• Users Online: 79
  • Home
  • Print this page
  • Email this page
Home About us Editorial board Ahead of print Current issue Search Archives Submit article Instructions Subscribe Contacts Login 


 
 Table of Contents  
ORIGINAL ARTICLE
Year : 2017  |  Volume : 4  |  Issue : 3  |  Page : 124-128

Traumatic brain injury predictive value of common intensive care severity scores


Department of Critical Care Medicine, El-Helal Hospital, Cairo, Egypt

Date of Submission05-Oct-2016
Date of Acceptance01-Mar-2017
Date of Web Publication5-Jul-2017

Correspondence Address:
Mahmoud Kenawi
Cairo University Hospitals, Cairo, 35855
Egypt
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/roaic.roaic_90_16

Rights and Permissions
  Abstract 

Background
Traumatic Brain Injury (TBI) causes a severe toll on society as a leading cause of mortality worldwide and the major cause of disability among young adults. The prognosis after TBI had been particularly challenging to predict, with limited availability of robust prognostic models.
Aim
To evaluate the usefulness of the acute physiology and chronic health evaluation II (APACHE II), simplified acute physiology score II (SAPS II) and sequential organ failure assessment (SOFA) scores compared to simpler models based on age and Glasgow coma scale (GCS) in predicting a six-month mortality of patients with moderate to severe traumatic brain injury (TBI) in the intensive care unit (ICU).
Methods
A prospective cohort study conducted on acute TBI patients admitted to ICU at EL-HELAL trauma Centre and KASR AL AINI university hospital, Egypt during the period from August 2014 to April 2015. All patients were followed-up for 6 months from the day of admission. Patients were divided into two groups (survivors and non-survivors).
Results
A total of 104 patients were enrolled. Mean age was 37±17.16 years. The overall six-month mortality was 25 patients (24.4%). The univariate analysis showed that APACHE II, SAPS II, SOFA, GCS, and age had a significant statistical difference regarding mortality between both groups (P-value < 0.05) and the optimal cut-off point as mortality indicator was 14, 26, 4, 9 and 49, respectively with area under the curve (AUC) 0.88, 0.87, 0.83, 0.80 and 0.79, respectively. Multivariate analysis using logistic regression found that only age and GCS had a statistically significant impact on outcome (P-value; 0.001, 0.022, respectively).
Conclusions
A simple prognostic model based only on GCS and age displayed good predictor for six-month mortality of ICU treated patients with TBI. The use of the more complex scoring systems (APACHE II, SAPS II and SOFA) added little to the prognostic performance.

Keywords: ICU scores, six-month mortality, traumatic brain injury


How to cite this article:
Kandil A, Kenawi M, Samir A, Hussein K. Traumatic brain injury predictive value of common intensive care severity scores. Res Opin Anesth Intensive Care 2017;4:124-8

How to cite this URL:
Kandil A, Kenawi M, Samir A, Hussein K. Traumatic brain injury predictive value of common intensive care severity scores. Res Opin Anesth Intensive Care [serial online] 2017 [cited 2017 Sep 20];4:124-8. Available from: http://www.roaic.eg.net/text.asp?2017/4/3/124/209672


  Introduction Top


Traumatic brain injury (TBI) is caused by an external mechanical force affecting the head and brain, possibly leading to permanent or temporary impairment of cognitive, physical, and psychosocial functions with an associated diminished or altered state of consciousness [1]. TBI is a leading cause of death, disability, and resource consumption [2].

TBI has traditionally been classified using injury severity scores; the most commonly used is the Glasgow Coma Scale (GCS). A GCS score of 13–15 is considered a mild injury, a score of 9–12 as moderate injury, and a score of 8 or less as severe TBI [3]. However, it is limited by confounding factors such as medical sedation, paralysis, endotracheal intubation, and intoxication [4].

