• LecturehallDiabetic Foot Infections - IDSA Guidelines
  • Lecture Transcript
  • Warren Joseph: Hello, my name is Warren Joseph. I’m a podiatrist infectious disease specialist at Roxborough Memorial Hospital in Philadelphia, Pennsylvania. And I’m editor of the Journal of the American Podiatric Medical Association. In this lecture we’re going to talk about the 2012 Infectious Diseases Society of America Updated Diabetic Foot Infection Guidelines. I think most of you are familiar with the problem of diabetic foot infection. It’s estimated that about 28 million people or about 8% of the US population in the United States do have diabetes. Diabetes is associated with abnormal healing and diabetes is associated with diminished white cell function and keloid. And that diminished white cell function and keloid that you see in diabetes tends to be directly related to the level of diabetes control. So if the patients are under fairly good control then their white cell function does improve. Approximately 15% of people with diabetes will develop a foot complication. The estimated prevalence of a foot ulcer is between 4 and 10% with a lifetime incidence being high as 25%. Basically, there’s a one in four chance that a patient with diabetes will develop a foot ulceration sometime during their life. Foot ulcerations lead to amputation. In fact, the American Diabetes Association estimates at about 70,000 or more lower extremity amputations per year as a result of diabetes. 85% are preceded with a foot ulceration. Diabetes is the number one nontraumatic cause of amputations. It used to be said that somebody in the world loses a limb to diabetes every 30 seconds. That’s not said anymore. Now, someone in the world loses a limb to diabetes every 20 seconds. So despite the advances we have in diabetes and care of diabetes we are still seeing significant number of amputations as a result of foot ulcerations leading to infection, leading to that amputation. And foot infection is the number one cause of hospital days. There are a lot of complications of diabetes, usually dysglycemia you think of or renal problems. But in fact, foot complications cause more hospital days than any of those other complications of diabetes. This is a major problem with significant morbidity and mortality. Speaking of the mortality of diabetic foot ulcerations, we know that patients who have diabetes, who undergo amputation, have a five-year mortality of about 48%. Now it used to be considered that the patient having the amputation would lead to that patient’s death and there was higher mortality in patients with amputation versus those without. However, if you actually look at the numbers you see on this graph, there’s only about a 2% difference. So, 48%, five-year mortality in patients who have had an amputation, 46% mortality in patients who have not have an amputation. But those numbers are still incredibly high. In fact, if you look at those numbers and put them into context of various cancers, what you see is that the five-year mortality rate of somebody presenting to their primary care physician for the first time with a diabetic foot ulceration is higher than the five-year mortality rate for breast cancer, prostate cancer, and Hodgkin’s lymphoma. If you add into that diabetic foot ulceration and peripheral arterial disease, the five-year mortality is higher even than colorectal cancer. Why don’t we talk about this? Why don’t our patients know about this? If we’ve got a patient who has cancer, they are happy to undergo, well not happy, that’s not a good way of saying it, but they will undergo radiation and chemo and surgery, everything necessary to get rid of that cancer. But if we have a patient who comes in with a diabetic foot ulceration we can’t get them to lay off their 64 rounds Coca-Cola, Big Gulps and their Twinkies to get their A1Cs down from 15 to whatever is supposed to be normal validation, the 6 or 7 range. So we need to rethink the way we talk to our patients about diabetes because diabetes and foot complications are going to kill a patient more readily than most common cancers. Of course, it’s not always the foot complication that leads to the mortality it’s the foot complication as a marker for systemic illness. Fortunately, there’s been a lot of interesting work going on in the world of diabetic foot infections.


