big and they can be very small.
So here's a really tiny one.
Again, you can see the mylohyoid defect
with this tissue, which looks just like the submandibular
glands herniating right through it.

Now, because this is salivary tissue, it's subject
to the same pathologic processes
that occur in submandibular tissue anywhere.
So this patient presented with relatively acute onset of
submandibular pain and fullness.
And you can note that in this
so-called accessory salivary tissue,
or in his sublingual gland, you can see the features
of acute sialadenitis in
that there's a greater degree of enhancement.
There's dilatation of the duct,
and there's actually a small air bubble in a patient with
sialadenitis in his accessory salivary tissue.

Number 4: Odontogenic Maxillary Sinusitis

Okay, we're down to number four,
odontogenic maxillary sinusitis.
Remember that there's a very intimate relationship
between the maxillary sinus
and the maxillary alveolar ridge containing the teeth, so
that it's not uncommon for inflammation to ascend
to the maxillary sinus from dental problems involving
the maxillary teeth.
Now, typically there will be a clue to this
because there'll be evidence of periodontal disease,
and that's typically this periapical cystic lesion
that surrounds the tooth.
And it's almost always the molar teeth,
because those are the ones
that get diseased the most frequently.
And you may or may not see an actual communication
between these periodontal cysts
and the actual maxillary antrum.
But you don't need to see that for this to occur.
And when should you have a strong suspicion for this,
when the disease is essentially isolated to
that maxillary sinus,
although certainly it will occur in patients
who have sinusitis for other reasons as well.

Here's a really classic case.
You can see that this patient has a little bit
of ethmoid disease, but predominantly right maxillary
sinus opacification.
And in this patient, you can actually see the very large cavity in his tooth.
I actually get physical pain when I see these like
really large things.
And then a actual fistula between the periapical cyst
and this maxillary sinusitis.
So you can treat the patient
for sinusitis from now till doomsday,
and they're not gonna feel any better
until somebody takes care of that tooth.

Here's just a series of three additional patients
to show you that the findings are pretty stereotypical.
All of these patients have disease that is predominantly present in that maxillary sinus.
In this patient, you can see there's a little bit
of dystrophic calcification above that alveolar ridge.
This is the classic appearance
of a periapical cyst without an obvious fistula.
And this patient has a smaller periapical cyst all
around these maxillary molar teeth.

Now this is another example
of dental disease causing maxillary sinus opacification,
but this patient doesn't even have technically maxillary
sinusitis because this patient's maxillary antrum is this
little tiny slit of air.
And the clue to this patient's diagnosis is the presence
of these two parallel lines.
You can see it nicely on the sagittal imaging.
One of these is the roof of the maxillary sinus,
and the other is the floor of the maxillary sinus,
which is being lifted up by this large dental cyst.
I think the presence of a tooth is the so-called smoking gun that we've heard about a lot when we have our
coffee and our well caffeinated.
And this is a very large dentigerous cyst,
which is opacifying the maxillary sinus,
but not technically a cause
of odontogenic maxillary sinusitis.

Number 3: Submandibular Sialadenitis

All right, we're up to number three.
Back to the salivary glands.
I think submandibular sialadenitis is a very challenging diagnosis,
for us, and that's
because the most frequent site
of stone impaction is at the level
of the punctum in the anterior floor of mouth.
And unfortunately, that's an area
that is frequently obscured by a dental artifact.
The stones can be really, really tiny
and very, very hard to see.
And even in the absence of dental artifact,
I think you really have to be looking for them in order
to make that diagnosis.

Here's a patient with left submandibular pain,
had a contrast enhanced scan,
and you can see the classic findings of sialadenitis.
The submandibular gland on the left is enlarged.
There's prominence of the intraglandular ductal system,
and then there's surrounding lytic change.
The platysma is thickened,
the subcutaneous fat is infiltrated.
And it's very easy to get focused on this process,
which is certainly fairly dramatic.
And then to not look very carefully at the anterior floor mouth
and to pick up that little millimeter
or two calculus in exactly the place that these things typically will impact.

You can play with your these scans,
and I just find it kind of cool to do.
I don't know that it adds very much,
but this patient with submandibular sialadenitis,
you can follow his dilated duct on the sagittal image all
the way along the floor of his mouth to the tiny calculus that's present in that location.
And you can do the same thing on the coronal scan.
You can see the dilated duct with a little bit
of peripheral enhancement,
and it ends at that little tiny calculus in that location.

