A recent paper has been published in Clinical Infectious Diseases by Brad Spellberg at UCLA and Ben Lipsky at the Puget Sound VA titled Systemic Antibiotic Therapy for Chronic Osteomyelitis in Adults (http://www.ncbi.nlm.nih.gov/pubmed?term=Spellberg%20B%20AND%20Lipsky%20BA).  I consider it a MUST READ for followers of this site.  This excellent review covers topics including the pharmacology of osteo therapy (i.e. parenteral vs. oral, bone penetration), animal models of osteo, human non-randomized clinical trials and randomized clinical trials.  It is THOROUGH yet quite readable at only 11 pages (there are, however, 172 references!) with 5 tables outlining all of the studies discussed. 

The authors arrive at 4 conclusions which I quote directly from the paper:

1.  “Oral antibiotic therapy with highly bioavailable agents is an acceptable alternative to parenteral therapy.”

2.   “Adding rifampin to a variety of antibiotic regimens has been shown to improve cure rates”

3.   “Clinicians must individualize the duration of antibiotic therapy based on the patient’s clinical and radiographic response…”

4.   “Surgical resection of necrotic and infected bone, in conjunction with antibiotic therapy, appears to increase the cure rate of chronic osteomyelitis. However, not all cases of chronic osteomyelitis require surgical debridement for cure, and we need studies to clarify which may and which may not.”

None of these conclusions should come as a surprise to regular readers of this blog or those who have heard me lecture on the topic, as I have discussed these very points in the past.  Dr. Lipsky and I collaborate frequently and I find it almost frightening how often we agree.  In this one paper he and Dr. Spellberg have eloquently laid out all of the evidence supporting these positions.  If I have said it once, I have said it a thousand times…We MUST rethink the universally pervasive dogma of 4-6 weeks of IV antibiotic therapy for osteomyelitis based on the best available evidence!

Pseudomonas colonizing a heel wound

A number of randomized, controlled clinical trials have actually substantiated this viewpoint.  Most have looked at the clinical outcome of patients who grew PA from a wound/infection when treated empirically with an antibiotic that was ineffective against PA vs. one which had anti-pseudomonal activity. In this entry I will present 3 such trials.  

In 2002 Graham, et.al. published the results of the pivotal phase III trials comparing ertapenem to piperacillin/tazobactam for complicated skin and skin structure infections (PubMed link: http://www.ncbi.nlm.nih.gov/pubmed/12015692).  Of these two antibiotics ertapenem has no inherent anti-PA activity while pip/tazo does.  Despite this difference in spectrum, the positive clinical response in patients growing PA was 70% for ertapenem and 60% for pip/tazo.  In a very similar trial comparing the same two antibiotics specifically for diabetic foot infection, Lipsky et. al. had similar outcomes with a positive clinical response in 76.9% of patients on ertapenem vs. 70% for pip/tazo. (The SIDESTEP trial http://www.ncbi.nlm.nih.gov/pubmed/16291062).  More recently, in September 2010, Corey, et. al.  published the results of the clinical trials on the new anti-MRSA cephalosporin ceftaroline .  This drug has no inherent anti-PA activity yet there was a positive clinical response in 80% of the patients from which the organism was isolated.   

None of these trials are meant to show that the study drug had some previously undiscovered anti-PA activity.  They just point to the role of PA as a notorious colonizer of lower extremity wounds/infections. Nannini stated it quite succinctly in his review paper of ceftaroline (http://www.ncbi.nlm.nih.gov/pubmed?term=nannini%20ceftaroline):

“The demonstration of efficacy in patients with P. aeruginosa receiving ceftaroline, a pathogen against which ceftaroline has little activity, most probably reflects the presumptive role of P. aeruginosa as a colonizer rather than a true pathogen in many of these infections.”

