Aust. N . Z . J . Surg. 1992.62, 759-162
750
ORIGINAL ARTICLES A SURVEY OF INTERHOSPITAL TRANSFER OF HEAD-INJURED PATIENTS WITH INADEQUATELY TREATED LIFE-THREATENING EXTRACRANIAL INJURIES A. HENDERSON,* T. COYNE,’ D. WALL+AND B. MILLER’ Departments of *Intensive Care and ‘Surgery, Princess Alexandra Hospital, Ipswich Road, Brisbane, Queensland, Australia A 12 month prospective study was undertaken to determine the frequency of untreated life-threatening extracranial injuries in patients transferred to a major trauma centre because of head injury. Of the 43 patients transferred (15 with isolated head injury and 28 with multiple injuries), four (9%) had an untreated lifethreatening extracranial injury, which caused death in two. All four patients with untreated extracranial injuries were transferred from hospitals with general surgical staff and facilities. In three of the patients (none with a major head injury), the extracranial injuries were recognized at the referring hospital, but were left untreated in the rush to transfer the patient to a neurosurgical facility. In the fourth patient, who had a severe head injury, recurrent hypotension from a ruptured spleen was mistakenly ascribed to a scalp wound. The series shows that the dangerous practice of hurriedly transferring patients to trauma centres because of actual or perceived head injuries, while leaving major extracranial injuries untreated, continues despite warnings in the literature and the efforts of the Royal Australasian College of Surgeons through the Early Management of Severe Trauma programme.
Key words: extracranial injuries, head injury, interhospital transfer, untreated.
Introduction The geography and demography of Australia make interhospital transfer of patients an integral part of health care. Although interhospital transfer is not without risk, early transfer of severely injured patients to major trauma centres confers a survival advantage and has become an important part of hospital practice. I-’ Patients with severe head injuries are especially likely to be urgently transferred, as it has been shown that early computerized tomography (CT) scanning of the head with expeditious evacuation of intracranial haematomas, reduces mortality and morbid it^.*,^ In multiply injured patients admitted to hospitals without CT scanning facilities, the occurrence of a head injury may distract the clinician to the extent that serious extracranial injuries may go undetected or untreated while the patient is urgently transferred to a trauma centre. Uncontrolled haemorrhage during transfer may cause severe hypotension in these patients and this is known to adversely affect the prognosis of the head injury. “ - I 2 Surveys from the UK indicate that between 5 and
”-’*
Correspondence: Dr A. Henderson, Intensive Care Unit, Princess Alexandra Hospital, Ipswich Road, Brisbane, Qld 4102, Australia. Accepted for publication 26 March 1992
9% of head-injured patients transferred to a neurosurgical centre had inadequately treated extracranial injuries. l 3 In Australia, deficiencies in cardiorespiratory management were identified in 7% of headinjured patients transferred to major centres from rural areas. l 4 The Royal Australasian College of Surgeons has evolved an education programme (Early Management of Severe Trauma, EMST) in an attempt to improve the overall care of trauma victims. This programme gives specific advice on matters relating to patient transfer. I s A survey was undertaken of all patients transferred to the intensive care unit because of head injuries. to determine the frequency with which head-injured patients arrived with untreated, life-threatening extracranial injuries.
Methods The Princess Alexandra Hospital (PA hospital) is a 1000 bed adult teaching hospital in Brisbane. The hospital serves as a major trauma centre for a large area of Southern Queensland and Northern New South Wales. Seriously injured patients are transferred to the intensive care unit (ICU) by road or air from numerous smaller hospitals within the geographical catchment area. A prospective computerized database was kept for all patients transferred
HENDERSON E T A L .
760
from outside Brisbane because of head injury using Dbase IV (Ashton Tate). This report is based upon a study of all head injury patients transferred to the ICU for the 12 months ending 30 June 1991. In every case, the prospective database was verified by inspection of the hospital notes. Age, sex, outcome and the severity and extent of extracranial injuries were noted. A record of any patient with untreated life-threatening extracranial injuries was kept and a detailed review of their case notes was undertaken.
