Burns Open (Apr 2022)

Comparison between ABSI and BOBI Score for Burns Mortality Prediction in Indonesia’s National Referral Burn Center: A 5-year study

  • Aditya Wardhana,
  • Michelle Valeria,
  • Regina Putri Apriza,
  • Adi Basuki

Journal volume & issue
Vol. 6, no. 2
pp. 92 – 96

Abstract

Read online

Objectives: The aim of this study is to test the validity of each burn mortality prediction model and to compare the accuracy between Abbreviated Burn Severity Index (ABSI) and Belgian Outcome for Burn Injury (BOBI) for mortality prediction in our burn unit. Methods: A retrospective cohort study was conducted enrolling burn patients during the period of 2016 to 2020 in burn unit of dr. Cipto Mangunkusumo Hospital. We specified our inclusion criteria to acute burn (≤48 h) within the age range of 17 to 58 years, and percentage of burned total body surface area (TBSA) 10–60%. Age, burn depth, gender, total TBSA and inhalation injury were variables accountable to ABSI and BOBI, thus were further analyzed statistically. Results: A total of 220 patients were enrolled (133 males and 87 females) with mean age of 37.5 years and mean percentage of burned TBSA was 33.2%. The overall rate of mortality in this study was 25.9%. Full-thickness burn was observed in 133 subjects (60.45%) and inhalation trauma was experienced by 29 subjects (13.18%). Univariate and multivariate analysis were done on each variables. Full-thickness burn (p = 0.039), inhalation injury (p = 0.041), and 35.75% TBSA (p < 0.001) remained as significant variables for mortality prediction. Goodness-of-fit statistics done using the Hosmer-Lemeshow (HL) test for ABSI and BOBI showed p = 0.718 and p = 0.001. Performance discrimination for both ABSI and BOBI were analyzed by area under curve (AUC) of 0.84 (CI 95%; 0.785–0.896) and 0,79 (CI 95%; 0.731–0.859), respectively. Discussion: Full-thickness burn, inhalation injury, and %TBSA were identified as significant mortality predictors. Further analysis showed that ABSI score model fit well to actual outcomes observed in our unit. Furthermore, ABSI had better AUC value when compared with BOBI score in predicting burn mortality. Conclusion: ABSI score is valid for predicting mortality in our burn unit and can be used as clinical adjuncts to escalate the quality of burn patients management. In addition, ABSI was more accurate than BOBI in predicting burn mortality in our burn unit.

Keywords