- Email: [email protected]
Children Have an Increased Risk of Periorbital Dog Bite Injuries
CRANIOMAXILLOFACIAL TRAUMA
Children Have an Increased Risk of Periorbital Dog Bite Injuries Philip J. Hurst, DDS, MD,* Marcus Ji Hoon Hwang, BS,y Thomas B. Dodson, DMD, MPH,z and Jasjit K. Dillon, DDS, MBBS, FDSRCSx Purpose:
Dog bites to the head, neck, and face (HNF) disproportionately affect children. It is unclear if specific facial regions are uniquely at risk in this population. Both the periorbital region and the central target area (CTA; nose, lips, and cheeks) have been proposed as being at increased risk. This study addressed the following question: Among individuals sustaining a dog bite injury to the HNF, are children, when compared with adults, at greater risk of injury to the periorbital region or CTA?
Patients and Methods:
Using a retrospective cohort design, we enrolled a sample composed of patients presenting with HNF dog bite injuries. The predictor variable was age category: pediatric or adult ($18 years). The primary outcome variable was injury location: isolated periorbital, isolated CTA, both periorbital and CTA, or other HNF location (neither periorbital nor CTA). Other variables included specific facial structures injured, demographic characteristics, injury circumstances, and clinical course. Descriptive and bivariate statistics were calculated.
Results:
The sample consisted of 183 pediatric (58.5%) and 130 adult (41.5%) patients. Isolated periorbital injuries were more common in the pediatric group (relative risk [RR], 2.2 [95% confidence interval (CI), 1.3 to 3.7]; P = .003), as was injury to both periorbital and CTA regions (RR, 2.0 [95% CI, 1.2 to 3.5]; P = .01). Isolated CTA injury was the most common injury pattern overall in both children and adults, but children were at significantly less risk than adults (RR, 0.7 [95% CI, 0.5 to 0.9]; P = .002). Children were also less likely to sustain an injury to other HNF regions (RR, 0.6 [95% CI, 0.3 to 0.97]; P = .038). Conclusions: Children were twice as likely as adults to incur a periorbital dog bite, but no such increased risk was observed for the CTA. This may reflect unique anatomic risks in children or targeting of their eyes by attacking dogs, contributing to higher rates of HNF injury in this population. Public health measures should recommend against children being at eye level with dogs, even if familiar or under adult supervision. Ó 2019 American Association of Oral and Maxillofacial Surgeons J Oral Maxillofac Surg -:1-9, 2019
The risk of being bitten by a domestic animal during one’s lifetime is estimated to approach 50%. Of these domestic animal bites, dog bites account for 80 to 90%,1 comprising approximately 1% of all emergency visits
to hospitals in the United States.2 Dog bites accounted for more than one third of all homeowner insurance liability claims paid in 2012, with an average paid claim of $29,752, totaling more than $489 million in claims.3
Received from University of Washington, Seattle, WA.
dation grant. Neither foundation has a conflict of interest with this
*Resident, Department of Oral and Maxillofacial Surgery.
manuscript.
yDental Student, School of Dentistry.
Address correspondence and reprint requests to Dr Hurst:
zProfessor and Chair, Department of Oral and Maxillofacial
Department of Oral and Maxillofacial Surgery, University of Washing-
Surgery.
ton, 1959 NE Pacific St, HSB Room B 241, Box 357134, Seattle, WA
xClinical Associate Professor and Program Director, Department
98195; e-mail: [email protected]
of Oral and Maxillofacial Surgery. This project was supported in part by the Department of Oral and
Received May 22 2019 Accepted August 22 2019
Maxillofacial Surgery Laboratory for Applied Clinical Investigation
Ó 2019 American Association of Oral and Maxillofacial Surgeons
and the Department of Oral and Maxillofacial Surgery Research
0278-2391/19/31046-8
and Training Fund.
https://doi.org/10.1016/j.joms.2019.08.021
Conflict of Interest Disclosures: Dr. Dillon is the recipient of an Oral and Maxillofacial Surgery Foundation and Osteoscience Foun-
1
2 Dog bite wounds are associated with a host of potential negative consequences including pain, infection, cosmetic disfigurement, psychological trauma, and financial burden.4 These consequences are amplified when the injuries involve the region of the head, neck, and face (HNF), which is the most common location for dog bite injuries in the pediatric population.5-18 Several researchers have identified a central target area (CTA) of the face, in which the nose, cheeks, and lips are most often injured by dog bites, particularly in children.5,17,19-24 Others have focused on ocular and periorbital injury and found children to be at greater risk.7,25-30 Proposed explanations have focused on the HNF region broadly and include smaller stature and convenient proximity of the head to the dog,22 inappropriate behaviors such as kissing,31 or instinctual targeting of these structures by an attacking dog.29 Some researchers have posited that because younger children are at eye level with dogs, their staring is perceived as threatening.32 Others have considered higher rates of periorbital injury in children as a byproduct of the aforementioned factors driving the higher frequency of HNF injury in the population.27 No study has directly compared injury to the periorbital region and CTA between age groups while accounting for children’s higher baseline risk of HNF injury. Thus, the possibility that children’s eyes or other facial structures are uniquely at risk of dog bite injury or play a role in their increased rates of HNF injury has not been well evaluated. The purpose of this study was to answer the following research question: Among individuals sustaining an HNF dog bite injury, are children, when compared with adults, at greater risk of injury to the periorbital region or CTA? The null hypothesis was that there is no difference in the rate of injury to the periorbital region or CTA between children and adults. The specific aims of the study were 1) to measure and compare the frequencies of periorbital and CTA injuries among individuals with HNF dog bites seeking medical attention and compare these rates between pediatric and adult populations; 2) to identify facial regions at increased risk of dog bite injury in children compared with adults; and 3) to discuss implications for the prevention and management of these injuries.
Patients and Methods We designed and implemented a retrospective cohort study and enrolled a study sample derived from patients who presented to the emergency department of Harborview Medical Center (HMC), Seattle, Washington, for the evaluation and management of
PERIORBITAL DOG BITE INJURIES
dog bite injuries to the HNF between January 1, 2000, and December 31, 2015. To identify study patients, we accessed the HMC trauma registry, which includes patients who have at least 1 trauma-specific diagnostic code and meet at least 1 of the following inclusion criteria: full or modified trauma team activation, admission to the hospital, dead on arrival, transfer from another acute or urgent care facility by emergency medical services, or flight to the scene by emergency medical services. This registry was queried for injuries with a mechanism of ‘‘bite,’’ and results were further refined using International Classification of Diseases, Ninth Revision codes 906.0 to 906.5 and bite codes in the International Classification of Diseases, Tenth Revision range of W53 to W59, as well as code W61. To be included in the sample, patients had to have sustained at least 1 dog bite injury to the HNF region. The HNF region was defined as including any anatomic structure superior to the clavicle. Injury was defined as any disruption of normal tissue architecture, appearance, or function as documented in the clinical record, including but not limited to laceration, puncture, abrasion, defect, scratch, contusion, avulsion, fracture, degloving, transection, bite, tear, or amputation. Patients with duplicate records, insufficient data, or without dog bite injuries to the HNF were excluded from the sample. This study followed the principles outlined in the Declaration of Helsinki and was approved by the University of Washington Institutional Review Board (No. 00003323). The primary predictor variable was age classification: pediatric or adult. Patients younger than 18 years at the time of injury were classified as pediatric patients, whereas those aged 18 years or older were classified as adults. Age 18 years was chosen given its use as a widely accepted definition of adulthood in the United States in medical and legal contexts and precedent in similar studies.23,33 The term ‘‘children’’ also was used to refer to the pediatric age group in this study. The primary outcome variable was the anatomic location of the dog bite injury into 1 of 4 categories: isolated periorbital, isolated CTA, both periorbital and CTA, and other HNF location (neither periorbital nor CTA). Injury to the periorbital region was defined as an injury to at least 1 of the following structures: eyebrow, eyelid, medial or lateral canthus, canalicular system, or orbit and its contents such as the extraocular muscles, conjunctiva, cornea, or globe. Injury to the CTA was defined as injury to at least 1 of the following: lip, nose, zygomatic region, or buccal region. Patients with an injury to at least 1 structure in both the periorbital region and CTA region were classified as having both periorbital and CTA injuries.
3
HURST ET AL
Patients with injuries to other HNF structures such as the neck, scalp, forehead, ear, or mandible but with no injury to the periorbital region or CTA were classified as having ‘‘other HNF’’ injuries. Patients with concomitant injuries outside the HNF region were not excluded from this study, but these injuries were not included in data analysis. Other study variables included patient age at the time of injury, gender, race, dog ownership, breed of dog, and location of injury event. Race was documented as listed in the medical record and was classified as follows: white, Hispanic or Latino, Black, Asian or Pacific Islander, or Native American. For the purposes of statistical analysis, race at times was grouped as white and ‘‘other,’’ consisting of the remaining 4 racial categories. The owner of the attacking dog was determined, when possible, from the medical record and was classified as family or other. ‘‘Family’’ was defined to include dogs owned by the patient or a relative. ‘‘Other’’ was defined to include the following categories: friend, neighbor, stranger, police, or stray. A friend was defined as someone referred to as a ‘‘friend’’ or ‘‘acquaintance’’ of the patient or patient’s family. A neighbor was defined as someone living near the patient who was not otherwise specified as a friend. A stranger was defined as someone unfamiliar or unknown to the patient. ‘‘Stray’’ was used in reference to dogs without an apparent owner. Injury event location was classified as home or other. ‘‘Home’’ was defined as occurring inside, in the yard, or within the property boundaries of any home or dwelling regardless of it being the patient’s home. ‘‘Other’’ was defined to include any location not meeting the definition of home. Breed of dog was recorded as reported in the medical record. The following breeds were grouped as ‘‘other’’: terrier, Doberman, retriever, bulldog, husky, Alaskan malamute, chow, Australian shepherd, Rhodesian ridgeback, Great Dane, Dalmatian, poodle, Saint Bernard, Portuguese water dog, Akita, boxer, border collie, beagle, Shar-Pei, dachshund, Chihuahua, Springer spaniel, cocker spaniel, German shorthair pointer, greyhound, and basenji. Cases of multiple attacking dogs were also deemed ‘‘other.’’ If no breed was reported or the breed was not known to witnesses, it was documented as unknown. Secondary outcome variables related to the clinical course were documented: transfer status, hospital admission status, whether the patient was treated in the operating room (OR), and length of stay. ‘‘Transfer status’’ referred to whether the patient was transferred from an outside facility rather than presenting directly to HMC. Patients were documented as having been admitted if they were admitted to HMC at any time for injuries related to a dog bite attack. Patients who were discharged directly from the emergency depart-
ment or postanesthesia care unit were not classified as having been admitted. Any patient who was treated in the OR at HMC for injuries related to a dog bite attack, as documented by a formal operation report and anesthesia record, was classified as having been treated in the OR. This was defined to include patients who were not treated in the OR on initial presentation but were treated in the OR during a subsequent encounter. Treatment in the emergency department, even with procedural sedation, was insufficient to qualify as having been treated in the OR. Length of stay was defined as 0 days for any patient who did not meet the criteria for having been admitted to the hospital. For a patient who was admitted, length of stay was defined as the number of midnights the patient remained an inpatient for any admission related to dog bite injuries. Data analysis was performed with Microsoft Excel software (Microsoft, Redmond, WA) and SPSS software (version 15; IBM, Armonk, NY). Univariate, bivariate, and multivariate statistics were computed. P < .05 was considered statistically significant.
