Turn around and listen to the bright minds of Drs. Annie Jacobs and Christopher Polk investigating two cases of pericardial effusion in the US and Kenya

Episodes | Consult Notes | Subscribe | Twitter | Merch | febrilepodcast@gmail.com

Hey everyone.

Welcome to Febrile, a cultured podcast about all things infectious disease.

We use consult questions to dive into ID clinical reasoning, diagnostics and antimicrobial management.

I'm Sara, your host, and a Med-Peds ID fellow.

I'll introduce our co-host today, Dr.

Annie Jacobs, who you may remember from the ID week 2022 episode.

Carolinas Medical Center in Charlotte, North Carolina.

She recently successfully completed the ID match process and will be starting her ID fellowship at UT Southwestern in July.

Hey, Sara.

Glad to have you back, Annie.

All right, and our guest discussant today is Dr.

Christopher Polk.

He is an Associate Professor of Medicine and Assistant Specialty Director of Infectious Diseases Research in the Division of ID at Atrium Health, Wake Forest School of Medicine in Charlotte, North Carolina.

His research interest include COVID-19, CRE producing infections, and HIV.

Good to be here.

So before we get to the cases today, we like to call Febrile a cultured podcast and ask if you would be willing to share a little piece of culture that you have enjoyed recently.

What do you guys got for today?

So I have been binging on true crime podcasts lately, and I have discovered one called Disappearances by Sarah Turney.

It takes you through a variety of people that have gone missing throughout history, sometimes with a very clear suspect and sometimes more ambiguous.

And I highly recommend it to anyone that enjoys stoking their anxiety with a little bit of true.

So I actually don't listen to podcasts much with one exception, and, and the time I listen

is during my limited hobbies, which are running and hiking and, and so I like to do long runs.

I've done some marathons and, and during that I will listen to febrile.

Oh, thanks.

My favorite podcast.

Well, we have some, we're doing things a little bit differently today, and not just having one case, but more than one.

Uh, so Annie, I'm gonna throw it over to you.

Perfect.

Our first case is a patient who presented to our urban hospital in North Carolina.

He's a 70 year old man with a medical history of hypertension and gastric cancer status post fresection andning chemotherapy well over 10 years ago.

He presented with one week of progressive cough and shortness of breath.

He tells us that two years ago he had a Covid 19 infection, and since then he's had some dyspnea on exertion, but it's gotten significantly worse over the past one week.

It is to the extent that he's unable to ambulate around his home without feeling significantly short of breath.

His cough started about two weeks ago and it was initially dry though in the past two days, he's been coughing up a small amount of clear sputum.

On further history, he shares that his breathing has been worse when laying down and he has to sleep with the head of his bed elevated.

He denies any other preceding or concomitant symptoms including nausea, diarrhea, rash, or fever.

The patient's only prescribed medication is losartan, though he admits that he is been out of this medication for over a month and has not been taking it.

He lives at home with his wife.

He has two adult sons, both of whom live in the area.

He's originally from Honduras and moved to this country four years ago to be closer to his family.

He previously worked in construction but has not worked for the last 10.

He denies a history of smoking, drinking, or drug use.

There's no pets in the home.

He denies any sick contacts or historical contacts with people known to have tuberculosis.

Upon presentation to the emergency department, he was tachycardic, normotensive, afebrile.

He was saturating 94% on room air with respiratory rate of 16 to 18.

Physical exam documented in the ED notes a prominent JVD to 13 centimeters, crackles in the bilateral lung fields, irregularly irregular heart rate with no notable murmurs or extra heart sounds, and two plus pitting edema in the lower extremities.

He underwent E K G and was noted to be in new onset atrial fibrillation with rapid ventricular response to the 160s.

Chest X-ray in the emergency department showed pulmonary edema and massive cardiomegaly.

He was started on diltiazem for rate control.

He was admitted and shortly after admission was started on a apixaban for anticoagulation.

While admitted, he undergoes diuresis and on hospital day two, he undergoes an echocardiogram as part of routine workup for new onset AFib.

It shows a large circumferential pericardial effusion.

We can pause here Dr.

Polk and can you talk about your differential in this?

Sure.

Although before I dive into that, one question would be, were there clues to the presence of his pericardial effusion prior to obtaining that echo?

You know, sometimes hindsight is really 220-200, and many times it's not completely obvious.

But I do wonder for this patient, on exam were their distant heart sounds.

