Severe open fractures of the lower extremity: a retrospective evaluation of the Mangled Extremity Severity Score (MESS)

Abstract

Recent reports using the Mangled Extremity Severity Score (MESS) suggest that a score of > or = 7 is 100% accurate in predicting the need for amputation of severely injured lower extremities. To further evaluate the value of the MESS in predicting amputation, specifically with respect to type IIIB and type IIIC (Gustilo and Anderson) open fractures of the tibia, we retrospectively evaluated 24 patients with these injuries. A significant difference (p = 0.001) between MESS values of 13 salvaged (6.36 +/- 0.35 SEM) and 11 amputated limbs (6.36 +/- 0.54 SEM) was found. A MESS value of > or = 4 was most sensitive (100%); a MESS value of > or = 7 was most specific, and a MESS value of > or = 7 was found to have a positive predictive value of 100%. Subsequently, we addressed recent criticisms of the MESS by including nerve injury in the scoring system and by separating soft-tissue and skeletal injury components of the MESS. We modified the MESS with a score called the NISSSA and applied it retrospectively to our cases. After careful statistical comparison we found both the MESS and NISSSA to be highly accurate (p < 0.005) in predicting amputation. The NISSSA was found to be more sensitive (81.8% versus 63.6%) and more specific (92.3 versus 69.2%).