Improving the Severity Classification System of Diabetes-Related Foot Infections: Necrosis is a Better Predictor of Amputation Than Systemic Inflammatory Response.

BACKGROUND

The International Working Group on the Diabetic Foot (IWGDF)/Infectious Disease Society of America (IDSA) classification system of diabetes-related foot infection (DFI) severity requires a severe infection to have the systemic inflammatory response syndrome (SIRS). Subsequent attempts to improve this classification system to include other clinically relevant characteristics (eg, necrosis and osteomyelitis) have not reliably predicted patient outcomes such as amputation. The purpose of this study was to propose an improved classification system based on clinically relevant predictors of amputation at our institution.

METHODS

This was a single-center, retrospective cohort study of 270 unique patients with 346 DFIs. We determined the association between clinical characteristics of each DFI (eg, cellulitis, abscess, necrosis, osteomyelitis, and SIRS) and amputation within 90 days of presentation. We derived a new severity classification system based on the presence of necrosis and compared the performance of this classification system to the IWGDF/IDSA classification system.

RESULTS

Our "necrosis classification system" predicted that mild, moderate, and severe DFI had a 20% (95% confidence interval [CI] 2%, 38%), 40% (95% CI [33%, 47%]), and 74% (95% CI [64%, 84%]) risk of amputation within 90 days of presentation, respectively, and discriminated between moderate and mild, and severe and moderate DFI. The IWGDF/IDSA classification system predicted that mild, moderate, and severe DFI had a 29% (95% CI [9%, 49%]), 43% (95% CI [36%, 50%]), and 54% (95% CI [44%, 63%]) risk of amputation within 90 days of presentation, respectively, and did not discriminate between moderate and mild, or severe and moderate DFI.

CONCLUSION

This necrosis classification system discriminated the risk of amputation at 90 days better than the IWGDF/IDSA classification system. Our new classification system may have greater clinical relevance but needs to be validated to confirm its generalizability.Level of evidence: Level III: Retrospective comparative study.