BRAIN INJURY,
1992, VOL. 6, NO. 1,53-58
Peripheral nerve injuries in children with traumatic brain injury PULIYODIL A . PHILIP and M E R S A M M A PHILIP
Brain Inj Downloaded from informahealthcare.com by Chinese University of Hong Kong on 12/26/14 For personal use only.
Northwestern University M e d d School and Rehabilitation Institute of Chicago, Chicago, Illinois 60611, USA (Received 20 November 1990; accepted 20 January 1991)
The relationship between traumatic brain injuries and associated peripheral nerve injuries has not been previously studied in children in the rehabilitation setting. One hundred and fifty-seven consecutive admissions of brain-injured children were included in this prospective study to determine the frequency and pattern of recovery of peripheral nerve injuries in such patients. Twelve patients met the clinical diagnostic criteria for peripheral nerve injury. Electrodiagnostic studies confirmed the clinical diagnosis in 11 (7%)patients. Grading of spontaneous activity and electrodiagnostic study of the contralateral limb were used to mtferentiate electromyographic abnormalities due to upper motor neuron lesion and nerve injury. Electrodiagnostic study was useful to confirm clinical diagnosis and predict prognosis of nerve injuries. Two patients developed preventable pressure neuropathy. Eight patients showed moderate or full recovery in 9-10 months post-injury. AU children with severe brain injuries should be evaluated for concomitant peripheral nerve injuries.
Introduction Multiple medical and musculoskeletal complications can occur in patients with brain injury [l]. A few severely brain-injured chldren sustain associated spinal cord or brachial plexus injuries [2]. Hoffer and associates documented the incidence of these complications as 4% and 2%, respectively [3]. Plexus and root avulsions, commonly seen in motorcycle accidents, usually result in flad extremity [4].Force sufficient to produce severe brain injuries in children could also produce peripheral nerve injuries. Direct external pressure over the nerve or pressure due to heterotrophic ossification or haematoma could lead to peripheral nerve lesions during the post-acute phase. Prolonged coma and severe spasticity are usually associated with such lesions [5]. There is a paucity of information in the literature regarding the types of peripheral nerve injuries, recovery and management of such lesions in brain-injured children. The purpose of thls prospective study was to document the frequency and pattern of recovery of peripheral nerve injuries in chldren with traumatic brain injury.
Methods One hundred and fifty-seven consecutive admissions, with the diagnosis of traumatic brain injury, to a paedatric rehabilitation unit in a rehabilitation hospital were studied. Age, sex, causes of head injury, length of stay in acute care and rehabilitation hospitals and time Address correspondence to: Dr Puliyodil A. Philip, Rehabilitation Institute of Chicago, 345 E. Superior Street, Chicago, IL 60611, USA. 0269-9052/92 $3.00 0 1992 Taylor & Francis Ltd.
P. A . Philip and M . Philip
Brain Inj Downloaded from informahealthcare.com by Chinese University of Hong Kong on 12/26/14 For personal use only.
