Outcome prediction in severe head injury: analyses of clinical prognostic factors

doi:10.1054/jocn.2000.0732

Journal of Clinical Neuroscience

Volume 8, Issue 2, March 2001, Pages 120-123

https://doi.org/10.1054/jocn.2000.0732 Get rights and content

Abstract

Retrospective analysis of 272 patients with severe head injury was performed. Patient age, Glasgow Coma Scale (GCS) score, pupillary abnormalities, impaired oculocephalic response, presence of subarachnoid haemorrhage, and multiplicity of parenchymal le sions on computerised tomography (CT) were examined. The CT findings were divided into 2 groups, diffuse brain injury (DBI) and mass lesion, according to the classification of the Traumatic Coma Data Bank. The DBI, basically, has no high or mixed density lesion more than 25ml on CT, and was classified into 4 subgroups: DBI I includes injuries where there is no visible pathology; DBI II includes all injuries in which the cisterns are present with a midline shift of less than 5mm; DBI III includes injuries with swelling where the cisterns are compressed or absent and the midline shift is less than 5mm; DBI IV includes injuries with a midline shift of more than 5mm. The mass lesions were categorised into 3 subgroups: epidural haematoma; acute subdural haematoma; and intracerebral haematoma. Outcomes were determined at 6 months following trauma using the Glasgow Outcome Scale. All DBI I patients recovered well. I n the DBI II group, age, GCS score and detection of multiple parenchymal lesions on CT were significantly correlated with outcome. For the DBI III and IV groups, the only significant prognostic factor was the GCS score. In patients with a mass lesion, th e GCS score was the only significant prognostic factor in the epidural haematoma group, but the GCS score and the presence of subarachnoid haemorrhage were predictive factors in the acute subdural haematoma group. Outcomes were unfavourable in the majori ty of patients with intracerebral haematoma. GCS score could predict outcome in all groups. The confidence of the outcome prediction ranged from 75.8 to 92.1%, depending on logistic regression analysis.

References (15)

G Teasdale et al.
Assessment of coma and impaired consciousness: a practical scale
Lancet
(1974)
KH Chan et al.
Multimodality monitoring as a guide to treatment of intracranial hypertension after severe brain injury
Neurosurgery
(1993)
TG Gennarelli et al.
Influence of the type of intracranial lesion on outcome from severe head injury
J Neurosurg
(1982)
RP Greenberg et al.
Prognostic implication of early multimodality evoked potentials in severely head-injured patients. A prospective study
J Neurosurg
(1981)
A Yamaura et al.
Prognostic value of CT in head injury
Neurol Med Chir
(1981)
A Marmarou et al.
NINDS Traumatic Coma Data Bank: intracranial pressure monitoring methodology
J Neurosurg
(1991)
TG Saul et al.
Effect of intracranial pressure monitoring and aggressive treatment on mortality in severe head injury
J Neurosurg
(1982)

There are more references available in the full text version of this article.

Cited by (102)