Results During the study period, 43 head-injured patients with a mean age of 34 years (range: 14-65 years) were transferred to the intensive care unit of the PA hospital from outside the Brisbane metropolitan area. There was a preponderance of males (M:F = 4 : 1). Fifteen (35%) had isolated head injuries while 28 (65%) had head and extracranial injuries. Eight (18.6%) died in ICU (three with isolated head injuries and five with multiple injuries). Four patients (9%) were identified as having life-threatening undiagnosed or untreated extracranial injuries. Their case histories were reviewed in detail.
Case histories
watershed infarcts consistent with profound hypotension but no evidence of intracranial haematoma. She died 2 days later. Autopsy confirmed the CT findings and revealed no further pathology within thel abdomen. PATIENT 2
A man of 29 was injured when his motor cycle struck a car. On arrival at a large urban hospital he had a systolic blood pressure of 70 mmHg and GCS of 1 1 . His injuries were a minor facial laceration, blood in the left external auditory meatus, bilateral closed radial fractures and a chest injury comprising a small left lower flail segment and pneumothorax. The patient was intubated and drug paralysed on admission. He was stabilized by the infusion of 6.5 L of colloid and crystalloid and drainage of the pneumothorax. A ruptured spleen was suspected, but a local decision was made to immediately transfer the patient to PA hospital because of a suspected basal skull fracture. He became hypotensive during the 1 h journey by road. Peritoneal lavage on arrival at PA hospital was positive. At laparotomy he had haemoperitoneum from a ruptured spleen. A subsequent CT scan of head was normal. The source of the blood in the external auditory meatus was shown to be from the nearby scalp laceration rather than from a basal skull fracture. He made a complete recovery.
PATIENT 1 PATIENT 3
A woman of 18 was injured when her car was struck by a lorry. She sustained a closed head injury with a minor scalp laceration. fractured pelvis and pneumothorax. She was admitted to a large urban hospital and had a systolic blood pressure of 60 mmHg and a Glasgow Coma Scale (GCS) of 8. The patient was intubated and ventilated using pancuronium muscle relaxation. Following resuscitation with 1 L of haemaccell and 1 L of Hartmann’s solution and drainage of the pneumothorax her systolic blood pressure rose to 121 mmHg. The systolic blood pressure fell to 70 mmHg during the following 30 min. Peritoneal lavage was positive and an ultrasound showed a ruptured spleen. She was eventually stabilized with 5 u haemaccell, 1 L normal saline and 3 u packed cells. It was decided locally that priority should be given to the head injury and that the patient required immediate transfer to PA hospital where her cranial and abdominal injuries could be treated together. During the hour long journey by road she became profoundly hypotensive and was asystolic on arrival. A laparotomy was carried out after resuscitation which revealed massive intraperitoneal bleeding from rupture of the liver and spleen. Haemostasis was rapidly secured and she was transferred to ICU. A subsequent CT scan of her head showed several
A man of 41 was injured when he was thrown from a car. He sustained a severe compound head injury. On arrival at a rural hospital he had a GCS of 5 and a systolic blood pressure of 75 mmHg. He was intubated, transfused 2.5 L of colloid and crystalloid and the scalp wound was sutured. He was then transferred, ventilated and drug paralysed, by road to the base hospital. On arrival he was pale and hypotensive (systolic blood pressure = 60 mmHg). Fbllowing stabilization with a further transfusion of crystalloid and 4 u of packed cells a CT head scan was done with showed a compound skull fracture and extensive frontal lobe contusion. In view of the serious head injury, the patient was flown to Brisbane, a journey of 2 h and 15 min, without further ifivestigation of the hypotension, which was ascribed to the scalp wound. On amval at PA hospital he was again hypotensive. Peritoneal lavage was positive and laparatomy revealed a ruptured spleen. He survived with a severe neurological deficit. PATIENT 4
A man of 20 was injured in a motor vehicle accident. He sustained a small compound depressed skull fracture, a large scalp wound and closed fractures of the right tibia and femur. On arrival at a
TRANSFER OF HEAD-INJURED PATIENTS
large rural hospital his systolic blood pressure was 85 mmHg and the GCS was 15. After transfusion of 2 L of colloid and crystalloid plus 2 u of packed cells, the systolic blood pressure rose to 130mmHg. Because of the head injury, the patient was immediately flown to the PA hospital, ajourney of 50min. On arrival his fractures were unsupported. Although the systolic blood pressure was 125 mmHg, the patient was cyanosed, intensely vasoconstricted with a heart rate of 140 beatshin and a respiratory rate of 40 beatdmin. Chest X-ray revealed diffuse alveolar shadowing typical of fulminant fat embolism. Arterial gases on air confirmed severe oxygenation failure (Pa02 =37 mmHg, PaCOz = 34mmHg). After resuscitation, including intubation and ventilation, the infusion of 1 L of haemaccell and 2 L of normal saline and splinting of the fractures, he had a CT scan of head which revealed a minor depressed skull fracture without evidence of further intracranial injury. He developed pulmonary hypertension (pulmonary artery systolic pressure =70 mmHg) with refractory hypoxia and died 7 h after admission. Autopsy confirmed massive fat embolism but showed no evidence of significant intracranial injury.
