Hi, 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 Dong, your host and a Med-Peds ID fellow. Here on Febrile, we use patient cases and chat with ID discussants to learn more about high yield topics. Our co-host today is Dr. Pratik "Tik" Patel. He is a second year pediatric ID fellow at Emory University and Children's Healthcare of Atlanta.
He also completed a Pediatric Hematology Oncology fellowship, also at Emory, and wishes to leverage his training in both fields to advance the ID care of immunocompromised children with a focus on those undergoing treatment of cancer and stem cell transplant. He also has a research interest in introduction and implementation of novel diagnostics for improved stewardship and clinical care.
Hey everyone.
Our discussant today is Dr. Joshua Wolf. Josh is a Pediatric Infectious Disease physician at St. Jude Children's Research Hospital, where he is the Division Director for Hematology and Oncology Infectious Diseases and Medical Director of Antimicrobial Stewardship.
He is an Associate Professor at the University of Tennessee Health Science Center, and he trained at the Royal Children's Hospital and Peter McCollum Cancer Center in Melbourne, Australia and St. Jude and Le Bonheur Children's Hospital in Memphis, Tennessee. His research interest is focused on novel approaches to prediction, prevention, and amelioration of life threatening infections in children with cancer.
Hi.
All right. I'm so glad you guys are here. Before we get to the case, we always ask one question, uh, meant to be non-medical. As everyone's favorite cultured podcast, I'd love to hear if you guys have any pieces of culture or things that you have enjoyed recently that you wanna share with the listeners.
Sure I can go first. Um, I enjoy travel hacking. Uh, so this stems from a passion of mine, which is I love to travel. Um, but I don't like to spend a lot of money . So with travel hacking, I use credit cards to maximize, uh, points and miles. So I travel on the cheap. Uh, I've churned, an official travel hacking term, through over 30 credit cards and been to over 30 countries.
Wow.
So I'm always plotting out my next trip.
Oh, so this is not like a casual, like you're serious travel hacker.
yep. As much as one can be. Yeah.
What about you, Josh?
Uh, mine couldn't be any more different and I have a seven month old daughter. And so I'm really into a book at the moment called, That's Not My Squirrel , which is a touchy feely board book.
Great, great. well, it's very pediatric appropriate. So today's consult question. Um, you get a call about a teenage boy undergoing treatment for cancer with respiratory distress, who unfortunately has been admitted to the ICU overnight. Can you please assist with evaluation and antibiotics? So I'll hand it over.
So we have a 17 year old male, uh, who has a history of a relapsed/ refractory brain tumor who came to the ED with three days of tachypnea and hypoxia. It started as a sore throat five days ago, and then seen, was a, was seen at the PCPs office at which time a rapid flu, COVID, Strep tests were negative. Unfortunately, continued to feel bad with tachypnea, mild sputum production. And so his mother used a home O2 pulse ox, and noticed he was saturating 87 to 89% on room air. So came.
So he came to the ER. In the ER, he's a febrile, mildly tachycardic and hypotensive. And so was given a fluid bolus, had respiratory distress with tachypnea and accessory muscle usage and he was setting 75% on room air. So he was placed on high flow nasal cannula at 12 liters with an FiO2 of 40%. Chest x-ray was done, which showed bilateral patchy infiltrates and a respiratory viral panel was positive for rhinovirus.
Upon other laboratory examinations, a complete blood count showed a white blood cell count of 3.8, a hemoglobin of 10 platelets of one 90, and a chemistry panel showed an elevated AST at 80, ALT of 45, normal bilirubin and normal renal function. Inflammatory markers were elevated with a CRP of 13 milligrams per deciliter, and an ESR of 81. And the patient did not have a central venous line. So peripheral blood culture was drawn.
He was started on empiric antibiotics of piperacillin-tazobactam and azithromycin and admitted to the oncology floor. And very quickly on this first day of hospitalization, he became febrile, had worsening hypoxia. So he was transferred to the pediatric intensive care unit for BiPAP and infectious disease was consulted. So here we have an immunocompromised critically ill teenager in the ICU with fever and respiratory distress.
