Management of Skin and Soft-Tissue Infections: Application of Antimicrobial Stewardship Principles


(upbeat music) – We’re gonna talk about
skin soft-tissue infections. I have worked with a few companies, although none related to
skin soft-tissue infections. These are the objectives. We wanna describe the opportunities. We wanna talk about pathogens. And, really the way I’m
gonna talk about this, you know, when we talk about
skin soft-tissue infections there are things we look at. Right? What does it look like? Well, it looks like this. What causes that is a really important thing to understand because if we know what it is, or we know what we think it is, and we know what bug it is, we can usually pick the right antibiotics. And, then we’re gonna
finish off at the end with a bunch of like interesting,
weird stuff, and mimics because I’ve been fooled a bunch of times by what I thought were infections which were not. This is just some data showing that skin soft-tissue infections are increasing and exceedingly common. It’s about three to 5% of all ED visits are due to skin soft-tissue infections. There’s a, as you get older, you develop more skin
soft-tissue infections. And, interestingly, during the summer there are more skin
soft-tissue infections. The approach we’re gonna take is really gonna be to work our way in from the superficial to the deep. Were gonna focus on the
microbial epidemiology because if you can sort out
the microbial epidemiology you can like look up the drugs, right? I think it’s Strep. What covers Strep? Oh, it’s green in the checkbox there. I can use that. And, then we’re gonna talk
about some uncommon things. And, then we’re gonna talk about mimics so, there is a real opportunity here. This is a study done at Denver General which is a big county hospital in Denver where they looked at about 300
skin soft-tissue infections. The culture shows Staph or
Strep in 97% in the cultures. What I would highlight here is that antimicrobial
use covered gram-negative 60 to 80% of the time, and anaerobe 70 to 80% of the time. Only a third of people, at best, got gram-positive therapy only. Over half got three or more antibiotics. And, the median duration of therapy was 10 to 14 days. And, when they looked at
another set of patients, in an outpatient setting, about half of patients
got totally unnecessary antibiotic exposures. And, so I think this shows that there’s some real opportunity in skin soft-tissue infections. The other thing I really
liked about this study was they looked at not only
skin soft-tissue infection antibiotic use, but they looked at imaging. And, when I’m on service, I see a lot of imaging for
skin soft-tissue infections. The thing I would point out here is that the overall yield for imaging in skin soft-tissue infection was 4%. Which means that we don’t find a lot. You know, imaging, it makes sense maybe to do an x-ray. If you find something in the x-ray it’s usually a big deal, but most of these other imaging, there’s really not a big
role in most patients. I see ultrasounds ordered all the time. I would say about .3% is about right for ultrasound yield in
skin soft-tissue infection. It’s pretty much useless. And, so I think there’s
real opportunities, not only for antibiotics. One of the things I often talk about is syndromic antibiotic stewardship. Antibiotic stewardship focused around an infectious syndrome, which is kinda what we’re
talking about here today. And, that’s where there’s
a lot of opportunity, not just to improve antibiotic use, but Dr. Roab talked about prevention. We can talk about imaging, diagnostic use. There’s just a lot of opportunities around not just antibiotic use. This is a study from
the dermatologists who, you know, I think one
wanted to show their value, but two also wanted to show you the problem that we have
with diagnosing cellulitis. They looked at 260 cases of cellulitis, and a third of them weren’t cellulitis. A lot of vascular disorders, other inflammatory things, and what they found was that
about half of those people who didn’t have cellulitis, didn’t even need to be
admitted to the hospital. And, when they extrapolated
that out for cost, they found it’s probably about 50 to 130 unnecessary hospitalizations
per year in the US, with lots of extra money, and lots of extra nosocomial infections caused by admitting all
those people to the hospital, and giving them antibiotics
when they don’t need it. And, so these are sort of my principles of skin soft-tissue infection treatment. And, these are important to think through when you’re looking at a
skin soft-tissue infection. And, these are kinda how
I think through things. Being a good steward is really