Machine learning and prediction of traumatic brain injury mortality
2022, Diagnosis and Treatment of Traumatic Brain Injury: The Neuroscience of Traumatic Brain Injury
Traumatic brain injury is a “pandemic” disease, with alarming numbers each year around the world that accounts for significant healthcare, socioeconomic, and psychological issues in developing countries. The main issue is associated with the disability of affected subjects and reduction of the quality of life, work capacity, and life expectancy. Ability to estimate prognosis at admission is an important decision-making factor. Creating a model for traumatic brain injury evaluation is desired by many researchers. Classical models are based on two important trials: International Mission for Prognosis and Analysis of Clinical Trials and Corticosteroid randomization after significant head injury trial. Each one investigated more than 10,000 patients and this database helped build these two important models. Both are based on admission characteristics, while some other models were created based on clinical courses. Construction of a model involves a rigid protocol: Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis statement recommendations for the reporting of prognostic models. The correct statistical tests must be correctly chosen to select appropriate variables. Machine learning for traumatic brain injury is an elaborate, methodological process that can be used as an example to build similar models for other neurological diseases.
Early death prediction in children with traumatic brain injury using computed tomography scoring systems
2022, Journal of Clinical Neuroscience
Marshall and Rotterdam are the most commonly used CT scoring systems to predict the outcome following traumatic brain injury (TBI). Although several studies have compared the performance of the two scoring systems in adult patients, none of these studies has evaluated the performance of the two scoring systems in pediatric patients. This study aimed to determine the predictive value of the Marshall and Rotterdam scoring systems in pediatric patients with TBI.
This retrospective study included 105 children with admission GCS < 12, with a mean age of 6.2 (±3.5) years. Their initial CT and status at hospital discharge (dead or alive) were reviewed, and both the Marshall and Rotterdam scores were calculated. We examined whether each score was related to the early death of pediatric patients.
The pediatric patients with higher Marshall and Rotterdam scores had a higher mortality rate. There was a good correlation between the Marshall and Rotterdam scoring systems (Spearman's rho = 0.618, significant at the 0.05 level).
Both systems demonstrated a high degree of discrimination when predicting early mortality. The Marshall scoring system had reasonable discrimination (AUC 0.782), and the Rotterdam scoring system had good discrimination (AUC 0.729). Comparing the two CT scoring systems, the Marshall scoring system provided a better positive predictive value (90%) for early mortality than the Rotterdam scoring system (78%).
Both the Marshall and Rotterdam scoring systems have good predictability for assessing mortality in pediatric patients with TBI. The performance of the Marshall scoring system was equal to or slightly better than that of the Rotterdam scoring system.
Quantitative analysis of brain herniation from non-contrast CT images using deep learning
2021, Journal of Neuroscience Methods
Citation Excerpt :
Higher MLS leads to severe damage in the brain. The situation becomes fatal when the observed MLS is more than 5 mm (Ono et al., 2001), (Englander et al., 2003). Among the above-listed features, MLS is the prominent feature for SOL and helps determine the patient's survival.
Brain herniation is one of the fatal outcomes of increased intracranial pressure (ICP). It is caused due to the presence of hematoma or tumor mass in the brain. Ideal midline (iML) divides the healthy brain into two (right and left) nearly equal hemispheres. In the presence of hematoma, the midline tends to shift from its original position to the contralateral side of the mass and thus develops a deformed midline (dML).
In this study, a convolutional neural network (CNN) was used to predict the deformed left and right hemispheres. The proposed algorithm was validated with non-contrast computed tomography (NCCT) of (n = 45) subjects with two types of brain hemorrhages - epidural hemorrhage (EDH): (n = 5) and intra-parenchymal hemorrhage (IPH): (n = 40)).
The method demonstrated excellent potential in automatically predicting MLS with the average errors of 1.29 mm by location, 66.4 mm² by 2D area, and 253.73 mm³ by 3D volume. Estimated MLS could be well correlated with other clinical markers including hematoma volume - R² = 0.86 (EDH); 0.48 (IPH) and a Radiologist-defined severity score (RSS) - R² = 0.62 (EDH); 0.57 (IPH). RSS was found to be even better correlated (R² = 0.98 (EDH); 0.70 (IPH)), hence better predictable by a joint correlation between hematoma volume, midline pixel- or voxel-shift, and minimum distance of (ideal or deformed) midline from the hematoma (boundary or centroid).
All these predictors were computed automatically, which highlighted the excellent clinical potential of the proposed automated method in midline shift (MLS) estimation and severity prediction in hematoma decision support systems.
Identification of clinical characteristics and factors predicting favorable treatment outcomes in elderly patients with traumatic brain injury
2019, Journal of Clinical Neuroscience