Please do not misinterpret me.  I am not saying that PA is always a colonizer and is never pathogenic.  If this bug is found from a reliable, deep surgical culture, consideration should be given to covering it. However, this is a ubiquitous organism that has been found in tap water and on plants and vegetables.  Just culturing it from a superficial wound swab will lead the clinician to treat an organism that can most likely be handled with thorough debridement and topical therapy, with an anti-PA antibiotic that may be unnecessary.  This may increase resistance of the organism against these antibiotics so that they are not available for the next time…when we may really need them.

“What is your opinion on the ‘standard of care’ of the need for serial imaging (x-rays in particular but also possibly MRI or nuclear scans can be considered) in a chronic plantar wound, in a patient with diabetes?  The wound does not probe and there are no signs of clinical infection.  The wound is probably not healing because of non-adherence with off loading instructions. How often would you order the different imaging studies?”

The reason I am bringing this up to you is that I think it is a really interesting question.  What do you think?  Please respond and let me know.  After I hear from you, I will try to give you my thoughts on it.

In the February 2011 issue of International Orthotpedics T. Rod-Fleury and colleagues looked at duration of post surgical antibiotic therapy in adult chronic osteomyelitis.  (http://www.ncbi.nlm.nih.gov/pubmed?term=Rod-Fleury%20T) This is from the same group at Geneva University Hospital that published on the utility of sinus tract cultures on which I commented back in November of this past year.  They retrospectively examined 49 episodes of chronic osteo with a minimum of 2 years follow-up.  I should point out that they specifically excluded diabetic foot osteo.   The patients underwent a median number of 2 surgical debridements.  These were bad cases with almost all being considered Cierny-Mader IV category disease with the patients having a minimum duration of symptoms of 3 months. They found that 80% were in remission at the two year point.  What I feel is important to stress is their conclusion right from the Abstract that, after multivariate logistic regression analysis “one week of intravenous therapy had the same remission as two to three weeks or ≥ 3 weeks.  More than six weeks of total antibiotic treatment equaled ≤ six weeks.”  To quote the conclusion of the paper “If our retrospective results are confirmed, a shorter or oral antibiotic treatment post-debridement could further decrease antibiotic consumption, as well as the related costs, adverse effects and selective pressure for resistant bacterial pathogens.”  Now, this is far from a perfect study and the authors do a commendable job citing the limitations of the study.  Also, they take the almost chauvinistic surgical attitude right in the beginning of the Introduction that “Chronic bacterial osteomyelitis is a surgical disease” (still not totally proven in my mind), but it does again bolster the argument that we need to re-examine this entire 4-6 week IV dogma.

Overall, this is an excellent piece of work and kudos goes out to lead author Dr. Catherine Liu of UCSF and the committee.  Sitting on an IDSA Guidelines committee, I understand the process.  It could not have been easy to come to consensus particularly in an area of ID that is changing so rapidly and where there is the potential for so much controversy.

Although much of the document pertains to infections other than skin/skin structure and bone/joint I still feel it is important to list those points from the Executive Summary which could have direct impact on a lower extremity infection practice.  With all of the new IDSA Guidelines the required format is “Question, Answer, Evidence Summary”.  I will pick out parts of those that apply to our practice and would ask you please read the entire document for the complete picture.  As you would expect, I will not resist adding my 2 cents! My comments will appear in italics

Question I:  What is the management of skin and soft tissue infections in the era of community-associated MRSA?

1. For a cutaneous abscess I&D is the primary treatment.  For simple abscesses or boils I&D alone is likely to be adequate. I have covered this subject in the past.  The treatment of a CA-MRSA simple (usually defined as <5 cm in diameter) abscess is I&D.  There is essentially no evidence that adjunctive antibiotics are warranted.