about being a good clinician. So, the first question is, is it an infection? And, you might say well
that’s really easy. I’ve been fooled many times. It’s actually not that
easy to look at something. Sometimes it’s really easy. I saw a guy with cellulitis yesterday. It was really obvious he had cellulitis. But, I’ve seen a lot of skin ulcers, and wounds that I’ve looked at and said, ahhh, I don’t know. Is that infected? But, it’s a really important question because if it’s not an infection, giving antibiotics clearly
aren’t going to make it better. Next, based on the appearance, what sort of skin
soft-tissue infection is it? Rather than classifying
as community onset, hospital onset, you know, COPD exacerbation. We look at a skin infection and say it looks like an abcess, it looks like a cellulitis. Then, we have to say based on what I see, what bacteria do I expect? Based on what it looks like, we can usually make a pretty good guess as to the microbial etiology. But, we need to factor in a few things like Dr. Rupp talked about. Are there exposures that might suggest an unusual pathogen? Something has happened that
makes me think it might be some other bug than the typical ones. Or, what’s the immune status, or something with the patient that makes me think it might
be a different pathogen. I’m gonna show you a
few examples at the end. And, then last, how bad is it? Right? Is it like so bad you’re like why did you come to clinic even to show this to me, or is it like call the surgeon right now, we have to do something. And, that again, is gonna influence our
antibiotic decisions. So, let’s work through
some of these pathogens. First, impetigo. You’ve probably, a lot of you have seen impetigo, Staph and Strep, it starts as a macule, then progresses to a pustule, but eventually ends up in
the honey colored crust here, which is usually what we see. It’s painless. You use topical therapy to cover Staph. Mupirocin, Retapamulin. If people aren’t getting better, or it’s really severe, like this bullous impetigo which I’m showing you here
in this larger picture, then you’re gonna wanna
use oral antibiotics again targeting Staph and Strep, to as a concern for MRSA you might use Clinda, or maybe add some Doxy or Bactrim. If it’s this bad, you might think about
doing a culture as well, but usually you don’t need
to do cultures in these. Folliculitis. You’ve all see inflamed hair follicles. These are often bacterial, although if you were to culture them, it would grow something like skin floras, which our lab will tell you, rarely they’d say Staph aureus. There is something called
hot tub folliculitis, so it’s hot tub use. It’s caused by Pseudomonas, it’s self-limited, it goes away. Usually you don’t need
to do much with these. Some warm compresses. Maybe some topical antibiotics. If they aren’t getting better you might think about culturing them because occasionally
dermatophytes can cause them, although there are a lot of things that can cause folliculitis as well. Erysipelas. You know, in infectious disease, you get infections that
you kinda think are fun. I kinda think erysipelas is kinda cool. Has a very characteristic,
clinical appearance. It’s an infection to the upper
dermis and the lymphatics. It’s well demarcated, and that’s the thing that
really sets it apart. You can see, and you can literally feel
the edge of the infection. In this gentleman’s face, you can feel the step off
where the infection ended. As opposed to cellulitis, we’ll look at it’s sort
of vague and indistinct where the infection ends. People with erysipelas are
usually systemically ill. They often are young or old. It’s often in people who have lymphedema. And, this is almost exclusively caused by beta-hemolytic Strep. And, when we have beta-hemolytic Strep we know they’re susceptible
to beta-lactams, and we can use penicillins, and early generation
cephalosporins to treat them. And, five to seven days is
really about all you need. Skin abscesses. You’ve probably all seen skin abscesses. So, I do have to warn you there are some gross pictures at the end. You couldn’t do a skin
soft-tissue infection talk without some gross pictures. So, they usually start in areas where there’s hair follicles, there’s inflammation, this develops and turns into an abscess in the subcutaneous tissue. If you have one it’s called a furuncle. If you have multiple
it’s called a carbuncle. They’re typically in areas
where there’s hair and friction. So, neck, buttocks, face, axillae. There’s perioral straining. This is a pathologic picture where there’s a lot of inflammatory cells sort of inside this capsule. Multiple ones in the back of the neck. This is a carbuncle with numerous draining sinus tracts. What do you need to do with these? Well, you need to drain them. But, you also need to think
