Infectious Diseases


This episode will help you recognize cellulitis and even differentiate it from erysipelas which is totally a different thing. You’ll also learn about treatment, whether or not a blood culture is necessary, and a whole lot more.


@PEMTweets on… sigh “X” (Twitter)

My Instagram

My Mastodon account @bradsobo


Chen AE, Carroll KC, Diener-West M, Ross T, Ordun J, Goldstein MA, Kulkarni G, Cantey JB, Siberry GK. Randomized controlled trial of cephalexin versus clindamycin for uncomplicated pediatric skin infections. Pediatrics. 2011 Mar;127(3):e573-80. doi: 10.1542/peds.2010-2053. Epub 2011 Feb 21. PMID: 21339275; PMCID: PMC3387913.

Daniel J. Pallin, William D. Binder, Matthew B. Allen, Molly Lederman, Siddharth Parmar, Michael R. Filbin, David C. Hooper, Carlos A. Camargo, Clinical Trial: Comparative Effectiveness of Cephalexin Plus Trimethoprim-Sulfamethoxazole Versus Cephalexin Alone for Treatment of Uncomplicated Cellulitis: A Randomized Controlled Trial, Clinical Infectious Diseases, Volume 56, Issue 12, 15 June 2013, Pages 1754–1762,

Liu C, Bayer A, Cosgrove SE, et al. Clinical practice guidelines by the infectious diseases society of america for the treatment of methicillin-resistant Staphylococcus aureus infections in adults and children. Clin Infect Dis 2011; 52:e18.

Stevens DL, Bisno AL, Chambers HF, et al. Practice guidelines for the diagnosis and management of skin and soft tissue infections: 2014 update by the Infectious Diseases Society of America. Clin Infect Dis 2014; 59:e10.


Note: This transcript was partially completed with the use of the Descript AI

 Welcome to another episode of PEM Currents, the Pediatric Emergency Medicine Podcast. As always, I’m your host, Brad Sobolewski, and today’s episode is all about cellulitis. What is it? Well when a break in the skin occurs, normal skin, flora, and bacteria can enter the subcutaneous tissue, where they do not belong, and they can also invade the lymphatic system.

And although this podcast episode is entitled cellulitis, I’m also going to talk about erysipelas. The two terms are not interchangeable. but both manifest as areas of skin, erythema, edema, and warmth. Cellulitis involves the deeper dermis and subcutaneous fat. Whereas erysipelas involves the upper dermis and there’s a more clear demarcation between the involved and uninvolved tissue.

There’s a fun fact, since the ear doesn’t have deep or dermal tissue, it’s always. ear-a-sipelas. I’ll pause for laughter. Anyway, a skin abscess, which is not the focus of this episode, is a collection of pus deep within the dermis or subcutaneous space. Impetigo, also not included in this episode, is a very superficial infection with that honey crusted drainage. There are also bullous versions. So cellulitis tends to develop in a bit more of an indolent fashion over a few to several days, whereas erys syphilis is more acute. You get systemic symptoms faster, such as fever. Chills, severe malaise, and headache. These can precede the onset of the local skin changes and start just in a matter of hours.

Clinically, for both, you’ll see areas of skin erythema. edema and warmth. You can also see petechiae and hemorrhage, as well as superficial bulla, vesicles, or even echemosis. Sometimes you also see regional lymphangitis or enlargement of the regional lymph nodes. If you’ve got a lot of edema surrounding the hair follicles, you can see some dimpling in the skin.

This creates an orange peel texture appearance, peau d’orange. I hope I pronounced that right. I took Spanish in high school. Anyway, the skin is warm to the touch, it’s uncomfortable, it hurts with movement, and some patients can describe an itchiness or a tight feeling in addition to the pain. You may see fever and other systemic symptoms, and cellulitis and erysipelas, especially in children, are nearly Always unilateral.

Bilateral red limbs? That’s probably something different. Complications that you should be aware of include bacteremia, endocarditis, septic arthritis, or osteomyelitis. Full blown sepsis and toxic shock syndrome. Fortunately, those are rare. So, in general, mild cellulitis has no systemic features in a patient with no significant comorbidities.

Moderate cellulitis has moderate swelling and tenderness with some systemic features like fever or tachycardia. Severe cellulitis has severe swelling and tenderness. really affecting function. It’s a larger body surface area, and you’ve got marked systemic features. So fever or hypothermia, extreme tachycardia, tachypnea, altered consciousness, a very unwell appearance, or even hypotension.