The outcome prediction scores were developed more than 30 years ago. The most commonly used scoring systems in intensive care are the acute physiology and chronic health evaluation II (APACHE II), simplified acute physiology score II (SAPS II), and sequential organ failure assessment (SOFA). These scores were used for predicting the risk for in-hospital death [5]. The role of the ICU scoring systems for long-term outcome prediction in patients with TBI treated in the ICU is uncertain; although TBI-specific prognostic models are likely to be more accurate compared with the ICU scoring systems in this patient group, they are not widely implemented [6].

Therefore, the aim of the current study was to evaluate the usefulness of the APACHE II, SAPS II, and SOFA scores in predicting 6-month mortality after TBI and to find out whether these scoring systems are of any additional value compared with a simple model based only on age and the GCS.


  Patients and methods Top


The present study was a prospective cohort one conducted on 104 patients admitted to the ICU at El-Helal Trauma Center and KASR Al Aini University Hospital during the period from August 2014 to April 2015. The study was approved by the ethics committee. A written consent was signed by the patient or his first-degree relatives. All studied populations were diagnosed with an acute TBI and GCS lower to 12 with age more than 15 years. Patients who had severe chest, abdominal, or orthopedic trauma (e.g. flail chest, cardiac tamponade, tears in abdominal organ, and fracture spine) were excluded.

Full clinical evaluation on admission was carried out, including full history taking, and if possible, clinical examination, radiological (radiography, brain computed tomography, and pelvic abdominal ultrasound) and laboratory investigations (arterial blood gases, complete blood count, kidney function tests, and bilirubin).

All patients were evaluated according to GCS, SOFA, SAPS II, and APACHE II score. Variables and scores were calculated from the worst score recorded during the first 24 h of admission, and then an estimated approximate mortality percentage was measured.

All patients subjected to an endotracheal tube and mechanical ventilation were studied. In addition, the plan of neurological management – whether conservative or surgical intervention – was recorded.

All patients were followed-up for 6 months from the day of admission to detect 6-month mortality (final outcome). Patients were divided into two groups: group 1 (survivors) and group 2 (nonsurvivors).

Statistical analyses

After we collected all data for 104 patients, statistical analysis was performed on a personal computer using State version 11.2 (StataCorp LP, College Station, Texas, USA) and IBM statistical package for the social sciences statistics version 21 (IBM Corp., Armonk, New York, USA).

The D’Agostino–Pearson test was used to test the normality of numerical data distribution. Non-normally distributed numerical data are presented as median and SD range, and intergroup differences are compared nonparametrically using the Mann–Whitney U-test. Categorical data are presented as number and percentage.

The predictive value of the results were determined through analysis of the receiver operating characteristic curve obtained by plotting the sensitivity and 1-specificity of the probability indices as estimated with the Cox model. P values are two-tailed. P less than 0.05 was considered statistically significant.


  Results Top


A total of 104 patients were enrolled. The mean age was 37±17.16 years and male sex constituted 85.6% of the studied population. Medical history of diabetes mellitus was present in 52% of patients in group 2 compared with 11.4% in group 1 (P<0.001). History of hypertension was prevalent in group 2 (48%) compared with group 1 (8.9%), with a significant difference (P<0.001). As regards associated cardiac diseases, 20% of patients in group 2 had associated cardiac diseases compared with 2.5% in group 1 (P<0.001) ([Figure 1]).
Figure 1 Prevalence of medical history in both groups

Click here to view


As regards laboratory variables, there was a significant difference in hemoglobin level between the two groups (10±1.56 g/dl in group 1 vs. 9±1.6 g/dl in group 2, P=0.01). Bilirubin level was significantly different in both groups as well (1.02±0.45 mg/dl in group 1 vs. 1.3±0.55 mg/dl in group 2, P<0.001) ([Table 1]).
Table 1 Effect of laboratory variables on outcome

Click here to view


As regards creatinine level, the results showed significantly different values between the two groups (1.32±0.68 mg/dl in group 1 vs. 1.96±0.83 mg/dl in group 2, P<0.001). In addition, PaO2/FiO2 ratio analysis showed significantly different levels in both groups (358±97.8 in group 1 vs. 278±113.16 in group 2, P=0.002). Acidosis was present in 44% of patients in group 2 compared with 11.4% in group 1, with a significant P-value of 0.001 ([Table 1]).

