Consults- Memorable/Dismal/Ridiculous/Unique

[HighPriest]

Is it because no one uses insufflation on the TM?

I mean, there are a variety of reasons. I think the way they (med school and residency) teach ear exams is bad. I think otoscopes are suboptimal. I think the way they teach the pathophysiology is bad. I think the anatomy is poorly understood. I think to get good at ear exams you have to start by looking in everyone’s ear even if they’re normal, and that takes time and effort and planning. I think people come out of primary care training with a very poor differential diagnosis for the causes of plugged ear or ear pain, so they don’t really ask any pertinent questions. But mostly I think it’s just way easier to call everything eustachian tube dysfunction and place a referral. The patient feels like you’ve figured it out and put them on the right path and you don’t have to see them again. Hands washed.

---

Members don't see this ad.

 

[OtoHNS]

I mean, there are a variety of reasons. I think the way they (med school and residency) teach ear exams is bad. I think otoscopes are suboptimal. I think the way they teach the pathophysiology is bad. I think the anatomy is poorly understood. I think to get good at ear exams you have to start by looking in everyone’s ear even if they’re normal, and that takes time and effort and planning. I think people come out of primary care training with a very poor differential diagnosis for the causes of plugged ear or ear pain, so they don’t really ask any pertinent questions. But mostly I think it’s just way easier to call everything eustachian tube dysfunction and place a referral. The patient feels like you’ve figured it out and put them on the right path and you don’t have to see them again. Hands washed.

In 4 years of med school, I think we spent a total of 1 hour learning how to do the "eye, ear, nose, and throat" exam during M2 physical exam class. Other than ENT rotation, I can't remember ever being taught how to look in an ear again. Any doctor who is not ENT probably has this level of training in the otoscopic exam.

That's problem #1. Problem #2 is that an otoscopic exam is not a reliable way to determine if there is a middle ear effusion. Sometimes it's visually obvious, and many times it is not. Sometimes, the eardrum looks very abnormal but there is no effusion and the pressure is normal.

I've never tried to insufflate an ear, but I use a tympanometer multiple times a day in clinic which reliably measures the middle ear pressure and tells me if there is actually fluid or not and whether the patient has ETD or not.

But yeah, ear pain/pathology is a total black box for most referring providers, especially doc-in-a-box types.

 

[TypeADissection]

Man, I'm not perfect. But it feels like the entire world of healthcare skipped learning about ears.

I'm gonna be real honest. I don't know jack **** about the ears. My wife will tell me about her day or some cool ear case she did and I just smile and nod. One of my favorite interactions:

Her: I mean seriously, any med student could diagnose this as ear wax and not a cholesteatoma.
Me: Well, 'aight, check this out, dawg. First of all, you throwin' too many big words at me, and because I don't understand 'em, I'm gonna take 'em as disrespeck.

 

[DoctwoB]

This is true for us too. 95% of ear infections and effusion are misdiagnosed. If you ask someone in urgent care to look in your ear for any reason, you’re going to have an effusion. Guaranteed. All pain on the side of the head (any area) - ear infection. Drum too white? Ear infection. To red? Infection. Can’t see past the wax? Infection. Little sclerosis on there? Infection.

I will point out that the VAST majority of these diagnoses are NPs or PAs.

And yeah, no one knows the difference between a swimmers ear and an otitis media. I love it when the patient has been to urgent care three times and treated with oral antibiotic, then drop, then oral antibiotics and drops. And the diagnosis code is “eustachian tube dysfunction.” Where we are, all ear issues are no longer the purview of primary care. At all. They’re a complete black box. About 10% of the time no ear exam is even documented when antibiotics are prescribed for an ear infection. I’ve gotten into the habit of asking urgent care referrals if someone looked in their ear when they went in for ear pain. The answer would make you angry.

I read a study a long time ago where they took like 200 kids, all of whom got ear tubes for whatever reason. On the morning of surgery they had each ear examined (around 400) by an ENT, a pediatric ENT and a pediatrician and the only question was: is there an effusion? Yes or no?
Then the kids got tubes, so you know for sure.
The peds ENT doc was right about 70% of the time which sounds low, but this was an otoscope exam in an awake kid. So they’re not perfect.
The ENT doc was about 60-something %.
The pediatrician was right 30% of the time.

These were all staff docs.

