The National Traumatic Coma Data Bank. Part 2: Patients who talk and deteriorate: implications for treatment

Avoidable factors contributing to death of children with head injury

OBJECTIVE

To assess the incidence of potentially avoidable complications contributing to death of children with head injuries.

DESIGN

Retrospective review of children who died with head injuries from 1979 to 1986 from data of the Office of Population Censuses and Surveys, Hospital Activity Analyses, case notes, coroners' records, and necropsy reports.

SETTING

District general hospitals and two regional neurosurgical centres in Northern region.

RESULTS

255 Children died from head injury in the region, the mortality being 5.3 per 100,000 children per year. Head injury was the single most important cause of death in children aged greater than 1 year, accounting for 15% of deaths in children aged 1-15 years and a quarter for those aged 5-15 years. 121 Potentially avoidable factors possibly or probably contributing to death occurred in 81 children (32%). Half the children (125) died before admission, 27 of whom (22%) had potentially avoidable factors possibly or probably contributing to death, and 130 died after admission, 54 of whom (42%) had 93 such factors, which included failure of diagnosis or delayed recognition of intracranial haemorrhage or associated injury, inadequate management of the airways, and poor management of the transfer between hospitals.

IMPLICATIONS

Regions should revise urgently their guidelines for optimal management and indications for neurosurgical referral to include children with severe head injuries and audit their systems of care for all patients with head injuries.

Brain shift, level of consciousness, and restoration of consciousness in patients with acute intracranial hematoma

Recently, Ropper reported that horizontal brain shift caused by acute unilateral mass lesions correlated closely with consciousness, and suggested that recovery of consciousness was unlikely to occur after surgical evacuation if the shift was insufficient to explain the observed diminution of consciousness. The authors have sought to confirm the correlation of pineal shift with level of consciousness and to assess the prognostic value of brain shift measurements in a prospective study. Forty-six patients (19 with subdural hematoma, 14 with intracerebral hematoma, and 13 with epidural hematoma) were accrued to the study group consecutively. A correlation was found between a decrease in the level of consciousness and a significant increase in the mean lateral brain displacement at the pineal gland (from 3.8 to 7.0 mm) and septum (5.4 to 12.2 mm). When outcome was examined in patients who were stuporous or comatose on admission, a significant increase in septal shift was found among patients with a poor outcome, but there was no significant relationship between outcome and degree of pineal or aqueductal shift. A poor outcome was more likely with effacement of both perimesencephalic cisterns or the ipsilateral cistern, but not the contralateral cistern, although this difference did not reach statistical significance. These results do not substantiate the value of brain shift as an independent prognostic factor after evacuation of an acute unilateral mass lesion. The decision to operate and the determination of prognosis should be based rather on established criteria such as the clinical examination, age of the patient, and the mechanism of injury.

Severe traumatic brain lesions in Sweden. Part 2: Impact of aggressive neurosurgical intensive care

During a 6 year period (1977-1982), 425 patients were treated in the Department of Neurosurgery, University Hospital of Lund, for severe traumatic brain lesions (coma greater than 6 hours). From 1983 a more aggressive management protocol was introduced including early recording of intracranial pressure (ICP) and 162 patients were included in the study 1983-1984. A dangerous increase in ICP in spite of adequate surgical treatment and moderately controlled hyperventilation was the incentive for barbiturate coma therapy in selected patients. In the first part of the study overall mortality was 48% whereas 39% of the patients reached good recovery/moderate disability 6 months after injury. During the second part of the study the corresponding figures were 35% and 54%, respectively (in both cases p less than or equal to 0.01). In the group of patients with focal intracranial mass lesions mortality decreased from 59% to 46% (p less than or equal to 0.05) and good recovery/moderate disability increased from 30% to 42% (p less than or equal to 0.05). Improvement in outcome was even more pronounced in patients with no-mass lesions, mortality decreased from 30% to 12% and good recovery/moderate disability increased from 56% to 80% (p less than or equal to 0.05 and p less than or equal to 0.01, respectively). No change occurred in age distribution or in the types of intracranial lesions that could explain these improvements. It is concluded that aggressive neurosurgical intensive care significantly improves outcome in patients with severe traumatic brain lesions.