What are you thinking, Josh, when you're scanning the chart before seeing the patient, what are you looking at? And what questions will you walk into the room already planning to ask.
Thanks Tik so this is an immunocompromised adolescent with fever and acute respiratory failure following a sore throat, but immunocompromised ain't immunocompromised. And so, I wouldn't expect him to be uh, profoundly immunosuppressed as we'd see in a patient post hematopoietic stem cell transplant or intensive therapy for leukemia. And that does affect the differential, but different people respond differently to chemo. So I'll keep an open mind.
Before I go in the room, I'll try and get a picture of his recent chemotherapy and review his blood count to get a feel for the depth and the characteristics of his immunosuppression. In terms of this possible differential diagnosis, it's in two groups, infectious or non infectious. And, um, the infectious differential here is quite broad. Uh, it can include viral, bacterial, fungal, or even parasitic infections.
And it's important to consider the common causes of pneumo- pneumonitis or lower respiratory tract infection in any host. Um, and then some that are special for immunocompromised hosts. And in here, I'm thinking about respiratory viruses like RSV, adeno, parainfluenza, reactivation of herpes viruses like CMV or HSV, uh, bacterial pneumonia with pneumococcus, mycoplasma, Staph aureus, or even Pseudomonas if neutropenic.
And then other bacteria like psittacosis, mycobacteria, uh, nocardia, legionella are all really rare in this setting, but with might consider them under certain circumstances. And then, um, locally here, fungi like Histoplasma and Blastomyces are considerations. And pneumocystic pneumonia needs to be on the, uh, on the differential.
Other opportunistic fungi like disseminated candidiasis or aspergillosis would be less likely in this patient because of his degree of immunocompromise and then strongyloides or toxoplasma usually affect much more immunocompromised hosts. We've certainly seen other parasites like toxocara in this. And then on the non-infectious diseases side, things like chemotherapy side effects, like, uh, gemcitabine bleomycin methotrexate, usually not acute or febrile.
And then there are other unusual things like vaping associated lung disease, um, which again, doesn't quite fit with the picture. And then lastly, and really importantly, I think it's difficult to pin this on a, a picorna RNA virus, like rhinovirus or enterovirus. That's a very rare cause of lower respiratory tract infection. And I keep looking.
great. Yeah, that's a great differential. So in terms of our exposure history, the patient was originally diagnosed with a brain tumor four years ago, and subsequently relapsed two years later, he was treated with surgery, radiation, and chemotherapy. He had progression a year af- a year prior to his presentation. And so now he's on a clinical trial with oral chemotherapy, which includes cyclophosphamide and etoposide. His vaccines are not up to date due to, uh, cancer therapy.
He lives in the Southern United States with his parents and a brother who recently experienced gastroenteritis. He's had no recent travel due to social distancing in quarantine, but he has gone to Europe in the past and he has no pets. He has gone swimming at a local lake and there has been some recent remodeling of his household. As far as his labs, specifically, his absolute neutrophil count is 2,500. And in review of his previous records, he has not been neutropenic anytime in the last year.
And then his absolute lymphocyte count is 300 and importantly, he has not been above 1000 in the last six months. He reports taking trimethoprim-sulfamethoxazole or Bactrim prophylaxis Saturday and Sunday without missed doses. And he's on no other infectious prophylaxis. So with this information, how does this change your differential and what recommendations will you make to the team about next steps in management?
I presume that the rest of the exposure history is negative. I'd also be asking about hobbies, hunting, vaping, animal contact, other than pets. It's so amazing how often they, uh, swept outta chicken coop, but didn't mention it because they're not pets. Uh, if he's had raw milk, game meat. Um, so, so I'll take a full exposure history. It's really essential in a case like this, to make sure you're not missing something. And there are also a few exam findings I be interested in.
Does he have lung crackles? Uh, does he have, whe the nature of these can sometimes help, like, fine crackles and paroxysmal coughing with deep inspiration, especially if it, they have desaturation might be more likely to be something like Pneumocystis.. Does he have neck tenderness to suggest jugular vein thrombosis? Does he have oral thrush or ulcerations? Those are all things that can give you clues to what's going on, uh, from the exam.