    This is just a graph that came from the Web of Science. If you look at the number of citations and publications in the past decade, published items from 1994 to 2012, the last date which I could find numbers, you will see that the number of published items per year went from about 20 per year up to about 150 per year. I will tell you that I signed up and I urge all of you to do the same thing, you go to pubmed.gov which is where you do your MEDLINE searches and you sign up for what’s called My NCBI. There’s a little tab to click on where it’ll get you to it. And from that tab you can sign up for daily, weekly, monthly, you set it anyway you want, alerts on all of the publications that have been published in a certain area or certain field in that period of time. So every Monday morning I get three emails from My NCBI, one of them on diabetic foot infection, one of them on osteomyelitis to the foot, and one of them on onychomycosis. And that’s really a great way to keep up with the science. But it also shows you that every week there are number of articles being published on diabetic foot infection. So the number of publications is getting much higher, there is better quality research from all over the world being done on this. Not only do we have more published items but with more published items we have more citations. Back in 1994, the number of citations of articles written on diabetic foot barely made a bleep. But as recently as 2012, you see over 3000 citations per year of the world literature in diabetic foot infection. So, there is good attention being paid to this area and there is good research being done on this area. We frequently talk about evidence-based guidelines or different subjects that come up, how useful are these guidelines? This is an interesting paper that was published out of France that looked at using the original guidelines that we published from the IDSA looking at diabetic foot infections and how to manage them. Basically, they did an audit of the microbiological assessment of diabetic foot infections. They found that there were many clinically uninfected wounds cultured. Most cultures were collected by suboptimal techniques. Frequently, they isolated multidrug-resistant organisms especially MRSA and they isolated many low-virulence organisms that probably were not even pathogens, they were colonizers. Then they instituted a program using the 2004, the original Infectious Disease Society of America Diabetic Foot Guidelines. They then re-audited their clinical and microbiological data on 405 patients between 2003 and 2008. And what they found is there was a decreased microbiological lab workload, there was decreased inappropriate use of antibiotics, and that lead to an overall cost savings of greater than 230,000 euros. By following evidence-based guidelines they were able to not only decrease workload they also decrease inappropriate antibiotic usage and save a significant amount of money. And that’s in a single institution. I mentioned the first set of guidelines were published in 2004, this is just a snapshot of the 2012 document. You can actually get this at idsociety.org, it was published in Clinical Infection Diseases and we reprinted the executive summary from this in Journal of the American Podiatric Medical Association. This is a guideline, there’s a large group of people who were involved with writing it. I’m honored to have been involved with both the 2004 and the 2012 document. But in 2004, it was primarily all infectious disease specialists. In this guideline we have orthopedic surgery, we have podiatric surgery, we have family medicine, we have a lot of the different constituencies that are involved in treating diabetic foot and they came from all over the world. Also in 2012, we had a major change in the format of the guideline. Back in 2004, when we wrote the guideline, it was just an issue of writing whatever we wanted to write and whatever we thought. IDSA standardized all guidelines that you can find in idsociety.org now. What they’ve done is they put all guidelines into a format where you ask a question, you get a recommendation based on that question, and then the evidence summary on which that recommendation was made. So it’s question-answer proof, basically it’s how it’s done. It makes these guidelines extremely readable and user-friendly. Not only that, you’ll see that the way the guidelines are written are as if we’re having a conversation. How would I do this? We think you should do that.


    It’s kept very personally to make it easier to understand and easier to follow. The other big change in the guidelines were that instead of using evidence grading that nobody can remember, like this is level 3B evidence, we’ve now standardized and gone through the grade system. The grade system was first proposed in the British Medical Journal and has been actually taken up by most organizations writing clinical practice guidelines and systematic reviews of the literature including Cochrane. What’s nice about the grade system, instead of being a combination of letters and numbers it’s actually human or English, it’s actually spoken English, it’s easy to understand. For instance, you combine a strength of recommendation with a quality of the evidence. Let’s say something is graded as strong, low. Strong, low, what that means, we really believe that you should