Severe traumatic brain lesions in Sweden. Part I: Aspects of management in non-neurosurgical clinics

This paper reports a study of 587 consecutive patients treated for severe traumatic brain lesions (coma greater than 6 hours) during 1977-1984. Epidemiology, management and outcome were documented in 425 patients during the first part of the study (1977-1982) as a basis for future efforts at improvements. A total of 70-80 patients with severe head injuries were admitted annually to the Department of Neurosurgery in Lund and 88.6% of these patients were referred from 14 local hospitals, most of which are situated more than 50 km from Lund. Half of the patients were older than 40 years and 25% older than 60. Focal intracranial mass lesions were diagnosed in 64% of the patients. In the total study 41% of the patients were described as 'talk and deteriorate' and 13% as 'talk and die'. In 1983 a protocol for primary management was introduced in all local hospitals in the region. The management protocol caused a significant decrease (p less than or equal to 0.05) in the number of explorative craniotomies in local hospitals and a virtual disappearance of late surgical procedures (greater than 6 hours after injury). A fall was observed in the number of patients arriving at the Department of Neurosurgery with respiratory insufficiency. The study illustrates the epidemiology of severe head injuries in Sweden and the present state of management of these patients in non-neurosurgical departments. It is concluded that an overall outcome comparable to other reported series is also feasible in regions with a relatively sparse population and large geographical distances provided that strict recommendations for initial management are given to the local hospitals.

Asymptomatic extradural haematomas. Results of a multicenter study of 158 cases in minor head injury

The authors report a study conducted in three Italian neurosurgical centres on 158 patients admitted after a minor head injury and with CT findings of a hitherto asymptomatic significant extradural haematoma. All patients were examined both prospectively by means of a computerized record containing 18 clinical and radiological parameters, and retrospectively by logistical regression analysis, in order to ascertain which factors influenced most the choice of surgical vs. conservative management. The size of the haematoma, rather than its location, and the degree of midline shift were the factors most influential in deciding in favour of surgical treatment, with a specificity of 0.83 and a sensitivity of 0.92. Conservative management of haematomas having a maximum thickness of less than 10 mm with a midline shift of less than 5 mm appears as safe. Outcome was "good recovery" in both the surgical and the nonsurgical patients, with only one death in the whole series, unrelated to the extradural lesion. This study focuses attention on a group of patients who are seldom examined by CT scan, but who can harbour potentially lethal lesions. Extension of CT scan examination to all adult patients with a minor head injury and a skull fracture can be recommended in order to identify significant haematomas in an asymptomatic phase.

Skull fracture as a factor of increased risk in minor head injuries. Indication for a broader use of cerebral computed tomography scanning

Two series of patients with a minor head injury (for a total of 182 cases), differing only in the presence and absence of a linear skull fracture, were admitted to a nonspecialized hospital and prospectively examined by computed tomography scanning. The presence of a fracture line proved to be significant, inasmuch as it was accompanied by approximately 38% of intracranial abnormalities versus 6% in the nonfracture cases. Early detection of any intracranial pathology that was still asymptomatic allowed prompt transfer of patients to the neurosurgical center, where operative treatment was carried out, when indicated, without mortality or morbidity. All operations (11 cases) were performed on patients with a fracture (105 cases) whereas none of the nonfracture patients (77 cases) required surgery. It is proposed that adult patients with minor head injuries with a skull fracture be submitted to computed tomography scanning in order that intracranial lesions may be detected, and treated, before the onset of clinical deterioration.

Current reporting of responsiveness in acute cerebral disorders. A survey of the neurosurgical literature

One hundred sixty-six papers published in seven neurosurgical journals from 1983 through 1985 have been surveyed to determine the methods used for assessment of overall patient responsiveness in acute cerebral disorders (coma grading). Fifty-one different coma scales or modifications were found. The Glasgow Coma Scale (GCS) sum score (that is, the sum of the scores of the individual eye, verbal, and motor scales) dominated (54%), and was used in 73 (76%) of 96 of the head-injury studies; in 56 (77%) of these 73 studies it was the single method of grading neurological status. The GCS sum score was used in 16 (23%) of 70 studies in patients with other etiologies. The Hunt and Hess scale was used in 26 (57%) of 46 reports of patients with subarachnoid hemorrhage. In 31 (55%) of the 56 studies of head injuries using the GCS alone, it was not obvious if the 12- or 13-grade scale was used. In 13 studies (23%) no reference to methodological investigations was made. In 44 papers (79%) the handling of untestable features, such as intubation or swollen eyes, was not reported. In the 56 studies using the GCS alone, coma was defined in many different ways and in 22 studies the definition of coma was not specified. In 63% of reports, the GCS sum score scale was combined in one to five groups of scores and this was done in 32 different ways. No information was available to describe the procedure of data aggregation or the reliability of the 13-grade GCS sum score. The lack of standardization makes it unnecessarily difficult to perform valid comparisons between different series of patients. Since the GCS sum score is the most widely used scale, it is suggested that the reporting of the GCS sum score should be standardized regarding pseudoscoring, coma definition, and use of combined scores. Further studies on the reliability of the GCS sum score are needed.

