Comment on: Lipsky et al. Developing and validating a risk score for lower-extremity amputation in patients hospitalized for a diabetic foot infection. Diabetes Care 2011;34:1695-1700.

We have carefully read the article by Lipsky et al. (1) who have developed a novel risk score for lower-extremity amputation (LEA) in patients hospitalized for a diabetic foot infection. For this purpose, the authors used data from a clinical research database of patients hospitalized in 97 acute-care hospitals in the U.S. that was compiled by CareFusion (CareFusion, Department of Clinical Research, Marlborough, MA). Lipsky et al. reported on a series of 3,018 patients who met the inclusion criteria for a diabetic foot infection that was culture-documented within 48 h of admission.

This current series seems to be the same series and database that Lipsky et al. used for a previous study dealing with skin and soft tissue infections in hospitalized patients with diabetes and positive results from a culture taken within 48 h of hospitalization (2). In this study, a total of 3,030 patients with diabetes and a secondary diagnosis indicating any type of soft tissue infection (including cellulitis, infected ulcers, and surgical site infections [SSIs]) involving the foot or any other anatomic site were included (2). The authors stated that since the number of patients with SSIs was relatively small (n = 114) and since the anatomic locations were not available, they had grouped these patients into the nonfoot infection group. The authors reported 2,220 foot infections and 810 nonfoot or SSIs using these criteria (2). It seems that 810 patients with infections in locations other than the foot have now been included in the current series because the figures are almost identical (with a difference of only 12 patients).

If the current series is the same as previously reported (2), no conclusions can be extracted concerning LEAs since 810 patients had infections in other sites. SSI has now been associated with the highest risk of LEA. However, since the number of patients with SSI located in the foot is unknown (2), the results become dubious. Furthermore, the rate of amputations in the current series, if nonfoot infections are excluded, is 29% (646 LEAs out of 2,220 patients with foot infections). The risk score would only be valid if just patients with diabetes and infections located in the feet were included.

Moreover, no patients with other types of foot infections were included in this study. The clinical presentation of diabetic foot infections is more complex than cellulitis, infected ulcers, or SSIs, as reported by other authors who used the ICD-9-CM in a retrospective series of 61,007 patients (3). Osteomyelitis has been reported to be the most frequent diagnosis in a series dealing with deep foot infections in patients with diabetes requiring hospitalization (4,5). Since other types of foot infections had been excluded, this score could only be applied when treating skin and soft tissue infections in the feet of patients with diabetes. This is not suggested by the title. It is possible that the risk score would have been totally different if nonfoot infections had been excluded, and other types of foot infections had been included.