do this. This is a strong recommendation. But we really don’t have good reason to say that, it’s low level of evidence. So it just makes it very easy to understand the recommendation and what sort of evidence is out there to back up that recommendation. I mentioned that these were all done in a question-answer format. Every committee is charged with coming up with the 5, 10, 15, whatever the number may be, most important questions in your specialty. This is what we came up with in the area of diabetic foot. Number one, in which patient should I suspect infection and how should it be classified? Number two, how should I assess a diabetic patient presenting with a foot infection? Question three, when should I request a consultation and from whom? Question four, which patient should I hospitalize and what criteria should be met for discharge? Question five, when and how should I obtain specimens for culture? Question six, how should I select and when should I modify an antibiotic regimen? Now, each member on the committee was given what’s called a primary responsibility to answer one question. This was mine. I love new antibiotic discovery, I love antibiotic usage and discussing antibiotic usage, so I was given this question. Of course, that’s what we’re going to talk the most about later in this lecture. Question seven, when should I consider imaging studies and which should I select? Question eight, how should I diagnose and treat osteomyelitis? Question nine, in which patient should I consider surgical intervention and what type of procedure? And the final question, what type of wound care techniques and dressings are appropriate? You see there’s a lot to cover. In fact, they document something like 38 pages long. If I were to go through 38 pages of all these recommendations, I think there are over 40 recommendations, we’d be here all day and I really only have a limited amount of time for this lecture. So I really thought the best way to present this would be to present a case. Here’s a patient who comes in, a 61-year-old male presents to our emergency department. He has an HPI of a few week history with an ulceration. He self-treated it with hydrogen peroxide dressings. Because his family doctor told him he had diabetes so he needed to exercise a little bit, he went out and started walking, including about a week before where he walked over a mile around the golf course and he found a recent increase in pain, redness and swelling. On admission to the ED, his white blood count was 17.3, his glucose was 220, his creatinine was 1.1, BUN 14, and temperature 101.5. X-ray showed local soft tissue gas, medial first MTP joint with no evidence of osteomyelitis. This is clinical picture. Now this was a couple of years ago but we’re going to follow him up through the years. This is how he looked on admission. You can see these two sinuses draining, and I will tell it was foul-smelling drainage and the extent of cellulitis. Here was a plantar ulceration that actually looks fairly clean. So most of this problem was now around the medial aspect of the foot. And here you see the dorsal view of the foot. You can see the edema up above the ankle and the erythema, at least to the level of the ankle. The first question is, how should this be classified? Now remember, that’s question number one, how should I classify a diabetic foot infection. We say, in our guidelines, a clinician should consider the possibility of infection in any wound in a patient with diabetes. That certainly makes sense. And factors that increase the risk of infection include probe-to-bone. Probe-to-bone does not mean that there’s osteomyelitis, different topic for a different day. The duration of the ulceration is longer than 30 days and that there have been previous ulcerations and amputation. That will increases the possibility that wound is infected.


    And we do believe that you should use a validated classification system. It’s actually one of the few strong recommendations with a high level of evidence. We’ll show you the classification system that we use and we devised through this guideline. Here comes the first question to you. How would you classify this? Would this be an IDSA mild infection, an IDSA moderate infection, or an IDSA severe infection? The answer to that is this would be an IDSA severe infection. That is because of the classification system we developed in 2004 and just tweaked for his 2012 update. Again, just like the grade system, we kept this in English. We basically look at the wound and say that it has no infection, a mild, a moderate or a severe infection. That’s it, noninfected mild, moderate, severe. I think everybody can pretty much remember those four topics or those four degrees. We don’t have to sit there and remember that this a great 2A or 3B ulceration. A noninfected wound is just that, there are no clinical signs of infection. And we take a fairly strong position in this. The way you diagnose infection is looking for clinical signs and symptoms of infection. If you don’t have clinical signs and symptoms of infection you don’t have an infection. If you don’t have an infection, you don’t culture, you don’t treat with antibiotics. We’ll talk about some of that. So noninfected wound, no sign of infections. A mild infection is an infection or is a wound that has at least two of the signs of