And then living in the rural Southeast Eastern US is theoretically a risk factor for Strongyloides but it's very rare in my experience. And we wouldn't expect hyper infection syndrome in someone with his degree of immunocompromised. Um, And then he isn't neutropenic. And so that does, um, push me away from unusual bacterial infections, but the prolonged, uh, lymphopenia, even though it's not profound, does raise, uh, his risk.
In HIV patients, we think about prophylaxis for pneumocystis below a CD4 count of about 200 and mycobacteria or CMV. Under CD4, kind of about 50. So is in that range of moderate T-cell or lymphocyte deficiency. And you might also have some antibody deficiency. See, this is really typical for solid tumor patients getting non-intensive chemotherapy.
And so I put a, uh, him in this, uh, pure area of immunocompromised from the standpoint of lymphocyte deficiency, moderate, um, neutrophil and neutropenia doesn't seem to exist right now. And hasn't been in the recent past and antibody deficiency. He may have some degree of antibody deficiency and, my workup for him is gonna be pretty broad upfront. This kid is sick and getting sicker and, um, it's happening quickly.
And so I'm thinking about respiratory viral, uh, panel, which I I think has already been done. And in this context, I'd send labs, Histoplasma, Blastomyces,, CMV, adenovirus. I'd love to see a chest CT. And this is one of the really rare times that I think about sending a beta-d-glucan test in general, BDG, is contraindicated in children for diagnosis of candidiasis or for diagnosis of other invasive fungal infection because of its very low positive and negative predictive value.
In one really well conducted study, the positive predictive value of beta D glucan was 0%, but an extremely high beta D glucan can, can be indicative of pneumocystic pneumonia. I'd also be scouting the discussion about bronchoalveolar lavage. I think this kid's probably gonna need a BAL at some point.
Yeah. So the rest of the exposure history was unremarkable. No reports of frolicking through chicken coops. The lung exam was notable for decreased breath sounds in bases with a few end expiratory wheezes and no neck tenderness or oral lesions. He had urine legionella antigen, urine histoplasma antigen sent and serum mycoplasma titers and blood viral PCR for CMV, E BV, and adenovirus.
He did have a 1,3-beta-D-glucan sent and his chest CT with contrast showed bilateral extensive airspace disease. Hmm.
The story is a good one for pneumocystis pneumonia, with his lymphopenia, bilateral airspace disease and rapid regression of hypoxia to respiratory distress to respiratory failure. But you said that the patient was taking prophylaxis. Is that correct? I really, I would make sure that someone from our team goes and asks.
Yeah. And actually on the chest CT, the radiologist was concerned too. So the team went and asked the family about prophylaxis and the family admits to frequently missing doses and that he had actually run out of tablets last month and hadn't had a refill. So given this news, what would you recommend to the team in terms of management and any further workup?
Wow. This really does push me towards a, a presumptive diagnosis of pneumocystis pneumonia. Although I wouldn't write off anything out. Um, so we are gonna start empiric therapy for pneumocystis while we figure this out. Um, and I, I think one really important take home point is that first, second and third line therapy for pneumocystis pneumonia is trimethoprim sulfamethoxazole. It's the best treatment far and away.
And although it's usually well absorbed after oral administration, I typically start with IV dosing until I start seeing some improvement, especially in someone as sick as this I'd give, uh, five milligrams per kilogram, three to four times a. lower doses might be effective in milder disease, but it's not that well investigated. Um, and the role of steroids in a setting like this, if a severe pneumocystis infection in patients without HIV is really poorly understood.
The big issue is that after you start treatment with antibiotics, the patients almost always get worse for about 48 hours before they turn the corner. And corticosteroids might prevent some of that deterioration. I, I almost always started into patients in the ICU with respiratory failure, but given the lack of evidence in a favor of it, if there's any contraindication or, uh, any changes in that suggest a steroid side effect, I have a low threshold for holding off or discontinuing it.