Even on a non-contrast scan, you can find findings that are suggestive of acute sialadenitis.
So this submandibular gland is certainly larger than the
other one, and it has an edematous appearance resulting in actual
decreased density on a pre contrast scan because it is boggy
and swollen.
But again, notice that there are findings of surrounding cellulitis and a little bit
of thickening of the platysma muscle.
And again, the culprit is this really teeny tiny
sub-millimeter calculus in the anterior floor of mouth.

With the patients that I've shown you up
to now have been ones without dental artifact,
but you can see how dental artifact really makes things very difficult, large enhancing boggy gland.
And luckily the stone was big enough to see despite the presence
of all this streak artifact from the patient's dental
amalgam on PACS workstations, now that we can adjust the window and level in real time
and kind of play with this, you can reduce the dental artifact
and have a better chance
of seeing these things if you actually sit
and fiddle with the dials and controls.
And you can kind of make this become
a little bit more obvious.

Another technique that can help,
and this is a courtesy of my friend Roy Hollander at New York Eye and Ear.
He has a thing where he has the patient lift
up their tongue.
A lot of ENT patients can't really do that
for a very long time, but he is apparently our ENT are facile
with their tongues, and that I think makes this really obvious
and it's sitting right in a perfect location.
And I think that's a pretty awesome thing if you can make that happen.

These issues are not just confined to the submandibular gland.
The anatomy around the terminus
of the parotid duct can be confusing as well.
And Dr. Somerville, I think, discussed
that this patient has left facial pain,
and you can see that his left parotid is enhancing again to a greater degree than on the right side.
And actually you can follow his dilated duct all the way
from inside of his gland over his masseter muscle,
and then ultimately to its terminus at the level
of the second maxillary molar tooth
where it inserts into the buccal mucosa there.
And there is the offending calculus.
But unfortunately there's a lot of complicated anatomy in this location.
And you can certainly look at this
and say, this all looks rather symmetric.
So I don't know, maybe that's not a stone.

Just to review again the anatomy, I think Dr.
Somerville went over it quite well.
The duct rises from the gland.
It passes over the masseter muscle inserts at the level of
that maxillary molar tooth.
But you can almost always find either a flow void
or enhancing round dot representing the facial vein.
And then you see some of the muscles
of facial expression, the zygomaticus muscles.
So there's really sort of two parallel lines
with a little bit in between them.
So if we put up our scan, we can locate all
of these structures and figure out where the extra white dot is.
So the more anterior structure that we see,
the line is a zygomaticus muscle.
We can see the facial vein.
This is a CT, so it's going to enhance.
And then we can see the parotid duct,
and there's our culprit, the extra white dot.

Some of these can be, again, difficult to see
because of dental artifact.
Again, Roy Hollander, let me, this case, this is a patient
who had right facial pain.
And I think the dental artifact is a problem,
but if you look very closely
and you pay attention to where you should be looking,
you could see this tiny calculus both on the axial
and reformatted coronal images.
And that was another distal parotid duct calculus.

Number 2: Fibrous Dysplasia

Number two, fibrous dysplasia.
And MRI, this is just a real problem, I think,
it's very, very hard to see this diagnosis.
And unfortunately, the head and neck, the skull
and facial bones is a very common site of fibrous dysplasia.
So it's affected in 25% of monostotic fibrous dysplasia,
and up to 50% in patients who have polyostotic disease.
Now on CT, it does have
that classic ground glass appearance,
but in many cases it can actually be hyperdense or cystic.
So in up to 25% of cases, the CT can even be confusing.
But the MR features are even less distinctive.
And these can really look like tumors, again, like contrary
to what seems intuitive to me, they enhance.
And therefore, that always makes you think of a tumor.
But the low T2 signal, I think has turned out
to be the best clue that I've come up with.
And also the fact that the enhancing soft tissue mass really
conforms to the shape of the parent bone.
So these are not growing off in all sorts of directions.
They're just expanding the native appearance of the bone.