inflammation, erythema, warmth, tenderness, induration. And it has to have at least two centimeters of surrounding cellulitis. It can’t go above two centimeter. Above two centimeters puts it into the realm of a moderate infection. I call this a slice of time, our mild infections. It’s that slice of time in a patient going from a noninfected wound to either a moderate or severe wound that’s probably going to need to be hospitalized with heavy duty systemic antibiotics and maybe surgery. So the mild is that little slice of time between those and has at least two signs of cellulitis or inflammation and two centimeters around the wound. Now moderate and severe, you can’t differentiate just by looking at the wound. They’re going to look exactly the same. They have greater than one of the following, cellulitis extending more than two centimeters, lymphangitis streaking, spread beneath superficial fascia, deep tissue abscess, gangrene. You can read this on the table as easily as I can read it to you, it’s just stated a little bit differently than I’m stating it. The difference between our moderate and our severe infections is not in how the foot looks. The difference between our moderate and severe infections strictly has to do with the patient’s systemic response. In a moderate infection the patient is systemically well and metabolically stable. In a severe infection the patient is systemically unwell, metabolically unstable. In fact, in 2012 we specifically say that these patients have to have evidence of SIRS, systemic inflammatory response syndrome also known as sepsis. And you can see the criteria for the diagnosis of sepsis. Moderate and severe look the same, severe infections the patient is actually septic. This classification system, again, first devised in 2004 has been validated. Lavery publishing in 2007 on clinical infectious diseases applied this classification over 1600 diabetic foot wounds and found that as you go through no infected mild and moderate and severe you have statistically significant increases in rates of hospitalization and amputation. It’s very gratifying having been one of the authors of this original classification system to see it being used in just about every paper that’s published in the world literature in diabetic foot infection today. Now one thing I have to show you is we talked a couple of seconds ago about that concept of slice in time, how it’s just very rapidly progressive sometimes between noninfected and significantly infected. This is a fascinating one-page article that came out of the New England Journal of Medicine in late 2013. Basically, this gentleman noticed on day one that his foot was a little bit red. Instead of going to see a physician about this, he just sat at home and watched the foot and then photographed it everyday. And you can see how rapidly this progressed from day one to day 10 to a significant foot with tissue loss, some evidence of gangrene. Amazingly, this gentleman did not lose his foot. You can see in photograph B there the end of this.


    But it’s just truly amazing to see how day by day you can see this progressing. We said for years that a diabetic foot ulceration can go from being clean and noninfective to having a limb-threatening infection literally overnight. This pretty much proves that point. So, what are the key steps in early management of diabetic foot infection? How would you prioritize the following test in the patient I presented? Not this gentleman who took pictures of his foot but our case. You could probe the wound. That was question two, how do you evaluate the wound? You could get a vascular consult. Question three, whom am I going to consult? You can initiate I.V. antibiotics. My question, question six, what drugs are you going to use and when? You could perform surgical drainage to the wound, that was question nine, what sort of surgical drainage are necessary? Or, you can always contact your malpractice insure. Now, that one we don’t actually have any categories or any recommendations but something to consider. In the meantime, look at these first four, which one would you consider takes primacy. I think most of you will agree, in this cases, it’s surgical drainage of the wound. You can do all of the probing and consulting and antibiotics in the world, but if you got a pocket of pus in there you need to drain it. I used to be taught by my mentor on infectious diseases, Dr. Jack Le Frock that the best antibiotic known to man was Drano, surgical incision and drainage. There’s no known resistance. Diabetics can’t tolerate pus, they can tolerate surgical incision and drainage. So which patient should I hospitalize? That’s question number four. What we say in the guidelines is certainly all severe infections are going to hospitalized, these patients are septic. Patients who have mild infections don’t need to be hospitalized. So what about the moderate infections? We actually say that a lot of moderate infections can be treated in the outpatient setting especially with highly orally bioavailable antibiotics. But there are some patients that should be hospitalized even with a moderate infection. These are the ones with severe peripheral arterial disease, maybe lack of home support, or inability to comply with an outpatient regimen. I would add one more that we do not actually have in the guideline but I think it’s important. If the patient requires