So what causes it? Well, bacteria, and the most common etiology. Staphylococcus aureus is actually an infrequent cause of cellulitis in children. But it can be seen more often in penetrating wounds. Methicillin resistant Staphylococcus aureus classically causes abscess formation. So you won’t really see that as the cause of isolated cellulitis or erysiplas in children.

So how do you make the diagnosis? Well, it’s clinical, right? Look for areas of skin that are erythematous, edematous, warm, and painful. Labs or imaging are not routinely necessary. If you think that there’s an abscess, you can diagnose it clinically by a localized area of induration or fluctuance or use an ultrasound.

Cellulitis can look like a cobblestone street on sonogram. And you should consider whether or not an abscess is present if you see significant induration, so thickening or hardening of the soft tissues, of greater than three centimeters or non uniform induration. The lesion’s been present greater than two to three days and changing or getting worse, and there’s a history of a previous incision and drainage in that patient.

And so do you need labs? Nah, not really. And the vast majority of patients of CBC or other labs will not aid in making the diagnosis of cellulitis. What about blood culture? Every febrile kid with cellulitis needs a blood culture, right? Not so fast, right? A blood culture can cost more than $200-300.

If you’re sending the kid home. Well, you definitely shouldn’t be sending a blood culture because if you’re worried about bacteremia and sepsis, that kid needs to stay in the hospital. And think about the risk of a false positive versus the risk of a true positive. So if the risk of a contaminant culture is greater than the risk of actually catching a bacteria, then don’t send it.

The cost of false positive cultures, repeat visits, length of stay, could be in the thousands of dollars. And so a lot of the previous studies on getting blood cultures were done in the immediate post Haemophilus influenzae B and Prevnar vaccine era. And we do now live in an era where these invasive organisms are fortunately not as big of a concern.

Vaccinate your children, people. But we do deal with MRSA. But still, for mild and moderate cellulitis, MRSA’s not really the etiology. And so even if a kid has a fever, But they look better after a dose of acetaminophen or ibuprofen. You don’t routinely need a blood culture. Now you could even admit a kid for IV antibiotics, maybe they’re dehydrated or they can’t take PO for some reason, without sending a blood culture.

So, admission does not mandate a blood culture. As always, you want to check with your local recommendations and follow algorithms present at your institution. So, what about disposition? So, with prompt identification and treatment with a correct antibiotic, which I’ll get to in a few minutes, patients can see an improvement in their signs and symptoms within about 48 hours.

The treatment failure rate is low, less than 1 out of 7, but probably a bit lower than that, with initial appropriate antibiotic treatment. Overall Cellulitis has a really good treatment prognosis. So when do you want to think about admission to the hospital or short stay unit versus discharge? Right, so you should probably admit a patient to the hospital if they have significant systemic symptoms.

You’re concerned about SIRS or sepsis. Again, fever alone does not necessitate admission. Especially if the kid looks better after antipyretics. If you are concerned that the child may need subsequent or future I& D, like they’re forming a phlegmon, they have a deeper infection, like necrotizing fasciitis, or they need sub sexually consultation, these are probably reasons to admit the patient to the hospital.

Now, some facilities have a short stay unit, like in their emergency department. So, if you don’t meet inpatient admission criteria, you’re likely to improve within 28 hours. Maybe the kid failed initial outpatient treatment with 48 hours of appropriate antibiotics, and they need just a day of IV. They’ve got a rapidly expanding lesion, but it’s probably IND.

The kid has a lot of pain. They can’t tolerate oral antibiotics, or they’re less than six months of age. You know, maybe that’s a patient you observe for just 24 hours in a short stay. And fortunately, cellulitis in children under the age of 2 months of age is rare, but those kids should probably be admitted for IV antibiotics as well, and you should get a blood culture in those situations. One example would be neonatal mastitis, a skin infection of the breast tissue.

Alright, so let’s talk about antibiotics. And generally, your best choice for the majority of children is cephalexin. You’d think because MRSA is everywhere. You want to avoid first generation cephalosporins, but it’s still mostly beta hemolytic strep.

And studies, including Chen et al., showed no difference between cephalexin and clindamycin. And what about length of treatment? Well, for mild cases, five days is probably just as good as ten days. But if you have moderate or severe symptoms, a week and a half is a good idea. You’ll see a lot of places start cephalexin plus clinda, or cephalexin plus trimethoprim sulfamethoxazole.

It’s still often done out in the community. A representative randomized control trial from Palin from 2013 showed that the addition of TrimSulfa did not improve outcomes in a very large cohort that contained lots of children. So the bottom line is, even in patients where you think they might have MRSA nasal carriage?

