Kowalski TJ, Matsuda M, Sorenson MD, Gundrum JD, Agger WA.  The effect of residual osteomyelitis at the resection margin in patients with surgically treated diabetic foot infection. J Foot Ankle Surg.  2011 Mar-Apr; 50(2): 171-5. (PubMed ID: 21354001)

WHY did the authors undertake this study?
One of the worst feelings that I regularly have as a foot and ankle surgeon comes a couple of days after a partial foot amputation with delayed primary closure, when I get either a pathology report or culture result back telling me that there is probably some remnant infection at the surgical site that I just felt confident enough to close. The authors of this study aimed to more closely examine this unfortunately common situation by retrospectively evaluating the effect of residual osteomyelitis after definitive partial foot amputation in diabetic patients.

HOW did they attempt to answer this question?
Inclusion criteria of the population cohort were diabetic patients who had surgically resected foot osteomyelitis with a histopathologic analysis of the residual bone margin (n=111). Thirty-nine patients (35%) had pathology reports returned with margins positive for osteomyelitis, whereas 72 patients (65%) had pathology reports returned with margins negative for osteomyelitis.

The primary outcome measure of the study was “definite treatment failure” defined as either histopathologic or microbiologic evidence of relapsed bone infection. Secondary outcome measures were also recorded, including clinician-diagnosed skin or soft tissue infection or the need for more proximal amputation.

WHAT were the specific results?
A statistically significant difference was found  between those patients with positive margins and negative margins with respect to “definite failure or need for proximal amputation” (44% vs. 15%; p=0.001). Patients with positive margins were also noted to be on antibiotic therapy for a significantly longer period of time (19 days vs. 14 days; p=0.01).

HOW did the authors interpret these results?
From these results, the authors concluded  that residual osteomyelitis at the resection margin was associated with a higher rate of treatment failure, even considering these patients underwent a longer duration of antibiotic therapy.

Jaung R, Cook P, Blyth P. A comparison of embalming fluids for use in surgical workshops. Clinical Anatomy.  2011 Mar; 24(3): 155-161. (Pubmed ID#: 21322038)

WHY did the authors undertake this study?
For better or for worse, medical and surgical training is very much based on a “learning curve”, and the more often a training surgeon practices a certain skill, the better at it they likely will become. This isn’t often practical in clinical practice, and a large part of surgical training now comes from cadavers, animal models and other anatomic specimens. The authors of this study aimed to compare three different cadaveric embalming methods with emphasis on which may be best for surgical training workshops. 

HOW did they attempt to answer this question?
Four primary outcome measures were evaluated on human cadaveric specimens: joint flexibility, tissue pliability, tissue color and resistance to fungal growth. Three different embalming preparations were utilized on eleven total cadavers: Graz (n=1), Dodge (n=4) and Genelyn (n=5).

WHAT were the specific results?
The Graz method provided joint flexibility, tissue pliability and tissue color most comparable to fresh tissue, as well as not growing any mold in this study. 

HOW did the authors interpret these results?
Based on these results, the authors concluded that although technically demanding, the Graz method of embalming produces a cadaver most similar to fresh tissue. 

Let’s take a closer look at the topic of how medical decision making guides medical research and vice-versa, specifically as it applies to the Kowalski et al osteomyelitis article. One line of this study in particular really got me thinking about how slow progress can be with respect to medicine, even though sometimes it feels as though everything we do is so fast and often instantaneous. The line is actually the very last sentence of the abstract which reads “Residual osteomyelitis at the pathologic margin was associated with a higher rate of treatment failure, despite the longer duration of antibiotic therapy”. 

After initially reading this sentence I didn’t agree that it was an appropriate conclusion, and still don’t after a couple weeks of thinking about it, but I do agree that it brings up a broader point.  The duration of antibiotic therapy was not a primary variable in this study and was really just one of the patient characteristics.  How long the patients received antibiotics was “at the discretion of the treating clinicians”, and it so happened that patients with positive margins were treated for a longer duration (median 19 days; range 10-134 days) than patients with negative margins (median 14 days; range 2-63 days). Although this difference is statistically significant (p=0.01 with a Wilcoxon rank sum test), I’m not sure how much clinical significance there is between 14 and 19 days of antibiotic therapy in this cohort of patients.

In fact, my mind immediately jumped to a set of guidelines that I usually use when presented with this situation (The Infectious Disease Society of America’s [IDSA] Diagnosis and Treatment of Diabetic Foot Infections). In this guideline, the IDSA recommends 4-6 weeks of antibiotic therapy if there is residual infected bone, 2-4 weeks of antibiotic therapy if there is residual infected soft tissue, and 2-5 days of antibiotic therapy if there is no residual infected soft tissue or bone. It doesn’t appear as though the clinicians in the Kowalski et al study were strictly utilizing these guidelines with a total median duration of therapy between 2-3 weeks and a total duration range of therapy from 2 days to 19 weeks. 

I think this example provides interesting insight into how cyclical our field of medicine can often be.  When faced with a medical decision (like what duration of antibiotic therapy to place a patient on), I like to use established guidelines in the literature as much as possible (like the IDSA guidelines). These guidelines aren’t set in stone however, are very often based more on expert opinion than Level I evidence, and change every couple of years (in fact, I think the IDSA is coming out with revised diabetic foot infection guidelines in the near future). They change based on new findings in the medical research, like this study which provides evidence on the successes and failures of current antibiotic prescribing patterns in real-life clinical situations. So although I am being critical of how a small section of this study didn’t match up with my own personal medical decision making, it is very possible that it will be used to guide my medical decision making in the future.