And then the last thing is we're gonna work hard to confirm this diagnosis.
Yeah. And, and speaking about that, why is it so important to confirm the diagnosis of pneumocystis?
Great question Tik. For one thing, you can stop alternative therapies and stop looking for alternative ideologies. But most importantly, high doses of trimethoprim sulfamethoxazole are really tough to take. It causes a lot of nausea. So in two weeks time, when he's throwing up or requiring a lot of anti emetics, it'll be good to be sure that we're treating the right thing.
Uh, I think for his workup, I would be pushing for a bronchoalveolar lavage, or BAL given that other ideologies are still possible. Um, I'd be sending it for bacterial, fungal, viral, uh, and, uh, pneumocystis testing. And then the other test that seems to be potentially useful is metagenomic sequencing directly from blood, which is now commercially available in cases where BAL is impossible to get or potentially dangerous. This confirms the diagnosis of Pneumocystis.
It's not first line because the sensitivity hasn't been that well evaluated yet, but it appears to be relatively specific.
Interesting. So, yeah, so he had a bronchoscopy with BAL, which was sent for the studies you mentioned, and the patient was started on IV trimethoprim-sulfa or Bactrim at treatment dosing with corticosteroids. And then three days later, his PJP PCR returned positive cytology done on the BAL was positive with cysts on the GMS stain or the silver stain and his one, three beta D glucan returned positive at greater than 500 picograms per ML.
So this is a really classic case of, of, uh, pneumocystic pneumonia. Often we get only a handful of those tests positive. The cytology is often negative. Um, it can be difficult to get a BAL and the one, three B glucan being positive, such a high value is, is strongly suggestive.
Yeah. And so I think this was a case of PJP pneumonia in an immunocompromised teenager. So to introduce the bug, it was initially thought to be a protozoa and named Pneumocystis carinii, and now reclassified as a fungus with a new name. So there's whole specific strains with Pneumocystis carinii. Now the rat pathogen and Pneumocystis jirovecii is the human one.
Well, everyone's ID favorite, when we have to change the name of a bug,
I've had to change from saying PCP to PJP or, and they promised us, this is something you don't know, when they changed the name. They promised us we could keep saying PCP, that it would stand for pneumocystis pneumonia. What happened instead was that young people started saying PJP and then old people who were saying PCP, the young people thought we just didn't know that it had changed and kept correcting us. Or they would just point pointedly say, well, PJP.
It's terrible being old it sucks man. Okay. Sorry.
No worries. So this infection was first identified in malnourished, premature infants in Europe, in the 1950s by Jirovec and Vanek. And it was the first opportunistic infection associated with the aids epidemic at the start of 1981 and of the first 1000 cases of aids, 50% had a diagnosis of PJP. PJP has an interesting life cycle. There's two forms, the cyst and trophic form. Interestingly the GMs or the silver stain stains for the cyst form.
And it seems to have a pretty, um, strong predominance or seropositivity in the, in the population. So if children, um, by age four, about 75% were seropositive for PJP. And a similar rate of colonization has been noted in autopsies. Of adults, uh, adult of adult lungs. So Josh, what are the risk factors for PJP related infection?
Pneumocystic pneumonia is an opportunistic infection. It, it almost always causes only pneumonia. Although infections that other sites have been, uh, described. The important risk factors are, um, being HIV positive with, um, progressive lymphopenia and especially with a CD four count under 200. but that's a rare risk factor in pediatrics now in in high income countries, because kids with HIV are usually relatively well controlled.
Other immunocompromised patients are also at risk and now make up the predominant source of patients with pneumocystic, pneumonia, uh, hematological malignancies, especially those acute lymphoblastic leukemia and solid or brain tumors within that population. Corticosteroid use is a really important risk factor. Transplant patients, either bone marrow transplant, or solid organ transplant.
Some patients with primary immune deficiency, especially severe combined immunodeficiency, and then other patients who are immunosuppressed, such as those with chronic arthritidies or inflammatory bowel disease, especially with the use of more immunomodulatory medications. Tcell active agents, like alemtuzumab are the highest risk.
