CLINICAL STUDIES / ETUDES CLINIQUES
SURVIVORS OF SELF-INFLICTED GUNSHOT HEAD INJURIES MANAGED AT A SINGLE NEUROSURGICAL UNIT: A 12-YEAR REVIEW
ETUDE DES SURVIVANTS A DES BLESSURES CRANIENNES, VOLONTAIRES PAR BALLES REÇU DANS UNE UNITE NEUROCHIRURGICALE : UNE REVUE SUR UNE PERIODE DE 12 ANS
E-Mail Contact - DOOKIE Sudhir : firstname.lastname@example.org
Background and objective:
Keywords: parasuicide, craniocerebral injury, gunshot wound
Civilian craniocerebral gunshot injuries (CGIs) result in severe traumatic brain injury, forming a significant public health affliction in South Africa (SA) and internationally [7, 12, 16]. The underlying aetiology of self-inflicted gunshot head injuries (SIGHI) following suicidal intent (SI) is multi-factorial. SA has a substantial history of violent trauma, fuelled by social strife and easy access to firearms [9, 13, 14]. Furthermore, recent reports suggest that SA has the 22nd highest suicide rate worldwide, with guns contributing to this phenomenon [4, 14].
SIGHIs impart a high pre-hospital mortality due to their destructive nature [1, 10, 12, 15, 18, 19]. A patient receiving treatment at a neurosurgical unit (NU) following this type of injury, thus offers a unique opportunity to gain insight into a survivor. Furthermore, in SA, there is paucity of data on management of these types of injuries, with most reports based on post mortem studies . The purpose of the current study was to report the prevalence, demographics and characteristics of these cranial injuries with regards to presentation, radiological findings, management, and outcomes at the NU in Inkosi Albert Luthuli Hospital (IALCH) which is well suited; being the only public neurosurgical service in the province of KwaZulu-Natal (KZN), catering to a population of over 10 million people .
This was a retrospective, observational, descriptive study performed at a single institution. We reviewed medical records of patients treated in the NU at IALCH with a diagnosis of a CGI during January 2003 to September 2014. From this cohort of patients we selected and included into the study sample those with a diagnosis of SIGHIs following SI. Those patients who were victims of inter-personal violence and those with accidentally-inflicted gunshot head injuries were excluded. The study approval was granted by the Biomedical Research Ethics Committee of the University of KwaZulu-Natal (reference number BE007/14).
Variables investigated included age, gender, neurological deficit, Glasgow Coma Scale (GCS) on admission and discharge, gunshot entry point, computerised tomography (CT) findings, management, complications, length of hospital stay (LOS), and in-patient mortality. Data was collected using a data collection sheet, captured in Excel® (Microsoft Inc., WA, USA) and analysed using Stata® version 14 (StatCorp LP, Texas, USA). Means (with standard deviations), medians and frequencies were used to describe the data. The Wilcoxon Mann-Whitney test was used to compare continuous variables between subgroups. Associations between categorical variables were tested using the Fisher’s exact test. Results with a P-value of <0.05 was considered statistically significant. Management of patients with CGIs in the NU at IALCH is guided by the ATLS® principles , followed by an urgent computerised tomography (CT) brain scan as a form of investigation. Angiography is performed if a vascular injury is suspected. Upon reviewing these investigations, patients are taken for operative management, which is based on the Brain Trauma Foundation guidelines  and the management principles of gunshot head wounds [1, 10, 12, 16, 18]. Tetanus toxoid is administered at admission whilst a course of empiric antibiotic therapy with gram positive cover is commenced and continued post-operatively for 7-days. Anti-seizure prophylaxis is routinely administered for 7 days post-injury and discontinued if seizures are not documented during this period. RESULTS
A total of 499 patients with the diagnosis of a CGI were identified during the study period. Thirty-one (6%) of these patients were survivors of SIGHIs following suicidal intent. Table 1 demonstrates their clinical characteristics. Majority of patients were male [28; 90%], with a male-to-female ratio of 9:1.
The gunshot wound entry sites are shown in Table 2. The clinical presentations included hemiplegia [6; 19%], pupillary abnormality [2; 6%], facial palsy [2; 6%], visual loss [2; 6%], cerebrospinal fluid (CSF) otorrhea [1; 3%], and seizures [1; 3%].
The CT brain scan findings and missile injury morphology are shown in Table 3. CT Angiography in one patient revealed a traumatic right middle cerebral artery (MCA) pseudoaneurysm (Figure 1).
All patients were managed surgically and the procedures performed were wound debridement craniectomy [19; 61%], superficial wound debridement and closure [7; 23%], craniotomy [4; 13%] and decompressive craniectomy [1; 3%]. Six (19%) patients required neurocritical care which included intracranial pressure (ICP) monitoring and ventilation.
Eleven (35%) patients developed early post-operative cranial complications. Septic morbidity occurring in ten (32%) patients comprised the majority of these cases and was due to meningitis [6; 19%], wound sepsis [4; 13%], and intracerebral abscesses [3; 10%]. The microbes responsible for these infective complications were acinetobacter baumani (2), staphylococcus aureus (1), and pseudomonas aeruginosa (1); with five patients not yielding an offending organism. Management of wound sepsis and intracerebral abscess included surgical debridement and excision of the abscesses in operating theatre, followed by directed antibiotic therapy guided by microscopy, culture and sensitivity results. Seven patients (23%) had CSF fistulae; with five of them developing meningitis. The CSF fistulae presented as rhinorrhea [3; 10%], otorrhea [1; 3%], contained pseudomeningocele of the scalp [2; 6%], and an open wound leak [1; 3%].