Results The study sample was composed of 313 patients with dog bite injuries to the HNF (Table 1). The mean age was 20.1 years (range, 0.7 to 94.4 years), and there were 174 male patients (55.6%). The study sample comprised 183 pediatric (58.5%) and 130 adult (41.5%) patients. A comparison of study variables against the primary predictor variable, age group, is presented in Table 2. The mean age was 5.7 years in the pediatric group and 40.3 years in adults. No significant differences were observed between the pediatric and adult groups for gender, race, transfer from an outside facility, and length of stay. Greater than 80% of patients in both age groups were white. For 26 patients, race was not indicated in the medical record; therefore, statistics were calculated with a reduced sample size for this variable. Most children and most adults were transferred from an outside facility. Pediatric patients were significantly more likely to be bitten by a family-owned dog (P = .002), to be bitten in a home setting (P = .001), to be admitted to the hospital (P < .0001), and to undergo treatment in the OR (P < .0001). The pit bull was the most commonly named breed to be involved in both pediatric (18%) and adult (26.2%) injuries. In many cases, the breed was not known or documented, and the ‘‘unknown’’ category was the largest category in both the pediatric and adult groups. In Table 3, study variables are compared with the primary outcome variable, injury location. No significant differences were found between injury location
4
PERIORBITAL DOG BITE INJURIES
Table 1. DESCRIPTIVE SUMMARY OF STUDY SAMPLE
Study Variable
Descriptive Statistics
Sample size Male gender Age, yr Age group Pediatric Adult Race (n = 293)* White Other Injury location Isolated periorbital Isolated CTA Both periorbital and CTA Other HNF region Injury event location (n = 289)* Home Other Transfer Hospital admission Treatment in OR Length of stay, days Dog owner (n = 285)* Family Other Breed Pit bull Labrador German shepherd Rottweiler Mix Mastiff Other Unknown
313 174 (55.6) 20.1 20 (0.7-94.4) 183 (58.5) 130 (41.5) 243 (82.9) 50 (17.1) 65 (20.8) 141 (45) 58 (18.5) 49 (15.7) 239 (82.7) 50 (17.3) 251 (80.2) 205 (65.5) 183 (58.5) 1.64 4.3 (0-45) 149 (52.3) 136 (47.7) 67 (21.4) 21 (6.7) 20 (6.4) 19 (6.1) 18 (5.8) 11 (3.5) 51 (16.3) 106 (33.9)
Note: Data are presented as number (percentage) or mean standard deviation (range). Abbreviations: CTA, central target area; HNF, head, neck, and face; OR, operating room. * If data are missing, the revised sample size is shown in parentheses. Hurst et al. Periorbital Dog Bite Injuries. J Oral Maxillofac Surg 2019.
and gender, race, injury event location, transfer status, dog owner, or breed. Significant differences were found for age, hospital admission, treatment in the OR, and length of stay. Younger age was associated with isolated periorbital injuries (15.6 20.7 years) and both periorbital and CTA injuries (13.9 16.7 years), whereas isolated CTA injuries (23.0 20.3 years) and other HNF injuries (24.9 19.7 years) were associated with a higher mean age (P = .002). Patients with isolated periorbital injuries had the highest rates of hospital admission, 76.9% (P = .049), and treatment in the OR, 78.5% (P < .0001). Patients with injuries to other
Table 2. STUDY VARIABLES VERSUS AGE GROUP
Pediatric Sample size Male gender Age, yr Race (n = 293)y White Other Injury event location (n = 289)y Home Other Transfer Hospital admission Treatment in OR Length of stay, days Dog owner (n = 285)y Family Other Breed Pit bull Labrador German shepherd Rottweiler Mix Mastiff Other Unknown
Adult
P Value
183 (58.5) 130 (41.5) 105 (57.4) 69 (53.1) .451 5.7 3.7 40.3 15.7 <.0001* .600 136 (81.9) 107 (84.3) 30 (18.1) 20 (15.7) .001*
158 (88.3) 21 (11.7) 141 (77) 140 (76.5) 134 (73.2) 1.6 3.8
81 (73.6) 29 (26.4) 110 (84.6) 65 (50) 49 (37.7) 1.7 5.2
104 (59.8) 70 (40.2)
45 (40.5) 66 (59.5)
33 (18) 16 (8.7) 9 (4.9) 15 (8.2) 15 (8.2) 8 (4.4) 33 (18) 54 (29.5)
34 (26.2) 5 (3.8) 11 (8.5) 4 (3.1) 3 (2.3) 3 (2.3) 18 (13.6) 52 (40)
.098 <.0001* <.0001* .146 .002*
.009*
Note: Data are presented as number (percentage) or mean standard deviation. Abbreviation: OR, operating room. * Statistically significant (P < .05). y If data are missing, the revised sample size is shown in parentheses. Hurst et al. Periorbital Dog Bite Injuries. J Oral Maxillofac Surg 2019.
HNF regions had the longest length of stay, 3.3 6.9 days (P = .038). Table 4 summarizes the primary analysis of interest, measuring the association between age and injury location. Isolated periorbital injuries were more common in the pediatric group (26.8%) than in adults (12.3%) (relative risk [RR], 2.2 [95% confidence interval (CI), 1.3 to 3.7]; P = .003). Injury to both periorbital and CTA regions also was more common in the pediatric group (23.5%) than in adults (11.5%) (RR, 2.0 [95% CI, 1.2 to 3.5]; P = .01). Isolated CTA injury was the most common injury pattern overall in both children (37.7%) and adults (55.4%), but children were at significantly less risk than adults (RR, 0.7 [95% CI, 0.5 to 0.9]; P = .002). Children also were less likely to sustain an injury to other HNF regions (RR, 0.6 [95% CI, 0.3 to 0.97]; P = .038).
5
HURST ET AL
Table 3. STUDY VARIABLES VERSUS INJURY LOCATION
Sample size Male gender Age, yr Race (n = 293)y White Other Injury event location (n = 289)y Home Other Transfer Hospital admission Treatment in OR Length of stay, days Dog owner (n = 285)y Family Other Breed Pit bull Labrador German shepherd Rottweiler Mix Mastiff Other Unknown
Periorbital Only
CTA Only
Both Periorbital and CTA
Neither Periorbital Nor CTA
65 41 (63.1) 15.6 20.7
141 73 (51.8) 23.0 20.3
58 31 (53.4) 13.9 16.7
49 29 (59.2) 24.9 19.7
51 (86.4) 8 (13.6)
114 (85.1) 20 (14.9)
42 (82.4) 9 (17.6)
36 (73.5) 13 (26.5)
59 (90.8) 6 (9.2) 57 (87.7) 50 (76.9) 51 (78.5) 1.5 5.6
100 (79.4) 26 (20.6) 110 (78) 82 (58.2) 71 (50.4) 1.4 3.4
46 (86.8) 7 (13.2) 48 (82.8) 38 (65.5) 38 (65.5) 1.0 1.2
34 (75.6) 11 (24.4) 36 (73.5) 35 (71.4) 23 (46.9) 3.3 6.9
39 (63.9) 22 (36.1)
57 (44.9) 70 (55.1)
29 (55.8) 23 (44.2)
24 (53.3) 21 (46.7)
8 (12.3) 3 (4.6) 4 (6.2) 4 (6.2) 5 (7.7) 3 (4.6) 12 (18.5) 26 (40)
32 (22.7) 9 (6.4) 11 (9.4) 8 (5.7) 6 (4.3) 3 (2.1) 26 (18.4) 46 (32.6)
16 (27.6) 8 (13.8) 8 (13.8) 2 (3.4) 4 (6.9) 1 (1.7) 6 (10.3) 21 (36.2)
11 (22.4) 1 (2) 1 (2) 5 (10.2) 3 (6.1) 4 (8.2) 7 (14.3) 13 (26.5)
P Value
.445 .002* .255
.104
.225 .049* <.0001* .038* .094
.186
Note: Data are presented as number (percentage) or mean standard deviation. Abbreviations: CTA, central target area; OR, operating room. * Statistically significant (P < .05). y If data are missing, the revised sample size is shown in parentheses. Hurst et al. Periorbital Dog Bite Injuries. J Oral Maxillofac Surg 2019.
Multiple comparisons of RR were performed between injury categories with adults as the reference group. The following comparisons showed significant findings: Compared with adults, pediatric patients were at greater risk of an injury to only the periorbital region than an injury to only the CTA (RR, 2.3 [95% CI, 1.4 to 3.7]; P = .001). Pediatric patients also were more likely to sustain an isolated periorbital injury than an injury to other HNF regions (RR, 1.9 [95% CI, 1.2 to
2.8]; P = .004). Compared with adults, pediatric patients were more likely to sustain an injury to both periorbital and CTA regions than an injury to only the CTA (RR, 2.2 [95% CI, 1.3 to 3.7]; P = .002) or an injury to other HNF regions (RR, 1.9 [95% CI, 1.2 to 2.9]; P = .006). Bivariate analysis of the frequency of injury to specific anatomic locations by age group is presented in Table 5. Among the periorbital structures, the eyelid,
Table 4. AGE GROUP VERSUS INJURY LOCATION
Pediatric (n = 183), n (%) Adult (n = 130), n (%) RR (95% CI) P value
Isolated Periorbital
Isolated CTA
Both Periorbital and CTA
Other HNF Region
49 (26.8) 16 (12.3) 2.2 (1.3-3.7) .003*
69 (37.7) 72 (55.4) 0.7 (0.5-0.9) .002*
43 (23.5) 15 (11.5) 2.0 (1.2-3.5) .01*
22 (12) 27 (20.8) 0.6 (0.3-0.97) .038*
Abbreviations: CI, confidence interval; CTA, central target area; HNF, head, neck, and face; RR, relative risk. * Statistically significant (P < .05). Hurst et al. Periorbital Dog Bite Injuries. J Oral Maxillofac Surg 2019.