about the epidemiology. You know? And, so this is pus. Right? And, so when I see pus, I think Staph aureus. And, that’s what you should think. Now, when you see pus
and think Staph aureus, you have to think is it MRSA or MSSA? This is some older literature, but I think it still holds to. These are perioral and skin lesions cultures in the ED. 60% of them were MRSA. Almost 20% of them were MSSA. Meaning that about 80% of all the pus in skin lesions in the ED is due to Staph aureus. And, a lot of that is MRSA. And, so knowing that, that adjusts how we think about therapy. These are painful, they’re tender. Obesity, steroids,
diabetes are risk factors. We talked about MRSA. My opinion on cultures is
usually you don’t need them. If they’re so bad they’re
coming in the hospital, I think it’s reasonable to get a culture. If it’s a weird place, you know, perirectal,
other places like that, you might think about other pathogens. The key treatment for these, incision and drainage. Right? It’s not even antibiotics. The most important thing you’re gonna do is incise and drain these. If they’re very small, you might get away with some moist heat. Carbuncles you’ll never
get away with that, you have to surgically debride them. What is the role of antibiotics? I’ll tell you this is an area
where my practice has changed in the last couple of years. As have our guidelines. So, three years ago I used to
have this slide in my talk. Three randomized trials, looking at the use of
antibiotics in skin abscesses, showing no improvement in
clinical cure with antibiotics. Although I’ll tell you this, KEFLEX study was basically
double placebo study because it was about 80% MRSA. There was maybe a suggestion of decreased recurrence of Bactrim. And, so there are
actually two large trials funded to look at the use of antibiotics in skin soft-tissue infections. And, what they found is
actually both of them showed improved clinical
cure with use of antibiotics. The first trial was in people over 12. They had to have a lesion
of at least two centimeters, and they got other placebo. They all got drainage, and then they got placebo or Bactrim. Increased clinical cure. And, actually decreased
subsequent infections. Decreased infections in household members in people who got antibiotics. This other trial randomized people, the placebo Bactrim or Clindamycin, and they saw basically
increased clinical cure with either antibiotic. A little better decrease
in subsequent infections with Clinda, but way more diarrhea with Clinda, about twice the rate of diarrhea. And, so my take from this, and our guidelines changed, to say boy, you know, if you have a skin abscess you probably do benefit from antibiotics. And, so while incision and
drainage is the primary therapy, most people are gonna
benefit from antibiotics, particularly when
there’s multiple lesions, anything over two centimeters. If it’s smaller it’s sort
of risk versus benefit. What’s the patient? Immunocompromised. Anything in a sensitive area. Anything that’s getting worse. And, when you pick an agent, you need to pick something
that’s active against MRSA, unless you know it’s not MRSA. And, so to me, that means Bactrim,
doxycycline, minocycline, you certainly could use linezolid, it’s more expensive, or tedizolid, or, you know, delafloxacin which is way more expensive which we’re not gonna talk about, but I’m not sure why you would. But, that’s just a personal opinion. We’ll talk about the role of
clinda here in a little bit. And, if it’s really bad you’re gonna use your IV anti-MRSA agents. And, again, five to seven days is all that you really need. What about clindamycin? I’m just gonna show you are antibiograms
susceptibility in Staph aureus to clindamycin? You know, it’s in the low 80’s. I think these last two years
we’ve changed our methods, it’s a little lower, but it’s not above 80%. And, so there is increasing
resistance to clindamycin, and so that’s why having those
antibiograms is very useful. But, know that Staph aureus
has increasing resistance to clindamycin. The other thing you should
know about clindamycin, there’s a lot more diarrhea. Probably two to three times the rate of any other antibiotic you’re gonna give, and the rate of C. diff is higher. And, so I’m not a big personal fan, but that’s just an opinion. All right. Cellulitis. Very common. It involves the deeper dermis and subcutaneous tissues. Painful, tender, swelling. It’s not elevated. The borders are very indistinct. You often have systemic symptoms. You may have this lymphangitis streaking, pretty characteristic from Streptococci. And, the etiology of this infection is almost exclusively Streptococcal. But, it’s really hard to know