Did he have low voltage onto E K G?

And we are given a chest x-ray that's described as cardiomegaly.

And you know, classically for a pericardial effusion or what is described is a water bottle sign of an enlarged cardiac silhouette.

And so I just wonder about that appearance of cardiomegaly on the chest x-ray and whether there were just some of those subtle clues to the presence of the effusion be before we got the echo.

But since we're talking about differential, the next question really would be, Were there any signs or symptoms of an acute inflammatory syndrome or acute pericarditis, and we're really not given any here.

So in the H P I, you talked about there really wasn't fever or chest pain or ST elevations on the EKG, suggestive of pericarditis.

He does have a cough, but it sounds like it might just be more from his pulmonary edema.

If he did have some of these acute inflammatory symptoms, then I might be more concerned for an infectious etiology, particularly for a pathogenic bacterial infection like Staph aureus, pneumococcus, or other streptococci diseases.

NNeisseria, Legionella, Mycoplasma.

In my experience, patients with those typical bacterial pathogens generally are fairly ill appearing and, and come in sick with if they have acute pericarditis.

In the absence of that, might think about viral pathogens, coxsackievirus, adenovirus, the one we see in adults is usually H I V, uh, which is something to think about.

Fungal pathogens might be on our differential, especially the endemic fungi.

Blastomycosis, Cocci(dioides), histoplasma.

And then of course we can't forget about tb, tuberculosis, and it's really possible for, uh, effusions from TB to really be more indolent with fewer acute inflammatory symptoms, although maybe patients might have weight loss or night sweats.

So thinking through some of the infectious etiologies of a pericardial effusion, I would think through that set of possibilities.

And then I would also think about non-infectious etiologies, especially given his lack of acute inflammatory symptoms.

You know, as ID clinicians, we are often asked to kind of act as a master diagnostician and really consider non-infectious etiologies, and I just often find it particularly satisfying to make a non-infectious diagnosis.

But just sort of thinking through those possibilities here.

You know, what might they be?

Well, he was recently initiated on anticoagulation, so I think we have to think about bleed a little bit, although on this echo he had, there's really no mention of an aortic dissection or leakage into the pericardial sac.

You know, always think about MI, which might have a complication of pericardial effusion at some point, although presumably maybe he had some cardiac enzymes done or was assessed for that given his new onset atrial fibrillation.

You know, we think about uremia in our adult patients as in etiology in this.

Then for this patient, he does have a history of malignancy and malignant effusions might be a consideration.

Although usually malignant effusions would complicate lung or breast cancer, leukemia, lymphoma, probably not as an isolated recurrence of prior gastric cancer.

That would seem to be unusual, but just sort of thinking through that, uh, mixed edema might cause a pericardial effusion, although, if anything with,

new onset atrial fibrillation with rapid ventricular rate.

You might think he's hyperthyroid.

And then always think about connective tissue disorders too in our differential, right?

Lupus, rheumatoid arthritis, granulomatosis with polyangiitis.

Sarcoid, maybe even familiar Mediterranean fever, all of which I would think would have systemic manifestations, uh, rather than an isolated effusion, but just sort of thinking through a full differential would consider those.

And then that sometimes medications or drugs can cause effusions such as hydralazine, but he really doesn't appear to be on those.

So that's a really big differential to both infectious and non-infectious etiologies.

But the truth is, a lot of times these effusions are idiopathic and we don't find a good answer

and maybe that's the answer here too.

So with that differential, Annie, Do you have some more information so that we can maybe narrow it down or focus on where we're going with this case?

Yes, and thank you.

That was very informative.

I do have more information, but I'd actually like to introduce another patient first, if that's okay.

Now you're really throwing some curve ball.

This time our patient is in Chogoria, Kenya.

Chogoria is a small town about 140 miles from Nairobi with somewhat limited availability of diagnostics.

Our patient here is a 23 year old man with no known medical history.

He presented to the hospital with shortness of breath of two weeks duration.

He has a slow but progressive onset of his shortness of breath.

At first he noticed it with extended activity, but over the past two to three days, he's been short of breath, even at rest.

He went to another hospital about a week ago and was told that he had pneumonia.

He was prescribed an oral antibiotic.

He isn't sure which one, and he didn't have improvement.

Further history revealed that he's had chest pain and hemoptysis for two weeks leading up to his current presentation.

The chest pain is sharp in nature.