Discussion Patients with actual or suspected major head injuries admitted to hospitals without immediate access to a CT scanner present a major clinical problem because it has been shown that those who are unconscious, or who have impaired consciousness in the presence of a skull fracture, have a 25% incidence of intracranial haematoma. l5The pressure to urgently transfer such patients to a neurosurgical unit is high because early CT scanning with expeditious drainage of intracranial haematomas has been shown to improve progno~is.'.~In patients with isolated head injury, hypoxia is the major transferrelated complication. "-I2 This serious complication can be avoided by intubation and ventilation for the journey. Patients with major head and extracranial injuries present a particular problem. In the present series, 65% of the referrals with head injuries had major extracranial injuries. This figure is similar to that reported from a neurosurgical unit in the West of Scotland.".13 In such patients, the presence of a head injury may distract the clinician from other injuries, or it may dominate the clinical picture to the extent that the head injury receives an inappropriately high treatment priority. ''.I3 The severity of the head injury can be difficult to assess, as factors other than brain injury, such as alcohol, hypoxia and hypotension, may depress the level of consciousness mimicking head injury or exaggeration of the severity of a head injury." While rapid resuscitation may improve cerebral func-
76 I
tion in patients with neurological deficits due to hypoxia or hypotension, such improvements may be masked by the use of drugs to aid mechanical ventilation, especially muscle relaxants. ' 7 , 1 8 In this series, the severity of the head injury was overestimated in three of the patients, two of whom later died. In patient 1, hypotension from intra-abdominal haemorrhage was the probable cause of the reduced level of consciousness. The neurological response to resuscitation was masked by the use of pancuronium to aid mechanical ventilation. With patient 2, the appearance of blood in the external auditory meatus coupled with a scalp laceration and reduced level of consciousness suggested a basal skull fracture. Again the early use of muscle relaxants prevented ongoing assessment of neurological status. Further review revealed no evidence of a major head injury. In patient 4 the dramatic scalp injury and minor depressed skull fracture led to concern about a severe head injury, even though the patient was fully conscious. This distracted the clinician away from the major limb fractures. Only one of the four patients with major untreated extracranial injuries had a severe head injury (patient 3). In that patient, recurrent hypotension from an undiagnosed ruptured spleen was wrongly ascribed to a scalp wound. In all four cases, the presence of an actual or suspected head injury distracted the surgeons to such an extent that a series of inappropriate decisions were made, whereby treatment of life-threatening extracranial injuries was neglected in the rush to transfer to a neurosurgical centre. The frequency of untreated major extracranial injuries in this series (9%) is similar to that reported from Scotland.l'-I3 It appears that, despite warnings in the literature over the past 10 and the efforts of the Royal Australasian College of Surgeons through the EMST training programme to improve trauma care," the dangers of transferring such patients are either still not fully appreciated or are being ignored and that undue priority is still being given to head injury in the face of life-threatening extracranial injuries. Although definitive advice regarding the proper triage and transfer of head-injured patients is beyond the scope of this report, many of the problems revealed in this series might be overcome by improved communications between the referring and the receiving hospitals. As soon as it becomes apparent that transfer is likely, direct communication between senior doctors in the refemng hospital and the trauma centre should be established. This not only speeds the process of transfer, but allows direct advice to be given about aspects of management before and during transfer.'".21 If the referring hospital has CT scanning and intensive care facilities, early discussions with the neurosurgeon may avoid the unnecessary
162
HENDERSON ETAL.