From the above results, patients who had acidosis, high serum creatinine, high bilirubin level, low PaO2/FiO2 ratio, and low hemoglobin were at high risk for 6-month mortality after TBI (P<0.05) ([Table 1]).

On comparing age in both groups, the mean age in group 1 was 32±14.29 versus 52.8±17.27 years in group 2 (P<0.001) ([Table 2]).
Table 2 Effect of age, Glasgow Coma Scale, and scoring system on outcome

Click here to view


As regards the ICU scoring systems, GCS score showed a highly significant difference between the two groups (10.1±1.87 in group 1 vs. 7.5±2.4 in group 2, P<0.001). The SOFA score showed a significant difference between the two groups (3.6±2.3 in group 1 vs. 7.92±4.45 in group 2, P=0.005). Moreover, group 2 patients showed higher APACHE II score compared with group 1 (19.5±6.3 in group 2 vs. 10.41±5.2 in group 1, P<0.001). Finally, the SAPS II score had similar results with significantly higher values in group 2 patients (19.56±10.08 in group 1 vs. 39.3±15.7 in group 2, P<0.001) ([Table 2]).

The univariate analysis showed that APACHE II, SAPS II, SOFA, GCS, and age had a significant statistical difference as regards mortality between the two groups (P<0.05) ([Table 2]).

On studying the correlation between age and outcome, it was found that the optimal cutoff point of age as a mortality indicator was estimated to be 49 years, which yielded a sensitivity of 81% and a specificity of 76%, with the area under the curve (AUC) at 0.79 [95% confidence interval (CI): 0.63–0.91]. As for GCS, the optimal cutoff point of GCS as a mortality indicator was estimated to be 9, which yielded a sensitivity of 70.9% and a specificity of 72%, with the AUC at 0.80 (95% CI: 0.71–0.89). As regards correlation between ICU scores and outcome through receiver operating characteristic curve, the optimal cutoff point of APACHE II score as a mortality indicator was estimated to be 14, which yielded a sensitivity of 82% and a specificity of 88, with the AUC at 0.88 (95% CI: 0.81–0.94). As for SOFA score, the optimal cutoff point as a mortality indicator was estimated to be 4, which yielded a sensitivity of 78% and a specificity of 80%, with the AUC at 0.83 (95% CI: 0.75–0.92). In addition, the optimal cutoff point of SAPS II score as a mortality indicator was estimated to be 26, which yielded a sensitivity of 86% and a specificity of 84%, with the AUC at 0.87 (95% CI: 0.78–0.96) ([Figure 2]).
Figure 2 Assessment of effect of age, Glasgow Coma Scale, and scoring systems on outcome through receiver operating characteristic curve. APACH II, acute physiology and chronic health evaluation II; GCS, Glasgow Coma Scale; ROC, receiver operating characteristic; SAPS II, simplified acute physiology score II

Click here to view


Multivariate analysis of logistic regression included age, GCS, APACHE II, SOFA, and SAPS II and showed that only age and GCS had a significant impact on the outcome (P=0.001 and 0.022) ([Table 3]).
Table 3 Logistic regression of age and scoring systems

Click here to view


The overall 6-month mortality was 24.4% (25 of 104) ([Figure 3]). In-hospital mortality constituted 60% (15 of 25) of dead patients compared with 40% (10 out of 25) for out-hospital mortality.
Figure 3 Outcome of the study

Click here to view



  Discussion Top


Predicting outcome after TBI is very difficult, as the nature of the primary brain injury is extremely heterogeneous; no two injuries will be exactly the same and the primary injury will be modified by secondary insults. Patient factors, premorbid state, and physiological reserve also affect patient response to the trauma [7].

The current study was conducted on 104 patients admitted to the ICU after moderate-to-severe acute TBI. In addition, all patients were followed-up for 6 months from the day of admission.