Age is an important prognostic factor for patients with traumatic brain injury (TBI), and worse treatment outcomes have been reported in elderly patients. Therefore, proper treatment is needed for the increasing number of elderly patients with TBI. We aimed to analyze predictive factors of favorable treatment outcomes (FTO) in elderly patients. Clinical and radiological data from 493 patients with TBI who visited a single institute were retrospectively collected from January 2014 to December 2015. We compared the characteristics of the elderly group (individuals above 65 years) and younger group (16–65 years). We analyzed the characteristics and outcomes in both groups and the factors related to the Glasgow outcome scale-extended score at 6 months after injury in 170 elderly patients. The treatment outcomes were dichotomized into favorable and unfavorable groups. In the elderly group, the proportion of female patients and the incidence of subdural hemorrhage (SDH) were higher than in the younger group. Among the 170 elderly patients, 62 had pure SDH, and 21 of the 62 with pure SDH had undergone surgical treatment. Compared with other types of intracranial hemorrhage, FTO was as high as 85.5%, and mortality was as low as 11.3% in patients with pure SDH. High initial Glasgow coma scale score, low injury severity score, and normal pupillary reflex were significantly related to FTO in multivariate analysis. Therefore, active therapeutic strategies, including surgery should be considered for elderly patients with pure SDH without intra-parenchymal injury.
Nonsurviving Patients with Severe Traumatic Brain Injury Have Maintained High Serum sCD40L Levels
2019, World Neurosurgery
Soluble cluster of differentiation 40 ligand (sCD40L) is a member of the tumor necrosis factor family with proinflamatory and procoagulant effects. A previous study found higher serum sCD40L levels at day 1 of traumatic brain injury (TBI) in nonsurviving than surviving patients. Thus the objective of this study was to compare serum sCD40L levels during the first week of a severe TBI between surviving and nonsurviving patients and to determine whether it could be used as a mortality predictor biomarker.
In this multicenter study severe TBI patients (with Glasgow Coma Scale score <9) with an Injury Severity Score in noncranial item <9 were included. Serum sCD40L concentrations at days 1, 4, and 8 of TBI were determined. We performed receiver operating characteristic analyses to determine the capacity of 30-day TBI mortality prediction by serum sCD40L levels at days 1, 4, and 8 of TBI.
We found that nonsurviving (n = 34) patients in comparison with surviving (n = 90) patients had higher sCD40L levels on days 1 (P < 0.001), 4 (P = 0.004), and 8 (P < 0.001) of TBI. We also found that the areas under curve of serum sCD40L concentrations at days 1, 4, and 8 of TBI to 30-day mortality prediction were 82% (P < 0.001), 72% (P = 0.01) and 83% (P < 0.001), respectively.
The existence of higher serum sCD40L levels in nonsurviving than surviving patients during the first week of TBI and fact that serum sCD40L levels during the first week of TBI can be used as a mortality predictor biomarker are the new findings of our study.
Prognostic Predictors of Early Outcomes and Discharge Status of Patients Undergoing Decompressive Craniectomy After Severe Traumatic Brain Injury
2019, World Neurosurgery
Although several prognostic factors for traumatic brain injury (TBI) have been evaluated, a useful predictive scoring model for the outcomes has not been developed for patients with severe TBI who undergo decompressive craniectomy (DC). The aim of the present study was to determine independent predictors and develop a multivariate logistic regression equation to predict the early outcome and discharge status for patients with severe TBI who have undergone DC.
A total of 13 different variables were evaluated. The data from all 278 patients with severe TBI who had undergone DC in the present study were retrospectively evaluated from July 2011 to June 2017. Using univariate, multiple logistic regression and prognostic regression scoring equations it was possible to draw receiver operating characteristic curves to predict the early outcomes and discharge status after TBI.
We found that younger age (P = 0.012), no significant medical history (P = 0.044), diameter of both pupils <4 mm (P = 0.032), higher admission Glasgow coma scale score (P = 0.004), no tracheotomy (P < 0.001), and DC for severe TBI were associated with a favorable early outcome and discharge status. Using receiver operating characteristic curves to predict the probability of a favorable outcome, the sensitivity was 80.0% and the specificity was 79.5%.
Our preliminary findings have shown that 5 variables can be used as independent predictors in assessing the early outcome and discharge status for patients with severe TBI after DC.

View all citing articles on Scopus

^f1: Correspondence to: Jun-ichi Ono, MD, Department of Neurosurgery, Chiba Cardiovascular Center, Tsurumai 575, Ichihara, 290-0512, Japan. Tel.: + 81 436 88 3111; Fax: + 81 436 88 3032.

View full text

MiscellaneousOutcome prediction in severe head injury: analyses of clinical prognostic factors

Abstract

Lancet

Neurosurgery

Influence of the type of intracranial lesion on outcome from severe head injury

J Neurosurg

Prognostic implication of early multimodality evoked potentials in severely head-injured patients. A prospective study

J Neurosurg

Prognostic value of CT in head injury

Neurol Med Chir

NINDS Traumatic Coma Data Bank: intracranial pressure monitoring methodology

J Neurosurg

Effect of intracranial pressure monitoring and aggressive treatment on mortality in severe head injury

J Neurosurg

Miscellaneous
Outcome prediction in severe head injury: analyses of clinical prognostic factors