a surgical incision and drainage more aggressive than what you’re able to do in an outpatient setting. Other than that, there’s no real reason to have to hospitalize these patients, even with a moderate infection. You can start a course of outpatient antibiotic therapy. So, what do the guidelines say about antibiotic selection? We recommend that clinically, uninfected wounds not be treated with antibiotic therapy. I mentioned that earlier, that if there are no signs of infection you don’t culture, you don’t treat with antibiotics. We also recommend prescribing antibiotics for virtually all infected wounds that causing you that that antibiotics, in and of themselves, are not going to be sufficient unless combined with appropriate wound care. So, you need to debride and you need to offload the feet also. What about the empirical selection of antibiotics, what are you going to choose? We recommend clinicians selecting empirical antibiotic regimen based on the severity of the infection and the likely etiologic agents. In most cases that is going to be the aerobic gram-positive cocci, in particular staph and strep. So from mild to moderate infections in patients who have not recently had a course of antibiotics we suggest that therapy is directed against those aerobic gram-positive cocci, in particular gram Staph aureus and group B streptococcus or Strep agalactiae. For most severe infections, we recommend starting broader. In the antibiotic stewardship realm this is considered de-escalation therapy. We start broad, and once we have the cultures back we narrow down therapy because the whole concept of stewardship is to use the narrow spectrum antibiotic you can for the shortest period of time. So for the severe infections, we’re going to start broad, wait for those culture results, wait to see the patient’s response and then de-escalate as possible. In the IDSA guideline we have a list of antibiotics that have been studied in diabetic foot and we decide whether they’re good for mild infections or moderate/severe infections. This table, I don’t expect you read it, just looking at the slide. What I would suggest is you look at it in the document itself. They will give you some guidelines on antibiotics to use. Again, these are all drugs that have been studied for diabetic foot. Even though all those drugs have been studied for diabetic foot only three have actually received an FDA indication for the treatment of diabetic foot infections all of them say without concomitant osteomyelitis.


    And they are ertapenem, usually use 1 gram q. 24 I.V., piperacillin/tazobactam either 3.375 q. 6 or 4.5 grams q. 8. Some people use continuous infusion nowadays. And linezolid 600 milligrams either I.V. or p.o. q. 12. That’s it. There are only three antibiotics in the United States, FDA approved for diabetic foot infection. And under new guidance that I’m touched on in the lecture on MRSA you will see that there are not going to be a lot of new antibiotics FDA approved for diabetic foot at this time because it changed in the way the FDA looks at antibiotics. So for now, this is about it. What about MRSA, when do you treat MRSA? What we say in the guideline is that we recommend definitive therapy be based on both the results of an appropriately obtained culture and sensitivity and the patient’s clinical response. What do I mean by that? Let’s say you have a patient, you start him on a drug that doesn’t work against MRSA. And you do an incision and drainage and they’re looking really great. Well then the culture comes back and it has MRSA in it. You treat the patient, not the culture result. If the patient is doing well without MRSA therapy then obviously the MRSA is not a pathogen in this case, you don’t need to add the MRSA coverage. So that’s what we mean by adding in the patient’s clinical response. There are some cases, however, in which you do empirically want to add MRSA coverage. There are basically three situations, a patient who had MRSA, there’s a saying we have, is that once MRSA, always MRSA, a patient where you have a local prevalence of MRSA that’s high, and we don’t define, we don’t define what high enough is, can only lead up to you. But if you’ve got 60, 70% of the patients in your community, in your office, in your practice, in your hospital, 60, 70% of staff is MRSA, then I think it’s a good idea to start empiric anti-MRSA therapy. And then finally, if the infection is clinically severe. That’s simply because we can’t afford to be wrong. If you have a patient who is septic, that’s how severe infection is going to be, maybe he’s got a bacteremia caused by MRSA. If you start therapy and there’s no MRSA coverage you have to wait a couple of days, maybe three, four days before those blood cultures come back. That could be too late for the patient. So on a severe infection start the patient empirically. Route of therapy? We suggest basing the route of therapy largely on infection severity, we prefer parenteral therapy for all severe infections and some moderate. But again, many moderate and all mild can be treated with oral antibiotics especially with some of the new highly orally bioavailable drugs. What about duration of therapy? We suggest continuing antibiotic therapy until but not beyond resolution of the findings of infection. Remember, antibiotics treat infection, they don’t heal wounds. So you just use