Characteristics and Complication Rates of Mandibular Fractures Caused by Violent Mechanisms Versus Nonviolent Mechanisms

 

Mandibular fractures are the second most commonly occurring maxillofacial fractures, comprising 36% to 59% of these injuries.¹ Each year, an estimated 2500 people in the US sustain a mandibular fracture.² These fractures are particularly common in young adult men, among whom violence and assault are the leading cause of mandibular trauma.^3-6 Violent crime continues to be prevalent in the US, with over 1.2 million violent crimes reported in 2019 and aggravated assault accounting for 68.2% of them.⁷ 

Management of mandibular fractures often requires specialty care and can present unique challenges in diagnosis and management.⁸ Mandibular injuries can result in complications, such as temporomandibular joint disorders, dental malocclusion, osteomyelitis, and facial asymmetry.¹ Complications are more likely to occur if injuries go untreated or are treated improperly.⁶ Despite the high rates of mandibular fractures that are caused by violent mechanisms, surgical outcomes of patients in this specific group are understudied.^5,6,9-11 In the US, patients of lower socioeconomic status are known to have higher rates of postoperative complications, mortality, failure to rescue, and readmission after discharge.¹² Previous studies of mandibular fracture surgical outcomes have established that a history of tobacco, alcohol, or drug use increases the risk of surgical site infections.⁹ Regions with high rates of interpersonal violence have been linked to higher rates of medical noncompliance, substance abuse, and inadequate nutrition for proper wound healing.⁵. This correlation implies a potential link between mechanism of fracture and treatment outcome. 

The current literature has extensively looked at the surgical outcomes, complications, and risk factors associated with patient specific data. However, there is limited evidence comparing the complication rates of patients with violent and nonviolent fractures and how potential differences in patient demographics might predict poor outcomes. The goal of our study was to compare the demographics and outcomes of adult mandibular fractures treated at our Level 1 trauma center caused by both types of mechanisms and to investigate potential health care disparities and opportunities for intervention. 

An institutional review board–approved, retrospective study of adult mandibular fractures in the Tampa General Hospital Level 1 Trauma Registry was performed. The International Classification of Diseases, Ninth Revision codes were used to identify patients who presented to Tampa General Hospital with 1 or more mandibular fractures between January 1, 2012, and January 1, 2022. Patients <18 years of age were excluded. A separate database was created using our institution’s electronic medical records.

Data abstracted from the medical record included patient demographics, fracture site, mechanism of injury, and presence of concomitant injuries, such as additional facial fractures, loss of consciousness (LOC), traumatic brain injury (TBI), cervical spine injury, and skull fracture, and whether the patient was admitted to the ICU. Also collected was the type of surgical intervention and whether there was a complication. Patients were diagnosed via computed tomography scan upon admission. Mechanism of injury was categorized as violent or nonviolent, with violence defined as facial trauma intended to hurt someone. Violent mechanisms were further subcategorized into a punch, kick, gunshot, or strike with a solid object. In some cases, the exact mechanism was unknown but resulted from some sort of interpersonal altercation. Nonviolent mechanisms in the study included motor vehicle accidents, cycling crash, motorcycle accident, pedestrian hit by a vehicle, fall, sports injury, pathologic fracture, and explosion. 

Data were collected in Microsoft Excel. Statistical analysis was performed using Microsoft Excel and SPSS software (IBM Corp). A P value of <.05 was used for statistical significance. 

During the study period, 694 adult patients met the inclusion criteria. Table 1 displays the demographic information, with subgroups containing 323 patients of violence (POVs) and 369 patients of nonviolence (PONVs). The average age for the entire population of mandibular fracture patients was 37.4 ± 16.3. The majority were male (79%), and the highest proportion were white (55%). Patients of violence tended to be significantly younger (age 34.4 ± 13.2 years vs 40.1 ± 18.1 years; P < .001), were more likely to be male (88% vs 71%; P < .001), and were predominantly from minority ethnic/race groups (P = .003) compared with patients of nonviolence. 