2. Antibiotic therapy is recommended for abscesses associated with antibiotic therapy is recommended for abscesses associated with the following conditions: severe or extensive disease (eg, involving multiple sites of infection) or rapid progression in presence of associated cellulitis, signs and symptoms of systemic illness, associated comorbidities or immunosuppression, extremes of age, abscess in an area difficult to drain (eg, face, hand, and genitalia), associated septic phlebitis, and lack of response to incision and drainage alone. This point is valuable in that it actually spells out those conditions where antibiotics should be considered adjunctive to the I&D

3. For outpatients with purulent cellulitis (eg, cellulitis associated with purulent drainage or exudate in the absence of a drainable abscess), empirical therapy for CA-MRSA is recommended pending culture results. Empirical therapy for infection due to beta-hemolytic streptococci is likely to be unnecessary. Five to 10 days of therapy is recommended but should be individualized on the basis of the patient’s clinical response. This is an interesting point in it differentiates “purulent” cellulitis from “nonpurulent” cellulitis (see next point).  Frankly, I am not certain that I have seen that in the past.  Most of what we are seeing in the lower extremity is the purulent variety but, as I write this, I have a patient in the hospital with venous insufficiency and a significant cellulitis of her leg, but with no pus.  Sure enough, cultures of the clear drainage just this morning grew Group B, beta-hemolytic Strep.  (See picture below)

4. For outpatients with nonpurulent cellulitis (eg, cellulitis with no purulent drainage or exudate and no associated abscess), empirical therapy for infection due to b-hemolytic streptococci is recommended. The role of CA-MRSA is unknown. For the patient I just mentioned above, she had been started empirically on vancomycin from the ED, she did not improve to my satisfaction so we added piperacillin/tazobactam and had remarkable improvement within 24 hours. I stopped the vanco. 

5. For empirical coverage of CA-MRSA in outpatients with SSTI, oral antibiotic options include the following: clindamycin, trimethoprim-sulfamethoxazole (TMP-SMX), a tetracycline (doxycycline or minocycline), and linezolid. If coverage for both beta-hemolytic streptococci and CA-MRSA is desired, options include the following: clindamycin alone or TMP-SMX or a tetracycline in combination with a beta-lactam (eg, amoxicillin) or linezolid alone. This point really just lists the oral antibiotic choices we knew we had all along.  I would direct you to the documents “evidence summary” to see how little actual data there is backing any of this though.  Furthermore, I am not as enamored with TMP/SMX as the recommendations in these Guidelines and tend to prefer using doxycycline or minocycline.

6. The use of rifampin as a single agent or as adjunctive therapy for the treatment of SSTI is not recommended. I agree and do not use rifampin.  As you will see in the osteomyelitis and prosthetic joint section these guidelines are very “pro” rifampin despite what I feel is a lack of any significant evidence.

7. For hospitalized patients with complicated SSTI (cSSTI; defined as patients with deeper soft-tissue infections, surgical/traumatic wound infection, major abscesses, cellulitis, and infected ulcers and burns), in addition to surgical debridement and broad-spectrum antibiotics, empirical therapy for MRSA should be considered pending culture data. This point it BIG!  They are recommending pretty much ALL patients admitted with cSSTI, which would translate into almost all admitted patients, be started on anti-MRSA therapy.  There is no leeway given for local/regional variations in incidence.  Sure, most of us have been doing this but, frankly, in the past 6 months to a year, most of our admitted patients are NOT growing MRSA at this point.  I have actually started going back to escalation (no MRSA empiric coverage) therapy in some cases. 

Given the importance of this document I would rather break down the discussion into shorter, hopefully more easily digestible portions.  Above I covered the first section on Skin and skin structure infections.  I also want to comment on the bone and joint recommendations which include osteomyelitis, septic arthritis and implant related infections along with my favorite part (see my previous postings on the topic) the recommendations for vancomycin dosing and monitoring.  Stay tuned…

Non-purulent cellulitis casued by Group B beta-hemolytic strep, not MRSA

The whole idea of this blog is to be interactive.  I want to hear others’ thoughts and ideas!  This is what will make it intellectually stimulating and fun for me and, hopefully, valuable to those monitoring the site

To this end I would like to start a “What Would You Do?” entry.  I will ask a clinical question or two, maybe present a case, and ask you for feedback on what you would do in this situation.  Give me your thought processes that lead you to your position.  Then, in a subsequent post I will review the responses and give you my approach along with any pertinent clinical evidence I have to back it up.

