Missile injuries of the spine
Between 1995 and 2000, 22 cases with low velocity missile injuries of the spine and spinal cord were treated in three service hospitals. All were adult males, with a mean age of 30.7 years. The wounds were caused by splinters in 18 (82%) and bullets in 4 (18%). Twelve patients received more than one splinter. The cervical and thoracic spines were most frequently involved. In 7 cases, there were injuries to other organs. There was extensive initial deficit (quadriplegia, paraplegia) in 18 (82%) cases, while 4 (18%) had partial deficits. The patients were evaluated by spine radiographs. Myelography was done in 4, CT myelography in 11 and MRI in 4 patients. Two patients had intramedullary hematoma without any skeletal injury, and were treated conservatively. Seventeen patients were treated operatively, and associated injuries of other organs received priority management. Surgery was in the form of debridement, exploration of the spinal cord, hemostasis, decompression and dural repair. Steroids and antibiotics were given routinely. Three patients (2 with cervical and 1 with thoracic spine injury) died preoperatively, and 1 (with dorsolumbar injury) died in the postoperative period due to multi-organ injury. Patients with complete injury remained completely paralyzed, while those with an incomplete injury showed improvement in their neurological grades. The initial neurological grade is the best prognostic indicator, and these injuries are often accompanied by multi-organ injuries. There was no instance of postoperative meningitis or CSF leak. These injuries should be explored for debridement and dural repair.
Penetrating wounds of the spine are the third most common cause of spinal cord injury (SCI). So far, there has been no data on missile injuries of the spine presented from India. There is a continuing debate on the management, with views ranging from conservatism to aggressive debridement. The Review of War Surgery and Medicine, prepared by Harvey Cushing and published in 1918 advised that a transected spinal cord should never be operated upon unless the lesion was below L1. Each war thereafter led to further refinements in the management of these injuries. In Vietnam, aggressive and meticulous surgical principles were applied which included debridement, dural exploration and repair, which resulted in improvement in mortality. It is thus important that the treating surgeon understands the exact role and scope of surgical intervention, so that neurological function can be preserved and further deterioration recognized and treated. We share our experiences in the management of 22 cases of low velocity missile injuries of the spine and spinal cord. The relevant literature is reviewed briefly.
Twenty-two patients with low velocity missile injuries to the spine were managed in three service hospitals between 1995 and 2001. Only 1 of these was a civilian gunshot wound sustained by a person from the service personnel in a village during a dacoity, while the rest sustained injury in military action. Four patients sustained injury due to a bullet (7.62 mm), and the rest had splinter injuries (exploding shrapnel from shells, grenades, improvised explosive devices, pellets from handguns, etc.).
Clinical Evaluation: All patients underwent a complete clinical and hemodynamic evaluation on arrival, and were neurologically examined once they were hemodynamically stable and the airway was ensured. All these patients were observed to have neurological deficit upon admission [Table - 1], [Table - 2], [Table - 3]. There were 6 cervical, 12 thoracic and 3 lumbar region injuries. Five patients had wounds of exit, the rest had lodgement of metallic fragment in the spine/surrounding soft tissues/spinal canal. These patients underwent a standard medical management protocol that included intravenous fluids, antibiotics, steroids, bladder catheterization, care of the skin and pressure points. Inj methyprednisolone could be administered to 3 patients who were received within 24 hours of the injury. Thereafter, patients were divided into 3 groups: -
Group A: Complete motor and sensory loss below the lesion
(Benzel & Larson Grades I & II): 15 patients
Group B: Incomplete sensory or motor loss below the injury
(Benzel & Larson Grades III & IV): 4 patients
Group C: Cauda equina injuries: 3 patients
Additional injuries: These are given in [Table - 2].
Imaging: Imaging studies included plain radiographs in all patients, myelography in 4, CT myelography in 11 and MRI in 4 patients [Figure - 1], [Figure - 2], [Figure - 4], [Figure - 4], [Figure - 5].