Anterior in location and exacerbated by inspiration and cough.

His cough is productive of white frothy sputum.

He also notes feeling febrile at home with night sweats.

Over the past two weeks though, he is been unable to take his temperature.

He has no other focal symptoms.

His vitals at presentation show sinus tachycardia, 110 to 120 beats per minute with a blood pressure within normal range.

He's febrile to 103.1 Fahrenheit.

He's tachypnic to 24 breaths per minute, but satting well on RA, he's ill appearing in diaphoretic.

Physical exam is significant for the tachycardia and soft heart sounds.

There's no murmurs, rubs, or gallops.

He's using accessory muscles for respiration with subcostal re retractions.

There's no notable rashes and no lower extremity edema.

He has palpable lymphadenopathy.

Mental status was intact.

On the initial assessment.

Labs showed elevated sedimentation rate of 124 millimeters per hour.

His chest x-ray shows blunted costophrenic angles with massive cardiomegaly.

Ekg showed sinus tachycardia with classical electrical alternans.

Point-of-care ultrasound confirms a massive pericardial effusion with RV collapse.

With this initial presentation, what are your initial thoughts, Dr.

Polk?

How is this similar and different than our patient in North Carolina and how does your patient's differential change with this specific history?

Yeah, so this patient really had very clear acute onset of inflammatory symptoms.

So he sounds sick, ill appearing.

He had fevers and chest pain and cough.

That's all new and onset.

And so I would really lean towards an infectious etiology here much more clearly than in our first patient, again, especially given his acuity.

So I would favor a pyogenic bacterial infection, maybe tuberculosis or fungal infection.

And of course, given the location, everyone is really thinking TB.

But I think we have to stop and really do due diligence and consider the the full infectious differential and think about staphylococcal disease, pneumococcal disease, meningococcal disease, all of which are certainly quite possible here.

Of note, where Chogoria is in Kenya is not in the meningitis belt.

It's a little south of that, but you know, certainly would remain a consideration.

Uh, histoplasma also can occur in Africa, so might consider that as well.

I also wanna comment briefly on this patient having cardiac tamponade, which I think is interesting and we didn't really discuss so much in the first case,

still was reported to have a large effusion, but is obviously a very important diagnosis to make and assess for.

And you know, in tamponade, maybe more related to the rapidity of fluid accumulation than the size, and maybe we suspected when there's J V D in the presence of hypotension, which the second patient really didn't have described interestingly, we might also suspect that if there's electrical alternans on the ekg, but if you don't have a handy point of care ultrasound to look for tamponade, the other quick bedside test you can do is a pulses paradoxus so here you check the patient's blood pressure doing inspiration and expiration and

systolic blood pressure drops at least 10 millimeters of mercury.

That's a concern and, and I will just briefly comment, the only time I was ever reprimanded as a resident is when I didn't do this, when admitting a patient and the program director came in the next morning and the patient was in tamponade.

So it's a quick, easy bedside test you can do that.

I'll certainly never forget if you don't have that point of care ultrasound.

Um, so are, are we going to find out more about either of these patients, Annie?

Yeah, we can do that.

Let's go back to the United States.

You'll recall our patient is a 70 year old with a history of gastric cancer who came in with shortness of breath and was found to be a new onset AFib.

And on echo, he was noted to have a massive pericardial effusion.

This patient was initially hemodynamically stable.

Recall that we talked about the differential, including malignancy, infection, and hemorrhagic sources of this effusion.

I'll add some basic labs at this point.

His C B C showed a white count of 12,000 with a hemoglobin of 10 and platelets of 400.

His electrolytes were all within normal limits, a Cr of one and normal BUN.

His a s t and a l t were unremarkabe.

He did undergo a CT chest in the emergency department.

The circumferential effusion is again noted, and he had no lesions in the lung parenchyma, cavitary, or otherwise.

He had trace plural, effusions, far too small to consider tapping.

It was actually at this point in the hospital course that infectious diseases was consulted.

As the ID consultant, what would you suggest in terms of diagnostics in this patient at this point in his hospital?

Well, you know, as an ID consultant, the first thing we do is go to bedside and take a better history, right?

We really try and tease out all of those crazy details that no one else probably cares much about.

So, you know, thinking back through this, these pa, this patient, some of the other questions I might ask would be related to further history of any TB exposures.

We told, we were told he really didn't have any, but was from, uh, central America.

Were there any experiences with incarceration or homelessness or other risk factors we can tease out?