transfer of head-injured patients who do not require neurosurgery. Such a system, which has proved effective in the UK, is particularly attractive in the context of patients with severe head and extracranial injuries.22
References I . SMITHR. F . , FRATSEDI L., SLOAN E. P. er a / . (1990) The impact of volume on outcome in seriously injured trauma patients: two years experience of the Chicago Trauma System. J . Trauma 30, 1066-76. 2. VON WAGONER F. H. (1961) Died in hospital: a three year survey of deaths following trauma. J . Trauma 1, 401-8. 3. WESTJ . G., TRUNKEY D. D. & LIM R. C. (1979) Systems of trauma care. Arch. Surg. 114, 455-60. 4. WESTJ . G . , CALES R. H . & GAZZANGIA A. B. (1983) Impact of regionalisation: the Orange County experience. Arch. Surg. 118, 740-4. 5 . TRUNKEY D. D. (1985) Toward optimal trauma care. Arch. Emerg. Med. 4, 181-95. R. (1984) Trauma mortality in Orange County: 6. CALES the effect of implementation of a regional trauma system. Ann. Emerg. Med. 13, 1-10. 7. GILLIGAN J . E., MCCLEAVE D., NICHOLSON B . et a / . (1977) Retrieval of the critically ill in South Australia: a coordinated approach. Med. J . Aust. 2, 849-55. 8. JENNETTB., MURRAY A., CARLIN J . , MCKEANM., MACMILLAN R. & S r R A N c I . (1979) Head injuries in three Scottish neurosurgical units. Br. Mrd. J . 2, 955-8. 9. TEASDA1.E G., GALBRAITH s., MURRAY L., WARD P., M. (1982) Management of G ~ ~ L ~D.M&AMCKEAN N traumatic intracranial haematoma. Br. Med. J . 285, 1695-7. S. & JENNET B. (1977) Avoidable 10. Rose J . , VALTONEN factors contributing to death after head injury. Br. Med. J . 2, 615-18.
1 I . GENTLEMAN D. & JENNETTB . (1981) Hazards of inter-
12.
13. 14. 15,
16. 17.
18.
19.
20. 21.
22.
hospital transfer of comatose head-injured patients. Lancet ii, 853-5. JEFFERIES R. V . & JONtS J . J . (1981) Avoidable factors contributing to death of head injury patients in general hospitals in the Mersey Region. Lancer ii, 459-6 I . GENTLEMAN D. & JENNETT B . (1990) Audit of transfer of unconscious head-injured patients to a neurosurgical unit. Lancet 335, 330-4. SELECKl B . R., BERRYG., DANN. G. et a / . (1986) Preventable causes of death and disability from neurotrauma. Ausr. N.Z. J . Surg. 56, 529-34. Guidelines for the initial management after head injury in adults. Suggestions from a group of neurosurgeons. (1984) Br. Med. J . 288, 983-5. TRUNKEY D. (1991) Initial treatment of patients with extensive trauma. N . Engl. J . Med. 324, 1259-63. D. P. & GARDENER G. (1985) Intensive manBECKER agement of head injury. In: Neurosurgery (Ed. R. H. Wilkins and S. S. Rengachary). McGraw-Hill Publishers, New York. GILDENBERG P. L. & MAKELA M. (1985) Effect O f early intubation and ventilation on outcome following head injury. In Trauma of the Nrrvous System (Ed. R. G. Dacy), pp. 79-90. Raven Press, New York. ROYAL AUSTRALASIAN COLLEGE OF SURGEONS (1989) Stabilization and transport. In: Early Manugenient of Severe Trauma; Provider Manual, Chapter 12. RACS, Melbourne. D ~ A NS.EA , , GRUNDY P. L., WOODSP. et a / . (1988) The management of injuries: a review of deaths in hospital. Aust. N.Z. J . Surg. 58, 463-9. AMERICAN COLLEGE OF SURGEONS COMMITTEE ON TRAUMA ( 1 987) Hospital and prehospital resources for optimal care of the injured patient. The College, Chicago. MILLER E. s., NEOFTOLEMOS J . P., AlTKENHhAD A. R. & FOSSARD D. P. (1985) Management of severe head injuries in a non-neurosurgical trauma centre. J . R . Coll. Surg. Edinb. 30, 82-7.