The sites fractured on the mandible were counted individually. A total of 1069 distinct mandibular fractures were documented. Some fractures were complex or comminuted, and so a specific anatomic site on the mandibular could not be assessed. Figure 1 displays the distribution of fracture sites depending on whether there was a violent or nonviolent mechanism of injury. One patient could have multiple fracture sites. Notably, there appeared to be significantly more mandibular angle fractures caused by violent than nonviolent mechanisms (32% vs 14%). Conversely, nonviolent mechanisms tended to cause more injuries in the subcondylar and condylar regions than violent mechanisms (34% vs 16%). 

Figure 2 illustrates the breakdown of violent mechanisms. Most of these injuries were the result of a punch. Severity of injury was assessed based on the number of mandibular fractures, presence of fractures in other facial bones, and presence of concomitant injuries as illustrated in Table 2. There were no differences in average number of mandibular fractures between the 2 groups. POVs were less likely to have fractures in additional facial bones compared with patients of nonviolence (30% vs 42%; P = .001). Concomitant injuries measured included reported loss of consciousness (LOC) from the blow, traumatic brain injury (TBI), cervical spine injury, and skull fractures. POVs were less likely to have LOC (24% vs 44%; P < .001), TBI (5% vs 26%; P < .001), C-spine injury (1% vs 10%; P < .001), and skull fracture (7% vs 16%; P < .001). Patients of violence were also less frequently admitted to the ICU because of their injuries (14% vs 43%; P < .001).

Treatment modality of the mandibular fractures included either open reduction, closed reduction (with or without maxillomandibular fixation), or conservative treatment. Table 3 displays the breakdown of which surgical intervention was performed for POVs and PONVs. It also shows the number of patients who had healing complications. POVs were more likely to be treated with open reduction (71% vs 57%; P < .001). PONVs were more often treated with no surgical intervention (21% vs 10%; P < .001). POVs had a higher complication rate than PONVs, although this was not statistically significant (16% vs 9%; P = .12). 

The complications were noted to be in one of the following categories: infection, bony misalignment, dehiscence, hematoma, seroma, neurovascular injury, sialocele, hardware exposure, hardware loosening, and malocclusion. Figure 3 contains a comparison of the occurrence frequency of each complication type between POVs and PONVs. Hardware exposure occurred much more frequently in patients with violent injury mechanisms than in those with nonviolent mechanisms (23% vs 9%). 

Mandible fractures are among the most common maxillofacial fractures, particularly vulnerable as the only movable maxillofacial bone and consisting of a mechanically weak composition.^13,14 Our study aligns with many large retrospective studies, finding interpersonal violence as a leading cause of mandibular fractures, most commonly in younger males in minority groups.^6,15 Mandibular angle fractures have been found in other studies to be more commonly caused by violence, the location a likely target for a punch to the face. This may account for the higher proportion of patients with violent mechanisms being treated with open reduction, as the angle is readily accessible for this type of repair. Condylar fractures, seen more often in PONVs, are more commonly treated conservatively. Angle fractures have also been associated with higher complication rates.^5,6,16 The angle bears the greatest force load, and so it is vulnerable to screw loosening and infection.¹⁶ A 17-year retrospective study showed that the mandibular angle is the most common site of recurrent fracture, with incidence higher in unmarried young men with a history of drug use. These fractures tended to occur in sites different from the previous fracture fixation, indicating a violent mechanism.⁸ 