Neurosurgical procedure: A total of 17 patients underwent surgery. Three patients from Group A (2 cervical and 1 thoracic injury) died preoperatively, and 2 patients with cervical root avulsion and cervical myelopathy were managed conservatively. The rest of the patients were operated upon once they were hemodynamically stable and were able to maintain SpO2 of more than 95% on spontaneous respiration. Visceral injuries were tackled first by the surgeon concerned.
The spinal cord was explored by midline incision in 12 patients. Lateral extracavitory decompression of the dorsal cord was done in 3 and anterior cervical corpectomy, debridement and fusion were done in 2 patients. Laminectomy was done and the dura was opened to inspect the cord. The cord was seen lacerated in 5, contused in 3 and transected in 1 patient. Three patients had transection of cauda equina rootlets. After debridement, irrigation and hemostasis, the dura was closed primarily or by duroplasty employing lumbodorsal fascia. The cord could not be inspected completely in the lateral extracavitory approach and only a patch of fascia was placed over the part and wound closed after hemostasis. Postoperatively, these cases had continuous lumbar drainage of CSF for 72 hours. All patients received antibiotics that cross the blood-CSF barrier, for a period of four weeks.
A total of 17 patients underwent surgery. There was no improvement in the neurological grade of Group A patients. There was distinct improvement in the neurological grade of Group B patients, within three to four weeks of surgery. Two of the 3 Group C patients with cauda equina injury showed improvement in their neurological grades. There were no instances of meningitis, wound infection or CSF fistula. The 2 patients with soft tissue injury to the neck with myelopathy who were not operated showed gradual improvement. [Table - 2] shows the functional outcome (Benzel & Larson) in 17 patients. There was consistent correlation of outcome with the initial neurological status at the time of arrival.
Continence was established in 6 patients, while the rest of the survivors continued to have neurogenic bladder dysfunction. Patients with soft tissue injury and those with incomplete injury experienced dysesthesiae in the recovery phase and responded to imipramine. There was no instance of delayed neurological deterioration. Four patients died in this series: 3 of these were from Group A (not operated group), and 1 in the postoperative period.
Missiles may damage the spinal cord by direct laceration, transection, blast wave or by nerve root avulsion leading to intramedullary hematoma. The velocity of the missile is the most important determinant of the severity of cord injury. High velocity military weapons produce extensive comminution and widespread damage due to high-energy transfer. These high velocity missiles may also produce neural damage by transmission of shock waves that penetrate the neural tissue. In such instances, the blast or concussive injuries may have potential for recovery, unlike those caused by direct laceration/transection of the cord. On the other hand, civilian injuries and those in insurgency prone regions often occur due to low velocity missiles, and direct injury to the cord and the adjacent vertebral structures produce a neurological deficit. Most of the injuries are due to exploding shells; hospital admission with bullet injuries is less common, possibly due to early fatality in the latter cases. Those seen in our series were possibly due to bullets at the end of their flight ('spent' bullets) since there was not much tissue damage.
There is often a delay in transporting these casualties from the combat zone to the neurosurgical center. Proper immobilization, care of the skin, bladder catheterization, and administration of steroids forms the interim management pending transfer. These casualties are best evacuated by air. Upon arrival, a thorough general evaluation is carried out, since the patient is likely to have suffered multi-organ damage due to multiple splinters. After hemodynamic stabilization and attention to visceral and thoracic injuries, neurological evaluation is carried out and a neurological grade (Frankel/Benzel Larson grade) is assigned.
Imaging studies comprise plain radiographs in two views, multiplanar imaging, and computed tomography in axial sections with 3D reformatting. Imaging by CT is the principal diagnostic procedure and MRI may be carried out in selected cases where the missile has exited, especially in cases with root avulsion and intramedullary hemorrhage and without associated vertebral injury.
