Big score presentation

Department of Neurosurgery
Tribhuvan University Teaching Hospital
Journal club
Sandesh Dahal
MCh resident

Introduction

 Journal- critical care
 Published: 11 July 2013
 Impact factor- 2.19

Background
 Early prediction of mortality in trauma is important.
 It often challenging but it is very important.
 Many tools are available, but they are complex and can not be assessed in initial phase of
examining the patient.
 Accurate early prediction of the risk of death might have the potential to
 inform triage decisions,
 inform treatment, or
 Stratify patients for further care.

 BIG score to predict outcome has
 Base deficit (that indicates metabolic acidosis)
 International normalized ratio and
 Glasgow coma score
 This score in pediatric population has been demonstrated to outperform all the
prediction models in various cohort studies.
 It is simple and easy score to calculate at bedside and very early stage of patient
management.

 BIG score has not been evaluated in adult population before.
 It also has not been compared with other prediction models like TRISS andPS09
scores.
 Hence the study aims to
 Assess whether the BIG score can predict mortality in an adult trauma
 Compare its prediction ability with the other scoring systems in adult
population.

BIG score calculation

TRISS calculation
 It is based on
 Age
 Revised trauma score
 Injury severity score

 Age
 Less than 55, score- 0
 More than 55, score- 1
 RTS

Probability of survival score (PS09 score)

Materials and methods
 Retrospective analysis
 Seven trauma centers In Europe and USA
 BIG score compared with TRISS and PS09 score.

 Data collection
 A data-collection template was developed to collect all needed parameters from
the participating sites
 All primary admitted trauma-team activation patients aged 18 years or older
during the period 2005 to 2010, inclusive, were eligible.
 Only patients with available and complete datasets for the calculation of the
analyzed scoring systems (BIG, TRISS, and PS09) were included in the study.
 only patients with an ISS ≥4 were included.

Data analysis
 one military and six civilian trauma centers and registries in Europe and the United
States were collected and retrospectively analyzed
 Primary outcome- 30 days mortality
 Comparison of the BIG score against the TRISS and PS09 score on a representative
population of trauma patients.
 Subgroup analysis on patients with blunt or penetrating trauma was additionally
done.
 Analysis on civilian and military trauma was also done.

 Demographic data are presented as means with standard deviation (SD) for
continuous variables and as percentages for incidence rates.
 The Mann Whitney U test was used for continuous variables, and the chi square χ2
test for categorical variables.
 Statistical significance was set at P values less than 0.05

 The quality of all scoring systems in predicting mortality was analyzed and presented
in terms of discrimination and precision.
 Discrimination measures the ability of a scoring system to separate survivors from
non-survivors.
 Discrimination ability of all scores compared via receiver operating characteristic
(ROC) curves and compared the expected mortality rate (precision) of all scores with
the observed mortality rate.

 The ROC curve summarizes the trade-off between sensitivity and specificity of a
predictive score by using all score values as potential cut-off values.
 Its value varies between 0.5 (no discrimination) and 1.0 (perfect discrimination).
 All statistical analyses were performed by using IBM SPSS 20 (IBM SPSS Inc,
Chicago, IL, USA).

Results
 Total 12206 patients included in the study.
 4,949 (61%) patients were included by civilian trauma centers, and
 7,257 (59%) patients were included by military trauma centers

Baseline characteristics

 AUROC values for military dataset
 BIG score 0.929 (95% CI, 0.909 to 0.949), and PS09 0.922 (95% CI, 0.904 to 0.940)
and TRISS of 0.915 (95% CI, 0.891 to 0.939) (all P > 0.31).
 On a civilian dataset (n = 4949),
 the PS09 score 0.901; (95% CI, 0.887 to 0.914)
 TRISS 0.896; 95% CI, (0.882 to 0.909; P = 0.24).
 The AUROC of the BIG score (0.849; 95% CI, 0.830 to 0.868) was significantly
lower (P < 0.001)

Discussion
 BIG score evaluated for first time in adult population.
 It performed well in predicting mortality in adult population.
 But unlike other systems, BIG score is easy to calculate and can be done in less time
with less effort.
 Less time consuming parameters used in BIG score, and hence can be very useful in
the triage phase till more complex parameters are available.

 In present analysis, BIG performed well in predicting mortality in blunt than in
penetrating injury.
 All predictors overpredicted mortality in penetrating trauma, less so by BIG score.
 BIG score overpredicted mortality only in penetrating group, but in blunt trauma
group, predicted and observed mortality were similar.
 The mortality prediction was better with BIG score because it might have included
two most important causes, CNS injury and exsanguination.

 The overall prediction could be good with BIG score because
 It calculates CNS injury with GCS
 Hemodynamic status, shock and resuscitation status with base deficit
 Bleeding tendency and coagulopathy with INR status.
 Besides, base deficit (BD), is shown to be a valuable indicator of shock, abdominal
injury, fluid requirements, efficacy of resuscitation, and a predictor of mortality after
trauma

Literature

Similar scores
 EMTRASS by Raum and Colleagues
 Contains
 Age
 GCS
 Base excess and
 Prothrombin time
 It outperformed TRISS, ISS, RTS. But further validation studies not done
prospectively.

Conclusion
 BIG score has been validated as having good reliability in predicting mortality,
functional outcome in children population.
 BIG score has good predictive ability of mortality in trauma cases in adults as well.
 It is rapid, simple and precise estimation that can be done readily.
 It also helps to emphasize the treatment in triage setting.

Critical appraisal
 Strength
 Multicenter study
 Large sample size
 Good statistical analysis done
 Comparison and validation with two well established scoring systems
 Data for both military and civilian settings available, with wide applicability of
the findings.

 Weakness
 Retrospective data from registry
 Treatment standard might differ in various centers
 Timing of blood gas not mentioned, pre and post resuscitation values might
differ.
 Duration of presentation after trauma not mentioned.
 Age is significant in adult population, which has significantly confounded in the
study.

Take home…..
 Many prediction scores are available for mortality prediction after TBI.
 Most of them are very time consuming and can not be done in emergency settings,
need lot of patient data.
 BIG score is simple and easily calculated scoring system even in emergency setting.
 It has almost similar predicting ability as other complex scoring system.
 It can be a good score to prognosticate the patient even in the emergency setting.

Big score presentation

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Big score presentation

Similar to Big score presentation (20)

Recently uploaded

Recently uploaded (20)

Big score presentation