750
ORIGINAL ARTICLES A SURVEY OF INTERHOSPITAL TRANSFER OF HEAD-INJURED PATIENTS WITH INADEQUATELY TREATED LIFE-THREATENING EXTRACRANIAL INJURIES A. HENDERSON,* T. COYNE,’ D. WALL+AND B. MILLER’ Departments of *Intensive Care and ‘Surgery, Princess Alexandra Hospital, Ipswich Road, Brisbane, Queensland, Australia A 12 month prospective study was undertaken to determine the frequency of untreated life-threatening extracranial injuries in patients transferred to a major trauma centre because of head injury. Of the 43 patients transferred (15 with isolated head injury and 28 with multiple injuries), four (9%) had an untreated lifethreatening extracranial injury, which caused death in two. All four patients with untreated extracranial injuries were transferred from hospitals with general surgical staff and facilities. In three of the patients (none with a major head injury), the extracranial injuries were recognized at the referring hospital, but were left untreated in the rush to transfer the patient to a neurosurgical facility. In the fourth patient, who had a severe head injury, recurrent hypotension from a ruptured spleen was mistakenly ascribed to a scalp wound. The series shows that the dangerous practice of hurriedly transferring patients to trauma centres because of actual or perceived head injuries, while leaving major extracranial injuries untreated, continues despite warnings in the literature and the efforts of the Royal Australasian College of Surgeons through the Early Management of Severe Trauma programme.
Key words: extracranial injuries, head injury, interhospital transfer, untreated.
Introduction The geography and demography of Australia make interhospital transfer of patients an integral part of health care. Although interhospital transfer is not without risk, early transfer of severely injured patients to major trauma centres confers a survival advantage and has become an important part of hospital practice. I-’ Patients with severe head injuries are especially likely to be urgently transferred, as it has been shown that early computerized tomography (CT) scanning of the head with expeditious evacuation of intracranial haematomas, reduces mortality and morbid it^.*,^ In multiply injured patients admitted to hospitals without CT scanning facilities, the occurrence of a head injury may distract the clinician to the extent that serious extracranial injuries may go undetected or untreated while the patient is urgently transferred to a trauma centre. Uncontrolled haemorrhage during transfer may cause severe hypotension in these patients and this is known to adversely affect the prognosis of the head injury. “ - I 2 Surveys from the UK indicate that between 5 and
”-’*
Correspondence: Dr A. Henderson, Intensive Care Unit, Princess Alexandra Hospital, Ipswich Road, Brisbane, Qld 4102, Australia. Accepted for publication 26 March 1992
9% of head-injured patients transferred to a neurosurgical centre had inadequately treated extracranial injuries. l 3 In Australia, deficiencies in cardiorespiratory management were identified in 7% of headinjured patients transferred to major centres from rural areas. l 4 The Royal Australasian College of Surgeons has evolved an education programme (Early Management of Severe Trauma, EMST) in an attempt to improve the overall care of trauma victims. This programme gives specific advice on matters relating to patient transfer. I s A survey was undertaken of all patients transferred to the intensive care unit because of head injuries. to determine the frequency with which head-injured patients arrived with untreated, life-threatening extracranial injuries.
Methods The Princess Alexandra Hospital (PA hospital) is a 1000 bed adult teaching hospital in Brisbane. The hospital serves as a major trauma centre for a large area of Southern Queensland and Northern New South Wales. Seriously injured patients are transferred to the intensive care unit (ICU) by road or air from numerous smaller hospitals within the geographical catchment area. A prospective computerized database was kept for all patients transferred
HENDERSON E T A L .
760
from outside Brisbane because of head injury using Dbase IV (Ashton Tate). This report is based upon a study of all head injury patients transferred to the ICU for the 12 months ending 30 June 1991. In every case, the prospective database was verified by inspection of the hospital notes. Age, sex, outcome and the severity and extent of extracranial injuries were noted. A record of any patient with untreated life-threatening extracranial injuries was kept and a detailed review of their case notes was undertaken.