As regards outcome, 79 (75.6%) patients survived and 25 (24.4%) patients died. Besides, intrahospital mortality constituted 60% of the nonsurvivors group and 14.4% of the whole study population. These results are in agreement with a study performed on 1652 patients with TBI in a study by Rahu et al. [8], who showed that intrahospital mortality constituted 64% of nonsurvivors and the overall 6-month mortality was 33%.

On analyzing age, there was a significant difference between the two groups (group 1 had a mean age of 32±14.29 vs. 52.8±17.27 years in group 2, P<0.001). These results are in agreement with those of Nejmi et al. [9], who studied 225 patients with moderate-to-severe head injury. The study showed that age had a significant effect on mortality (survivors had a mean 28.9±18.6 vs. 48.1±16.3 years in nonsurvivors, P=0.02) [9]. In contrast, Tjahjadi et al. [10] found that age had no significant effect on mortality after TBI as survivors had a mean age of 27.48±10.5 versus 36.30±15.9 years in nonsurvivors (P=0.139). This is probably attributed to the smaller sample size of patients studied by them (61 patients) [10].

The current study showed that decreasing GCS is a risk for mortality in TBI (the mean value in group 1 was 10.1±1.87 vs. 7.5±2.4 in group 2, P<0.001). These results are in agreement with those of Nejmi et al. [9], who found that the mean GCS was 9 in survivors versus 6 in nonsurvivors (P=0.002). In contrast, a study performed by Majdan et al. [11] showed that addition of pupil size and reactivity to GCS was better in predicting 6-month mortality compared with GCS alone, as GCS motor score had an AUC of 0.754; pupillary assessment at admission had AUC of 0.66 and combination of them showed best performance (AUC=0.876).

Univariate analysis of APACHE II and SAPS II score showed a significant correlation between increasing score and 6-month mortality (19.5±6.3 and 39.3±15.7 for nonsurvivors vs. 10.41±5.2 and 19.56±10.08 for survivors, respectively, P<0.05). The results are in agreement with those of Nejmi et al. [9], who found that APACHE II and the SAPS II of the nonsurviving patients are higher than those of the survivors (20.4±6.8 and 31.2±13.6 for nonsurvivors vs. 15.7±5.4 and 22.7±10.3 for survivors, respectively) with a statistically significant difference (P=0.0032 for APACHE II and P=0.0045 for SAPS II) [9]. However, Brinkman et al. [12] found that APACHE II-based and SAPS II-based models are better for predicting in-hospital mortality than 6-month mortality.

As regards SOFA score, there was a highly significant difference between the two groups as regards mortality (mean was 3.6±2.3 in group 1 vs. 7.92±4.45 in group 2, P=0.005). A study performed by Dubendorfer et al. [13] found that nonsurvivors after TBI showed higher SOFA scores combined survivors in the combined injury group (P=0.001) and also in isolated TBI (P=0.023).

Another study performed by Rahu et al. contradicted these findings. Comparison between SOFA and APACHE II score found that SOFA has limited value in predicting long-term mortality in critically ill patients after TBI (APACHE II had AUC=0.79, and the SOFA had AUC=0.68) [8].

In the current study, multivariate analysis of logistic regression, which included age, GCS, APACHE II, SOFA, and SAPS II score, showed that only age and GCS had a statistically significant impact on the outcome. These results are in accordance with data collected by Rahu et al. [8], who found that the use of the more complex scoring systems added little to the prognostic performance.

Limitations

  1. Small number of patients. A larger number may improve validity.
  2. Limitation of using 6-month mortality as the primary outcome measure; some long-term outcome data were missing (e.g. amnesia or disability).
  3. Scores were measured once during the first 24 h of admission.