6
PERIORBITAL DOG BITE INJURIES
Table 5. INJURY TO SPECIFIC STRUCTURES VERSUS AGE GROUP
Periorbital Eyebrow Eyelid Medial or lateral canthus Canaliculus Orbit and contents CTA Zygoma Buccal region Nose Lip
Pediatric, n (%)
Adult, n (%)
RR (95% CI)
P Value
17 (9.3) 78 (42.6) 42 (23) 38 (20.8) 5 (2.7)
7 (5.4) 22 (16.9) 12 (9.2) 4 (3.1) 3 (2.3)
1.5 (0.8-2.8) 2.3 (1.6-3.4) 2.1 (1.2-3.4) 4.9 (1.9-12.5) 1.1 (0.4-2.7)
.201 <.0001* .002* <.0001* .815
58 (31.7) 68 (37.2) 45 (24.6) 54 (29.5)
19 (14.6) 20 (15.4) 31 (23.8) 64 (49.2)
1.9 (1.3-2.9) 2.2 (1.4-3.2) 1.0 (0.7-1.5) 0.6 (0.5-0.8)
.001* <.0001* .880 <.0001*
Abbreviations: CI, confidence interval; CTA, central target area; RR, relative risk. * Statistically significant (P < .05). Hurst et al. Periorbital Dog Bite Injuries. J Oral Maxillofac Surg 2019.
medial or lateral canthus, and canaliculus were more commonly injured in pediatric patients (P = .0001 to .002). The eyelids were the most commonly injured structure in the periorbital region among both pediatric patients (42.6%) and adults (16.9%). No significant difference between age groups was seen for injuries to the eyebrow or orbit and orbital contents. We identified 3 cases of open globe injury, involving 1 pediatric patient and 2 adults. One of the adult patients ultimately lost vision in the eye. Additional injuries to the orbital contents included 2 cases of conjunctival laceration and 1 case of corneal abrasion. In 3 cases, patients sustained a fracture of the bony orbit. Within the CTA, buccal-region injuries were most common in pediatric patients (37.2%) and lip injuries were most common in adults (49.2%). Injuries to the zygomatic (P = .001) and buccal (P < .0001) regions were more common in the pediatric group, and lip injuries were more common in the adult group (P < .0001). No significant difference in the frequency of nasal injury was observed between children and adults, 24.5% and 23.8%, respectively, with an RR of 1.0 (95% CI, 0.7 to 1.5; P = .880). In Figure 1, the frequency of periorbital injury is further evaluated by age. Most patients with a periorbital injury were younger than 12 years, with the highest frequency among those younger than 4 years. The frequency of periorbital injury progressively decreased until age 12 years, where it remained low through adulthood.
Discussion The purpose of this study was to compare the risk of injury to the periorbital region or CTA between children and adults, with the null hypothesis that there
is no difference between groups. The specific aims of the study were to measure rates of periorbital and CTA injuries among pediatric and adult patients with HNF dog bite injuries, identify facial regions at increased risk of dog bite injury in children, and discuss implications for the prevention and management of these injuries. The results of this study refute the null hypothesis. Pediatric patients had twice the risk of periorbital injury compared with adults, even after considering the increased rates of HNF injury in children. This effect appears to be amplified in patients younger than 12 years. Conversely, the CTA was the most commonly injured facial region overall, but adults were at greater risk than children. These findings suggest that the increased risk of HNF injury in children may be more closely tied to increased rates of periorbital rather than CTA injury. Analysis of specific facial structures provides additional insight into why children may be at increased risk of periorbital injury. The nose is a particularly cogent point of comparison given its proximity to the eyes; its presence in the same vertical third of the face; and its apparent suitability as a target, owing to physical prominence and its central location. Although children were twice as likely as adults to have a periorbital injury, frequencies of nasal injury were nearly identical (RR, 1.0). If injury patterns were driven primarily by their shorter stature, pediatric patients also might be expected to have higher rates of injury to the nose, a prominent structure in the same vertical third as the eyes. This study’s findings build on findings in a similar population from the authors’ institution: Prendes et al27 compared pediatric and adult rates of periorbital injury across all patients with dog bite injuries, including those without HNF injuries. They found
7
HURST ET AL
FIGURE 1. Histogram of frequency of periorbital dog bite injury by patient age at time of injury. Hurst et al. Periorbital Dog Bite Injuries. J Oral Maxillofac Surg 2019.
that children were 4.2 times more likely than adults to sustain a periorbital injury. We found them twice as likely after including only those with HNF injuries. This isolated risk is unique to the periorbital region, independent of the likelihood of sustaining HNF injury. We found periorbital injury in 50.3% of pediatric patients with a dog bite to the HNF, greater than the 16% rate reported in 2 studies from tertiary pediatric centers.26,34 This difference could reflect a tendency for a trauma center to select for the most acute pediatric injuries. One study reported eyelid injuries as the most common ophthalmic injury and globe injury as rare,26 consistent with our findings. Low rates of injury to the globe have been attributed to the corneal reflex,25 and the protective architecture of the facial skeleton also is likely instrumental. Higher observed rates of injury to the eyelid and midface structures possibly reflect this protective role. Periorbital injury was associated with measures of injury severity such as transfer from an outside facility and treatment in the OR (Table 3). This finding speaks to the increased cost and complexity of care associated with these injuries and is consistent with prior studies showing that more complex injuries in younger patients were more likely to undergo repair in the OR.35 The CTA was the most commonly injured facial region in both adults and children, consistent with previously published reports.5,6,8,22-24,36 Children were not found to be uniquely at risk of injury to this area and were in fact at lower risk than adults. This is a new finding from our study. Dog behavior, child behavior, and anatomic factors may play a role in the higher risk of periorbital injury. Some authors have described an instinctual predilection for dogs to attack the face; the term ‘‘central target area’’ implies intentionality on the part of the dog to
target specific structures.5,7,21,22,24,25,29 Others have proposed that being at eye level could present a child as a potential threat or playmate and subject the child to face- and mouth-biting behavior that has been observed between dogs.6-8,31,32 Newborns have been described as being at risk of apparent predatory behavior by dogs.37 It is possible that the periorbital region is targeted in an attempt to blind or otherwise incapacitate a perceived threat. Child behavior has been referenced, including lack of education or awareness that could cause a child to approach a dog improperly, inadvertently provoke it, or interfere with food or other resources.16,38,39 In addition, a willingness to bring the face within a close distance to a dog or kiss a dog could be perceived as threatening or submissive.8,14,26,31,36,39,40 In fact, bending over a dog, putting the face close to the dog’s face, or gazing between a human and dog has been shown to immediately precede a dog bite to the face more often than stepping on the dog, tugging the dog’s hair or body, falling on the dog, or punishing the dog by hitting.19 Physical differences such as shorter stature and a larger ratio of head to body could make the HNF more available to attack in children.12,14,22,38,40-42 In addition, the eyes and orbits are proportionally larger in children,43,44 making them more exposed and possibly more threatening in appearance. Our finding that children younger than 12 years are at particularly increased risk of injury is consistent with previously published reports27,45 and could reflect important physical and behavioral changes associated with preadolescence. In addition, the peak incidence of periorbital injury between ages 0 and 4 years is consistent with previously published data regarding the HNF.15,16,41,46
8 Understanding that children are at increased risk of periorbital injury from dog bites should inform both the prevention and management of these injuries. From the perspective of management, the evaluation of a pediatric dog bite patient should include a thorough examination for evidence of ophthalmic injury as an integral part of the head and neck trauma examination. Injuries such as globe rupture or retrobulbar hematoma are rare but represent true ophthalmic emergencies. Specific emphasis should be placed on the presence or absence of canalicular involvement given the high rate of eyelid injury and possible complications, such as epiphora, if these injuries are not probed and intubated when appropriate.47 Failure to recognize and resuspend a canthal laceration could lead to telecanthus, and improper repair of a laceration involving the eyelid margin could lead to cicatrization, resulting in ectropion, lagophthalmos, or exposure keratopathy.48 The availability of oculoplastic, ophthalmologic, oral and maxillofacial, or other appropriately trained facial trauma consultant is therefore integral in the management of these patients. In addition, given the nature of the pediatric population, as well as examination techniques such as canalicular duct exploration, an examination under anesthesia may be indicated. Thus, access to the OR or pediatric sedation capabilities are often required for proper evaluation of these injuries, not to mention their ultimate repair. Our findings have important implications for the prevention of dog bite injuries in the pediatric population. The increased frequency of periorbital injury in children and the possible targeting of these structures create a powerful narrative among parents and dog owners and could help raise awareness of the issue. Children were more likely to be bitten in the HNF by a familiar dog at home, a finding consistent with previously published reports, which have characterized such attacks as happening to younger, provocative children by quiet, familiar dogs indoors with parents present.14,19,25,37 Public health and preventive measures should stress that young children not be left at eye level on the floor with a dog, with or without immediate adult supervision and regardless of the familiarity with the dog. In addition, potentially risky behaviors such as interfering with food, hugging, kissing, staring, or playing roughly should be discouraged. Educational programs including instructional videos, websites, and training with live dogs have been shown to be effective interventions to improve safety knowledge and behavior in children.38,39,49 Additional preventive efforts include National Dog Bite Prevention Week, sponsored by the American Veterinary Medical Association. This campaign features educational video content, social media messaging, and provider toolkits to promote awareness and skills
PERIORBITAL DOG BITE INJURIES
for the prevention of dog bites at a community level.50 The American Veterinary Medical Association Task Force on Canine Aggression and Human-Canine Interactions has called for a multidisciplinary approach to identify potential partners in these efforts.51 In this spirit, we propose a possible collaboration between National Dog Bite Prevention Week and National Facial Protection Month, sponsored by the American Association of Oral and Maxillofacial Surgeons and other professional groups.52 Both campaigns take place during the month of April. Bridging the interdisciplinary divide between veterinary, medical, and dental providers creates opportunities for synergy through shared experience and expanded reach. This study has several limitations. First, it was retrospective and relied on extraction of heterogeneous documentation of injuries and other clinical data input by a variety of providers at the time of the encounter. This was moderated by the fact that multiple services and providers examined and evaluated the patients, providing us with the ability to cross-check findings. Another limitation was the grouping of all patients younger than 18 years into a single pediatric age group, which was used interchangeably with the term ‘‘children.’’ This is a developmentally and physically heterogeneous group. As illustrated in Figure 1, there were few patients aged between 12 and 18 years, making this less likely to significantly impact data analysis. In addition, the study population was limited to patients presenting to a level 1 trauma center with trauma code activation, hospital admission, or transfer from an outside facility. This likely selected for more acute or extensive injuries. Some patients with more straightforward injuries likely did not seek medical attention, did not qualify for trauma code activation, or were treated in the community. In conclusion, our study found that pediatric patients were at greater risk of periorbital injury than adults, even after controlling for increased rates of HNF injury in children. No such increased risk was observed in the CTA, although this was the most frequently injured region in both the adult and pediatric groups. Although the explanation is likely multifactorial, the unique physical and behavioral characteristics of children, as well as preferential targeting of the periorbital region by dogs, could play a role in the increased rates of HNF injury observed in children. Further studies are indicated to better characterize the etiology and prevention of these injury patterns. Acknowledgments The authors thank Shauna L. Carson, Program Support Supervisor, Harborview Medical Center–Trauma Program, for helping us obtain the data from Harborview Medical Center.
