|
While acknowledging several limitations of the SIDESTEP study, the panelists discuss how the accumulated data will be used as a springboard to future publications involving pharmacoeconomic observations, risk factors for treatment failure and other areas of interest.
Dr. Armstrong: Dr. Lipsky, in your opinion, what were some of the limitations of the SIDESTEP study? What would you do differently were you designing the study today versus several years ago?
Dr. Lipsky: We knew there were some inherent limitations to the study when we designed it. Perhaps the most obvious was the lack of an exact oral equivalent to either ertapenem or piperacillin/tazobactam. Therefore, when we switched patients over to an oral antibiotic drug, it was difficult to differentiate between the effects on outcome of the initial parenteral therapy versus the later oral therapy.
Secondly, we excluded mild infections. This was because most patients with mild infections do not need parenteral therapy. Accordingly, we limited the kinds of patients we could enroll, and therefore the broader application of the results. We also excluded patients who had osteomyelitis, unless all of the infected bone was surgically removed. Therefore, we cannot say anything about the efficacy of either of these drugs for treating osteomyelitis. We also excluded patients who had critical limb ischemia that required revascularization so we cannot speak to whether the drugs might be effective in that setting.
The numbers of patients enrolled at the different sites varied quite a bit, causing us to drop some sites and add new ones. In addition, some of the sites did not track all of the minor surgical procedures as well as we had asked them to. We also had no long-term follow-up for these patients so we cannot speak to what happened beyond the 10 days after they concluded their antimicrobial therapy.
Dr. Armstrong: In doing another study then, we would probably want to include or evaluate patients with more severe ischemia and also track any adjunctive procedures that are done. Are these the things you are getting at in terms of looking at future studies in this area?
Dr. Lipsky: I would be interested to hear opinions from the other panelists, but the questions we have raised today are among the ones that I would love to see addressed in future studies. There are a number of companies looking at new antimicrobial products, both approved antibiotics that are already on the market and are being evaluated for a specific indication for diabetic foot infections as well as newly developed agents, particularly ones with anti-MRSA activity. Companies are also looking at some unique products, such as antimicrobial peptides, that may be administered by unusual routes, such as topical therapy or direct installation into the wound.
|  | | “There are still questions about diabetic foot infections. … What is the
optimal
duration of
therapy with
antimicrobials?” – Dr. Lipsky |
We still have many questions to answer with diabetic foot infections. What are the true pathogens and which among them do we have to specifically target with therapy? How do we approach the treatment of patients who have bone infection? Can we use antimicrobials alone or is it necessary to have some form of surgical resection of necrotic, if not infected, bone? How well do antimicrobials get to infected sites in patients who have varying degrees of ischemia of the limb? What is the minimal duration or the optimal duration of therapy with antimicrobials in treating patients with diabetic foot infections? Do clinically uninfected wounds benefit from antibiotic therapy? Which among the available adjunctive therapies are useful and for which patients? Those are just some of the questions that still need to be answered
Dr. Lavery: One of the things that has not been discussed is stratifying moderate and severe infections, and looking at amputations and limb salvage. That is the quality of life side of the story. If there longer-term follow-ups from the SIDESTEP study, it would speak to recidivism and those are the folks who are probably going to have more wounds in the future.
A Few Thoughts About The Diabetic Foot Wound Score
And Potential Risk Factors For Treatment Failure
Dr. Armstrong: What are some of the other studies that are currently in the works using the SIDESTEP data set?
Dr. Lipsky: We developed a wound score that was modified from one that we had used in previous trials. In this study, we were able to show that the score, which included measurements of the size and depth of the wound as well as semi-quantitative scores for the degree of various signs of inflammation, correlated well with the clinical response to therapy.36
We are also very interested in the clinical risk factors for treatment failure. A great majority of the patients in the SIDESTEP study were successfully treated with antibiotic therapy. However, in looking at those that failed, we asked the question: Are there factors that were present at the time of their initial evaluation that could predict who was going to do poorly? We ask because if a clinician knew that a patient had a risk factor for failure, he or she might be more aggressive about hospitalizing the patient, getting a surgical consultation or considering other forms of aggressive therapy.
While there were nine univariate factors that appeared to have a statistically significant association with failure, only two were significant independent risks in the multivariate analysis. These were an elevated white blood cell count and a severe wound as classified by the University of Texas scheme.37
We also have some interesting microbiology data that we obtained from the study. Diane Citron and Ellie Goldstein, MD, have presented some of these data in abstracts at the recent ICAAC and IDSA meetings.38,39 In an elegant analysis of the organisms isolated from patients with diabetic foot infections, they noted that anaerobic bacteria can frequently be isolated using optimal methods. In studies comparing ertapenem and piperacillin/tazobactam, they also noted that the anaerobic organisms that are isolated are more often gram-positive cocci (e.g., peptostreptococci) in diabetic foot infections as compared with gram-negative rods (e.g., Bacteriodes) in intraabdominal infections.