Let’s start with this; a recent very well done study was published on dental procedures and prosthetic joint infections (I will not give you the reference yet. That will hold until the answers come in) that may alter the way we look at antibiotic prohylaxis in these situations.  So, here are the questions/clinical scenarios I have for you:

1.  You placed a prosthetic joint into a patient’s foot.  Or, let’s even expand it out since there aren’t as many joints being placed anymore, you placed some hardware into a patient’s foot while doing surgery on that foot.  Let’s say the surgery was 18 months ago.  The patient calls you because she is going to the dentist for some dental hygiene and to have a crown placed on a tooth.  DO YOU, OR DON’T YOU, RECOMMEND ANTIBIOTIC PROPHYLAXIS FOR THIS PATIENT BEFORE THE DENTAL WORK?

2.  A patient comes to see you for the chief complaint of a bunion.  You discuss surgical correction.  After the patient agrees to have the surgery scheduled she informs you that she had a prosthetic knee implanted 18 months ago.  DO YOU PLACE THE PATIENT ON PROPHYLACTIC ANTIBIOTICS TO PREVENT A PROSTHETIC JOINT INFECTION FOLLOWING YOUR FOOT SURGERY?

OK, have at it.  There are really no right or wrong answers (or, maybe there are…it depends on the answers I get).  Once I receive some input I will give you my thoughts on the subject and give a summary of this recent study to which I referred above.

This comment came in under my entry on the VULCAN trial but I actually covered the question in a post almost exactly a year ago.  I am reprinting my response here for more recent readers who may not have reviewed all of the earlier posts I will also summarize my current thinking at the bottom:

The question of whether or not the patient needed to be off of all antibiotic therapy prior to the culture was always less clearly delineated.  It was almost an empirical belief that if a patient was on antibiotics at the time of the culture then the results would be unreliable.  After all, if a culture is reported as showing “no growth” was this because the specimen was poor, the patient never had osteomyelitis in the first place, or the antibiotics had prevented the organism from growing?

A recent study by Louis Bernard and colleagues out of University Hospital of Geneva Switzerland, and published in the September 6, 2009 issue of the International Journal of Infectious Diseases challenges much of what we currently hold as sacrosanct.  In this non-randomized, prospective trial 141 patient with 154 episodes of osteomyelitis each underwent 4 microbiological samplings.  Sample “A” consisted of 2 consecutive sinus tract cultures with bone contact (“A1” & “A2”) after only local cleansing of the sinus tract with 0.9% saline.  Sample “B” consisted of a surgical bone biopsy through the sinus.  Sample “C” was a surgical biopsy obtained through an uninfected sited, referred to as the “gold standard”.  Their results showed that when both sinus tract cultures “A1 & A2” revealed the same organism the concordance between “A” and gold standard “C” was 96%.  They calculated the sensitivity as 91%, specificity as 86% and accuracy as 90%.  On top of that, they found that the result was not affected depending on whether the patient was on antibiotics prior to the cultures or not.

In their Discussion the authors go out of their way to state that “in no way should these consecutive deep sinus tract specimens replace bone culture in situations where a biopsy can be readily obtained because bone culture remains the gold standard for the microbiological diagnosis of osteomyelitis”.  There are sometimes, however, when a surgical culture is just not practical and that patient has already been started on antibiotic therapy, as the authors call it “clinical reality”.

Since this original post, my thought process has not changed much.  I still believe that, if they have not yet been started, it is best to HOLD antibiotics if at all possible, until after the patient is taken to the OR for deep bone (or, for that matter, soft tissue) cultures. The “real world” gets involved, however, when you admit a patient through the Emergency Dept for the stated diagnosis of “infection” and do not plan on taking the pt to the OR until the next day. Unfortunately, in a case like this, you cannot realistically hold antibiotics or the hospital runs the risk of losing the admission.  If the patient is going to the OR later the same day then I believe in holding the drug unless the pt is sick i.e. metabolically unstable, systemically unwell or septic.