And even though we typically think of T-cells as the most important for preventing Pneumocystis infection, agents that profoundly deplete B cells like rituximab or blinotumumab also increase the risk because of the interaction between B cells and T cells.
We talk about a prednisone equivalent of two milligrams per kilogram per day, or greater than 20 milligrams per day total as sufficient to create a risk scenario, how it's unclear, whether corticosteroids alone are important risk factors in Pneumocystis in children. We do usually recommend prophylaxis, but there are some populations that have not received routine prophylaxis and seem to do okay.
One example is boys with muscular dystrophy who receive corticosteroids for prolonged periods of time and don't seem to get pneumocystis. It might be a question of dose or of the complimentary immunosuppression from different agents that are given as well.
Great. So I know St. Jude has a long history in the study of PJ P and the investigation of treatments and prophylaxis. Can you tell us more about this?
So certainly St. Jude, uh, has a long history of being involved with discovery around Pneumocystis. Um, in the 1960s, when we were just starting to use combination therapy for acute lymphoblastic leukemia, the first successes in, uh, cure of leukemia came with the cost of Pneumocystis in a very high proportion of cases.
By the time we were in the 1970s and combination therapy for acute leukemia became routine, we were seeing very high rates of morbidity and mortality from pneumocystis infection and at St. Jude, Walter Hughes, who was the Chief of Infectious Diseases here started looking into this, both in the lab and in the clinic, he discovered that
rats affected with pneumocystis, you could prevent pneumocystis infection and prevent death from pneumocystis infection with either Pentamidine or trimethoprim sulfamethoxazole. And he ran a clinical trial here, the only real clinical trial that has been, uh, done for trimethoprim-sulfamethaxazole against placebo for pneumocystis prevention in, uh, the mid seventies where he randomized patients with acute lymphoblastic leukemia to receive, uh, trimethoprim-sulfamethaxazole or placebo.
And in the, after the first year of the study, it was clear. There was a benefit in terms of reduction of Pneumocystis. And a discussion took place about whether to unblind the study or to complete the two years that they had planned to run it for. He argued very strongly that this was our only chance to run a blinded placebo controlled study for pneumocystis prevention.
And he was concerned that potential harms from the drug wouldn't become apparent until the second year of the study, he knew that the, uh, chemotherapy was gonna go on for a long time and he didn't wanna give this drug just for a short time and observe it without, uh, seeing whether there was long term harm. And a decision was made to complete the study. They didn't unblind until, uh, two years into the study.
And at that point they showed that, um, Pneumocystis was almost 100% prevented with trimethoprim-sulfamethaxazole. He then went on to do, um, different randomized studies comparing, TMP-SMX against pentamidine for prophylaxis and two different regimens of TMP-SMX showing that, um, three times a week was as effective as daily TMP-SMX, but with reduced toxicity.
Well let me ask you about that. What are your thoughts on optimal trimethoprim-sulfa (Bactrim) prophylaxis for PJP nowadays? I see people doing daily, three times a week, either consecutive days versus Monday, Wednesday, Friday, two times a week. Like our patient was doing on the weekends. Is there a correct regimen or do they all work? And then how long should prophylaxis be continued for different populations?
The randomized controlled evidence really best supports three consecutive days per week, uh, over daily as providing a balance between extremely high efficacy and low toxicity. But a lot of other regimens such as once or twice weekly or three non consecutive days are all supported by observational studies and probably work. At St. Jude, we give three consecutive days a week and our feeling is that it allows for some non-adherence, it's well tolerated and very effective.
How long do you need to continue? We typically continue until after completion of therapy and steroids. And we continue about six weeks after this and we'll often give longer therapy if they're receiving therapies that deplete immune cells. For transplant patients, it's a little more complicated.
They'll continue as long as they're receiving immunosuppressive medication and there's some interest in this population using T-cell subset analysis to stop prophylaxis, but it's really not well established outside the HIV world where a stable CD4 count above 200 is used as a cutoff to discontinue prophylaxis in adults and older children, at least.