Uh, we might think about that in the context of risk factors for endemic fungi as well of.

He was here in Charlotte, where we have a little bit of Histoplasma from time to time, but he was from Central America where there's sort of increasing rates of histoplasma.

So we might think about that.

We might ask about family history of rheumatologic illness, which might.

Provide some clues and, and then we would really wanna embark on a further workup.

We really have some basic labs here, which is great.

He probably doesn't have a uremic effusion.

We know that now.

But what about an H I V test?

It was negative.

Okay.

And rheumatologic serologies, presumably those might be.

Yep.

They were all sent by the admitting team and they are in lab being processed.

Of course.

Um, we do have a CT chest.

You gave us some imaging and it doesn't seem to suggest any malignancy or recurrence of malignancy, which I think is, uh, helpful here.

You would think there would be some other abnormality in the chest.

Uh, to suggest that as there isn't about a third of the cases of pericardial effusion from TB or from malignancy, sorry.

I presume maybe he had blood cultures at some point since we were talking about bacterial pathogens.

Yeah, they were collected in the emergency department, no growth after 24 hours.

Okay.

And, and then the other question is, do we send testing for TB or latent tb, such as an interferon gamma release assay.

Right, so an IGRA doesn't diagnose active tb and there's always this question about what do you do if it is it's positive or negative, particularly with patients from a highly endemic area, but it certainly might raise your suspicion if it was, uh, positive.

I'm not sure it would completely exclude the diagnosis if it was negative.

You know, just looking at some case series from tuberculous peritonitis, a positive igra has a sensitivity and specificity of around 85%.

So that's a positive predictive value, about 90% negative predictive value, 70%.

So it might raise your suspicion for tuberculosis, um, if it's positive, but again, Really excluded from the diagnosis, either if it's negative.

Other things we might do really boil down to obtaining fluid or tissue for diagnosis here.

Right?

And pericardial fluid, while helpful is only diagnostic and about 40% of the time pericardial effusion with from tb.

So something to keep in mind and we can't really use lights criteria on pericardial fluid the way we can for pleural effusions.

That's part is not helpful.

But what we can send, obviously our AFB smears and cultures, and then an adenosine deaminase test.

Which maybe is worth talking a little bit about.

In addition to TB pcr, adenosine deaminase is part of purine metabolism that is elevated in TB and given off by lymphocytes.

It's not exclusive to tuberculosis, but is helpful

to make a diagnosis of a ppleurall or pericardial effusion from tb.

So I think it might be really helpful here to send that.

The other gold standard we can think about is doing a pericardial biopsy, and we classically think about this as increasing yield for making a diagnosis of TB from in the pericardial space.

Unfortunately, it's maybe a little less sensitive than we like to think of it as.

So an AFB smear from pericardial fluid, fluid is only about 5% sensitive, culture is about 50%, and a pericardial biopsy does increase your sensitivity, but only to about 65%.

So making the diagnosis here is a little bit of a challenge no matter what modality you use, but you probably do wanna obtain some fluid and try and think through sending some of those tests again, particularly the adenosine deaminase or the TB pcr, which may have a sensitivity of up to 90%.

And then finally, you could also think you about sending a T-SPOT from your fluid, which also is pretty sensitive, but about 90%.

So that's some of the things I would sort of think through in workup of the patient from the us What, what about that patient in Kenya?

Yeah, well, I will pass those recommendations onto the primary team.

But going back to our patient in Kenya, with that massive pericardial effusion, our patient has become hypotensive to seventies over fifties, in tachycardic to the 130s.

He's also become increasingly somnolent.

He's definitely not stable.

The patient undergoes an emergent bedside pericardiocentesis with placement of a temporary dialysis catheter to serve as a drain.

One liter of dark bloody fluid was drained in the first 30 minutes after placement.

The patient's hemodynamics improved.

Soon thereafter, the patient was started on empiric RIPE therapy.

Over the next three days, a total of three and a half liters of fluid was drained from the pericardial drain throughout the hospital course, the patient clinically improved with a complete resolution of that altered mental status.

Gene expert testing of the pericardial fluid did confirm a diagnosis of tuberculosis.

Dr.

Polk, what are your thoughts about this?

Yeah.

No one's surprised here, right?

That that's what everyone was kind of thinking a little bit, especially given the location.

That being said, I would go back to, don't jump to conclusions, do the workup.