54
elapsed from injury to definitive diagnosis of peripheral nerve injury were documented. Patients whose loss of consciousness was 6 h or more, and Glasgow Coma Scale 5 or less, were categorized as having sustained severe head injury. Possible causes of nerve injury such as fracture, dislocation, haematoma, heterotrophic ossification, pressure and direct trauma were also documented. The clinical diagnosis of peripheral nerve injury was based on the following criteria: (a) decreased muscle tone in the distribution of a nerve or plexus; (b) hyporeflexia or areflexia; (c) varying patterns of motor recovery. Muscle atrophy was considered an additional objective findlng, but was not used in all cases due to the short period of time elapsed since the initial injury in some patients. Electrodiagnostic studies including electromyography (EMG) and nerve conduction studies (NCS) were performed on all patients who met the clinical criteria for the diagnosis and were done on all such children within 7 days of the diagnosis of nerve injury. Positive waves and fibrillation potentials can be observed in upper motor neuron lesions [6, 71. In order to avoid this possible confounding factor, EMG studies were performed on muscles supplied by the injured nerve, and muscles of the ipsilateral and contralateral limbs. Muscles supplied by the nerve counterpart in the contralateral limb were specifically tested for comparison. EMG abnormalities were quantified as 0 = none; I + = occasional; 2+ = few; 3+ = moderate and 4+ = extensive. Nerve conduction studies were performed bilaterally. Electrodiagnostic studies were used to confirm the clinical diagnosis and evaluate recovery. The studies were repeated on all patients a t follow-up 6 months after discharge. Recovery of nerve injury was documented as a reflection of functional status of the muscle groups supplied by the injured nerve and such evaluations were performed at admission, discharge and follow-up. Recovery was classified as 3 = full, 2 = moderate, 1 = minimal or 0 = n o recovery. Full recovery was defined as complete return of motor function, moderate recovery as incomplete return requiring no orthotic treatment, minimal recovery as return of function but requiring orthosis and no recovery as absence of any return of motor function. Patient records from acute care hospitals were reviewed to determine whether nerve injury occurred at the time of head injury or as a complication during the post-trauma period. Consciousness level of all patients with nerve injury was determined at admission to the rehabilitation facility using a classification [2] modified from Brink and associates [8] and Onlmaya [9]. This includes five categories: (I) oriented to time and place, end of post-traumatic amnesia; (11) responsive to environment, end of coma; (111) localized response; (IV) generalized response; and (V) no response. Results
One hundred and fifty-seven children admitted as inpatients to a paediatric rehabilitation unit, with the diagnosis of traumatic brain injury, were included in this study. Twelve were clinically diagnosed as having sustained peripheral nerve injuries and 11 showed abnormal EMG findings. Age ranged from 5 to 17 years, with a mean of 12 years. There were 7 males and 4 females. All were involved in motor vehicle accidents as pedestrians. Acute care hospitalization ranged &om 14 to 46 days, with a mean of 29 days. Average rehabilitation stay was 84 days, with a range of 26-134 days. Ten patients suffered severe brain injuries and average length of coma was 21 days. In addition, 10 patients scored 5 or below in the Glasgow Coma Scale and 1 patient scored 8. Level of consciousness at admission ranged from I to 111.
Brain Inj Downloaded from informahealthcare.com by Chinese University of Hong Kong on 12/26/14 For personal use only.
Nerve injuries in brain-injured children
55
Clinical diagnosis of peripheral nerve injury was made in 12 out of 157 brain-injured children involving a total of 14 nerves. Eleven (7%) patients showed abnormal EMG involving 13 nerves. Clinical diagnosis was made between 18 and 57 days (mean 34 days) following the initial trauma. Presence of nerve injury was detected within 1 week of rehabilitation admission. Eleven nerve injuries occurred due to trauma sustained at the initial impact, and 2 were due to pressure neuropathy during the acute care hospital stay. All patients with the clinical diagnosis of peripheral nerve injury underwent electrodiagnostic studies between 25 and 64 days following injury or within 7 days of rehabilitation admission. Eleven patients showed abnormal spontaneous muscle fibre discharges (positive waves and fibrillation potentials) in the muscles supplied by the nerve or plexus in question. EMG and NCS studm were repeated in 5 patients due to poor patient co-operation at the initial study. Thirteen peripheral nerve or plexus injuries were confirmed in 11 patients using electrodiagnostic studies. Varying degrees of decrease in amplitude of evoked compound muscle action potentials (CMAP) were noted in NCS. Stimulation of nerves which supplied the muscles with absent voluntary motor unit action potentials (MUAP) failed to elicit CMAP. Two such patients suffered