Let me show you some examples.
This is a 25-year-old patient who presented
with right visual loss,
and it's easy to see what the problem is.
These are the optic nerves bilaterally affected
by this enhancing process, but more so on the right side.
So you can see again, there's expansion
and there is enhancement,
and it certainly is scary looking.
But note that on the T2 weighted images,
it's really quite dark.
Could this be lymphoma or a meningioma?
It's certainly possible, but the patient went on
to have a CT, and that's what I would recommend as the tiebreaker in any
of these cases in which you're really
suspecting a process in the bone.
I mean, as in the temporal bone, as Dr. Phillips mentioned,
I think if you're suspecting that you're dealing
with a bone lesion CT is really the way to go.
And this is the classic ground glass appearance
of fibrous dysplasia affecting the anterior clinoid
process and the body of the sphenoid
and narrowing the right carotid canal, which as we know,
runs medial to that anterior clinoid process.

This actually is kind of a,
not quite every day, but it's a common clinical problem.
This patient has a history of lung cancer
and complaints of headache.
So he was scanned to rule out brain mets.
And somebody noticed the fact
that he has a lesion in his clivus,
and the lesion is of intermediate signal intensity on T1
and does enhance quite brightly.
And somebody rightfully raised the suspicion
that this could be a metastatic bone lesion from the patient's lung cancer.
But notice again, it's really quite dark
on T2 weighted images.
There's really very little disruption
of the overall architecture of the bone.
And this patient, again, went to CT,
and this is another kind of classic ground glass appearance
of fibrous dysplasia of the clivus
and a patient in whom there was a significant clinical
suspicion that there might be something neoplastic going on.

This was a kind of weird one.
This 46-year-old also had a headache
and had a what I thought was a very bizarre lesion in
this sphenoid sinus.
This was just a non-contrast sort of screening MR
and it looks almost like there's a cyst.
The sphenoid bone looks abnormal,
but you've got this thing projecting into the lumen
of the sinus, which is relatively dark on T1.
We expected it to be very bright on T2,
but it stayed very dark on T2, so not a cystic lesion.
And again, you can see it sort
of encapsulated thing projecting into the back
of the sphenoid sinus.
Patient went on to have a CT
and another focus of fibrous dysplasia.
This one with a somewhat atypical sort
of hypostatic or exophytic appearance, not
entirely typical of fibrous dysplasia,
but a proven by endoscopic biopsy to be fibrous dysplasia.

This can be a very deforming condition,
and typically these are people
with the polyostotic form of the disease.
Nobody could be faulted
for considering the presence of a tumor on this image marked expansion of the sphenoid bone
with contrast enhancement involving the ethmoid
as well, severe narrowing of the region
of the orbital apex and optic canal.
This patient had significantly impaired vision.
But note again, it's very dark on T2
and also note that really the shape
of the bones is maintained.
They're expanded, but really
this is not growing off into any directions distinct
from where the bone would normally live.
So a bit of an imaging clue.
I'm not saying that this is one of these things
that you can really hang your hat on
and feel entirely comfortable about,
but I think it does potentially raise a suspicion
that this would be fibrous dysplasia.
And again, on CT, a little bit more difficult
'cause this patient had a mixed cystic appearance as well
as the classic ground glass appearance.
But this was another case of fibrous dysplasia.

And then the final case, Dr. Phillips, I think,
he stole my thunder on this one, but this is his case.
So this I think confounded everybody, an entirely lytic appearance to this process.
And I don't think anybody realized what it was, particularly
after the prominent degree of contrast enhancement.
But the very low T2 signal proved
that this was, I believe,
after biopsy proven to be a case of fibrous dysplasia.
Again, the low T2 signal, the expansion without really all that much distortion of the anatomy.

Number 1: Fenestral Otosclerosis

And finally, the number one frequently missed: fenestral otosclerosis.
I've been at my new job for about seven months now.
But when I got to my previous job and the first day I was there, there was a case
of otosclerosis
and my neuroradiology colleagues,
they had never had a head and neck person there.
I showed them that I had all these people standing
around me going, oh, ah, you know, they had literally somebody told me they had not made that diagnosis once in the six years the group had been in
that practice before I got there.
So not to beat my own drum,
but to show you that a very, very good neuroradiology group.
If you're not attuned to this diagnosis,
I think you're really not gonna make the diagnosis in a lot, in most cases.