HURST ET AL
References 1. Nygaard M, Dahlin LB: Dog bite injuries to the hand. J Plast Surg Hand Surg 45:96, 2011 2. Weiss HB, Friedman DI, Coben JH: Incidence of dog bite injuries treated in emergency departments. JAMA 279:51, 1998 3. Insurance Information Institute. Spotlight on: dog bite liability. Available at: https://www.iii.org/article/spotlight-on-dog-biteliability. Published 2017. Accessed March 17, 2018 4. Peters V, Sottiaux M, Appelboom J, Kahn A: Posttraumatic stress disorder after dog bites in children. J Pediatr 144:121, 2004 5. Palmer J, Rees M: Dog bites of the face: A 15 year review. Br J Plast Surg 36:315, 1983 6. Karlson TA: The incidence of facial injuries from dog bites. JAMA 251:3265, 1984 7. Gonnering RS: Ocular adnexal injury and complications in orbital dog bites. Ophthalmic Plast Reconstr Surg 3:231, 1987 8. Brogan TV, Bratton SL, Dowd MD, Hegenbarth MA: Severe dog bites in children. Pediatrics 96:947, 1995 9. Kahn A, Bauche P, Lamoureux J: Child victims of dog bites treated in emergency departments: A prospective survey. Eur J Pediatr 162:254, 2003 10. Kahn A, Robert E, Piette D, et al: Prevalence of dog bites in children: A telephone survey. Eur J Pediatr 163:424, 2004 11. Schalamon J, Ainoedhofer H, Singer G, et al: Analysis of dog bites in children who are younger than 17 years. Pediatrics 117:e374, 2006 12. Dwyer JP, Douglas TS, van As AB: Dog bite injuries in children— A review of data from a South African paediatric trauma unit. S Afr Med J 97:597, 2007 13. Steele MT, Ma OJ, Nakase J, et al: Epidemiology of animal exposures presenting to emergency departments. Acad Emerg Med 14:398, 2007 14. Kaye AE, Belz JM, Kirschner RE: Pediatric dog bite injuries: A 5year review of the experience at the Children’s Hospital of Philadelphia. Plast Reconstr Surg 124:551, 2009 15. Chen HH, Neumeier AT, Davies BW, Durairaj VD: Analysis of pediatric facial dog bites. Craniomaxillofac Trauma Reconstr 6: 225, 2013 16. Chiam SC, Solanki NS, Lodge M, et al: Retrospective review of dog bite injuries in children presenting to a South Australian tertiary children’s hospital emergency department. J Paediatr Child Health 50:791, 2014 17. Chhabra S, Chhabra N, Gaba S: Maxillofacial injuries due to animal bites. J Maxillofac Oral Surg 14:142, 2015 18. O’Brien DC, Andre TB, Robinson AD, et al: Dog bites of the head and neck: An evaluation of a common pediatric trauma and associated treatment. Am J Otolaryngol 36:32, 2015 19. Rezac P, Rezac K, Slama P: Human behavior preceding dog bites to the face. Vet J 206:284, 2015 20. Gurunluoglu R, Glasgow M, Arton J, Bronsert M: Retrospective analysis of facial dog bite injuries at a Level I trauma center in the Denver metro area. J Trauma Acute Care Surg 76:1294, 2014 21. Wei LA, Chen HH, Hink EM, Durairaj VD: Pediatric facial fractures from dog bites. Ophthalmic Plast Reconstr Surg 29:179, 2013 22. Eppley BL, Schleich AR: Facial dog bite injuries in children: Treatment and outcome assessment. J Craniofac Surg 24:384, 2013 23. Monroy A, Behar P, Nagy M, et al: Head and neck dog bites in children. Otolaryngol Head Neck Surg 140:354, 2009 24. Lackmann GM, Draf W, Isselstein G, Tollner U: Surgical treatment of facial dog bite injuries in children. J Craniomaxillofac Surg 20:81, 1992 25. Erickson BP, Feng PW, Liao SD, et al: Dog bite injuries of the eye and ocular adnexa. Orbit 38:43, 2019 26. Bratton EM, Golas L, Wei LA, et al: Ophthalmic manifestations of facial dog bites in children. Ophthalmic Plast Reconstr Surg 34: 106, 2018 27. Prendes MA, Jian-Amadi A, Chang SH, Shaftel SS: Ocular trauma from dog bites: Characterization, associations, and treatment patterns at a regional level I trauma center over 11 years. Ophthalmic Plast Reconstr Surg 32:279, 2016
9 28. Wladis EJ, Dewan MA: Periorbital trauma from pit bull terrier attacks. Orbit 31:200, 2012 29. Savar A, Kirszrot J, Rubin PA: Canalicular involvement in dog bite related eyelid lacerations. Ophthalmic Plast Reconstr Surg 24: 296, 2008 30. Slonim CB: Dog bite-induced canalicular lacerations: A review of 17 cases. Ophthalmic Plast Reconstr Surg 12:218, 1996 31. Iazzetti L: Anticipatory guidance: Having a dog in the family. J Pediatr Health Care 12:73, 1998 32. Overall KL: Understanding and treating canine dominance aggression: An overview. Vet Med 94:976, 1999 33. Schultz RC, McMaster WC: The treatment of dog bite injuries, especially those of the face. Plast Reconstr Surg 49:494, 1972 34. McHeik JN, Vergnes P, Bondonny JM: Treatment of facial dog bite injuries in children: A retrospective study. J Pediatr Surg 35:580, 2000 35. Wu PS, Beres A, Tashjian DB, Moriarty KP: Primary repair of facial dog bite injuries in children. Pediatr Emerg Care 27:801, 2011 36. Toure G, Angoulangouli G, Meningaud JP: Epidemiology and classification of dog bite injuries to the face: A prospective study of 108 patients. J Plast Reconstr Aesthet Surg 68:654, 2015 37. Reisner I. The epidemiology & public health issue of dog bites: Who, where & why, Presented at the Western Veterinary Conference, Las Vegas; February 17, 2013 38. Shen J, Rouse J, Godbole M, et al: Systematic review: Interventions to educate children about dog safety and prevent pediatric dog-bite injuries: A meta-analytic review. J Pediatr Psychol 42: 779, 2017 39. Dixon CA, Mahabee-Gittens EM, Hart KW, Lindsell CJ: Dog bite prevention: An assessment of child knowledge. J Pediatr 160: 337, 2012 40. Bernardo LM, Gardner MJ, Rosenfield RL, et al: A comparison of dog bite injuries in younger and older children treated in a pediatric emergency department. Pediatr Emerg Care 18:247, 2002 41. Jaindl M, Grunauer J, Platzer P, et al: The management of bite wounds in children—A retrospective analysis at a level I trauma centre. Injury 43:2117, 2012 42. Hon KL, Fu CC, Chor CM, et al: Issues associated with dog bite injuries in children and adolescents assessed at the emergency department. Pediatr Emerg Care 23:445, 2007 43. Wheeler J, Phillips J: Pediatric facial fractures and potential longterm growth disturbances. Craniomaxillofac Trauma Reconstr 4: 43, 2011 44. Farkas LG, Posnick JC, Hreczko TM: Growth patterns of the face: A morphometric study. Cleft Palate Craniofac J 29:308, 1992 45. Foster MD, Hudson JW: Contemporary update on the treatment of dog bite: Injuries to the oral and maxillofacial region. J Oral Maxillofac Surg 73:935, 2015 46. Chun YT, Berkelhamer JE, Herold TE: Dog bites in children less than 4 years old. Pediatrics 69:119, 1982 47. Kennedy RH, May J, Dailey J, Flanagan JC, Canalicular laceration. An 11-year epidemiologic and clinical study. Ophthalmic Plast Reconstr Surg 6:46, 1990 48. Nerad J: Eyelid and orbital trauma, in Techniques in Ophthalmic Plastic Surgery—A Personal Tutorial. Amsterdam, Elsevier, 2010, pp 355–390 49. Meints K: Test yourself—A first assessment of the dog bite prevention project "Blue Dog". Part I: Education—The "Blue Dog" is born; Part II: The "Blue Dog"—Do children really get the message?. Presented at the World Small Animal Veterinary Association World Congress, Prague, Czech Republic; October 11, 2006 50. American Veterinary Medical Association. National Dog Bite Prevention Week. Available at: https://www.avma.org/Events/ pethealth/Pages/Dog-Bite-Prevention-Week.aspx. Published 2018. Accessed March 17, 2018 51. American Veterinary Medical Association Task Force on Canine Aggression and Human-Canine Interactions. A community approach to dog bite prevention. J Am Vet Med Assoc 218: 1732, 2001 52. American Association of Oral and Maxillofacial Surgeons. April is National Facial Protection Month. Available at: https://www. aaoms.org/media/april-is-national-facial-protection-month. Published 2018. Accessed March 17, 2018
Children Have an Increased Risk of Periorbital Dog Bite Injuries Philip J. Hurst, DDS, MD,* Marcus Ji Hoon Hwang, BS,y Thomas B. Dodson, DMD, MPH,z and Jasjit K. Dillon, DDS, MBBS, FDSRCSx Purpose:
Dog bites to the head, neck, and face (HNF) disproportionately affect children. It is unclear if specific facial regions are uniquely at risk in this population. Both the periorbital region and the central target area (CTA; nose, lips, and cheeks) have been proposed as being at increased risk. This study addressed the following question: Among individuals sustaining a dog bite injury to the HNF, are children, when compared with adults, at greater risk of injury to the periorbital region or CTA?
