# Assessing the Predictive Accuracy of Open Injury Scores in Determining the Salvageability of Type IIIB Tibial Fractures

## TL;DR

This study compares three scoring systems to determine which best predicts whether a severely injured leg can be saved or needs amputation.

## Contribution

The study validates and compares the predictive accuracy of three injury scores for Type IIIB tibial fractures in a large patient cohort.

## Key findings

- GHOISS demonstrated the highest predictive accuracy (AUC = 0.997) for limb salvageability.
- Higher scores in MESS, LSI, and GHOISS were significantly associated with amputation.
- Limb salvage was achieved in 93.5% of cases, with amputations more common in the grey zone of scores.

## Abstract

Severe open tibial fractures pose a major challenge in orthopaedic trauma care. The decision to salvage or amputate a limb requires objective assessment tools. The Mangled Extremity Severity Score (MESS), Limb Salvage Index (LSI), and Ganga Hospital Open Injury Severity Score (GHOISS) are widely used, but their accuracy in Type IIIB open tibial fractures remains under evaluation. This study aimed to validate these scoring systems and determine the most reliable predictor of limb salvageability. A prospective observational study was conducted on 307 patients with Type IIIB open tibial fractures at the Department of Orthopaedic Surgery of a tertiary care centre in North India. Patients were assessed using MESS, LSI, and GHOISS after the first surgical debridement. Limb salvageability was defined as limb retention at six months. Statistical analysis included chi-square tests and receiver operating characteristic (ROC) curve analysis. The majority of patients were male (85%), with a mean age of 28.67 ± 9.28 years. Road traffic accidents (72.3%) were the most common cause of injury. Limb salvage was achieved in 93.5% of cases, while 6.5% underwent amputation, particularly in cases within the grey zone (scores 15-16). Higher MESS, LSI, and GHOISS scores were significantly associated with amputation (p < 0.001). ROC analysis showed that GHOISS (AUC = 0.997) had superior predictive accuracy compared with MESS (AUC = 0.958) and LSI (AUC = 0.953). GHOISS demonstrated higher accuracy and specificity in predicting limb salvageability within this cohort, suggesting that it may be a useful tool to support clinical decision-making in Type IIIB open tibial fractures.