Increasingly, attention has been focused on the epidemic of firearm-related injuries in the pediatric population, which carries a mortality rate of 12% to 19%.¹ What often escapes mention are the estimated 32 000 injuries attributable to nonpowder firearms (ie, BB gun, pellet gun, and air rifle) which occur each year in the United States.¹ The majority of victims (more than three quarters) are under the age of 15. This is particularly disturbing because most states have minimal or no restrictions on the sale and use of these weapons to children. In fact, only 14 states in the United States have legislation governing the sale or possession of these weapons.¹

Many case reports of serious and even fatal nonpowder firearm injuries have been published describing ocular, intracranial, abdominal, and thoracic wounds,⁴ yet there is little data regarding the epidemiology and outcome in a defined population. We reviewed all nonpowder firearm injuries treated at an urban pediatric hospital during an 11-year period to ascertain demographic characteristics, specific injuries, and outcomes.

METHODS

Abstraction of information included the following: demographic information such as month/year of injury; patient age, race, and gender; circumstances of injury; anatomic location of injury; diagnostic studies; treatment; and outcome including delayed complications, functional deficits (defined as persistent alteration in the ability to perform activities of daily living), and mortality. A random sample (approximately 15%) of medical records were reabstracted to assure quality of the abstraction process.

Data were stored and analyzed using EpiInfo (Centers for Disease Control and Prevention, Atlanta, GA) version 5 statistical software. All data are reported as means ± standard deviation for continuous variables or as frequencies for categorical variables. Tests of significance using either ², Student's t, or Fisher's exact tests were used when appropriate. Odds ratios with 95% confidence intervals are reported, and a P value of <.05 was considered statistically significant.

RESULTS

The highest rate of injury (45%) occurred during the summer and early fall months (July through October) (Fig 1). The mean age was 12 ± 3.7 years (range, 2 to 18 years) and the highest prevalence of injury was observed in the 10- to 14-year age group (54%). Eight percent of the children injured were under 5 years of age (Table 1). Most children sustained a single wound; however, up to four wounds were reported. Overall, 49% of injuries were intentional and were the direct result of an assault. Most injuries (77%) occurred outside the home.

Table 1. Injuries Caused by Nonpowder Firearms, by Age and Gender [View Table]

The most common sites of injury were the extremities (39%), head and neck (33%), thorax (13%), and eye (8%) with insignificant variability based on age (Table 2). Evaluation and treatment of the patients included: laboratory tests (17%), radiographic imaging (either plain radiography or computed tomography scan) (59%), antibiotic prophylaxis (22%), and tetanus booster (25%). Specific wound therapy included wound irrigation (30%), removal of foreign body (17%), and operative intervention (12%).

Table 2. Age-specific Injuries by Anatomic Site [View Table]

The majority of patients (74%) were treated and discharged from the emergency department (P < .001) (Table 3). Of the 27% of patients admitted to the hospital for their injuries, 45% required an operative procedure. The most common injuries requiring either hospitalization or operative intervention were injuries to the eye (40%), abdomen (15%), and neck (15%) (Table 4). There was a significant association among ocular and abdominal injuries requiring hospitalization or operative intervention when compared with all other injuries (ocular odds ratio, 44.5; 95% confidence interval 6, 951; P < .001; abdomen odds ratio, 11.8; 95% confidence interval 1.2, 285; P = .02). Seven percent of patients sustained some degree of functional deficit, including loss of vision and severe neurologic impairment, and there was a single late complication (wound infection). No deaths (0% with 95% confidence interval, 0 to 2%) occurred in this series. Three patients returned with a second nonpowder firearm-related injury within the study period.

Table 3. Disposition and Outcome of Children With Nonpowder Firearm Injuries [View Table]

Table 4. Disposition Based on Body Part Injury [View Table]

DISCUSSION

The potential for serious injury, however, mandates that all nonpowder firearm wounds be thoroughly evaluated to avoid missing underlying severe injury. This should include localization of the foreign body, if present, in three dimensions using imaging techniques (typically roentgenograms), determination of the trajectory to postulate potential organs injured, and assessment of the need for operative intervention. Wounds determined to be minor should receive local wound care (irrigation, removal of foreign body if superficial), and tetanus prophylaxis. Antibiotics are not required routinely, but their use should be at the discretion of the treating physician. Antibiotics are typically reserved for patients with additional risk factors for wound infection (ie, tissue devitalization, delay in treatment, or gross contamination).⁷ Patients with potentially serious wounds should be admitted for observation and, if indicated, operative intervention.

In this series, extremity injuries were most common and the least likely to require admission. Typically these wounds required only local wound care, careful neurovascular evaluation, and tetanus prophylaxis if boostering is necessary. Ocular injuries, in contrast, occur less frequently but account for 28% of admissions, have frequent residual functional deficits (42%), and require operative intervention in 62%. This is in accordance with the consequences of nonpowder firearm ocular injuries previously reported.⁸ Although abdominal injury occurred in only 3% in this population, 80% of those patients required hospitalization and/or operative intervention. In our population, the odds of requiring admission for ocular or abdominal injuries were substantial.

Thoracic injury occurred in 13% of patients and required operative intervention in 5%. The potential lethal nature of these wounds is evidenced by the child who sustained right ventricular penetration and presented with pericardial tamponade. Others have had similar experiences including cardiac penetration with pellet embolization and aortic penetration with delayed cardiac arrest and death.^9,10 Intracranial penetration of the pellet is also a possibility with its attendant risk of morbidity and mortality.^11,12 Our series included a child with intracranial penetration and associated subarachnoid hemorrhage and residual aphasia related to injury to the temporal lobe of the brain.

Although deaths attributable to nonpowder firearms have been reported in previous case reports, there were no deaths in this series. The lethal potential of these wounds is difficult to quantify due to the paucity of large reported series; however, this should not diminish the concern in the evaluation of these patients. Projectiles fired from modern carbon dioxide-powered rifles can achieve a muzzle velocity of more than 900 foot-pounds per second, comparable to most handguns.^5,13

Education of the public as to the wounding capacity of these weapons is essential. Traditionally, these firearms are owned by adolescents who believe that nonpowder firearms are little more than toys. Wounds created by these weapons should be regarded as potentially serious as powder firearm injuries. As has been suggested by Powell et al,²¹ unless we embrace and integrate a public health approach to reduce these violent injuries and participate proactively in the legislative process, we will continue to witness preventable injuries and death from nonpowder firearms in children.