Patients and Methods:
Using a retrospective cohort design, we enrolled a sample composed of patients presenting with HNF dog bite injuries. The predictor variable was age category: pediatric or adult ($18 years). The primary outcome variable was injury location: isolated periorbital, isolated CTA, both periorbital and CTA, or other HNF location (neither periorbital nor CTA). Other variables included specific facial structures injured, demographic characteristics, injury circumstances, and clinical course. Descriptive and bivariate statistics were calculated.
Results:
The sample consisted of 183 pediatric (58.5%) and 130 adult (41.5%) patients. Isolated periorbital injuries were more common in the pediatric group (relative risk [RR], 2.2 [95% confidence interval (CI), 1.3 to 3.7]; P = .003), as was injury to both periorbital and CTA regions (RR, 2.0 [95% CI, 1.2 to 3.5]; P = .01). Isolated CTA injury was the most common injury pattern overall in both children and adults, but children were at significantly less risk than adults (RR, 0.7 [95% CI, 0.5 to 0.9]; P = .002). Children were also less likely to sustain an injury to other HNF regions (RR, 0.6 [95% CI, 0.3 to 0.97]; P = .038). Conclusions: Children were twice as likely as adults to incur a periorbital dog bite, but no such increased risk was observed for the CTA. This may reflect unique anatomic risks in children or targeting of their eyes by attacking dogs, contributing to higher rates of HNF injury in this population. Public health measures should recommend against children being at eye level with dogs, even if familiar or under adult supervision. Ó 2019 American Association of Oral and Maxillofacial Surgeons J Oral Maxillofac Surg -:1-9, 2019
The risk of being bitten by a domestic animal during one’s lifetime is estimated to approach 50%. Of these domestic animal bites, dog bites account for 80 to 90%,1 comprising approximately 1% of all emergency visits
to hospitals in the United States.2 Dog bites accounted for more than one third of all homeowner insurance liability claims paid in 2012, with an average paid claim of $29,752, totaling more than $489 million in claims.3
Received from University of Washington, Seattle, WA.
dation grant. Neither foundation has a conflict of interest with this
*Resident, Department of Oral and Maxillofacial Surgery.
manuscript.
yDental Student, School of Dentistry.
Address correspondence and reprint requests to Dr Hurst:
zProfessor and Chair, Department of Oral and Maxillofacial
Department of Oral and Maxillofacial Surgery, University of Washing-
Surgery.
ton, 1959 NE Pacific St, HSB Room B 241, Box 357134, Seattle, WA
xClinical Associate Professor and Program Director, Department
98195; e-mail: [email protected]
of Oral and Maxillofacial Surgery. This project was supported in part by the Department of Oral and
Received May 22 2019 Accepted August 22 2019
Maxillofacial Surgery Laboratory for Applied Clinical Investigation
Ó 2019 American Association of Oral and Maxillofacial Surgeons
and the Department of Oral and Maxillofacial Surgery Research
0278-2391/19/31046-8
and Training Fund.
https://doi.org/10.1016/j.joms.2019.08.021
Conflict of Interest Disclosures: Dr. Dillon is the recipient of an Oral and Maxillofacial Surgery Foundation and Osteoscience Foun-
1
2 Dog bite wounds are associated with a host of potential negative consequences including pain, infection, cosmetic disfigurement, psychological trauma, and financial burden.4 These consequences are amplified when the injuries involve the region of the head, neck, and face (HNF), which is the most common location for dog bite injuries in the pediatric population.5-18 Several researchers have identified a central target area (CTA) of the face, in which the nose, cheeks, and lips are most often injured by dog bites, particularly in children.5,17,19-24 Others have focused on ocular and periorbital injury and found children to be at greater risk.7,25-30 Proposed explanations have focused on the HNF region broadly and include smaller stature and convenient proximity of the head to the dog,22 inappropriate behaviors such as kissing,31 or instinctual targeting of these structures by an attacking dog.29 Some researchers have posited that because younger children are at eye level with dogs, their staring is perceived as threatening.32 Others have considered higher rates of periorbital injury in children as a byproduct of the aforementioned factors driving the higher frequency of HNF injury in the population.27 No study has directly compared injury to the periorbital region and CTA between age groups while accounting for children’s higher baseline risk of HNF injury. Thus, the possibility that children’s eyes or other facial structures are uniquely at risk of dog bite injury or play a role in their increased rates of HNF injury has not been well evaluated. The purpose of this study was to answer the following research question: Among individuals sustaining an HNF dog bite injury, are children, when compared with adults, at greater risk of injury to the periorbital region or CTA? The null hypothesis was that there is no difference in the rate of injury to the periorbital region or CTA between children and adults. The specific aims of the study were 1) to measure and compare the frequencies of periorbital and CTA injuries among individuals with HNF dog bites seeking medical attention and compare these rates between pediatric and adult populations; 2) to identify facial regions at increased risk of dog bite injury in children compared with adults; and 3) to discuss implications for the prevention and management of these injuries.
Patients and Methods We designed and implemented a retrospective cohort study and enrolled a study sample derived from patients who presented to the emergency department of Harborview Medical Center (HMC), Seattle, Washington, for the evaluation and management of
PERIORBITAL DOG BITE INJURIES
dog bite injuries to the HNF between January 1, 2000, and December 31, 2015. To identify study patients, we accessed the HMC trauma registry, which includes patients who have at least 1 trauma-specific diagnostic code and meet at least 1 of the following inclusion criteria: full or modified trauma team activation, admission to the hospital, dead on arrival, transfer from another acute or urgent care facility by emergency medical services, or flight to the scene by emergency medical services. This registry was queried for injuries with a mechanism of ‘‘bite,’’ and results were further refined using International Classification of Diseases, Ninth Revision codes 906.0 to 906.5 and bite codes in the International Classification of Diseases, Tenth Revision range of W53 to W59, as well as code W61. To be included in the sample, patients had to have sustained at least 1 dog bite injury to the HNF region. The HNF region was defined as including any anatomic structure superior to the clavicle. Injury was defined as any disruption of normal tissue architecture, appearance, or function as documented in the clinical record, including but not limited to laceration, puncture, abrasion, defect, scratch, contusion, avulsion, fracture, degloving, transection, bite, tear, or amputation. Patients with duplicate records, insufficient data, or without dog bite injuries to the HNF were excluded from the sample. This study followed the principles outlined in the Declaration of Helsinki and was approved by the University of Washington Institutional Review Board (No. 00003323). The primary predictor variable was age classification: pediatric or adult. Patients younger than 18 years at the time of injury were classified as pediatric patients, whereas those aged 18 years or older were classified as adults. Age 18 years was chosen given its use as a widely accepted definition of adulthood in the United States in medical and legal contexts and precedent in similar studies.23,33 The term ‘‘children’’ also was used to refer to the pediatric age group in this study. The primary outcome variable was the anatomic location of the dog bite injury into 1 of 4 categories: isolated periorbital, isolated CTA, both periorbital and CTA, and other HNF location (neither periorbital nor CTA). Injury to the periorbital region was defined as an injury to at least 1 of the following structures: eyebrow, eyelid, medial or lateral canthus, canalicular system, or orbit and its contents such as the extraocular muscles, conjunctiva, cornea, or globe. Injury to the CTA was defined as injury to at least 1 of the following: lip, nose, zygomatic region, or buccal region. Patients with an injury to at least 1 structure in both the periorbital region and CTA region were classified as having both periorbital and CTA injuries.
3
HURST ET AL
Patients with injuries to other HNF structures such as the neck, scalp, forehead, ear, or mandible but with no injury to the periorbital region or CTA were classified as having ‘‘other HNF’’ injuries. Patients with concomitant injuries outside the HNF region were not excluded from this study, but these injuries were not included in data analysis. Other study variables included patient age at the time of injury, gender, race, dog ownership, breed of dog, and location of injury event. Race was documented as listed in the medical record and was classified as follows: white, Hispanic or Latino, Black, Asian or Pacific Islander, or Native American. For the purposes of statistical analysis, race at times was grouped as white and ‘‘other,’’ consisting of the remaining 4 racial categories. The owner of the attacking dog was determined, when possible, from the medical record and was classified as family or other. ‘‘Family’’ was defined to include dogs owned by the patient or a relative. ‘‘Other’’ was defined to include the following categories: friend, neighbor, stranger, police, or stray. A friend was defined as someone referred to as a ‘‘friend’’ or ‘‘acquaintance’’ of the patient or patient’s family. A neighbor was defined as someone living near the patient who was not otherwise specified as a friend. A stranger was defined as someone unfamiliar or unknown to the patient. ‘‘Stray’’ was used in reference to dogs without an apparent owner. Injury event location was classified as home or other. ‘‘Home’’ was defined as occurring inside, in the yard, or within the property boundaries of any home or dwelling regardless of it being the patient’s home. ‘‘Other’’ was defined to include any location not meeting the definition of home. Breed of dog was recorded as reported in the medical record. The following breeds were grouped as ‘‘other’’: terrier, Doberman, retriever, bulldog, husky, Alaskan malamute, chow, Australian shepherd, Rhodesian ridgeback, Great Dane, Dalmatian, poodle, Saint Bernard, Portuguese water dog, Akita, boxer, border collie, beagle, Shar-Pei, dachshund, Chihuahua, Springer spaniel, cocker spaniel, German shorthair pointer, greyhound, and basenji. Cases of multiple attacking dogs were also deemed ‘‘other.’’ If no breed was reported or the breed was not known to witnesses, it was documented as unknown. Secondary outcome variables related to the clinical course were documented: transfer status, hospital admission status, whether the patient was treated in the operating room (OR), and length of stay. ‘‘Transfer status’’ referred to whether the patient was transferred from an outside facility rather than presenting directly to HMC. Patients were documented as having been admitted if they were admitted to HMC at any time for injuries related to a dog bite attack. Patients who were discharged directly from the emergency depart-
ment or postanesthesia care unit were not classified as having been admitted. Any patient who was treated in the OR at HMC for injuries related to a dog bite attack, as documented by a formal operation report and anesthesia record, was classified as having been treated in the OR. This was defined to include patients who were not treated in the OR on initial presentation but were treated in the OR during a subsequent encounter. Treatment in the emergency department, even with procedural sedation, was insufficient to qualify as having been treated in the OR. Length of stay was defined as 0 days for any patient who did not meet the criteria for having been admitted to the hospital. For a patient who was admitted, length of stay was defined as the number of midnights the patient remained an inpatient for any admission related to dog bite injuries. Data analysis was performed with Microsoft Excel software (Microsoft, Redmond, WA) and SPSS software (version 15; IBM, Armonk, NY). Univariate, bivariate, and multivariate statistics were computed. P < .05 was considered statistically significant.
