INTRODUCTION

Brain injury is one of the most common problems that neurosurgeons deal with and constitutes a major problem. Despite the advent of rapid air transport, emergency medical service system, computerized tomography and aggressive surgical treatment, injuries are the leading cause of death and disability under the age of 45 years in most countries. (1-5) Up to 50% of trauma deaths are due to brain damage.(1) Recognition that Trauma Brain Lesions (TBL) is a major health problem that has led to several studies. Because of well coordinated emergency systems, some of the patients are treated at the accident site, with sedation, intubation and ventilation. Consequently, these patients are admitted to the care of the neurosurgeon in a condition that does not allow neurological examination. (2) In such situations the major role for estimation of prognosis and indication of surgery depend on Computed Tomography (CT) scan images. CT-scan and its evaluation thus have increased importance because it demonstrates the real extent of the injury. Some hospitals are equipped with a CT-scan but have no Neurosurgical department. The consultant neurosurgeon therefore depends increasingly on the interpretation of the CT-scan transmitted via medical information systems. This paper addresses the prediction of bad outcome from TBL, in Dubai, by a ball system of brain CT-scans designed by M. Haj Ali and called Haj CT score (HCTS). The scope of this study is to analyse the CT findings and their correlation with the clinical state and with bad outcome.

In this study, under the term of "Severe TBL," we classified the patients who had on admission recorded clinical and pathological data by neuroexamination and CT-scan imaging that led to bad prognosis i.e. death, vegetative state or severe disability. Severe disability means the patient is dependent on others. Despite differences in most cases only 5-10 % of admissions categorised as severe. (1,2)

We studied CT scan findings and clinical data of patients who had bad outcome from a TBL in Dubai. Dubai is an active quick developing trade center with high level of travel and tourism. There are 700,000 residents most of them active and young. Large number of visitors entered Dubai from other cities of United Arab Emirates and abroad.

Patients and Methods: In Rashid Hospital, the only Neurosurgical department in Dubai 2123 admissions of head-injury victims were admitted during the last 5 years. Bad outcome was noted in 156 isolated severe head injury patients: Death 107, vegetative 18, severe disability 31.

In this study we will present the CT scan data of 156 patients who had bad outcome. Fourty-two patients died on the first day, 34 between 1 and 3 days after trauma. Sixteen between 4 to 10 days. Fifteen others died 1 month and later. Thirty-one patients were severely handicapped and 18 vegetative survivors. Follow up period 6 months to 5 years.

Classification of Trauma: Road traffic accident: 101, fall: 28, sports trauma 5, camel racing 9, assault 3. The ages ranged from 2 months to 73 years. Mean: 39.7 years, M:131 and F:25. On admission, these patients were categorised according to Glasgow Coma Scale (GCS).

Table

GCS 3 3-5 5-8 8-13 Patients 49 77 24 6

Haj CT scan Score of patients with TBL: From prospective review of head-CT features, a total CT -grade was assigned to each abnormal finding in each CT-scan cut, in relation to the type and location of the lesion consisting of the following features: Basal cistern effacement localised or generalised edema, thickness and extent of epidural \ subdural haematoma, subarachnoidal haemorrhage, presence of mass effect, shift of midline, location and extent of contusion: cortical, subcortical or deeply seated in basal ganglia, midbrain or brain stem.

Computed tomography scans were studied cut by cut. Each cut was divided into three regions of the anterior posterior aspect and three regions in the depth of the brain.

Each side left or right should be calculated(See Figure 1). Any abnormal traumatic lesion found in the CT- cuts were calculated and recorded accordingly. Then a sum ball of all cuts was recorded and all were added together to get Haj CT Score of each patient. Anterior area: extradural damage 0.01, subdural, cortical 0.02 deeply seated 0.03. Totally damaged one frontal lobe in one cut = 0.06. Medial area: extradural damage 0.02, subdural, cortical 0.04 deeply seated 0.06. Totally damaged cut in temporal or medial fossa 0.12. and in the Posterior area: Posterior fossa or occipitoparietal: extradural damage 0.03, subdural, cortical 0.06 deeply seated 0.09. Totally damaged cut in this area: 0.18. Damage in midbrain and brain stem 0.14. Damage of full half of the CT-scan cut 0.5. In the hole cut 1.0.

	Figure 1 — Scheme for calculation of one Cut Computed Tomography scanning for Haj CT score.

Treatment: The management was limited mainly by hyperventilation and dehydration. Operative treatment was used as a generally accepted rule when there was a considerable compressing factor with clear mass effect and midline shift (Subdural or epidural haematomas or deeply depressed bone fragment) indicating surgical intervention. Patients who had fixed dilated pupils, without corneal reflexes and not responding to painful stimuli were not decompressed if they were in such a condition for more than 1 hour before the diagnosis was made and if the patient had an apnea for more than 10 minutes.

Results: Whatever form of treatment was used we found that 98 patients who had a HCTS > 5 died. Among these, 6 cases were operated upon but all died. Among 25 patients who had HCTS of 4-5, six died, 16 became vegetative and 3 severely disabled.