the etiology of cellulitis because we get blood cultures, and they’re positive in
like one to 2% of people. And, people have tried
lots of different ways to diagnose this. Be it, you know, biopsies, eluting culture aspirates, and skin swabs, and there was a study of PCR, all of which have been really unfruitful. And, so I do wanna show you
one study that looked at this. This is a prospective study
of nonculturable cellulitis. No pus present. And, what they did was
they wanted to find out what percent were due
to beta-hemolytic Strep. And, they found using Strep serologies and blood cultures, that about 73% were clearly
beta-hemolytic Strep. The problem with Streptococcal serology is you gotta like do
them when they present, and do them four weeks later, and so that’s not very
useful as a clinician, right? But, as somebody who
wants to sort of sort out on the back end what
percent were due to Strep, it’s somewhat useful. They treated most of the
patients with beta-lactam, and had a 96% response
rate to beta-lactams. And, in those patients who
tested negative for Strep, they had a 91% response
rate to beta-lactams, that being Ancef, cefazolin, narrow spectrum beta-lactam. And, so their conclusion was, boy cellulitis, non purulent cellulitis, it’s due to beta-hemolytic Strep. There’s a second study from Scandinavia that replicated that very similar. And, so the recommendation is for cellulitis we use first
generation cephalosporins, or anti-Staphylococcal penicillins, either IV or oral. The guidelines would
recommend five to seven days. If people don’t respond, well you could treat them, but I’ll tell you this paper from 2016 is a really interesting paper. The factors that influence
response to treatment had nothing to do with antibiotics. They were all patient factors. And, so their lack of response
at day one or day three, didn’t have anything to do with whether you picked the right antibiotic or not, it was really if whether
they had lymphedema, how long they’d had symptoms, what their sex was, what their BMI was. Couple questions about cellulitis. Should we add anti-MRSA
therapy for cellulitis? Should we add Bactrim to our beta-lactam? Our guidelines used to recommend that. They do not anymore. There have been two randomized trials looking at the addition of Bactrim to a beta-lactam in cellulitis. Both of which showed no
improvement in clinical cure rate. Absolutely none. Actually, no difference
in adverse events either. So, we’re not really hurting, but boy we really aren’t helping. And, so I would not add an anti-MRSA agent to a beta-lactam for cellulitis. What about clindamycin? There have been some studies suggested maybe including clindamycin
helps you get better faster. This was a randomized
trial that looked at this where they had beta-lactam
plus clindamycin in nonculturable cellulitis, and their outcome was
improvement at day five because that’s a thought as maybe clindamycin makes
people better faster. The answer was in this
study, which was randomized, it did not. It did improve diarrhea, though. You want more diarrhea? Again, clindamycin, more diarrhea. So, again, I don’t add clindamycin. In typical cellulitis cases, doesn’t look like it has benefits. So, to summarize, non-purulent cellulitis we treat for Strep using a beta-lactam. In our very large patients, or lots of lymphedema patients, it’s reasonable to go with very high doses because it is hard to get
the drugs into the tissue. If they have an abscess, you must drain it, but most people need antibiotics, and generally you’re gonna treat for MRSA, which means Bactrim, or doxy/minocycline, maybe clindamycin, it will work but it has some downsides, and then for severe you got your Vancos. Complicated infections which I’m not gonna talk a lot about, burns, wounds, et cetera, generally we treat those
like purulent infections because Staph aureus is a common cause of those complicated infections. So, in that whole talk, I’ve not mentioned a single gram-negative, or a single anaerobe, and I will tell you I see lots of people admitted to my hospital
who get vanco and cefepime, or vanco and Zosyn, for their cellulitis with sepsis. When should we cover with gram-negatives? There are situations where we should use anti-gram-negative therapy. One of them is diabetic foot infection, but not all diabetic foot infections. Generally, from mild to moderate
diabetic foot infections it’s Staph and Strep. For our severe, you know, my toe is black and necrotic, okay that’s where we need
gram-negative therapy, that’s where we need
anti-anaerobic therapy. We can often use oral
therapy in these patients once we’ve sort of sorted
