M. Elhadi1, A. Khaled1, A. Msherghi1 1University of Tripoli, Faculty Of Medicine, Tripoli, Libya
Introduction: The Shock Index (SI) and Assessment of Blood Consumption (ABC) score, with cut-off values of >1 and ≥2 respectively, are frequently utilized to anticipate the need for massive blood transfusion in trauma patients. Nevertheless, the comparative diagnostic accuracy of these tools remains a subject of ongoing discussion. This study aims to conduct a diagnostic accuracy meta-analysis comparing Shock Index (SI) and Assessment of Blood Consumption (ABC) scores to predict the need for massive transfusion in trauma patients.
Methods: A systematic review of studies published up to December 2022 in MEDLINE, EMBASE, and Scopus was conducted, focusing on those assessing the diagnostic accuracy of SI (>1) or ABC score (≥2) in predicting massive transfusion need in adult trauma patients. Statistical analyses were performed using R software (version 4.0.3) using mada package to pool diagnostic accuracy parameters, each reported with a 95% Confidence Interval (CI).
Results: From the identified studies, 4 studies involving 18,254 patients evaluated SI, while 11 studies with 7,589 patients evaluated the ABC score. The pooled sensitivity was 68.4% (95% CI: 60.5-75.4%, I2=92.9%) for SI and 54.8% (95% CI: 43.9-65.2%, I2=85.3%) for ABC score. The pooled specificity was 82.8% (95% CI: 82.2-83.4%, I2=50.8%) for SI and 88.7% (95% CI: 82.7-92.7%, I2=96.2%) for ABC score. The diagnostic odds ratio was 10.44 (95% CI: 7.38-14.78) for SI and 9.49 (95% CI: 4.92-18.29) for ABC score. The positive likelihood ratio (LR) was 3.98 (95% CI: 3.55-4.46) for SI and 4.84 (95% CI: 3-7.80) for ABC score, while the negative LR was 0.38 (95% CI: 0.30-0.48) for SI and 0.51 (95% CI: 0.39-0.65) for ABC score. The false-positive rate estimates were 17.2% (95% CI: 16.6-17.8%) for SI and 11.3% (95% CI: 7.3-17.3%) for ABC score.
Conclusion: This study compares the Shock Index and ABC score in predicting massive transfusion needs, underscoring the ABC score's higher specificity and lower false positives despite reduced sensitivity, particularly in populations with low outcome prevalence. The results emphasize the importance of considering patient demographics and outcome prevalence in score selection and call for additional studies to enhance these tools.