The point of the article was that we need better methods for evaluating effusion. My take away is the pediatrician is misdiagnosing your kid 2x more often then they get it right when it comes to ear infections.

I’m at a point now where if I ask the patient if the referring doc did an ear exam and they say “yes,” my first thought is “alright, man! Good on you for trying!” And that’s pretty sad.

The good news is that we're probably not far off from NP/PAs just plugging the digital otoscope into their phone, snapping a picture, and then GPT bot tells them if there is an ear infection or not.

Artificial Intelligence Model Outperforms Clinicians in Diagnosing Pediatric Ear Infections

A machine-learning model developed at Mass Eye and Ear accurately diagnosed pediatric ear infections at a rate 30 percent higher than doctors surveyed in head-to-head testing.

masseyeandear.org

I'm inherently skeptical about AI in medicine, but when comparing to a very low bar as in this case it's probably an appropriate use case.

 

[HighPriest]

The good news is that we're probably not far off from NP/PAs just plugging the digital otoscope into their phone, snapping a picture, and then GPT bot tells them if there is an ear infection or not.

Artificial Intelligence Model Outperforms Clinicians in Diagnosing Pediatric Ear Infections

A machine-learning model developed at Mass Eye and Ear accurately diagnosed pediatric ear infections at a rate 30 percent higher than doctors surveyed in head-to-head testing.

masseyeandear.org

I'm inherently skeptical about AI in medicine, but when comparing to a very low bar as in this case it's probably an appropriate use case.

22 test images isn’t necessarily a revelation, and I can say that there’s some significant bias in diagnosing an ear infection based upon a still image versus a program that’s designed specifically to look for predetermined findings on said images, but it can’t possibly be worse than what we’re doing now. 600 images with 80% accuracy is pretty good.

 

[DoctwoB]

22 test images isn’t necessarily a revelation, and I can say that there’s some significant bias in diagnosing an ear infection based upon a still image versus a program that’s designed specifically to look for predetermined findings on said images, but it can’t possibly be worse than what we’re doing now. 600 images with 80% accuracy is pretty good.

Yeah it's interesting that MEEI couldn't pull together a larger training set. There are millions of infected ears a year but acquiring the data is 99% of the battle.

The training data and what is the gold standard to compare to are always issues. Of course the context of the pictures and metadata are also problems. The classic example is the model that could identify melanoma was trained to identify rulers. You could omagine a similar situation where a model recognizes from metadata that the picture is of a 2 year old at an urgent care center in winter and assigns an auto AOM dx.

 

[HighPriest]

You could omagine a similar situation where a model recognizes from metadata that the picture is of a 2 year old at an urgent care center in winter and assigns an auto AOM dx.

Basically what we’re doing now. Except it’s any patient with an ear in any season.

 

[eyeeye_captain]

I feel for you ENT folks. If you want at least as big a black box as ears, try eyes. My favorite from training:

Call at 3 AM:
ER resident: “Hey, I’ve got this non-verbal, combative patient sent in from his facility because he’s pointing at his eye and yelling. I checked his pressure and it’s sky high, I think he’s got angle closure glaucoma.”

See patient.

ER resident: “So it’s angle closure, right?”
Me: “Technically his angle is closed, but I’m more worried that it’s closed because half of his iris is coming out of his cornea.”

At least the resident looked I guess, but that’s not exactly a subtle finding. Also not usually a good idea to poke on an open eyeball.

 

[HighPriest]

I feel for you ENT folks. If you want at least as big a black box as ears, try eyes. My favorite from training:

Call at 3 AM:
ER resident: “Hey, I’ve got this non-verbal, combative patient sent in from his facility because he’s pointing at his eye and yelling. I checked his pressure and it’s sky high, I think he’s got angle closure glaucoma.”

See patient.

ER resident: “So it’s angle closure, right?”
Me: “Technically his angle is closed, but I’m more worried that it’s closed because half of his iris is coming out of his cornea.”

At least the resident looked I guess, but that’s not exactly a subtle finding. Also not usually a good idea to poke on an open eyeball.

I have no doubt that the globe is in league with the ear when it comes to misdiagnosis. Did they try to just poke the iris back in there? Pop a contact lens over it for a bandage?

 

[eyeeye_captain]

I have no doubt that the globe is in league with the ear when it comes to misdiagnosis. Did they try to just poke the iris back in there? Pop a contact lens over it for a bandage?