the antibiotic until the infection signs are gone. We suggest an initial course for a soft tissue infection of about one to two weeks for mild and two to three weeks for moderate to severe. And that’s only a weak low recommendation, there’s not a lot of good evidence to support that. Do we need to cover pseudomonas? I get this question all the time. Whenever I ask, what’s the most common organism we see in diabetic foot, and it was just, pseudomonas. Actually pseudomonas is wearily, if ever, a pathogen in diabetic foot infections at least in North America. Now, in Southeast Asia, in the southern hemisphere, there is more pseudomonas, but not in North America. Let’s look at some of these studies to show this. There’s a study that was published in 2002 by Graham in clinical infectious diseases that compared ertapenem to piperacillin/tazobactam. Now ertapenem has not have pseudomonal activity, piperacillin/tazobactam does. Yet in patients who grew pseudomonas, 70% had a positive clinical response in ertapenem, 60% on pip/tazo. So patients actually did better on the antibiotic that’s not effective against pseudomonas. That doesn’t mean that ertapenem works against pseudomonas, it means that it’s a colonizer, it’s not a pathogen. This was followed up by the largest study ever on moderate to severe diabetic foot infections, the SIDESTEP study that Lipsky’s work, published in 2005. Very similar except the other was all-comers of bacterials complicated skin and skin structure infections. This was just diabetic foot. 76.9% of patients with pseudomonas did well on ertapenem, only 70 on pip/tazo. Again, patients did better on the drug not active against pseudomonas. Just shows pseudomonas is not a pathogen. Let’s forget pip/tazo and ertapenem, relatively newer drugs ceftaroline came out.


    Now ceftaroline has no activity against pseudomonas yet 80% of patients who grew pseudomonas had a positive clinical response in ceftaroline. In fact that lead to this opinion the demonstration of efficacy in patients with Pseudomonas aeruginosa receiving ceftaroline, a pathogen against which ceftaroline has little activity most probably reflects the presumptive role of pseudomonas as a colonizer rather than a true pathogen in many of these infections. So let’s finish up just looking at our patient and his course of action or his course of therapy. Here is the first incision and drainage. This is a day or two after the first incision and drainage. And I don’t like the way this looks. It’s still erythematous. It’s still swollen. In fact, there was a little area, right there, that just was a little friable when I touched it. It just kind of drained a little bit. So I called the surgeon back and he took the patient back to the OR the next day and that’s how it looked. Now this is great. I love the look of this one. The wound is clean, you can see the cellulitis is down, there’s no erythema, there is no edema. You can see dressing creases, you can see skin lines. This really looks good. Three weeks later, I happened to walk by the wound care center and saw this patient. And this is how the foot looked. So this is a success, by all accounts, this patient did really well. Unfortunately, we all know that the recidivism rate on these patients tends to be a little bit high and they always come back. Almost exactly a year later, the patient came back with a new infection. Now what happened, this is a different location, his first MTP joint wasn’t there anymore because it have been resected, but now he was wearing special orthotics and sure enough the orthotic just rubbed the hole into the plantar aspect of the foot and that became infected, with the exact same organisms we had a year before. A year after that, this patient was hospitalized, and about a year after that he came in again. And this gentleman has been in at least half dozen times since this point. And he was my concern. In fact, I just saw him the other day. He is now status post to transmet amputation. Here is my point on this. When this gentleman first started coming in he grew methicillin-susceptible Staph aureus and some gram-negatives that we could treat with cefazolin and stuff. Now, every time he comes in he’s growing organisms resistant to all known antibiotics including carbapenem-resistant Enterobacteriaceae. Because of all these courses of antibiotics and every time he’s been discharged to a SNIF the patient just becomes more and more resistant to antibiotic therapy. After a while you start having to wonder when does it become time to maybe go ahead with a definitive procedure to one of these patients to prevent him from possibly becoming systemically infected. In conclusion, diabetic foot infection is devastating both to the patient and to the healthcare system. The new 2012 IDSA Diabetic Foot Infection Guidelines present 44 evidence-based recommendations, answering 10 of the most common questions on the diagnosis and treatment of diabetic foot. I highly recommend you all look at them. Again, you can find them at idsociety.org. Click the tab that says Clinical Practice Guidelines. They are open source so you can download them, you can share them with colleagues. You can spread them around, spread the good word around. And then preliminary evidence suggests that following evidence-based guidelines can improve patient outcomes. Thank you very much for your attention.