The traditional approach for the management of missile injury of the spine has been towards conservatism, arising out of pessimism associated with poor functional outcome, and spinal cord decompression was advised only if there was neurological deterioration during observation. Yashon et al in a series of 65 cases of bullet injuries of the spine found no difference in the return of neurological function between surgical and non-surgical groups, while others have recommended a policy of selection for surgery. There are several studies carried out with contradicting results about the value of neurosurgical intervention in these cases.,, While there is no debate on the need for surgery in patients with partial deficits, whether all patients with complete deficits should be operated upon is the subject of controversy. In military neurosurgical practice, the trend is towards surgical exploration as has been brought out in the experiences of the Korean and Vietnam conflicts. During the civil war in Croatia, patients were reported to have shown significant neurological improvement and operative mortality of 1 %. From India, the data from the Armed Forces is sketchy, and there is only 1 report pertaining to civilian injuries. The present series also consists of a small number of cases, and no different conclusions can be drawn at present.
Cauda equina lesions too are an important indication for surgical exploration since these injuries carry a better functional prognosis due to the involvement of the nerve roots and not the spinal cord. Surgical exploration, decompression and removal of a bullet or bone fragments decompresses the nerve roots, allowing for better recovery of neurological function. Hence we recommend removal of intraspinal bullets in the region of cauda equina. In the cervical spine, surgical exploration assumes importance, as the recovery of even 1 nerve root makes a significant difference in the rehabilitation. In spinal cord injuries, the final neurological outcome in such cases is directly correlated with the initial neurological grade.,,,, The indications for surgery should not be as much for neurological recovery as for avoiding the complications of CSF leak from the wound. Moreover, retained intraspinal fragments in the region of cauda equina can lead to chronic pain and should be removed.
The aims of surgery can be summarized as improvement in the neurological grade: a desirable goal, often achievable in incomplete injuries but neurological improvement in cases with grade 0/5 power is still a mirage; avoidance of complications pertaining to external CSF fistulae; avoidance of pyogenic complications such as meningitis, spinal abscess, etc.; and facilitation of early rehabilitation.
The scope of surgery in missile injuries of the spine and spinal cord includes debridement of the missile track and removal of bone and missile fragments from the spinal canal. (The need for the removal of intramedullary splinters is debatable); visualization of the cord to determine prognosis; surgical decompression of the cord; and dural closure with or without duroplasty
The choice of surgical approach has to be individualized for every patient and no single standard approach can be recommended. Injury restricted to the posterior elements would warrant a posterior midline approach; laminectomy, debridement, dural reapir are carried out. Intraspinal fragments can be extricated if visible or if the patient is completely paralyzed below the injury. Judgment has to be exercised in the presence of preserved neurological function with intramedullary splinter; the aim being to not worsen the neurological status by trying to dissect the splinter. Injury to the disco-corporeal components of the cervical spine would warrant anterior decompression with fusion with/without instrumentation.
Injury to the dorsal spine may be accompanied by chest injury. In these cases, the spine can be exposed by a thoracotomy, or by costotransversectomy approach and debridement carried out. Lumbar spine injuries are generally stable due to the large size of the vertebral bodies, and debridement and dural exploration can be carried out by the posterior midline approach and laminectomy; we did not carry out instrumentation in any of the cases.
The exact timing of surgery depends upon associated injuries, and these cases should be tackled electively once systemic stability is achieved.
Missile injuries of the spine are not commonly seen in Indian neurosurgical practice. Proper immobilization, resuscitation, bladder catheterization and evacuation to a specialized center constitute the initial management. Neurological evaluation, assessment of collateral damage and imaging by CT is done, followed by decision regarding surgery. Surgical decompression in the presence of complete neurological deficit is debatable, although in military neurosurgical practice, there is a tendency for aggressive surgical decompression. Incomplete injuries, and injury to cauda equina constitute definite indications for surgery. Retained fragments rarely lead to infection or delayed neurological deficit, and prophylactic removal is not warranted except in cauda equina injury. The overall prognosis for recovery in the presence of complete neurological deficit remains poor, and initial neurological deficit remains the best prognostic indicator.