Results During the study period, 43 head-injured patients with a mean age of 34 years (range: 14-65 years) were transferred to the intensive care unit of the PA hospital from outside the Brisbane metropolitan area. There was a preponderance of males (M:F = 4 : 1). Fifteen (35%) had isolated head injuries while 28 (65%) had head and extracranial injuries. Eight (18.6%) died in ICU (three with isolated head injuries and five with multiple injuries). Four patients (9%) were identified as having life-threatening undiagnosed or untreated extracranial injuries. Their case histories were reviewed in detail.
Case histories
watershed infarcts consistent with profound hypotension but no evidence of intracranial haematoma. She died 2 days later. Autopsy confirmed the CT findings and revealed no further pathology within thel abdomen. PATIENT 2
A man of 29 was injured when his motor cycle struck a car. On arrival at a large urban hospital he had a systolic blood pressure of 70 mmHg and GCS of 1 1 . His injuries were a minor facial laceration, blood in the left external auditory meatus, bilateral closed radial fractures and a chest injury comprising a small left lower flail segment and pneumothorax. The patient was intubated and drug paralysed on admission. He was stabilized by the infusion of 6.5 L of colloid and crystalloid and drainage of the pneumothorax. A ruptured spleen was suspected, but a local decision was made to immediately transfer the patient to PA hospital because of a suspected basal skull fracture. He became hypotensive during the 1 h journey by road. Peritoneal lavage on arrival at PA hospital was positive. At laparotomy he had haemoperitoneum from a ruptured spleen. A subsequent CT scan of head was normal. The source of the blood in the external auditory meatus was shown to be from the nearby scalp laceration rather than from a basal skull fracture. He made a complete recovery.
PATIENT 1 PATIENT 3
A woman of 18 was injured when her car was struck by a lorry. She sustained a closed head injury with a minor scalp laceration. fractured pelvis and pneumothorax. She was admitted to a large urban hospital and had a systolic blood pressure of 60 mmHg and a Glasgow Coma Scale (GCS) of 8. The patient was intubated and ventilated using pancuronium muscle relaxation. Following resuscitation with 1 L of haemaccell and 1 L of Hartmann’s solution and drainage of the pneumothorax her systolic blood pressure rose to 121 mmHg. The systolic blood pressure fell to 70 mmHg during the following 30 min. Peritoneal lavage was positive and an ultrasound showed a ruptured spleen. She was eventually stabilized with 5 u haemaccell, 1 L normal saline and 3 u packed cells. It was decided locally that priority should be given to the head injury and that the patient required immediate transfer to PA hospital where her cranial and abdominal injuries could be treated together. During the hour long journey by road she became profoundly hypotensive and was asystolic on arrival. A laparotomy was carried out after resuscitation which revealed massive intraperitoneal bleeding from rupture of the liver and spleen. Haemostasis was rapidly secured and she was transferred to ICU. A subsequent CT scan of her head showed several
A man of 41 was injured when he was thrown from a car. He sustained a severe compound head injury. On arrival at a rural hospital he had a GCS of 5 and a systolic blood pressure of 75 mmHg. He was intubated, transfused 2.5 L of colloid and crystalloid and the scalp wound was sutured. He was then transferred, ventilated and drug paralysed, by road to the base hospital. On arrival he was pale and hypotensive (systolic blood pressure = 60 mmHg). Fbllowing stabilization with a further transfusion of crystalloid and 4 u of packed cells a CT head scan was done with showed a compound skull fracture and extensive frontal lobe contusion. In view of the serious head injury, the patient was flown to Brisbane, a journey of 2 h and 15 min, without further ifivestigation of the hypotension, which was ascribed to the scalp wound. On amval at PA hospital he was again hypotensive. Peritoneal lavage was positive and laparatomy revealed a ruptured spleen. He survived with a severe neurological deficit. PATIENT 4
A man of 20 was injured in a motor vehicle accident. He sustained a small compound depressed skull fracture, a large scalp wound and closed fractures of the right tibia and femur. On arrival at a
TRANSFER OF HEAD-INJURED PATIENTS
large rural hospital his systolic blood pressure was 85 mmHg and the GCS was 15. After transfusion of 2 L of colloid and crystalloid plus 2 u of packed cells, the systolic blood pressure rose to 130mmHg. Because of the head injury, the patient was immediately flown to the PA hospital, ajourney of 50min. On arrival his fractures were unsupported. Although the systolic blood pressure was 125 mmHg, the patient was cyanosed, intensely vasoconstricted with a heart rate of 140 beatshin and a respiratory rate of 40 beatdmin. Chest X-ray revealed diffuse alveolar shadowing typical of fulminant fat embolism. Arterial gases on air confirmed severe oxygenation failure (Pa02 =37 mmHg, PaCOz = 34mmHg). After resuscitation, including intubation and ventilation, the infusion of 1 L of haemaccell and 2 L of normal saline and splinting of the fractures, he had a CT scan of head which revealed a minor depressed skull fracture without evidence of further intracranial injury. He developed pulmonary hypertension (pulmonary artery systolic pressure =70 mmHg) with refractory hypoxia and died 7 h after admission. Autopsy confirmed massive fat embolism but showed no evidence of significant intracranial injury.