through this disaster. How long we treat really
depends on a host of factors like what debridement did we do, was there underlying osteo, how is the foot doing? I’ll just point out some data that one of our fellows collected. He looked at diabetic foot
infections at our institution, and looked at the culture data. 30% of our diabetic foot infections had a surface swab which
is actually recommended not to be done in the guidelines. But, they say don’t do a swab because it will grow a
whole bunch of bacteria that don’t bear any
relation to what’s actually causing the infection. Only about 57% actually
had a surgical debridement with tissue cultures, but when we did that, this is what we found. Beta-hemolytic Strep,
MSSA, Strep viridans, only 10% MRSA, and only 2% Pseudomonas. And, so we looked at what percent got anti-MRSA therapy, it was 90%, and 80% got anti-pseudomonal therapy related to the percent who
actually had those pathogens. Very out of proportion there. And, so we revised our diabetic
foot infection guidelines to sort of help people figure out what’s mild, moderate, and severe. And, so for mild we’re recommending anti-Staphylococcal and
Streptococcal therapy, and if they have a history of MRSA you might wanna cover for MRSA. That’s sort of our rule. If you’ve had MRSA before in your foot it’s probably reasonable to
treat for MRSA in your foot. I point out in our severe
recommendation with therapy is ceftriaxone plus metronidazole. It’s not Zosyn, although I wouldn’t say
that’s a terrible choice, but it’s not generally Zosyn, it’s not, you know, cefepime, unless you have sort of water exposure, or a history of Pseudomonas. And, so it’s just not
a very common pathogen. And, so we’ve really tried
to change our guidelines. We changed our order sets. Here’s another situation you might use anti-gram-negative therapy, necrotizing fasciitis, right? Can be mixed, or monomicrobial by Strep pyogenes, could be clostridial. These are infections
where when you see them you want broad spectrum covering MRSA, covering gram-negatives, covering anaerobes until you know differently. Surgical debridement, again, is the key you need to
go from here to here when there’s lots of infection and then narrow it down. These adjunctive therapies I think are not very useful. Surgical site infection. This actually our most
costly nosocomial infection because there are a lot
of surgeries that go on even though it’s only a small
percent that get infected. Pretty obvious there’s pain, swelling, drainage at the wound site. How do you treat these? Often opening the wound
is all that’s needed for the mild ones. Open the wound, pack it, it’ll often get better on its own. The guidelines recommend
if people have fever, significant surrounding erythema, signs of systemic infection, then you should use an antibiotic, and I think if you’re
gonna use an antibiotic, in this case it’s
reasonable to get a culture. The antibiotic you use depends
on what the surgery was. If it’s a clean site, trunk, head, neck, something like that, Staph and Strep are your bug, and so you’re gonna use
antibiotics for that. If it’s a contaminated site, GI site, female genital tract, oropharynx, then you wanna cover
gram-positive, gram-negatives, anaerobes those are the
bacteria that are encountered at the surgical wound. All right, we’ll finish
off with about five minutes of fun stuff. So, which animal bite is
most likely to get infected? Is it a dog, a cat, a human,
a zombie, or a chupacabra? So, right, I think most
people would say zombie. It’s like 100% of people bitten by zombies turn into zombies, there is no prophylaxis. If you were to look at
actual animals that exist it’s actually human followed by cat followed by dog which is way down there. Human bites are the most
likely to get infected. There’s a whole host of
bacteria in our mouths. Strep, Staph, anaerobes. We always prophylax human bites, or human mouth wounds for three
to five days with Augmentin. For animals, cats much more likely to get infected. I just saw a guy who was
admitted to the hospital with a cat bite yesterday. Pasteurella is the bacteria
you always think about, but there’s Staph and
Strep and other things. When do you need prophylaxis? Anything deep. That’s why cats often need it because they have very sharp teeth that are small that can puncture deep. Anything that is bad on
the hand or genitals, something sensitive, or immunocompromised patient. But, again it’s easy, it’s amox/clav. Easy to remember. All right. Some final fun things. This is where we talk
about unusual exposures. So, a 56 year-old guy, chronic Hep C, severe cellulitis of his left leg, and septic shock. He just came back from the Gulf Coast where he scraped his leg on