Another group of 33 patients who had HCTS between 3-4, majority of them (28 patients) became severely disabled three died and 2 became vegetative. Patients who had HCTS < 3 had a favourable outcome and were not included in this study.

Initially we started our study with retrospective analysis of 100 files patient who had already bad outcome followed up for 2 years after the incident of trauma. We found the most from evident factor existing on all death cases, vegetative and severe disabled patients. Then we developed HCTS. The admission criteria in preventable bad outcome of severe TBL in Dubai were found as follows:

1.)	GCS < 5.
2.)	HCTS > 3 which includes:
	Temporal lobe contusion more than 3 cm in diameter.
	Two lobes contusion more than 2 cm in diameter each lobe.
	Diffuse axonal brain injury in more than two locations: spot of haemorrhagic contusion in the basal ganglion, midbrain, brain stem or pare\ intraventricular bleeding.
	Marked subarachnoid bleeding.
	Severe brain oedema including: a. Obliteration of perimesencephalic cistern. b. Thinning of the lateral ventricles. c. Obscuring of third and/or fourth ventricle.
	Effacement of gyrii, sulci including disappearance of Sylvan fissure.
	Ischemic brain changes(Brain stem or hemisphere).

Predisposing factor for Bad Outcome of severe TBL:

1.)	Shock condition. Long bone fracture.
2.)	Brain hypoxia, hypercapnia, and hyperglycemia.

We found 26 dead cases were of traumatic brain lesions, but unfortunately there was no postmortem section and they were certified as death of head injury nature.

	Table 1 — Outcome of patients in relation to the HCTS
	HCTS Patients Death Veg. SD >5 98 98     4-5 25 6 16 3 3-4 33 3 2 28 Total 156 107 18 31
	Veg.-: vegetatives, SD-severe disability

	Table 2 — Haj CTS and Glascow Coma Scale
	HCTS Patients GCS 3 3-5 5-8 8-13
	>5 98 40 58 - - 4-5 25 7 13 3 2 3-4 33 2 6 21 4 Total 156 49 77 24 6

DISCUSSION

The role of CT- scan imaging in diagnosis and prognosis of TBL is well established. Several authors studied the role of single or group of pathological traumatic findings in the CT- scan images on the outcome of TBL. (4,7) On the basis of these studies and on our clinical observation with head injured patients, Haj CT Score was developed. Well known depth and extent of the lesion was a predictive level of recovery. This term was taken into consideration so in our study we took the CT scan images as a whole. It gives a real impression of the lesion and gives a real possibility for estimation of the correct prognosis of the outcome depending on facts and mathematical calculation. Our findings correlate with the studies mentioned above about the value of subarachnoid haemorrhage, subdural haematoma, extent and location of contusion, as was the case for the role of GCS scores, and the neurological findings in prediction of outcome.(5,6)

Our data suggests that Hai CT scan score References system in brain injured patients is certainly indicative of the severity of injury and correlates with the outcome. The most unfavourable outcome found when HCTS reaches > 5 correlates with the types of lesion such as multiple contusions, marked oedema, shearing injury, intracerebral haematoma, and subdural haematoma.

CONCLUSION

Although CT data is an important determinant of the final outcome in brain injured patients, the CT picture alone did not allow prediction of an absolute prognosis in all cases. It can do so in patients with isolated severe head injury. The presence of extracranial lesions, complications and differences in management can influence the outcome of some cases. Closed isolated brain injury can be fully prognosticated by Haj CT-scan Score. HCTS provides the most accurate identification of the type and extent of brain damage, which permits estimation of the outcome on pathological basis. Utilisation of neuroimaging information with neurological data enabled investigators and clinicians to predict the outcome with high degree of confidence. In future this will be enhanced and determined by functional brain imaging.

REFERENCES

1.	Head injury, Graham M Teasdale, J. of Neurology, Neurosurgery, and Psychiatry, 1995, 58: 526-539.
2.	Bryan Jennett, Epidemiology of head injury, J.of Neurology, Neurosurgery, and Psychiatry, 1996, 60: 362-369.
3.	Raj K. Naryan, Development of Guidelines for the Management of Severe Head Injury, J. Nuerotrauma, 1995, NS, 12: 907-912.
4.	Karl A. Green, Fredrick F. Marciano, Blake A. Jonson et all. Impact of traumatic subarachnoid hemorrhage on outcome in nonpentrating head injury, J Neurosurgery, 1995, 83: 445-453.
5.	H. S. Levin , Prediction of recovery from Traumatic Brain Injury, J of Neurotrauma, 1995, 12, N5.
6.	T. Tomberg, U. Rink, E. Pikkoja and A. Tikk, Computerized Tomography and Prognosis in Pediatric Head Injury, Acta Neurochir (Wien), 1996, 138: 543-548.
7.	Matthia's Zumkeller, Renate Behrman, Hans Egmont Heissler et all. Computed tomographic criteria and survival rate for patients with acute subdural hematoma, Neurosurgey, 1996, 39: 4.