Monotherapy with cephalexin is fine. So, if you’re doing outpatient treatment or you’re transitioning patient to oral treatment after IV, the first line therapy is oral cephalexin. The dose is 50 mg per kg per day, divided every 8 hours or 3 times a day with a max of 500 mg per dose. If the patient has a true allergy to cephalosporins, you do oral clindamycin, 10 mg per kg per dose, every 8 hours, or 3 times a day, with a max of 1, 800 mg per day, or 600 mg per dose.

If you see treatment failure, that’s another reason to do clinda. So if the kid’s not getting better in 48 to 72 hours on cephalexin, you can do the same dosing of clinda that I just mentioned a moment ago. For inpatient treatment, first line therapy is intravenous cefazolin. So 20mg per kg per dose every 8 hours with a max of 1g per dose.

Cephalosporin allergy, you would use clindamycin. And yes, PO and IV clindamycin are bioavailable. But if you’re admitting somebody, there’s probably a reason that they may not be able to take PO. So if it’s IV, it’s 10mg per kg per dose every 8 hours with a max of 900mg. So it’s higher than the max for PO.

And so again, I will reiterate that 5 days for very mild cellulitis is totally okay. But you could do 10 if that’s what you do locally. Moderate and severe, you need 10 days. Some specific scenarios that you might want to deviate from cephalexin, so if there’s a mammalian bite, some don’t need prophylactic antibiotics, and I’ll admit, There’s no randomized controlled trials of dog bites.

I don’t think the IRB would approve that, but if you are concerned about infection, amoxicillin clavulonate, 22. 5 milligram per kilogram with a max of 875 milligram per dose, oral twice daily. If the kid looks well, And the degree of redness or symptoms is very mild, you could do five days, that’s totally okay.

If the child got an infection in a seawater or freshwater environment, there’s some different organisms like Aeromonas and Vibrio and other things that you want to consider. So you would still start with Cephalexin, but you would add Ciprofloxacin, 10 mg per kg, max 500 mg per dose, twice a day. Or Trimethoprim sulfamethoxazole twice a day with an overall treatment length of 5 to 10 days.

And if you definitely think that it’s methicillin resistant staphylococcus aureus that you’re treating, mild cellulitis, you could do trimethoprim sulfamethoxazole or clindamycin. If you’re sure it is moderate MRSA cellulitis, you could do a trial of oral antibiotics with close follow up or vancomycin IV.

And if they have severe cellulitis, Vancomycin IV, and if you have Staphylococcal Scalded Skin Syndrome, consider adding clindamycin. Switch to oral antibiotics as soon as the patient looks better. And there are some cases of cellulitis where you might need a specific subspecialist. So a general surgeon should be consulted if you have cellulitis of the breast, perianal tissues, perineal tissues, a complex of recurrent pilonidal abscess, Though you could drain these and have them follow up as an outpatient.

Or the cellulitis is just large and complex, or you think it’s going to need drainage in a day or two. ENT should see invasive neck cellulitis, especially with significant symptoms, or you’re concerned that it might be a deeper infection. ophthalmology and or ENT for orbital or periorbital cellulitis, orthopedics for septic arthritis, tenosynovitis or osteomyelitis, and dental or oral maxillofacial surgery for facial cellulitis due to dental infection where the kid might need to be admitted.

All right, so that’s all for this episode on cellulitis. Remember, most cases are mild or moderate, and the child will do incredibly well with oral antibiotic therapy, which should generally be cephalexin. You do not need labs, and especially a blood culture, for the vast majority of children with infections like cellulitis or erysipelas.

A fever does not mean you have to get a blood culture or admit the kid. If they haven’t started antibiotic therapy yet, and they look better after they defer vest with antipyretics, Start oral antibiotic therapy and develop a close follow up plan. Well, I hope this episode is useful. Ultimately, my goal is to deliver succinct evidence based information to help you on your next shift in the ED.

If there’s other topics that you want to hear about, send them my way. I’ll take an email. I’ll take a comment on the blog. I’ll take a direct message on Twitter or X or whatever it’s called. Leave a review on your favorite podcast site. That helps other people find it. And if more people can learn, I’m going to be happy about it.

Encourage your colleagues to listen and subscribe to the podcast. And thank you so much for your time and attention. I know you’re all busy out there. For PEM Currents, the Pediatric Emergency Medicine Podcast, this has been Brad Sobolewski. See you next time.