Nah, emergent surgery with iris repositioning and a ton of 10-0 nylon sutures. Not a fun case. Patient had an old corneal transplant that he managed to dehisce.

More along the lines of poorly diagnosed common ear stuff:

ER NP: “Hey, I’ve got a patient here for something completely unrelated, but she has a red painful eye and I’m not discharging her until you come in. *chuckle* Called your co-resident in for 5 patients last night. *chuckle*”

See patient, NP is now off shift.

Me to ER doc who came on: “Hey, that lady you’re waiting for dispo on has a stye. Nothing to do.”
ER doc: *heavy sigh* “Sorry.”

 

[HighPriest]

Nah, emergent surgery with iris repositioning and a ton of 10-0 nylon sutures. Not a fun case. Patient had an old corneal transplant that he managed to dehisce.

More along the lines of poorly diagnosed common ear stuff:

ER NP: “Hey, I’ve got a patient here for something completely unrelated, but she has a red painful eye and I’m not discharging her until you come in. *chuckle* Called your co-resident in for 5 patients last night. *chuckle*”

See patient, NP is now off shift.

Me to ER doc who came on: “Hey, that lady you’re waiting for dispo on has a stye. Nothing to do.”
ER doc: *heavy sigh* “Sorry.”

At least he didn’t drop the “hey I got an interesting one for you” line. He knew he was just dumping his work on you so he could get out. In a way I almost prefer the self insight.

So did the provider taking care of the dehised cornea at least ask and then mention to you that the patient had a history of a corneal transplant? I’m betting “no,” because why would you ask someone with a suspected eye problem if they’d ever had eye surgery
Although sometimes that backfires. Like: any patient who has ever had sinus surgery, even if it was 30 years ago, if they get a sinus infection they get sent to ENT without treatment because no one is “sure” how to treat a patient with a sinus infection if they’ve had surgery before.

Hint: you can treat them exactly the same way and it’ll probably work.

Had a guy yesterday who say his NP for a swimmers ear. She looked in his ear, diagnosed a swimmers ear, then referred him to me with no treatment because in 2019 I put an ear tube in his ear after a post-flight effusion. Tube had been out for a few years. She was the one who noted that it fell out and that his ear looked normal after. He had had no other ear issues since 2019. Not to mention you don’t treat swimmers ear with tubes. So he had a draining ear for a month so that I could order drops for him.

 

[eyeeye_captain]

At least he didn’t drop the “hey I got an interesting one for you” line. He knew he was just dumping his work on you so he could get out. In a way I almost prefer the self insight.

So did the provider taking care of the dehised cornea at least ask and then mention to you that the patient had a history of a corneal transplant? I’m betting “no,” because why would you ask someone with a suspected eye problem if they’d ever had eye surgery
Although sometimes that backfires. Like: any patient who has ever had sinus surgery, even if it was 30 years ago, if they get a sinus infection they get sent to ENT without treatment because no one is “sure” how to treat a patient with a sinus infection if they’ve had surgery before.

Hint: you can treat them exactly the same way and it’ll probably work.

Had a guy yesterday who say his NP for a swimmers ear. She looked in his ear, diagnosed a swimmers ear, then referred him to me with no treatment because in 2019 I put an ear tube in his ear after a post-flight effusion. Tube had been out for a few years. She was the one who noted that it fell out and that his ear looked normal after. He had had no other ear issues since 2019. Not to mention you don’t treat swimmers ear with tubes. So he had a draining ear for a month so that I could order drops for him.

Unfortunately the only self insight was from the ER MD who relieved her. I dream that she got written up for that garbage, but I know it would never happen.

To be fair to the resident on the cornea issue, the patient was non-verbal and I don’t think I’ve ever seen a long term care facility actually bring useful records anywhere. My gripe is that it’s not tough to see half of the inside of an eyeball coming out. Hell, I used to get calls on head trauma with brain plainly visible - maybe we can work on priorities, folks?

Totally feel you on the “someone operated on it years ago so now I can’t touch it“ ridiculousness. There’s a big time hot potato mentality in a lot of settings. Hopefully your referral sources don’t know buzzwords to force emergent/urgent referrals. I’m pretty sick of people calling everything a retinal detachment so it’s my problem immediately and they can wash their hands of whatever is going on.