Outcome from head injury related to patient's age. A longitudinal prospective study of adult and pediatric head injury

A series of 8814 head-injured patients admitted to 41 hospitals in three separate metropolitan areas were prospectively studied. Of these, 1906 patients (21.6%) were 14 years of age or less. This "pediatric population" was compared to the remaining "adult population" for mechanism of injury, admission Glasgow Coma Scale score, motor score, blood pressure, pupillary reactivity, the presence of associated injuries, and the presence of subdural or epidural hematoma. The relationship of each of these factors was then correlated with post-traumatic mortality. Except for patients found to have subdural hematoma and those who were profoundly hypotensive, the pediatric patients exhibited a significantly lower mortality rate compared to the adults, thus confirming this generally held view. This study indicates that age itself, even within the pediatric age range, is a major independent factor affecting the mortality rate in head-injured patients.

Changes produced by CT scanning in the outlook of severe head injury

This is a study of the changes that have occurred in the field of severe head injury since the advent of CT scanning, comparing two homogeneous series of patients selected by clinical status (Glasgow Coma Scale less than or equal to 8), namely a series of 1,000 cases admitted to our Department between 1973 and 1976, already published in this Journal, and one of 385 cases cared for between 1979 and 1980, when CT scanning had become generally available. The two series of patients compare very closely in many respects, particularly in the incidence of surgical cases. In the more recent series the overall outcome was better both in surgical and in non-surgical cases. Among patients in the CT scan series the incidence of brain contusion associated with haematoma was greater than that of pure subdural haematomas. In non-surgical patients the CT scan, unlike cerebral angiography, afforded better identification of traumatic lesions and the grouping of patients into homogeneous categories correlating with a given outcome. On admission, cerebral angiography and CT scanning were equally effective in detecting lesions of surgical import; later in the course of the illness, however, CT scanning proved far more effective in detecting changes, with fully 15% of the patients being referred for surgery in the light of repeat CT scan findings as opposed to only 4% undergoing surgery on the indications of repeat angiography. Also, in the new series the mean interval from injury to surgery was shorter, with 64% of patients being operated on within 6 hours of the injury.(ABSTRACT TRUNCATED AT 250 WORDS)

Epidural monitoring of the intracranial pressure in severe head injury characterized by non-localizing motor response

Recent studies on prognostic variables and the intensive care of head injuries enabled us to select 64 patients and administer a standard treatment protocol to prevent secondary brain injury. All the patients were in coma with a flexor motor pattern as the best response between 6 and 24 hours after the accident and/or decompressive surgery. Continuous epidural intracranial pressure (ICP) monitoring was used in all patients to control the effect of positioning, analgetics, hyperventilation and osmotherapy. None of the patients with a normal initial ICP (15 mm Hg or less) developed an ICP increase leading to brain tamponade. A subsequent further rise to 40 mm Hg signified a very high risk of progression towards brain tamponade. The majority of the patients (71%) with a maximum ICP increase of less than 40 mm Hg had an acceptable recovery. After 6-12 months, the outcome in this series of patients was 48% with a good/moderate recovery, 14% with severe deficits and 38% dead/vegetative.

Review of 1,000 consecutive cases of severe head injury treated before the advent of CT scanning

This is a review of 1,000 consecutive cases of severe head injury admitted to our Neurosurgical Department between January 1973 and August 1976, before the advent of CT scanning. All patients were comatose following head injury (GCS less than or equal to 8) and were treated homogeneously by the same neurosurgical team by a protocol that included immediate resuscitation on arrival, diagnosis of intracranial lesions by angiography, early surgery when needed, mechanical ventilation, steroids, and mannitol. Extracranial lesions, even if preponderant, were treated by various specialists in the Neurosurgical Department, which for all practical purposes operated as an Emergency Department. Admission criteria were very broad with no preadmission selection. The overall mortality for this series was 45%. A little less than half the patients made good recoveries or remained moderately disabled (47%); 6% were severely disabled, and 2% survived in a persistent vegetative state. More than two-thirds of the patients were brought to our Neurosurgical Department after a short stay at a general hospital; 72% were admitted within 6 hours of injury; 71% were traffic accident victims; and 34% had significant associated extracranial injuries. Carotid angiography was performed in 78% of the patients and indicated the presence of an intracranial haematoma requiring surgery in 36% of the whole series. Mortality was significantly higher in operated than in unoperated patients (56% versus 39%); those treated surgically, however, were older, in worse clinical condition, and showed a higher incidence of acute subdural haematomas associated with brain contusion. Carotid angiography proved very effective in revealing the presence of an expansive lesion but failed to reflect the severity of brain damage, since the group with "negative" angiograms showed a high mortality (52%). Patients with a lucid interval had a higher percentage of surgical lesions than those with immediate coma (58% versus 26%); but fully 42% of them did not require surgery, and 25% had negative angiograms. From the prognostic point of view the clinical data elicited after initial resuscitation were highly predictive of the outcome: some individual neurological signs, such as mydriasis, posturing and eye movements, were not inferior to the GCS score in that respect. Age also proved a strong predictor, since elderly patients are more likely to have severe subdural and parenchymal lesions and their clinical severity is accordingly greater.(ABSTRACT TRUNCATED AT 400 WORDS)