What if the patient has already been started on antibiotics?  Conventional wisdom still calls for the pt to be off antibiotics for at least 48 hours.  I don’t believe there is any evidence to support that thinking.  The Bernard study cited above does give us some hope that it doesn’t matter if the patient is still on antibiotics or not but I would still recommend, again if at all possible, taking the patient off the drugs for as long as it is feasible. Heck, I would go for a week if you can.  There is no magic to 48 hours.  The problem is that if the culture is negative, you don’t know whether it is negative because it is not osteo or if it is negative just because of the antibiotics.  Also, the International Working Group on the Diabetic Foot (www.iwgdf.org) has shown in their diabetic foot osteomyelitis guidelines that the only independent factor leading to a positive response to antibiotic therapy for osteo is bone culture directed antibiotic therapy.  We could really use those deep reliable cultures!

McCarthy NL, et al.  Health care-associated and community associated MRSA infections: A comparison of Definitions.  American Journal of Infection Control, 2010:  This retrospective chart review attempts to determine risk factors for CA-MRSA and HA-MRSA based on two commonly used definitions; the epidemiology of the infection and the phenotypic antibiotic sensitivity without looking at the genotype (i.e. looking for the SCCmecIV gene to determine a USA300 strain).  Not surprisingly, to quote their Conclusion from the Abstract “We found few differences between CA- and HA-MRSA infections regardless of how health care association was found. We believe that the migration of CA-MRSA into health care settings and the recent increasing antibiotic resistance of CA-MRSA strains contribute to the lack of factors associated with HA (vs CA) MRSA.”  One finding I did think was particularly relevant was that CA-MRSA accounted for the overwhelming majority of skin and soft tissue infections (p <.0001).

For the past year or two I have been lecturing and writing that the use of classic “risk factors” to determine HA vs CA was a technique of the past. This was recently driven home to me when I saw 3 patients in the hospital with MSSA, 2 of whom were 80+ years old and lived in long term care facilities while I had a perfectly healthy 30 something with no pertinent medical history whatsoever present with a CA-MRSA.  I still continue to use the phenotypic method of looking at antibiotic susceptibilities, particularly toward Bactrim and tetracyclines since we have not seen much CA-MRSA resistance to these drugs at our facility.  It is becoming clearer that perhaps we will need to do the genetic typing, such as the TEM-PCR I discussed in a recent post to really know what we have.  Of course, it may be a moot point in skin and skin structure infections since, as validated by this report, I don’t remember seeing an HA-MRSA in over 2 years.

Nagoba BS et al. A Simple and effective approach for the treatment of diabetic foot ulcers with different Wagner grades.  International Wound Journal, April 23, 2010:  This is a retrospective case series from India that collected data from 1998 into 2009.  They looked at the application of a 3% citric acid gel preparation onto 115 diabetic foot ulcerations of various Wagner classification severities up to, and not including grades IV and V.  Of interest to me was that they took cultures of all of the wounds and performed broth dilution sensitivities of all the organisms against the citric acid.  They found that in grade I-II ulcerations they had 94%+ healing in 5-34 applications.  They also found that in deeper wounds there was marked decrease in slough and drainage with increases in healthy granulation.  They concluded that citric acid was safe and maybe even promoted granulation tissue while killing the bacteria found on the wounds.   

I find various problems with the design of this study and am not reporting it because I consider it an earth shattering breakthrough.  Rather, I bring it to your attention because, currently, one of the “Holy Grails” of wound healing is to find an safe, effective, affordable topical antimicrobial to, perhaps, assist in wound healing.  Heck, I am still not even convinced that bioburden reduction is all that many claim it is, as I have blogged before, however, we are all programmed to kill bacteria on a wound and maybe this is an option.  It should probably at least be looked at in more detail.