Seven (23%) patients with SIGHIs demised during the study period and they had a significantly lower post resuscitation GCS ≤ 8 (P = 0.018) and age ≥ 39 years (P = 0.026). Four (22%) of the eighteen patients that sustained transaxial injuries demised, yet this did not show a statistically significant contribution to mortality (P = 0.999).
SIGHIs impart destructive injuries following maximal energy transmission to the brain due to the close range of the weapon. Survivors of SIGHIs provide a small proportion to the overall CGIs in our unit when compared to the experience of other centres . This could partially be explained by the particularly high pre-hospital mortality, and with our centre being the only NU in the province, delays in transfer further influence the number of salvageable survivors making it to the unit. Also, firearm legislature and socioeconomic factors vary between countries, affecting gun usage and resulting in differing patterns of CGIs.
Suicide attempts are reported to be more common in females and young adults . Females though are less likely to possess firearms and use them in suicide attempts. Victims of such injuries are frequently reported to be males , and this has been mirrored in our study. The predominance of males and the median age in the current study fall within a similar range to victims of traumatic brain injury following interpersonal violence treated in our unit . The underlying factors leading to SI are many, amongst them being work related stress, depression, relationship discontent, alcohol and drug abuse.
The admission post-resuscitation GCS has been reproduced in numerous studies as the single strongest clinical predictor of outcome in gunshot head injuries [1, 7, 16, 18]. Our findings did suggest that patients presenting in coma (post-resuscitation GCS ≤ 8) are more prone to mortality.
A CT brain scan is a vital investigative tool in the acute management of these patients. It displays the injury morphology, is used to plan surgical intervention, and used as a baseline for follow-up. It can also be used to prognosticate the injury, as presence of tranventricular and bihemispheric injury is associated with a poor outcome [1, 16]. Transaxial CGIs though known to be associated with high mortality did not show a statistical significant relationship to mortality in our study . This atypical finding could be explained by the small patient numbers and that majority of these injuries were bifrontal, possibly not affecting absolute critical structures in the brain. Regarding the entrance site, the frontal regions predominated and together with the temporal area formed the majority of the sites of injury, which is congruent with the possible gun orientation during a suicide attempt . Furthermore, with self-inflicted close range gun orientation leaving little room for error, it is not unsurprising that a tangential injury would not be likely and was the least common type of gunshot head wound in our series.
Angiography is an ancillary imaging modality employed in those where there is concern of significant neurovascular injury. A middle cerebral artery pseudoaneurysm was discovered in one patient. The said individual had a heavy subarachnoid blood load in the sylvian fissure adjacent to the bullet tract and CT hypodensities suspicious of ischaemia, which prompted this further investigation. However, this patient had a GCS of 6 and was deemed a poor grade for intervention.
Emergent surgical management via a craniotomy or craniectomy is life saving in patients with intracranial haematomas causing raised intracranial pressure. Surgical debridement involves removal of necrotic brain tissue, loose bone fragments and retained missiles when they are easily accessible. This is followed by repair of the dura in order to prevent post-operative CSF leaks. Infective complications remain a problematic cause of morbidity and mortality with CSF fistulae greatly increasing this risk; reaffirming the value of the operative debridement and dural repair as a vital component in this regard . Every means of reducing septic sequelae should be adhered to, thus during the post-operative period judicious use of antibiotics is practiced . Unfortunately, septic complications still occurred in 32% of our patients, which was higher than the rate reported by other authors .
An early post traumatic seizure occurred in one (3%) patient while admitted in the NU. These have been reported between 1.3% and 24% of CGIs , lending support to our protocol practiced for seizure prophylaxis. The mortality rate in our study of 23 % was within the reported range in other series .
Key to reducing the incidence of SIGHIs in SA lies in prevention strategies. This may involve identifying those at risk and intervening early through primary preventative psychosocial interventions, coupled with stricter gun control legislation to limit access to firearms.
The limitations of this study are its retrospective nature, a small sample size, and no long-term follow-up. The study only analysed patients within the public health sector and those patients treated in the private sector have not been represented in our numbers. However, the study represents the experience of a single NU centre in KZN and thus gives us insight into the presentation and outcomes of survivors of SIGHIs who receive treatment at our institution.
Survivors of SIGHIs following an attempt at suicide account for a small contribution to the total CGIs in our practice. Early surgical treatment is critical in providing the best chances for a favorable outcome in those patients that survive the initial onslaught. Unfortunately though, those presenting in coma have poor outcomes.
Holistic patient care involving a multidisciplinary team which includes rehabilitation specialists, social workers and psychologists is essential if these patients are to be re-integrated into society. However, due to constrained resources in our environment, this rehabilitative and psychological component is entrusted to step-down facilities upon discharge.
To our knowledge this study is the first neurosurgical review of SIGHIs describing patient presentation and outcomes in KZN. This information can be useful in exploring the burden imparted on resources, may provide an impetus for preventative measures to be applied, assist in producing clinical recommendations, and help guide policies to institute more holistic patient management.
CONFLICT OF INTEREST
The authors declare no conflicts of interest.