The results of our study regarding severity and concomitant injuries were expected. Most nonviolent mechanisms of mandibular fractures involve motor vehicle collisions, which have a high impact velocity and thus lead to more severe injuries.⁶ Mandibular fractures from violence, most commonly assault, are typically focused and low velocity, and consequently are less severe. Gunshot wounds are the exception among violent mechanisms; they are very high energy injuries. “Severity” of a mandibular fracture is difficult to define. The Mandibular Injury Severity Score (MISS) was introduced to quantify this, considering the following factors: fracture type, location, occlusion, soft tissue involvement, infection, and interfragmentary displacement.¹⁷ Most of these variables could not be directly measured in the study. Number of constituent fractures had the highest correlation to MISS, but our study showed no differences in number of mandibular fractures between POVs and PONVs.¹⁷ Some studies have demonstrated an increase in inflammatory complications, such as infection and exposed hardware, with increased MISS.^18-20 Such complications as soft tissue involvement, occlusion, and displacement are functions of impact velocity, which can be approximated by rates of TBI, skull fracture, C-spine injury, and LOC. In addition, MISS score has been positively correlated with need for extended hospital stay.¹⁷ Thus, our measures of injury severity appear to be reliable.  

Our results demonstrating higher complication rates for POVs despite lesser severity of injuries implies an influence of other factors. Although this difference did not reach statistical significance, many more patients with violent injury mechanisms were likely lost to follow-up, and so this complication rate is likely an underestimate. Victims of violence are more likely to have such risk factors as alcohol and drug abuse, homelessness, poor social support, and medical noncompliance.^5,21 The literature has posed several theories for these complication risk factors. Haug and Schwimmer proposed that abuse of alcohol and drugs could impair bone healing because of poor nutrition and circulation.²² Hall and Ofodile published a retrospective study of patients treated at Harlem Hospital in which they found that a high prevalence of drug and alcohol abuse contributed to a higher incidence of poor oral hygiene and delay in seeking medical care.²³ 

Homelessness has been identified as an independent risk factor for surgical site infections after open reduction of mandibular fractures. These patients possess a higher risk of oral and dental disease as well as limited access to dental and medical care.²⁴ Hardware exposure caused a significantly higher proportion of complications in POVs. Contributing factors for hardware exposure are multifaceted; poor oral hygiene, collection of food and saliva in the vestibule, smoking, extent of soft tissue damage, tooth infection, and trauma caused by repeated masticatory forces are all implicated.²⁵ It is critical for patients with signs of hardware exposure or infection to follow up promptly, as timing is essential in the feasibility of hardware salvage.²⁶ 

Patients who undergo treatment for mandibular fractures have many responsibilities in their healing process. This includes following antibiotic regimens, adhering to non-chew diets while in maxillofacial fixation, maintaining good nutrition status with calorie supplementation, keeping fixation wires intact except in emergencies, maintaining good oral hygiene with chlorhexidine rinse, and promptly following up if complications arise.²⁷ Given that patients injured by violent mechanisms tended to have more complications, eliciting this in the history should prompt extra attention to engaging them in how to successfully carry out their discharge instructions, as we often overestimate patient understanding. In a study of 180 patients who underwent minor oral surgery and received verbal and written instructions postoperatively, only 60% remembered both sets at 1 week after surgery. Twelve percent did not comply with postoperative mouthwash instructions, and 67% did not comply with antibiotic prescriptions.²⁸ 

Many studies have demonstrated that patients who increasingly possess the ability and willingness to manage their own health care have better health outcomes.^29,30 The Patient Activation Measure is a survey that has quantified this as having higher follow-up rates, better knowledge of their treatment guidelines, and avoidance of health-damaging behaviors, such as smoking or illegal drug use.²⁹ Many health care outreach interventions exist that serve to improve patient activation and thus improve outcomes. Digital patient engagement platforms have demonstrated effectiveness in decreasing complication and readmission rates among hip and knee arthroplasty patients by providing guidance and remote monitoring perioperatively.^30,31 Telephone outreach programs have been shown to boost patient self-management behaviors when patients answer the phone calls, a characteristic associated with high patient activation and improved outcomes.^29,32 Our institution could improve patient compliance by incorporating the teach-back method to check understanding during the preoperative and discharge process, providing written instructions for the patient and asking the patient about their support system and encouraging family members to join discussions. Compliance rates may improve if staff assisted patients in registering with the hospital patient portal if they have a mobile device; if possible, a medical staff member could also reach out by phone a few days after discharge to assess the patient’s recovery process. 