Traunmuller F, et al. Linezolid concentrations in infected soft tissue and bone following repetitive doses in diabetic patients with bacterial foot infections.  International Journal of Antimicrobial Agents, 2010:  This study assayed fluid from interstitial spaced, inflamed subcutaneous adipose tissue and metatarsal bone from patients with inflamed diabetic foot infections who were at steady state condition after 600 mg twice daily dosing of linezolid.  What they found was not particularly surprising; in all tissues linezolid penetrated to a level that was similar, if not slightly higher than what was found in plasma.  To quote their Abstract “We conclude that linezolid administered at 600 mg twice daily may be considered an effective treatment in diabetic patients suffering from bacterial foot infections complicated by osteomyelitis”. 

We all know that linezolid is not FDA indicated for the treatment of osteomyelitis.  In fact, NO antibiotic marketed in the past 10-12 years has been indicated for osteo, not because they don’t work but because the FDA has not developed “Guidance for Industry” on how to perform an osteomyelitis study.  I have also never been a believer in the concept of “bone penetration” as being an important indicator of osteo efficacy.  They techniques have never been standardized and really, to my mind, penetration numbers mean nothing. That all being said, at least this paper gives those of you who remain stuck on the concept of bone penetration evidence that linezolid does get into, not only bone but also frankly infected diabetic foot tissues.  This is not unlike data published by Gary Stein, PharmD and recently reviewed in the Jan/Feb 2010 issue of the Journal of the American Podiatric Medical Association (Nicolau DP, Stein GE. Therapeutic options for diabetic foot infections: A review with an emphasis on tissue penetration characteristics)

Prior to my session, I will be in the Exhibit Hall from 9:30 – 11:30 at the Data Trace Booth to sign copies of the Handbook, talk about this blog and just answer questions anyone might have about treating infections.

I hope to see folks there!

In the 3^rd Edition of the Handbook I review the evidence that shows that perhaps we should rethink this “Standard”.  The entire concept of 4-6 weeks of IV antibiotics as being some magical bullet that will treat osteomyelitis is based on 30-40 year old animal studies.  There is really no human evidence to support that this is the optimum therapy.  In fact, much of the recent data that has been presented, particularly out of Europe, suggests that ORAL antibiotics alone with either NO surgery or just “conservative” debridement of infected bone may be just as effective therapy.  The literature supporting this approach is actually quite a bit more hefty that that supporting the old traditional “standard”.  I explore this question in detail in a lecture I frequently present entitled (thanks to David Armstrong for the catchy title) Antibiotics for Osteomyelitis: Boneheaded or brilliant? One of these presentations can be viewed for free at the website for the always stimulating Diabetic Foot Global Conference (to be held this coming year in Hollywood, CA March 18-20, 2010): http://www.dfcon.com/presentations2009/sat_joseph.html

In no way do I mean to say that surgery is unnecessary.  In fact, if ablative surgery can be performed, it is curative without the use of any antibiotic therapy.  But, sometimes, patients either do not want, or can’t undergo a surgical procedure.  In those cases I am relieved to see that we do have data to support the non surgical treatment of these cases with oral antibiotics alone.  Of course, this approach is not perfect and there are still unanswered questions.  Probably the greatest is the question of duration of therapy.  Do we only need the orals for 4-6 weeks or longer?  One study, referenced in the lecture, by John Embil incorporated 41 weeks +/- 31 weeks of therapy!  This needs to be addressed.  The other literature I would direct you to review is the consensus statement put out by the International Working Group on the Diabetic Foot (IWGDF) http://iwgdf.org  whose recent systematic reviews of the diagnosis and treatment of diabetic foot osteomyelitis should be reviewed by everyone treating this condition. 

To me, this is one of the most fascinating and important areas that we need to explore in the future.

Osteomyelitis of the 1st MTP Joint