complete nerve injuries involving one tibial and one peroneal nerve with no MUAP. Table 1 shows the EMG findings. There were occasional or few abnormahties in the ipsilateral or contralateral limbs in 6 patients. The abnormahties in the muscle groups supplied by the injured nerves were at least 2 grades higher. AU patients who showed MUAP and CMAP in the initial study showed improvement in function at discharge and follow-up, with decrease in EMG abnormahties at follow-up. Six (55%) patients had associated spontaneous activity in ‘unaffected’ muscle groups of ipsilateral and contralateral limbs at the initial study, whereas only 4 (36%) showed such findings at follow-up. Table 1 shows the nerves injured, causes of nerve injury and pattern of recovery. At initial evaluation 2 nerves showed moderate, 9 minimal and 2 no recovery. Moderate recovery was noted in all patients whose injuries were due to haematomas. At discharge, 2 nerves fully recovered, 3 showed moderate, 6 minimal and 2 no recovery. When patients were re-evaluated at 6 month after discharge, 5 nerves showed full recovery, 4 moderate, 2 minimal and 2 no recovery. The nerves which did not show any recovery were those with extensive EMG abnormahties, absent MUAP and absent CMAP in the initial electrodiagnostic study. All patients with minimal or no recovery required orthosis for positioning of the affected limb or functional activities. Patients with tibial or peroneal nerve injuries received ankle-foot orthosis (AFO) and the patient with radial nerve injury was given wrist-hand orthosis with extension assist. At discharge, all except 2 patients required the use of orthosis. O f the 10 patients with nerve injuries in the lower limbs, 2 were fully or partially ambulatory without orthosis, 4 with orthosis and 4 were non-ambulatory due to severe generalized motor deficits at the time of dlscharge. At follow-up, 3 patients required orthosis and 5 required no orthosis for partial or full ambulation. Three patients remained non-ambulatory in spite of full nerve recovery in 2 patients. The level of functional ambulation and recovery of nerve injuries did not match due to associated multiple functional deficits in these patients. AU patients with haematoma as a cause of nerve injury continued to show improvement at discharge, and by follow-up there was b l l functional recovery. Of the 4 patients who showed no recovery at follow-up, 3 sustained tibial nerve injuries due to fractures and 1 suffered peroneal nerve injury due to pressure.
Sciatic Sciatic Tibial
Tibial I'croncal Tibial Tibial Peroncal Pcroneal I'eroncal Brachial plcxur Radial
11 9 12
x
I
1992, VOL. 6, NO. 1,53-58
Peripheral nerve injuries in children with traumatic brain injury PULIYODIL A . PHILIP and M E R S A M M A PHILIP
Brain Inj Downloaded from informahealthcare.com by Chinese University of Hong Kong on 12/26/14 For personal use only.
Northwestern University M e d d School and Rehabilitation Institute of Chicago, Chicago, Illinois 60611, USA (Received 20 November 1990; accepted 20 January 1991)
The relationship between traumatic brain injuries and associated peripheral nerve injuries has not been previously studied in children in the rehabilitation setting. One hundred and fifty-seven consecutive admissions of brain-injured children were included in this prospective study to determine the frequency and pattern of recovery of peripheral nerve injuries in such patients. Twelve patients met the clinical diagnostic criteria for peripheral nerve injury. Electrodiagnostic studies confirmed the clinical diagnosis in 11 (7%)patients. Grading of spontaneous activity and electrodiagnostic study of the contralateral limb were used to mtferentiate electromyographic abnormalities due to upper motor neuron lesion and nerve injury. Electrodiagnostic study was useful to confirm clinical diagnosis and predict prognosis of nerve injuries. Two patients developed preventable pressure neuropathy. Eight patients showed moderate or full recovery in 9-10 months post-injury. AU children with severe brain injuries should be evaluated for concomitant peripheral nerve injuries.
Introduction Multiple medical and musculoskeletal complications can occur in patients with brain injury [l]. A few severely brain-injured chldren sustain associated spinal cord or brachial plexus injuries [2]. Hoffer and associates documented the incidence of these complications as 4% and 2%, respectively [3]. Plexus and root avulsions, commonly seen in motorcycle accidents, usually result in flad extremity [4].Force sufficient to produce severe brain injuries in children could also produce peripheral nerve injuries. Direct external pressure over the nerve or pressure due to heterotrophic ossification or haematoma could lead to peripheral nerve lesions during the post-acute phase. Prolonged coma and severe spasticity are usually associated with such lesions [5]. There is a paucity of information in the literature regarding the types of peripheral nerve injuries, recovery and management of such lesions in brain-injured children. The purpose of thls prospective study was to document the frequency and pattern of recovery of peripheral nerve injuries in chldren with traumatic brain injury.