Homeless patients are at particularly high risk for poor follow-up. A retrospective study of 2221 adults with facial fractures demonstrated that homeless patients with facial fractures were nearly 3 times more likely to have been assaulted than housed patients and twice as likely to have mandible fractures.³³ Homeless patients often lack health insurance and therefore tend to utilize acute hospital services more and have poorer ambulatory follow-up.³³ Resource allocation for homeless people would be most efficient if delivered while they are still in the hospital. This should involve assessing for issues with access to health care, such as cost or transportation concerns, or lack of support at home. Further research is needed in follow-up outcomes to assess for other risk factors of poor-follow up and how this metric can be improved. This population is inherently difficult to study since it can be difficult to track in retrospective studies whether patients were truly lost to follow-up, or had simply presented to a different hospital system or were completely recovered from their injury. 

The most definitive action in the community to prevent complications in POVs is to prevent violent injuries in the first place. In 2019, the American College of Surgeons developed a multidisciplinary working group called Improving the Social Determinants of Health to Attenuate Violence (ISAVE) to address root causes of this issue.³⁴ Trauma-informed care is emphasized to build trust between patients and the health care system. It works by prioritizing understanding and responsiveness to the impact of trauma, acknowledging the emotional and psychological component, and allowing survivors to rebuild a sense of control and empowerment. Social care should be integrated into trauma care because addressing unmet social needs, such as housing instability, food insecurity, poor health literacy, and lack of transportation, has been shown to improve health outcomes.³⁵ Advocacy is an underlying theme in addressing the violence epidemic. 

One important avenue is investing in disadvantaged communities, as health disparities are demonstrably evident at the geographic level. For example, in 2017, Boston Medical Center pledged $6.5 million in affordable housing initiatives over a 5-year period.³⁴ Advocacy can also come in the form of support for governmental funding of violence prevention initiative programs. Project Safe Neighborhoods is an evidence-based initiative developed by the US Department of Justice that aims to identify prolific violent offenders and deter or incapacitate them from offending. Its implementation in Tampa, Florida, resulted in significant decreases in rates of violence and gun crime.³⁶ Public health is a valuable avenue for stopping traumatic injury at its source and ensuring optimal outcomes by removing socioeconomic barriers.

There were several limitations to our study. Given the retrospective nature, some parameters were not consistently recorded, and we can only record what is present in the electronic medical record. Complications are sometimes not picked up if the patient never follows up or if they follow up at a different institution. As previously mentioned, the POV group is likely the most undercounted in this category. Furthermore, the patients in this study were treated at a Level I trauma center, and so the data may not be generalized to all communities. Further studies should focus on follow-up rates of POVs and efficacy of the aforementioned interventions to improve treatment compliance and follow-up. Future studies can also study trends in patients living in different zip codes, using average income to compare patient demographics, mechanisms, and outcomes for different socioeconomic conditions. 

This study provides a large-scale comparison of mandibular fractures caused by violent and nonviolent mechanisms at a Level I trauma center. Patients presenting with violent mechanisms were more likely to be younger men of minority background, with the angle as the most common anatomic site fractured. These patients were more likely to be treated with open reduction than those with nonviolent mechanisms. They were also more likely to have a complication, especially hardware exposure or infection, than those with nonviolent mechanisms despite having less severe injuries. Patients presenting with injuries caused by violence are more likely to have socioeconomic barriers and other risk factors in place that are resulting in poorer outcomes. Thus, there is an important opportunity to address this disparity through interventions at the level of discharge and follow-up as well as public health initiatives to stop the injury at its source.