Yeah. Thank you for that. So I think that's very important to keep in mind, especially in this case, because there was suboptimal prophylaxis. Now, what are the options if there is TMP-SMX intolerance, adverse events or a contraindication?
I think there are a few things to think about in this setting. Firstly, all of the alternative prophylaxis regimens are likely to be inferior protection against pneumocystis. Secondly, contraindications for TMP-SMX are frequently over called. Marrow suppression from three days a week therapy is overdiagnosed and the drugs should be re trialed if that's ever a possibility. And then thirdly, a break from prophylaxis for up to 14 days is safe.
So it's okay to stop the drug and not start an alternative for a couple of weeks while you sort out whether to re challenge. But if we need an alternative aerosolized or IV pentamidine, oral dapsone or oral atovaquone are all similarly available and effective.
and now million dollar question. Do you think of the alternatives in terms of priority? Like would you go to one versus the other or are they all the same?
Yeah, my personal approach is to use Pentamidine and I, I used IV or aerosolized without a clear preference for one over the other, unless one's contraindicated. As the next line of prophylaxis, we recently looked back at the St Jude experience with IV and inhaled pentamidine and found that it was well tolerated. And the breakthrough infections, even using a very broad definition to catch all possible cases was really rare.
Dapsone has cross reactivity with trimethoprim-sulfa so we don't use it in allergy. And atovaquone solution tastes terrible. So adherence is pretty tough in kiddos, but any of these are acceptable and I don't think there's a clear winner.
Awesome. Well, how about we get into something a little bit more controversial. So let's say our patient getting therapy did not improve significantly in the ICU, continuing to acquire mechanical ventilation after starting treatment for a few days. Is there such thing as PJP resistance to TMP-SMX? And if so, are there other treatments you would consider a second line or salvage therapy?
There's definitely no clear evidence of clinically relevant resistance to TMP-SMX. There are certainly mutations that, uh, decreased the, um, in vitro effect of TMP-SMX on Pneumocystis but it's not translated into clinical failure. And so there's no role for resistance testing. If there's no response to therapy after four to seven days, I might consider switching or reconsidering my diagnosis. Have I got the wrong diagnosis or is there a co-infection.
Second line therapy is with clindamycin plus primaquine or with IV pentamidine, but it's really rarely required in clinical practice. They usually start turning the corner within that period.
Great. Thank you. And lastly, I've seen some literature describing the use of echinocandins with PJP. Can you talk a little bit about that?
Yes. The cyst form of Pneumocystis does have 1,3-beta-D-glucan in the cell wall, and there's in vitro and animal data suggesting that it might improve outcomes. There's certainly case report data and some retrospective study that suggests that caspofungin plus TMP-SMX might be superior to TMP-SMX alone. But really I think that the evidence is very weak right now. I don't use it routinely.
It's something I might consider, uh, adding in a case of poor clinical response and certainly a echinocandin is a well tolerated and a reasonable thing to try. Uh, it's not part of my usual practice yet. . Yeah.
Well, I am so grateful that you guys both came on to talk about this, cuz I think that I maybe often we don't realize how little sometimes we know about translating some of the data from PJP into other immunocompromised host. Um, and I think a lot of times we actually don't see this that much anymore, but at the end I always open it up to check to see if there's anything you guys think we missed or any sort of final, final parting words.
Often we focus on neutropenia for our patients with cancer and risk of infection. But as demonstrated in this case, I urge folks to pay attention to the rest of the differential as well, including absolute lymphocyte count.
I agree Tik, getting to a, a nuanced understanding of the nature and the degree of immunocompromise is essential for an individual patient or a group of patients thinking about which components of the immune system are compromised, how profoundly and for how long should influence your management, it makes you a better doctor and it can help you pass boards.
Well what an ending. That's a perfect description of what I hope a lot of the Febrile episodes can do. Um, so thanks to Josh and Tik for joining Febrile today. Don't forget to check out the website, febrile podcast.com, where you can find the Consult Notes, which are written complements of 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 suggestions for future shows or want to be more involved with febrile.
Thanks for listening. Stay safe, and I'll see you next time.