Fenestral otosclerosis is an idiopathic bone remodeling
disorder that involves the otic capsule.
And it is actually quite a common cause of conductive
or mixed hearing loss,
and it's found to be present in 10% of autopsy specimens.
So obviously it is subclinical a significant percentage
of the time, but it's going to be a cause
of clinical hearing loss, not uncommonly,
typically it's the hearing loss starts from early adulthood.
It's more common in females with a Caucasian predominance,
and it's frequently bilateral.

The two forms of otosclerosis, the fenestral form is considered to be the early form.
And you get a formation of these plaques
of abnormal spongy vascular bone.
It typically starts along the anterior margin
of the oval window at the embryonic fissula ante fenestram.
The retrofenestral form, which typically follows,
is considered to be a more advanced form,
actually spreads to involve the dense otic capsule
bone surrounding the labyrinth.
And what you'll see is an abnormal halo
of lucency surrounding the cochlea and or the vestibule.
Typically, you don't have retrofenestral disease
without fenestral disease.

Here's what is actually a nice correlation between a histologic
specimen, just to orient, here's the cochlea on the histology specimen, and here's our CT.
And this is the spongy plaque.
This is the oval window.
You can see the stapes sitting on the oval window,
so the stapes foot plate and the oval window,
and the characteristic location along the anterior
margin of that oval window.
And on the CT exam,
I think really nice histologic correlation.
But this is all you're gonna see in many, many cases.
And this is actually not a subtle case.
Overgrowth of lytic loosened bone,
abnormal loosened bone along the anterior
margin oval window.
So what happens is that the stapes foot plate
or the stapes crura get fixated by the plaque,
and that results in hearing loss.

Here's a couple more cases.
These are two different patients.
This one, these are both the left ear on axial imaging.
So here's our oval window,
and here is a somewhat obvious plaque overgrowth
of bone abnormally loose compared
to the dense otic capsule bone.
And then this is a very subtle case,
but it is present
nonetheless, obviously with good history,
you are more likely to make this diagnosis.
But you know, scans are frequently called
and for obvious reasons.
Normal, very subtle plaque, just a very, very subtle area
of lucency along the anterior margin of that oval window.

The retrofenestral form is an easier diagnosis to make the otic capsules of the densest bone in the body.
It typically appears ivory white.
And seeing this loose halo is
pretty much pathognomonic.
Osteogenesis imperfecta is an uncommon cause.
Occasionally Paget disease can involve the otic capsule
as well, but much more uncommonly,
this is a much more common diagnosis.

Honorable Mentions

Just to finish up with honorable mention
of some other frequent misses, labyrinthitis, I think Dr.
Phillips showed you how subtle the finding
of acute viral labyrinthitis.
It could be a really minor degree of contrast enhancement.
And patients get scanned
for rule out vestibular schwannoma.
People focus on the IAC
and not so much on the structures of the membranous labyrinth,
which we see quite well on current MR scanning,
the large vestibular aqueduct syndrome,
which you just saw as the most common cause
of hearing loss in children.
So it's not an uncommon diagnosis,
but you know, you have to specifically be looking
for those findings of the dilated vestibular aqueduct
and the Mondini defect in the cochlea, looking for the nasal pharyngeal mass in an adult who develops a mastoid effusion and does not have a history of otitis.
Very important. And something that we really stress
with our radiology residents, retropharyngeal chain adenopathy,
because it lies outside the normal nodal chains
that I think we're all used
to looking at levels 1, 2, 3, 4, 5.
I think retropharyngeal lymph nodes are easy to miss
because it just sometimes doesn't become part of your scan review technique.
Optic neuritis involving the intracanalicular segment
of the optic nerve.
So we tend to focus on the portion
of the optic nerve in the orbit is easier to see.
It's easier to find,
but you can sometimes have optic neuritis isolated to
that intracanalicular segment.
So here's the anterior clinoid process
and the hyperintensity on T2 weighted imaging in a
patient with left optic neuritis.
And then of course, perineural tumor spread is very challenging.
It requires a high level of suspicion.
It requires really good imaging technique, and is
therefore difficult diagnosis to make.

Closing

With that, we're finished with our program
and on behalf of the other speakers,
we had so much fun.
And you guys were so great.
You sat here
and had great questions during the break.
We really appreciate your attention. Thanks for coming.
And feel free to contact any of us.