  Conclusion Top


Prognostic model based only on GCS and age displayed a good predictor for 6-month mortality of ICU treated patients with TBI. The use of the more complex scoring systems APACHE II, SAPS II and SOFA added little to the prognostic performance.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

 
  References Top

1.
Thompson HJ, Takacs NC, Saatman KE, Raghupathi R, McIntosh TK. Hyperthermia following traumatic brain injury: a critical evaluation. Neurobiol Dis 2011; 12:163–173.  Back to cited text no. 1
    
2.
Haddad SH, Arabi YM. Critical care management of severe traumatic brain injury in adults. Scand J Trauma Resusc Emerg Med 2012; 20:12–19.  Back to cited text no. 2
[PUBMED]    
3.
Stocchetti N, Pagan F, Calappi E, Canavesi K, Beretta L, Citerio G et al. In accurate early assessment of neurological severity in head injury. J Neurotrauma 2004; 21:1131–1140.  Back to cited text no. 3
    
4.
Balestreri M, Czosnyka M, Chatfield DA, Steiner LA, Schmidt EA, Smielewski P et al. Predictive value of Glasgow Coma Scale after brain trauma: change in trend over the past ten years. J Neurol Neurosurg Psychiatry 2004; 75:161–162.  Back to cited text no. 4
    
5.
Minne L, Abu-Hanna A, de Jonge E. Evaluation of SOFA-based models for predicting mortality in the ICU: a systematic review. Crit Care 2008; 12:161–187.  Back to cited text no. 5
[PUBMED]    
6.
Vincent JL, Moreno R. Clinical review: scoring systems in the critically ill. Crit Care 2010; 14:207–224.  Back to cited text no. 6
    
7.
Reynolds S. Predicting outcome after traumatic brain injury. JICS 2011; 3:12–24.  Back to cited text no. 7
    
8.
Rahu R, Skrifvars M, Bendel S, Selander T, Kivisaari R, Siironen J, Reinikainen M. Predicting six-month mortality of patients with traumatic brain injury: usefulness of common intensive care severity scores. Crit Care 2014; 18:60–73.  Back to cited text no. 8
    
9.
Nejmi H, Rebahi H, Ejlaidi A, Abouelhassan T, Samkaoui M. The ability of two scoring systems to predict in-hospital mortality of patients with moderate and severe traumatic brain injuries in a Moroccan intensive care unit. Indian J Crit Care Med 2014; 18:369–375.  Back to cited text no. 9
[PUBMED]  [Full text]  
10.
Tjahjadi M, Arifin M, Gill A, Faried A. Early mortality predictor of severe traumatic brain injury: a single center study of prognostic variables based on admission characteristics. Indian J Neurotrauma 2013; 10:38–45.  Back to cited text no. 10
    
11.
Majdan M, Steyerberg E, Nieboer D, Mauritz W, Rusnak M, Lingsma H. Glasgow coma scale motor score and pupillary reaction to predict six-month mortality in patients with traumatic brain injury: comparison of field and admission assessment. J Neurotrauma 2015; 32:101–108.  Back to cited text no. 11
    
12.
Brinkman S, Abu-Hanna A, de Jonge E, de Keizer NF. Prediction of long-term mortality in ICU patients: model validation and assessing the effect of using in-hospital versus long-term mortality on benchmarking. Intensive Care Med 2013; 39:1925–1931.  Back to cited text no. 12
[PUBMED]    
13.
Dubendorfer C, Billeter AT, Seifert B, Keel M, Turina M. Serial lactate and admission SOFA scores in trauma: an analysis of predictive value in 724 patients with and without traumatic brain injury. Eur J Trauma Emerg Surg 2013; 39:25–34.  Back to cited text no. 13
    


    Figures

  [Figure 1], [Figure 2], [Figure 3]
 
 
    Tables

  [Table 1], [Table 2], [Table 3]



 

Top
 
 
  Search
 
Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

 
  In this article
Abstract
Introduction
Patients and methods
Results
Discussion
Conclusion
References
Article Figures
Article Tables

 Article Access Statistics
    Viewed65    
    Printed0    
    Emailed0    
    PDF Downloaded22    
    Comments [Add]    

Recommend this journal


[TAG2]
[TAG3]
[TAG4]