Results The study sample was composed of 313 patients with dog bite injuries to the HNF (Table 1). The mean age was 20.1 years (range, 0.7 to 94.4 years), and there were 174 male patients (55.6%). The study sample comprised 183 pediatric (58.5%) and 130 adult (41.5%) patients. A comparison of study variables against the primary predictor variable, age group, is presented in Table 2. The mean age was 5.7 years in the pediatric group and 40.3 years in adults. No significant differences were observed between the pediatric and adult groups for gender, race, transfer from an outside facility, and length of stay. Greater than 80% of patients in both age groups were white. For 26 patients, race was not indicated in the medical record; therefore, statistics were calculated with a reduced sample size for this variable. Most children and most adults were transferred from an outside facility. Pediatric patients were significantly more likely to be bitten by a family-owned dog (P = .002), to be bitten in a home setting (P = .001), to be admitted to the hospital (P < .0001), and to undergo treatment in the OR (P < .0001). The pit bull was the most commonly named breed to be involved in both pediatric (18%) and adult (26.2%) injuries. In many cases, the breed was not known or documented, and the ‘‘unknown’’ category was the largest category in both the pediatric and adult groups. In Table 3, study variables are compared with the primary outcome variable, injury location. No significant differences were found between injury location
4
PERIORBITAL DOG BITE INJURIES
Table 1. DESCRIPTIVE SUMMARY OF STUDY SAMPLE
Study Variable
Descriptive Statistics
Sample size Male gender Age, yr Age group Pediatric Adult Race (n = 293)* White Other Injury location Isolated periorbital Isolated CTA Both periorbital and CTA Other HNF region Injury event location (n = 289)* Home Other Transfer Hospital admission Treatment in OR Length of stay, days Dog owner (n = 285)* Family Other Breed Pit bull Labrador German shepherd Rottweiler Mix Mastiff Other Unknown
313 174 (55.6) 20.1 20 (0.7-94.4) 183 (58.5) 130 (41.5) 243 (82.9) 50 (17.1) 65 (20.8) 141 (45) 58 (18.5) 49 (15.7) 239 (82.7) 50 (17.3) 251 (80.2) 205 (65.5) 183 (58.5) 1.64 4.3 (0-45) 149 (52.3) 136 (47.7) 67 (21.4) 21 (6.7) 20 (6.4) 19 (6.1) 18 (5.8) 11 (3.5) 51 (16.3) 106 (33.9)
Note: Data are presented as number (percentage) or mean standard deviation (range). Abbreviations: CTA, central target area; HNF, head, neck, and face; OR, operating room. * If data are missing, the revised sample size is shown in parentheses. Hurst et al. Periorbital Dog Bite Injuries. J Oral Maxillofac Surg 2019.
and gender, race, injury event location, transfer status, dog owner, or breed. Significant differences were found for age, hospital admission, treatment in the OR, and length of stay. Younger age was associated with isolated periorbital injuries (15.6 20.7 years) and both periorbital and CTA injuries (13.9 16.7 years), whereas isolated CTA injuries (23.0 20.3 years) and other HNF injuries (24.9 19.7 years) were associated with a higher mean age (P = .002). Patients with isolated periorbital injuries had the highest rates of hospital admission, 76.9% (P = .049), and treatment in the OR, 78.5% (P < .0001). Patients with injuries to other
Table 2. STUDY VARIABLES VERSUS AGE GROUP
Pediatric Sample size Male gender Age, yr Race (n = 293)y White Other Injury event location (n = 289)y Home Other Transfer Hospital admission Treatment in OR Length of stay, days Dog owner (n = 285)y Family Other Breed Pit bull Labrador German shepherd Rottweiler Mix Mastiff Other Unknown
Adult
P Value
183 (58.5) 130 (41.5) 105 (57.4) 69 (53.1) .451 5.7 3.7 40.3 15.7 <.0001* .600 136 (81.9) 107 (84.3) 30 (18.1) 20 (15.7) .001*
158 (88.3) 21 (11.7) 141 (77) 140 (76.5) 134 (73.2) 1.6 3.8
81 (73.6) 29 (26.4) 110 (84.6) 65 (50) 49 (37.7) 1.7 5.2
104 (59.8) 70 (40.2)
45 (40.5) 66 (59.5)
33 (18) 16 (8.7) 9 (4.9) 15 (8.2) 15 (8.2) 8 (4.4) 33 (18) 54 (29.5)
34 (26.2) 5 (3.8) 11 (8.5) 4 (3.1) 3 (2.3) 3 (2.3) 18 (13.6) 52 (40)
.098 <.0001* <.0001* .146 .002*
.009*
Note: Data are presented as number (percentage) or mean standard deviation. Abbreviation: OR, operating room. * Statistically significant (P < .05). y If data are missing, the revised sample size is shown in parentheses. Hurst et al. Periorbital Dog Bite Injuries. J Oral Maxillofac Surg 2019.
HNF regions had the longest length of stay, 3.3 6.9 days (P = .038). Table 4 summarizes the primary analysis of interest, measuring the association between age and injury location. Isolated periorbital injuries were more common in the pediatric group (26.8%) than in adults (12.3%) (relative risk [RR], 2.2 [95% confidence interval (CI), 1.3 to 3.7]; P = .003). Injury to both periorbital and CTA regions also was more common in the pediatric group (23.5%) than in adults (11.5%) (RR, 2.0 [95% CI, 1.2 to 3.5]; P = .01). Isolated CTA injury was the most common injury pattern overall in both children (37.7%) and adults (55.4%), but children were at significantly less risk than adults (RR, 0.7 [95% CI, 0.5 to 0.9]; P = .002). Children also were less likely to sustain an injury to other HNF regions (RR, 0.6 [95% CI, 0.3 to 0.97]; P = .038).
5
HURST ET AL
Table 3. STUDY VARIABLES VERSUS INJURY LOCATION
Sample size Male gender Age, yr Race (n = 293)y White Other Injury event location (n = 289)y Home Other Transfer Hospital admission Treatment in OR Length of stay, days Dog owner (n = 285)y Family Other Breed Pit bull Labrador German shepherd Rottweiler Mix Mastiff Other Unknown
Periorbital Only
CTA Only
Both Periorbital and CTA
Neither Periorbital Nor CTA
65 41 (63.1) 15.6 20.7
141 73 (51.8) 23.0 20.3
58 31 (53.4) 13.9 16.7
49 29 (59.2) 24.9 19.7
51 (86.4) 8 (13.6)
114 (85.1) 20 (14.9)
42 (82.4) 9 (17.6)
36 (73.5) 13 (26.5)
59 (90.8) 6 (9.2) 57 (87.7) 50 (76.9) 51 (78.5) 1.5 5.6
100 (79.4) 26 (20.6) 110 (78) 82 (58.2) 71 (50.4) 1.4 3.4
46 (86.8) 7 (13.2) 48 (82.8) 38 (65.5) 38 (65.5) 1.0 1.2
34 (75.6) 11 (24.4) 36 (73.5) 35 (71.4) 23 (46.9) 3.3 6.9
39 (63.9) 22 (36.1)
57 (44.9) 70 (55.1)
29 (55.8) 23 (44.2)
24 (53.3) 21 (46.7)
8 (12.3) 3 (4.6) 4 (6.2) 4 (6.2) 5 (7.7) 3 (4.6) 12 (18.5) 26 (40)
32 (22.7) 9 (6.4) 11 (9.4) 8 (5.7) 6 (4.3) 3 (2.1) 26 (18.4) 46 (32.6)
16 (27.6) 8 (13.8) 8 (13.8) 2 (3.4) 4 (6.9) 1 (1.7) 6 (10.3) 21 (36.2)
11 (22.4) 1 (2) 1 (2) 5 (10.2) 3 (6.1) 4 (8.2) 7 (14.3) 13 (26.5)
P Value
.445 .002* .255
.104
.225 .049* <.0001* .038* .094
.186
Note: Data are presented as number (percentage) or mean standard deviation. Abbreviations: CTA, central target area; OR, operating room. * Statistically significant (P < .05). y If data are missing, the revised sample size is shown in parentheses. Hurst et al. Periorbital Dog Bite Injuries. J Oral Maxillofac Surg 2019.
Multiple comparisons of RR were performed between injury categories with adults as the reference group. The following comparisons showed significant findings: Compared with adults, pediatric patients were at greater risk of an injury to only the periorbital region than an injury to only the CTA (RR, 2.3 [95% CI, 1.4 to 3.7]; P = .001). Pediatric patients also were more likely to sustain an isolated periorbital injury than an injury to other HNF regions (RR, 1.9 [95% CI, 1.2 to
2.8]; P = .004). Compared with adults, pediatric patients were more likely to sustain an injury to both periorbital and CTA regions than an injury to only the CTA (RR, 2.2 [95% CI, 1.3 to 3.7]; P = .002) or an injury to other HNF regions (RR, 1.9 [95% CI, 1.2 to 2.9]; P = .006). Bivariate analysis of the frequency of injury to specific anatomic locations by age group is presented in Table 5. Among the periorbital structures, the eyelid,
Table 4. AGE GROUP VERSUS INJURY LOCATION
Pediatric (n = 183), n (%) Adult (n = 130), n (%) RR (95% CI) P value
Isolated Periorbital
Isolated CTA
Both Periorbital and CTA
Other HNF Region
49 (26.8) 16 (12.3) 2.2 (1.3-3.7) .003*
69 (37.7) 72 (55.4) 0.7 (0.5-0.9) .002*
43 (23.5) 15 (11.5) 2.0 (1.2-3.5) .01*
22 (12) 27 (20.8) 0.6 (0.3-0.97) .038*
Abbreviations: CI, confidence interval; CTA, central target area; HNF, head, neck, and face; RR, relative risk. * Statistically significant (P < .05). Hurst et al. Periorbital Dog Bite Injuries. J Oral Maxillofac Surg 2019.