Discussion Patients with actual or suspected major head injuries admitted to hospitals without immediate access to a CT scanner present a major clinical problem because it has been shown that those who are unconscious, or who have impaired consciousness in the presence of a skull fracture, have a 25% incidence of intracranial haematoma. l5The pressure to urgently transfer such patients to a neurosurgical unit is high because early CT scanning with expeditious drainage of intracranial haematomas has been shown to improve progno~is.'.~In patients with isolated head injury, hypoxia is the major transferrelated complication. "-I2 This serious complication can be avoided by intubation and ventilation for the journey. Patients with major head and extracranial injuries present a particular problem. In the present series, 65% of the referrals with head injuries had major extracranial injuries. This figure is similar to that reported from a neurosurgical unit in the West of Scotland.".13 In such patients, the presence of a head injury may distract the clinician from other injuries, or it may dominate the clinical picture to the extent that the head injury receives an inappropriately high treatment priority. ''.I3 The severity of the head injury can be difficult to assess, as factors other than brain injury, such as alcohol, hypoxia and hypotension, may depress the level of consciousness mimicking head injury or exaggeration of the severity of a head injury." While rapid resuscitation may improve cerebral func-
76 I
tion in patients with neurological deficits due to hypoxia or hypotension, such improvements may be masked by the use of drugs to aid mechanical ventilation, especially muscle relaxants. ' 7 , 1 8 In this series, the severity of the head injury was overestimated in three of the patients, two of whom later died. In patient 1, hypotension from intra-abdominal haemorrhage was the probable cause of the reduced level of consciousness. The neurological response to resuscitation was masked by the use of pancuronium to aid mechanical ventilation. With patient 2, the appearance of blood in the external auditory meatus coupled with a scalp laceration and reduced level of consciousness suggested a basal skull fracture. Again the early use of muscle relaxants prevented ongoing assessment of neurological status. Further review revealed no evidence of a major head injury. In patient 4 the dramatic scalp injury and minor depressed skull fracture led to concern about a severe head injury, even though the patient was fully conscious. This distracted the clinician away from the major limb fractures. Only one of the four patients with major untreated extracranial injuries had a severe head injury (patient 3). In that patient, recurrent hypotension from an undiagnosed ruptured spleen was wrongly ascribed to a scalp wound. In all four cases, the presence of an actual or suspected head injury distracted the surgeons to such an extent that a series of inappropriate decisions were made, whereby treatment of life-threatening extracranial injuries was neglected in the rush to transfer to a neurosurgical centre. The frequency of untreated major extracranial injuries in this series (9%) is similar to that reported from Scotland.l'-I3 It appears that, despite warnings in the literature over the past 10 and the efforts of the Royal Australasian College of Surgeons through the EMST training programme to improve trauma care," the dangers of transferring such patients are either still not fully appreciated or are being ignored and that undue priority is still being given to head injury in the face of life-threatening extracranial injuries. Although definitive advice regarding the proper triage and transfer of head-injured patients is beyond the scope of this report, many of the problems revealed in this series might be overcome by improved communications between the referring and the receiving hospitals. As soon as it becomes apparent that transfer is likely, direct communication between senior doctors in the refemng hospital and the trauma centre should be established. This not only speeds the process of transfer, but allows direct advice to be given about aspects of management before and during transfer.'".21 If the referring hospital has CT scanning and intensive care facilities, early discussions with the neurosurgeon may avoid the unnecessary
162
HENDERSON ETAL.