Furthermore, they found that the incidence of resistance to fluoroquinolones among the anaerobes was significantly higher in patients who had diabetic foot infections compared to those who had intraabdominal infections. This was probably due to a higher rate of exposure to fluoroquinolones among diabetic patients with foot infections prior to their presentation than for the patients who had intraabdominal infections.39
Dr. Sheehan: In regard to the presentation at the IDSA this year on the diabetic foot infection risk factors for treatment failure, the two factors, as Dr. Lipsky has mentioned, are the white blood cell count and the wound severity score.37 Just to clarify the wound severity score for the clinicians, it is essentially the depth of the wounds. We call it the University of Texas (UT) wound severity score based on grade — for grade 0 or 1, it is moderate, for grades 3 and 4, it is severe. So it is really the depth of the wound. A deep wound is less likely to respond to therapy and this brings up a number of possible implications. It will be interesting to evaluate this in the future.
Dr. Armstrong: That is critical.
Dr. Lipsky: I would also add one other thing about that study, on which Dr. Sheehan was the lead author.37 There were many risk factors that did not correlate with a worse outcome. Accordingly, it is important that studies in this area not just concentrate on what we find to be independent risk factors. We all have biases about which factors would lead us to decide to hospitalize a patient, put him or her in the ICU, give the patient broad spectrum therapy or get the surgeons involved. However, it was important that a whole variety of things, including C-reactive protein levels, erythrocyte sedimentation rates, serum albumin and serum glucose levels, did not correlate with outcome.
|  | | Here is a view of osteomyelitis. Patients with osteomyelitis were excluded from the SIDESTEP study unless all of the infected bone was surgically removed. |
A Preview Of Early Pharmacoeconomic Findings
Dr. Armstrong: There are a number of other studies that are being contemplated. One will look at the potential value of dermal thermography as a surrogate marker for inflammation.40 Alan Tice, MD, will also be leading a study that will be looking at the usefulness of serial digital photographs. These photographs may ultimately play a role in how we assess patients from afar and in making determinations of when to switch therapy and resolution.
In the near future, we will also see the emergence of pharmacoeconomic evaluations of the data from the SIDESTEP study.
Dr. Lipsky: I recently received a copy of an abstract of a study based on data from the SIDESTEP study that will be presented soon.41 The authors looked at the subset of about 100 patients whose entire course was with intravenous antibiotic therapy. In other words, they were not switched to oral amoxicillin/clavulanate and they were all seen entirely as inpatients.
They looked at costs based on the actual dosing of these patients with antibiotics. Given that ertapenem is a once-a-day drug and piperacillin/tazobactam is dosed four times a day, it is not surprising that the mean dose of active drug given for treatment was significantly lower for ertapenem (7.6 versus 25.7).
Interestingly enough, the difference in total costs was significantly lower for the patients who received ertapenem compared to those treated with piperacillin/tazobactam ($358 versus $511). The $153 difference between groups accounts for ~3% of total hospital DRG reimbursements for a diabetic foot infection for Medicare patients. These data emphasize an aspect of the importance of a once-a-day drug. Acquisition costs are not the only issue in terms of treating these patients. The frequency of dosing becomes very important.
Dr. Joseph: Did those numbers cover the actual cost of drug acquisition? Also, do those numbers include pharmacy time, bags, tubes, solutions, nursing, etc.?
Dr. Lipsky: What was included was the actual average hospital acquisition price, the average U.S. wage and benefit for labor based on time and motion studies for administering antibiotic, and the consumable supplies of the manufacturer for all of the supplies needed for giving intravenous drugs. For each patient, the actual doses were multiplied by a total cost per dose. That was the cost per dose for the drugs.
This pharmacoeconomic data is an example of the enormous treasure trove of data that we have from the SIDESTEP study that will produce many papers in the future.
Final Notes
Dr. Armstrong: Dr. Joseph, given all the information we have covered here and the emerging study results, what is the appropriate role of ertapenem in our armamentarium for treating diabetic foot infections? We have a lot of data as well as clinical experience with this agent. How now should we use this once-a-day agent in the treatment of diabetic foot infections?
Dr. Joseph: Given the data we have now with both SIDESTEP and the Graham study, which included almost 100 diabetic foot infections, and taking into account the pharmacoeconomics involved, we are at a point now where ertapenem could be considered a first-line therapy for moderate to severe diabetic foot infection in patients who do not have MRSA as a primary pathogen.6,19 |