6
PERIORBITAL DOG BITE INJURIES
Table 5. INJURY TO SPECIFIC STRUCTURES VERSUS AGE GROUP
Periorbital Eyebrow Eyelid Medial or lateral canthus Canaliculus Orbit and contents CTA Zygoma Buccal region Nose Lip
Pediatric, n (%)
Adult, n (%)
RR (95% CI)
P Value
17 (9.3) 78 (42.6) 42 (23) 38 (20.8) 5 (2.7)
7 (5.4) 22 (16.9) 12 (9.2) 4 (3.1) 3 (2.3)
1.5 (0.8-2.8) 2.3 (1.6-3.4) 2.1 (1.2-3.4) 4.9 (1.9-12.5) 1.1 (0.4-2.7)
.201 <.0001* .002* <.0001* .815
58 (31.7) 68 (37.2) 45 (24.6) 54 (29.5)
19 (14.6) 20 (15.4) 31 (23.8) 64 (49.2)
1.9 (1.3-2.9) 2.2 (1.4-3.2) 1.0 (0.7-1.5) 0.6 (0.5-0.8)
.001* <.0001* .880 <.0001*
Abbreviations: CI, confidence interval; CTA, central target area; RR, relative risk. * Statistically significant (P < .05). Hurst et al. Periorbital Dog Bite Injuries. J Oral Maxillofac Surg 2019.
medial or lateral canthus, and canaliculus were more commonly injured in pediatric patients (P = .0001 to .002). The eyelids were the most commonly injured structure in the periorbital region among both pediatric patients (42.6%) and adults (16.9%). No significant difference between age groups was seen for injuries to the eyebrow or orbit and orbital contents. We identified 3 cases of open globe injury, involving 1 pediatric patient and 2 adults. One of the adult patients ultimately lost vision in the eye. Additional injuries to the orbital contents included 2 cases of conjunctival laceration and 1 case of corneal abrasion. In 3 cases, patients sustained a fracture of the bony orbit. Within the CTA, buccal-region injuries were most common in pediatric patients (37.2%) and lip injuries were most common in adults (49.2%). Injuries to the zygomatic (P = .001) and buccal (P < .0001) regions were more common in the pediatric group, and lip injuries were more common in the adult group (P < .0001). No significant difference in the frequency of nasal injury was observed between children and adults, 24.5% and 23.8%, respectively, with an RR of 1.0 (95% CI, 0.7 to 1.5; P = .880). In Figure 1, the frequency of periorbital injury is further evaluated by age. Most patients with a periorbital injury were younger than 12 years, with the highest frequency among those younger than 4 years. The frequency of periorbital injury progressively decreased until age 12 years, where it remained low through adulthood.
Discussion The purpose of this study was to compare the risk of injury to the periorbital region or CTA between children and adults, with the null hypothesis that there
is no difference between groups. The specific aims of the study were to measure rates of periorbital and CTA injuries among pediatric and adult patients with HNF dog bite injuries, identify facial regions at increased risk of dog bite injury in children, and discuss implications for the prevention and management of these injuries. The results of this study refute the null hypothesis. Pediatric patients had twice the risk of periorbital injury compared with adults, even after considering the increased rates of HNF injury in children. This effect appears to be amplified in patients younger than 12 years. Conversely, the CTA was the most commonly injured facial region overall, but adults were at greater risk than children. These findings suggest that the increased risk of HNF injury in children may be more closely tied to increased rates of periorbital rather than CTA injury. Analysis of specific facial structures provides additional insight into why children may be at increased risk of periorbital injury. The nose is a particularly cogent point of comparison given its proximity to the eyes; its presence in the same vertical third of the face; and its apparent suitability as a target, owing to physical prominence and its central location. Although children were twice as likely as adults to have a periorbital injury, frequencies of nasal injury were nearly identical (RR, 1.0). If injury patterns were driven primarily by their shorter stature, pediatric patients also might be expected to have higher rates of injury to the nose, a prominent structure in the same vertical third as the eyes. This study’s findings build on findings in a similar population from the authors’ institution: Prendes et al27 compared pediatric and adult rates of periorbital injury across all patients with dog bite injuries, including those without HNF injuries. They found
7
HURST ET AL
FIGURE 1. Histogram of frequency of periorbital dog bite injury by patient age at time of injury. Hurst et al. Periorbital Dog Bite Injuries. J Oral Maxillofac Surg 2019.
that children were 4.2 times more likely than adults to sustain a periorbital injury. We found them twice as likely after including only those with HNF injuries. This isolated risk is unique to the periorbital region, independent of the likelihood of sustaining HNF injury. We found periorbital injury in 50.3% of pediatric patients with a dog bite to the HNF, greater than the 16% rate reported in 2 studies from tertiary pediatric centers.26,34 This difference could reflect a tendency for a trauma center to select for the most acute pediatric injuries. One study reported eyelid injuries as the most common ophthalmic injury and globe injury as rare,26 consistent with our findings. Low rates of injury to the globe have been attributed to the corneal reflex,25 and the protective architecture of the facial skeleton also is likely instrumental. Higher observed rates of injury to the eyelid and midface structures possibly reflect this protective role. Periorbital injury was associated with measures of injury severity such as transfer from an outside facility and treatment in the OR (Table 3). This finding speaks to the increased cost and complexity of care associated with these injuries and is consistent with prior studies showing that more complex injuries in younger patients were more likely to undergo repair in the OR.35 The CTA was the most commonly injured facial region in both adults and children, consistent with previously published reports.5,6,8,22-24,36 Children were not found to be uniquely at risk of injury to this area and were in fact at lower risk than adults. This is a new finding from our study. Dog behavior, child behavior, and anatomic factors may play a role in the higher risk of periorbital injury. Some authors have described an instinctual predilection for dogs to attack the face; the term ‘‘central target area’’ implies intentionality on the part of the dog to
target specific structures.5,7,21,22,24,25,29 Others have proposed that being at eye level could present a child as a potential threat or playmate and subject the child to face- and mouth-biting behavior that has been observed between dogs.6-8,31,32 Newborns have been described as being at risk of apparent predatory behavior by dogs.37 It is possible that the periorbital region is targeted in an attempt to blind or otherwise incapacitate a perceived threat. Child behavior has been referenced, including lack of education or awareness that could cause a child to approach a dog improperly, inadvertently provoke it, or interfere with food or other resources.16,38,39 In addition, a willingness to bring the face within a close distance to a dog or kiss a dog could be perceived as threatening or submissive.8,14,26,31,36,39,40 In fact, bending over a dog, putting the face close to the dog’s face, or gazing between a human and dog has been shown to immediately precede a dog bite to the face more often than stepping on the dog, tugging the dog’s hair or body, falling on the dog, or punishing the dog by hitting.19 Physical differences such as shorter stature and a larger ratio of head to body could make the HNF more available to attack in children.12,14,22,38,40-42 In addition, the eyes and orbits are proportionally larger in children,43,44 making them more exposed and possibly more threatening in appearance. Our finding that children younger than 12 years are at particularly increased risk of injury is consistent with previously published reports27,45 and could reflect important physical and behavioral changes associated with preadolescence. In addition, the peak incidence of periorbital injury between ages 0 and 4 years is consistent with previously published data regarding the HNF.15,16,41,46
8 Understanding that children are at increased risk of periorbital injury from dog bites should inform both the prevention and management of these injuries. From the perspective of management, the evaluation of a pediatric dog bite patient should include a thorough examination for evidence of ophthalmic injury as an integral part of the head and neck trauma examination. Injuries such as globe rupture or retrobulbar hematoma are rare but represent true ophthalmic emergencies. Specific emphasis should be placed on the presence or absence of canalicular involvement given the high rate of eyelid injury and possible complications, such as epiphora, if these injuries are not probed and intubated when appropriate.47 Failure to recognize and resuspend a canthal laceration could lead to telecanthus, and improper repair of a laceration involving the eyelid margin could lead to cicatrization, resulting in ectropion, lagophthalmos, or exposure keratopathy.48 The availability of oculoplastic, ophthalmologic, oral and maxillofacial, or other appropriately trained facial trauma consultant is therefore integral in the management of these patients. In addition, given the nature of the pediatric population, as well as examination techniques such as canalicular duct exploration, an examination under anesthesia may be indicated. Thus, access to the OR or pediatric sedation capabilities are often required for proper evaluation of these injuries, not to mention their ultimate repair. Our findings have important implications for the prevention of dog bite injuries in the pediatric population. The increased frequency of periorbital injury in children and the possible targeting of these structures create a powerful narrative among parents and dog owners and could help raise awareness of the issue. Children were more likely to be bitten in the HNF by a familiar dog at home, a finding consistent with previously published reports, which have characterized such attacks as happening to younger, provocative children by quiet, familiar dogs indoors with parents present.14,19,25,37 Public health and preventive measures should stress that young children not be left at eye level on the floor with a dog, with or without immediate adult supervision and regardless of the familiarity with the dog. In addition, potentially risky behaviors such as interfering with food, hugging, kissing, staring, or playing roughly should be discouraged. Educational programs including instructional videos, websites, and training with live dogs have been shown to be effective interventions to improve safety knowledge and behavior in children.38,39,49 Additional preventive efforts include National Dog Bite Prevention Week, sponsored by the American Veterinary Medical Association. This campaign features educational video content, social media messaging, and provider toolkits to promote awareness and skills
PERIORBITAL DOG BITE INJURIES
for the prevention of dog bites at a community level.50 The American Veterinary Medical Association Task Force on Canine Aggression and Human-Canine Interactions has called for a multidisciplinary approach to identify potential partners in these efforts.51 In this spirit, we propose a possible collaboration between National Dog Bite Prevention Week and National Facial Protection Month, sponsored by the American Association of Oral and Maxillofacial Surgeons and other professional groups.52 Both campaigns take place during the month of April. Bridging the interdisciplinary divide between veterinary, medical, and dental providers creates opportunities for synergy through shared experience and expanded reach. This study has several limitations. First, it was retrospective and relied on extraction of heterogeneous documentation of injuries and other clinical data input by a variety of providers at the time of the encounter. This was moderated by the fact that multiple services and providers examined and evaluated the patients, providing us with the ability to cross-check findings. Another limitation was the grouping of all patients younger than 18 years into a single pediatric age group, which was used interchangeably with the term ‘‘children.’’ This is a developmentally and physically heterogeneous group. As illustrated in Figure 1, there were few patients aged between 12 and 18 years, making this less likely to significantly impact data analysis. In addition, the study population was limited to patients presenting to a level 1 trauma center with trauma code activation, hospital admission, or transfer from an outside facility. This likely selected for more acute or extensive injuries. Some patients with more straightforward injuries likely did not seek medical attention, did not qualify for trauma code activation, or were treated in the community. In conclusion, our study found that pediatric patients were at greater risk of periorbital injury than adults, even after controlling for increased rates of HNF injury in children. No such increased risk was observed in the CTA, although this was the most frequently injured region in both the adult and pediatric groups. Although the explanation is likely multifactorial, the unique physical and behavioral characteristics of children, as well as preferential targeting of the periorbital region by dogs, could play a role in the increased rates of HNF injury observed in children. Further studies are indicated to better characterize the etiology and prevention of these injury patterns. Acknowledgments The authors thank Shauna L. Carson, Program Support Supervisor, Harborview Medical Center–Trauma Program, for helping us obtain the data from Harborview Medical Center.