transfer of head-injured patients who do not require neurosurgery. Such a system, which has proved effective in the UK, is particularly attractive in the context of patients with severe head and extracranial injuries.22
References I . SMITHR. F . , FRATSEDI L., SLOAN E. P. er a / . (1990) The impact of volume on outcome in seriously injured trauma patients: two years experience of the Chicago Trauma System. J . Trauma 30, 1066-76. 2. VON WAGONER F. H. (1961) Died in hospital: a three year survey of deaths following trauma. J . Trauma 1, 401-8. 3. WESTJ . G., TRUNKEY D. D. & LIM R. C. (1979) Systems of trauma care. Arch. Surg. 114, 455-60. 4. WESTJ . G . , CALES R. H . & GAZZANGIA A. B. (1983) Impact of regionalisation: the Orange County experience. Arch. Surg. 118, 740-4. 5 . TRUNKEY D. D. (1985) Toward optimal trauma care. Arch. Emerg. Med. 4, 181-95. R. (1984) Trauma mortality in Orange County: 6. CALES the effect of implementation of a regional trauma system. Ann. Emerg. Med. 13, 1-10. 7. GILLIGAN J . E., MCCLEAVE D., NICHOLSON B . et a / . (1977) Retrieval of the critically ill in South Australia: a coordinated approach. Med. J . Aust. 2, 849-55. 8. JENNETTB., MURRAY A., CARLIN J . , MCKEANM., MACMILLAN R. & S r R A N c I . (1979) Head injuries in three Scottish neurosurgical units. Br. Mrd. J . 2, 955-8. 9. TEASDA1.E G., GALBRAITH s., MURRAY L., WARD P., M. (1982) Management of G ~ ~ L ~D.M&AMCKEAN N traumatic intracranial haematoma. Br. Med. J . 285, 1695-7. S. & JENNET B. (1977) Avoidable 10. Rose J . , VALTONEN factors contributing to death after head injury. Br. Med. J . 2, 615-18.
1 I . GENTLEMAN D. & JENNETTB . (1981) Hazards of inter-
12.
13. 14. 15,
16. 17.
18.
19.
20. 21.
22.
hospital transfer of comatose head-injured patients. Lancet ii, 853-5. JEFFERIES R. V . & JONtS J . J . (1981) Avoidable factors contributing to death of head injury patients in general hospitals in the Mersey Region. Lancer ii, 459-6 I . GENTLEMAN D. & JENNETT B . (1990) Audit of transfer of unconscious head-injured patients to a neurosurgical unit. Lancet 335, 330-4. SELECKl B . R., BERRYG., DANN. G. et a / . (1986) Preventable causes of death and disability from neurotrauma. Ausr. N.Z. J . Surg. 56, 529-34. Guidelines for the initial management after head injury in adults. Suggestions from a group of neurosurgeons. (1984) Br. Med. J . 288, 983-5. TRUNKEY D. (1991) Initial treatment of patients with extensive trauma. N . Engl. J . Med. 324, 1259-63. D. P. & GARDENER G. (1985) Intensive manBECKER agement of head injury. In: Neurosurgery (Ed. R. H. Wilkins and S. S. Rengachary). McGraw-Hill Publishers, New York. GILDENBERG P. L. & MAKELA M. (1985) Effect O f early intubation and ventilation on outcome following head injury. In Trauma of the Nrrvous System (Ed. R. G. Dacy), pp. 79-90. Raven Press, New York. ROYAL AUSTRALASIAN COLLEGE OF SURGEONS (1989) Stabilization and transport. In: Early Manugenient of Severe Trauma; Provider Manual, Chapter 12. RACS, Melbourne. D ~ A NS.EA , , GRUNDY P. L., WOODSP. et a / . (1988) The management of injuries: a review of deaths in hospital. Aust. N.Z. J . Surg. 58, 463-9. AMERICAN COLLEGE OF SURGEONS COMMITTEE ON TRAUMA ( 1 987) Hospital and prehospital resources for optimal care of the injured patient. The College, Chicago. MILLER E. s., NEOFTOLEMOS J . P., AlTKENHhAD A. R. & FOSSARD D. P. (1985) Management of severe head injuries in a non-neurosurgical trauma centre. J . R . Coll. Surg. Edinb. 30, 82-7.