Methods One hundred and fifty-seven consecutive admissions, with the diagnosis of traumatic brain injury, to a paedatric rehabilitation unit in a rehabilitation hospital were studied. Age, sex, causes of head injury, length of stay in acute care and rehabilitation hospitals and time Address correspondence to: Dr Puliyodil A. Philip, Rehabilitation Institute of Chicago, 345 E. Superior Street, Chicago, IL 60611, USA. 0269-9052/92 $3.00 0 1992 Taylor & Francis Ltd.
P. A . Philip and M . Philip
Brain Inj Downloaded from informahealthcare.com by Chinese University of Hong Kong on 12/26/14 For personal use only.
54
elapsed from injury to definitive diagnosis of peripheral nerve injury were documented. Patients whose loss of consciousness was 6 h or more, and Glasgow Coma Scale 5 or less, were categorized as having sustained severe head injury. Possible causes of nerve injury such as fracture, dislocation, haematoma, heterotrophic ossification, pressure and direct trauma were also documented. The clinical diagnosis of peripheral nerve injury was based on the following criteria: (a) decreased muscle tone in the distribution of a nerve or plexus; (b) hyporeflexia or areflexia; (c) varying patterns of motor recovery. Muscle atrophy was considered an additional objective findlng, but was not used in all cases due to the short period of time elapsed since the initial injury in some patients. Electrodiagnostic studies including electromyography (EMG) and nerve conduction studies (NCS) were performed on all patients who met the clinical criteria for the diagnosis and were done on all such children within 7 days of the diagnosis of nerve injury. Positive waves and fibrillation potentials can be observed in upper motor neuron lesions [6, 71. In order to avoid this possible confounding factor, EMG studies were performed on muscles supplied by the injured nerve, and muscles of the ipsilateral and contralateral limbs. Muscles supplied by the nerve counterpart in the contralateral limb were specifically tested for comparison. EMG abnormalities were quantified as 0 = none; I + = occasional; 2+ = few; 3+ = moderate and 4+ = extensive. Nerve conduction studies were performed bilaterally. Electrodiagnostic studies were used to confirm the clinical diagnosis and evaluate recovery. The studies were repeated on all patients a t follow-up 6 months after discharge. Recovery of nerve injury was documented as a reflection of functional status of the muscle groups supplied by the injured nerve and such evaluations were performed at admission, discharge and follow-up. Recovery was classified as 3 = full, 2 = moderate, 1 = minimal or 0 = n o recovery. Full recovery was defined as complete return of motor function, moderate recovery as incomplete return requiring no orthotic treatment, minimal recovery as return of function but requiring orthosis and no recovery as absence of any return of motor function. Patient records from acute care hospitals were reviewed to determine whether nerve injury occurred at the time of head injury or as a complication during the post-trauma period. Consciousness level of all patients with nerve injury was determined at admission to the rehabilitation facility using a classification [2] modified from Brink and associates [8] and Onlmaya [9]. This includes five categories: (I) oriented to time and place, end of post-traumatic amnesia; (11) responsive to environment, end of coma; (111) localized response; (IV) generalized response; and (V) no response. Results
One hundred and fifty-seven children admitted as inpatients to a paediatric rehabilitation unit, with the diagnosis of traumatic brain injury, were included in this study. Twelve were clinically diagnosed as having sustained peripheral nerve injuries and 11 showed abnormal EMG findings. Age ranged from 5 to 17 years, with a mean of 12 years. There were 7 males and 4 females. All were involved in motor vehicle accidents as pedestrians. Acute care hospitalization ranged &om 14 to 46 days, with a mean of 29 days. Average rehabilitation stay was 84 days, with a range of 26-134 days. Ten patients suffered severe brain injuries and average length of coma was 21 days. In addition, 10 patients scored 5 or below in the Glasgow Coma Scale and 1 patient scored 8. Level of consciousness at admission ranged from I to 111.