HURST ET AL
References 1. Nygaard M, Dahlin LB: Dog bite injuries to the hand. J Plast Surg Hand Surg 45:96, 2011 2. Weiss HB, Friedman DI, Coben JH: Incidence of dog bite injuries treated in emergency departments. JAMA 279:51, 1998 3. Insurance Information Institute. Spotlight on: dog bite liability. Available at: https://www.iii.org/article/spotlight-on-dog-biteliability. Published 2017. Accessed March 17, 2018 4. Peters V, Sottiaux M, Appelboom J, Kahn A: Posttraumatic stress disorder after dog bites in children. J Pediatr 144:121, 2004 5. Palmer J, Rees M: Dog bites of the face: A 15 year review. Br J Plast Surg 36:315, 1983 6. Karlson TA: The incidence of facial injuries from dog bites. JAMA 251:3265, 1984 7. Gonnering RS: Ocular adnexal injury and complications in orbital dog bites. Ophthalmic Plast Reconstr Surg 3:231, 1987 8. Brogan TV, Bratton SL, Dowd MD, Hegenbarth MA: Severe dog bites in children. Pediatrics 96:947, 1995 9. Kahn A, Bauche P, Lamoureux J: Child victims of dog bites treated in emergency departments: A prospective survey. Eur J Pediatr 162:254, 2003 10. Kahn A, Robert E, Piette D, et al: Prevalence of dog bites in children: A telephone survey. Eur J Pediatr 163:424, 2004 11. Schalamon J, Ainoedhofer H, Singer G, et al: Analysis of dog bites in children who are younger than 17 years. Pediatrics 117:e374, 2006 12. Dwyer JP, Douglas TS, van As AB: Dog bite injuries in children— A review of data from a South African paediatric trauma unit. S Afr Med J 97:597, 2007 13. Steele MT, Ma OJ, Nakase J, et al: Epidemiology of animal exposures presenting to emergency departments. Acad Emerg Med 14:398, 2007 14. Kaye AE, Belz JM, Kirschner RE: Pediatric dog bite injuries: A 5year review of the experience at the Children’s Hospital of Philadelphia. Plast Reconstr Surg 124:551, 2009 15. Chen HH, Neumeier AT, Davies BW, Durairaj VD: Analysis of pediatric facial dog bites. Craniomaxillofac Trauma Reconstr 6: 225, 2013 16. Chiam SC, Solanki NS, Lodge M, et al: Retrospective review of dog bite injuries in children presenting to a South Australian tertiary children’s hospital emergency department. J Paediatr Child Health 50:791, 2014 17. Chhabra S, Chhabra N, Gaba S: Maxillofacial injuries due to animal bites. J Maxillofac Oral Surg 14:142, 2015 18. O’Brien DC, Andre TB, Robinson AD, et al: Dog bites of the head and neck: An evaluation of a common pediatric trauma and associated treatment. Am J Otolaryngol 36:32, 2015 19. Rezac P, Rezac K, Slama P: Human behavior preceding dog bites to the face. Vet J 206:284, 2015 20. Gurunluoglu R, Glasgow M, Arton J, Bronsert M: Retrospective analysis of facial dog bite injuries at a Level I trauma center in the Denver metro area. J Trauma Acute Care Surg 76:1294, 2014 21. Wei LA, Chen HH, Hink EM, Durairaj VD: Pediatric facial fractures from dog bites. Ophthalmic Plast Reconstr Surg 29:179, 2013 22. Eppley BL, Schleich AR: Facial dog bite injuries in children: Treatment and outcome assessment. J Craniofac Surg 24:384, 2013 23. Monroy A, Behar P, Nagy M, et al: Head and neck dog bites in children. Otolaryngol Head Neck Surg 140:354, 2009 24. Lackmann GM, Draf W, Isselstein G, Tollner U: Surgical treatment of facial dog bite injuries in children. J Craniomaxillofac Surg 20:81, 1992 25. Erickson BP, Feng PW, Liao SD, et al: Dog bite injuries of the eye and ocular adnexa. Orbit 38:43, 2019 26. Bratton EM, Golas L, Wei LA, et al: Ophthalmic manifestations of facial dog bites in children. Ophthalmic Plast Reconstr Surg 34: 106, 2018 27. Prendes MA, Jian-Amadi A, Chang SH, Shaftel SS: Ocular trauma from dog bites: Characterization, associations, and treatment patterns at a regional level I trauma center over 11 years. Ophthalmic Plast Reconstr Surg 32:279, 2016
9 28. Wladis EJ, Dewan MA: Periorbital trauma from pit bull terrier attacks. Orbit 31:200, 2012 29. Savar A, Kirszrot J, Rubin PA: Canalicular involvement in dog bite related eyelid lacerations. Ophthalmic Plast Reconstr Surg 24: 296, 2008 30. Slonim CB: Dog bite-induced canalicular lacerations: A review of 17 cases. Ophthalmic Plast Reconstr Surg 12:218, 1996 31. Iazzetti L: Anticipatory guidance: Having a dog in the family. J Pediatr Health Care 12:73, 1998 32. Overall KL: Understanding and treating canine dominance aggression: An overview. Vet Med 94:976, 1999 33. Schultz RC, McMaster WC: The treatment of dog bite injuries, especially those of the face. Plast Reconstr Surg 49:494, 1972 34. McHeik JN, Vergnes P, Bondonny JM: Treatment of facial dog bite injuries in children: A retrospective study. J Pediatr Surg 35:580, 2000 35. Wu PS, Beres A, Tashjian DB, Moriarty KP: Primary repair of facial dog bite injuries in children. Pediatr Emerg Care 27:801, 2011 36. Toure G, Angoulangouli G, Meningaud JP: Epidemiology and classification of dog bite injuries to the face: A prospective study of 108 patients. J Plast Reconstr Aesthet Surg 68:654, 2015 37. Reisner I. The epidemiology & public health issue of dog bites: Who, where & why, Presented at the Western Veterinary Conference, Las Vegas; February 17, 2013 38. Shen J, Rouse J, Godbole M, et al: Systematic review: Interventions to educate children about dog safety and prevent pediatric dog-bite injuries: A meta-analytic review. J Pediatr Psychol 42: 779, 2017 39. Dixon CA, Mahabee-Gittens EM, Hart KW, Lindsell CJ: Dog bite prevention: An assessment of child knowledge. J Pediatr 160: 337, 2012 40. Bernardo LM, Gardner MJ, Rosenfield RL, et al: A comparison of dog bite injuries in younger and older children treated in a pediatric emergency department. Pediatr Emerg Care 18:247, 2002 41. Jaindl M, Grunauer J, Platzer P, et al: The management of bite wounds in children—A retrospective analysis at a level I trauma centre. Injury 43:2117, 2012 42. Hon KL, Fu CC, Chor CM, et al: Issues associated with dog bite injuries in children and adolescents assessed at the emergency department. Pediatr Emerg Care 23:445, 2007 43. Wheeler J, Phillips J: Pediatric facial fractures and potential longterm growth disturbances. Craniomaxillofac Trauma Reconstr 4: 43, 2011 44. Farkas LG, Posnick JC, Hreczko TM: Growth patterns of the face: A morphometric study. Cleft Palate Craniofac J 29:308, 1992 45. Foster MD, Hudson JW: Contemporary update on the treatment of dog bite: Injuries to the oral and maxillofacial region. J Oral Maxillofac Surg 73:935, 2015 46. Chun YT, Berkelhamer JE, Herold TE: Dog bites in children less than 4 years old. Pediatrics 69:119, 1982 47. Kennedy RH, May J, Dailey J, Flanagan JC, Canalicular laceration. An 11-year epidemiologic and clinical study. Ophthalmic Plast Reconstr Surg 6:46, 1990 48. Nerad J: Eyelid and orbital trauma, in Techniques in Ophthalmic Plastic Surgery—A Personal Tutorial. Amsterdam, Elsevier, 2010, pp 355–390 49. Meints K: Test yourself—A first assessment of the dog bite prevention project "Blue Dog". Part I: Education—The "Blue Dog" is born; Part II: The "Blue Dog"—Do children really get the message?. Presented at the World Small Animal Veterinary Association World Congress, Prague, Czech Republic; October 11, 2006 50. American Veterinary Medical Association. National Dog Bite Prevention Week. Available at: https://www.avma.org/Events/ pethealth/Pages/Dog-Bite-Prevention-Week.aspx. Published 2018. Accessed March 17, 2018 51. American Veterinary Medical Association Task Force on Canine Aggression and Human-Canine Interactions. A community approach to dog bite prevention. J Am Vet Med Assoc 218: 1732, 2001 52. American Association of Oral and Maxillofacial Surgeons. April is National Facial Protection Month. Available at: https://www. aaoms.org/media/april-is-national-facial-protection-month. Published 2018. Accessed March 17, 2018