the sea wall while swimming. So, what’s he got? It’s a classic medicine board question. So, he’s got Vibrio vulnificus which causes severe necrotizing infections associated with salt water. 44 year old guy presents with infection of left knee laceration, he cut it while swimming
in a local Iowa lake, is infection’s really bad, and progressing rapidly. The last one you probably wouldn’t see. This guy I have seen. This is not his knee, but I did see this. So, it’s a water associated bacteria, this is Aeromonas. So, also associated with leech therapy. You probably don’t see a lot of that. This is a gal I saw in clinic. 28 year-old female from Mexico. She had progressive skin lesions over the last three to six months. They were non-tender, she said maybe my dad had
something similar in Mexico, she had these nodules in her arms, you could also, on her legs, you could also feel them
on her arms, her face, she otherwise looked pretty well. She hadn’t been visiting any nail salons to give you a hint. So this is a mycobacterial infection, she had a skin biopsy which was swimming with acid-fast bacilli, and she had lepromatous leprosy. Which is a skin infection, right? Showed up in my clinic here. So, this is 28 year-old guy, who recently came back from Iraq, he has multiple skin lesions like this, he’s gotten a bunch of antibiotics, they aren’t getting better. He’s not really sick, they’re not really tender. Saw this guy in my VA clinic. So, he has, this is what the biopsy would show. He has cutaneous leishmaniasis, right? And, so again, we talked about a good history. Well, I spent the last
four months in Iraq. Oh, well, hmmmm. Maybe this isn’t a bacterial infection. So, this was another patient I saw. 37 year-old non-breastfeeding female. She presented with
bilateral breast infections that weren’t getting better despite antibiotics to treat Staph. So, again, we took a
little bit of history, and she admitted to injecting IV drugs into her breast veins, and so she had a Mycobacterium
fortuitum infection. This gentleman was up
in our bone marrow unit, had been neutropenic for 23 days, and developed this
progressive black lesion on his forearm. So, progressive black lesions, in people who don’t have
neutrophils are always bad, so he had mucormycosis, and had to have a big
wide tissue debridement. All right. Finish off with a mimic. This won’t take long. Patient admitted two days ago for community acquired pneumonia. Now we’re called to see her
for left deltoid cellulitis. She has a painful, red,
hot, swollen left deltoid. I asked one question
and made the diagnosis. Did you get a flu shot? And, she said oh yeah, they gave that to me yesterday. Okay, well this is not deltoid cellulitis, it’s a flu vaccine reaction. But, they had her on
antibiotics for cellulitis. This was a patient with chronic
liver and kidney disease. She came in. These lesions had been progressive over the last three weeks. She otherwise looked okay, but these are pretty tender and sore. They were very firm, very hard. So, she has calciphylaxis
due to renal failure. Also doesn’t get better with antibiotics. This patient had a high white count, a high fever, left shift, high eosinophil, they had this blanching rash. This is a drug reaction
due to allopurinol. This patient had pain, swelling, redness at a peripheral IV. Thrombophlebitis, right? This patient has a
right leg that’s swollen after coming back from a trip. They have a DVT. This patient was referred
for recalcitrant cellulitis that nobody could make better. He looked all right. It was kinda painful, kinda red, didn’t have a fever, CRP was normal. That’s his leg. I lifted it up above his heart, and it suddenly went away. This is dependent rubor. This is probably a quarter
of the patients I see for refractory cellulitis. Bilateral cellulitis? Venous stasis, right? And, then this last one, patient with a rash, he’d had lots of antibiotics, super painful, indurated and tender, didn’t get better with elevation. He had something I’d never heard of. Lipodermatosclerosis. Which is a fibrosing panniculitis due to chronic venous stasis that
looks like cellulitis, but ain’t. So, it’s a dermatologic condition. There are lots of other
things that can cause and look like infections, but aren’t. And, so I’d encourage you to
be familiar with those as well. So, these are my conclusions. We make a lot of misdiagnoses. We can improve our antibiotics. You guys can see what we
should do with our antibiotics. Think about patient factors, and then mimics when
you’re looking at these. With that I will stop. (audience applauding) (upbeat music)

Leave a Reply

Your email address will not be published. Required fields are marked *