Brain Inj Downloaded from informahealthcare.com by Chinese University of Hong Kong on 12/26/14 For personal use only.
Nerve injuries in brain-injured children
55
Clinical diagnosis of peripheral nerve injury was made in 12 out of 157 brain-injured children involving a total of 14 nerves. Eleven (7%) patients showed abnormal EMG involving 13 nerves. Clinical diagnosis was made between 18 and 57 days (mean 34 days) following the initial trauma. Presence of nerve injury was detected within 1 week of rehabilitation admission. Eleven nerve injuries occurred due to trauma sustained at the initial impact, and 2 were due to pressure neuropathy during the acute care hospital stay. All patients with the clinical diagnosis of peripheral nerve injury underwent electrodiagnostic studies between 25 and 64 days following injury or within 7 days of rehabilitation admission. Eleven patients showed abnormal spontaneous muscle fibre discharges (positive waves and fibrillation potentials) in the muscles supplied by the nerve or plexus in question. EMG and NCS studm were repeated in 5 patients due to poor patient co-operation at the initial study. Thirteen peripheral nerve or plexus injuries were confirmed in 11 patients using electrodiagnostic studies. Varying degrees of decrease in amplitude of evoked compound muscle action potentials (CMAP) were noted in NCS. Stimulation of nerves which supplied the muscles with absent voluntary motor unit action potentials (MUAP) failed to elicit CMAP. Two such patients suffered complete nerve injuries involving one tibial and one peroneal nerve with no MUAP. Table 1 shows the EMG findings. There were occasional or few abnormahties in the ipsilateral or contralateral limbs in 6 patients. The abnormahties in the muscle groups supplied by the injured nerves were at least 2 grades higher. AU patients who showed MUAP and CMAP in the initial study showed improvement in function at discharge and follow-up, with decrease in EMG abnormahties at follow-up. Six (55%) patients had associated spontaneous activity in ‘unaffected’ muscle groups of ipsilateral and contralateral limbs at the initial study, whereas only 4 (36%) showed such findings at follow-up. Table 1 shows the nerves injured, causes of nerve injury and pattern of recovery. At initial evaluation 2 nerves showed moderate, 9 minimal and 2 no recovery. Moderate recovery was noted in all patients whose injuries were due to haematomas. At discharge, 2 nerves fully recovered, 3 showed moderate, 6 minimal and 2 no recovery. When patients were re-evaluated at 6 month after discharge, 5 nerves showed full recovery, 4 moderate, 2 minimal and 2 no recovery. The nerves which did not show any recovery were those with extensive EMG abnormahties, absent MUAP and absent CMAP in the initial electrodiagnostic study. All patients with minimal or no recovery required orthosis for positioning of the affected limb or functional activities. Patients with tibial or peroneal nerve injuries received ankle-foot orthosis (AFO) and the patient with radial nerve injury was given wrist-hand orthosis with extension assist. At discharge, all except 2 patients required the use of orthosis. O f the 10 patients with nerve injuries in the lower limbs, 2 were fully or partially ambulatory without orthosis, 4 with orthosis and 4 were non-ambulatory due to severe generalized motor deficits at the time of dlscharge. At follow-up, 3 patients required orthosis and 5 required no orthosis for partial or full ambulation. Three patients remained non-ambulatory in spite of full nerve recovery in 2 patients. The level of functional ambulation and recovery of nerve injuries did not match due to associated multiple functional deficits in these patients. AU patients with haematoma as a cause of nerve injury continued to show improvement at discharge, and by follow-up there was b l l functional recovery. Of the 4 patients who showed no recovery at follow-up, 3 sustained tibial nerve injuries due to fractures and 1 suffered peroneal nerve injury due to pressure.
Sciatic Sciatic Tibial
Tibial I'croncal Tibial Tibial Peroncal Pcroneal I'eroncal Brachial plcxur Radial
11 9 12
x
I