I, I'd also say that this is a great demonstration of how epidemiology matters, right?

So in case series of pleural effusions from the US and Europe, Then most are either idiopathic or from malignancy with really the minority being from infection.

Alternatively, from South Africa, a case series published on pericardial effusions identified TB as the most common cause.

So epidemiology always matters.

Absolutely.

That was a really satisfying textbook case from Kenya.

But let's jump back to the US.

With our 70 year old, with our pericardial effusion of unknown etiology, the patient did undergo a pericardiocentesis with one liter of bloody fluid removed cultures and a f B were sent.

Notably that TB P C R that you recommended was not sent in this patient.

No organisms were seen on the initial gram stain.

What treatments, if any, are you starting on this patient at this point in the hospital course and at what point did we consider empiric tuberculosis therapy in this patient in the United States?

Well, this is interesting because now we're given information that this was really a large hemorrhagic effusion, and when we think through our differential as far as the etiologies, Of pericardial effusions.

There are only a few things that probably give you a large hemorrhagic effusion most likely, and those are bleed, malignancy and tb.

And we already kind of excluded bleed.

And the fact that he really didn't have trauma here or mention of dissection or concern on echo.

So we're down to thinking about cancer or tb and we've already talked about how it would be unusual for gastric cancer to recur just in the pericardium and from the chest imaging we have.

We don't see a suggestion of other chest cavity malignancy that might go with a malignant effusion.

So we're down to thinking about tuberculosis again, and the fact that he did have a prior residence in a country with higher endemicity of tb.

I'd also say we previously talked about sending an interferon gamma release assay here, and I'm guessing that's still not back yet.

Still in lab

and, and that ADA is still in lab?

Yep, still in lab.

But hopefully it was at least sent unlike the TB PCR.

You know, at my institution, those usually come back within a week or so, and he's relatively clinically stable.

So to your question, as far as empiric therapy, I'm not sure there's a huge rush.

That being said, if he were sick, certainly would consider starting therapy for a TB and regardless, given where it is on our differential, I presume he's in airborne isolation and we're trying to get three sputums for a f b smear at this

point.

Yes, we are working on it.

Mm-hmm.

So our patient did undergo some workup that you asked for earlier in this conversation.

His rheumatologic workup came back with a negative a n a rheumatoid factor in c c p, and complements were within normal.

For oncologic workup, he underwent a total body PET scan without hypermetabolic activity anywhere in his body.

His infectious workup was a covid and respiratory pathogen panel that came back negative.

The H I V that we already mentioned came back negative.

He had a negative A f B sputum smears.

Three were collected, specifically negative blood cultures, but his QuantiFERON did return.

The pleural fluid was sent for analysis and the ADA was elevated at 55 milliliters.

What do we make of diagnosis at this point?

So we're back to tuberculosis just in a different location, and of course we're gonna start him on therapy at this point with what we term is ripe.

Rifampin, isoniazid, pyrazinamide, ethambutol with a little pyridoxine thrown in for.

Prevention of toxicity and you know, certainly want to still get those smears.

He'll, he'll need referral to the health department and ongoing treatment.

But just thinking about TB pericardial disease in general, since we have these two cases, it's interesting because it's really rare.

It's less than 5% of cases of TB present this way and unfortunately, Classically, it had a fairly high mortality rate prior to sort of modern, effective therapy, and still we worry about the complication of constrictive pericarditis because in the later stages of pericardial.

Uh, TB disease, the fluid gets reabsorbed, and then there's scarring of the pericardium, granuloma formation, and constrictive heart physiology can occur.

So that's a concern.

Certainly treatment, uh, with ripe is indicated there.

And then sometimes steroids have been given and the, this is maybe a little controversial, and the, the data.

Inconclusive, but they might be helpful early.

Although studies really haven't completely demonstrated definitive benefit in preventing constrictive pericarditis, but it is a concern.

And if there's progression in some patient, they may need surgical pericardiectomy for constructive pericarditis, but two really interesting cases in different locales

leading to the same diagnosis.

Thank you so much to Annie and Christopher for making this awesome episode spanning two cases across two continents.

Don't forget to check out the website febrilepodcast.com to find the consult notes, which are written complements to the show, with links to references, our library of ID infographics, and a link to our merch store.

Please reach out if you have any questions, suggestions for future shows, or wanna be more involved with Fbri.

Thanks for listening.

Stay safe and I'll see you