Airway management in self-inflicted gunshot wounds to the face

Highlights

• •Self-inflicted GSWs to the face may compromise a patient's airway and mental status.
• •Patients that can protect their airway may be managed with non-invasive support.
• •Supraglottic airways may be an alternative to endotracheal intubation in select cases.

Abstract

Background

This study characterizes the prehospital and trauma bay airway management of self-inflicted gunshot wounds (SI-GSWs) to the face.

Methods

We reviewed SI-GSWs to the face from 2017 to 2021. Patients with isolated temporal GSWs were excluded. Emergency Medical Services (EMS) and trauma bay information on airway management were collected.

Results

38 patients presented with a SI-GSW to the face. 20 of the 29 transported by EMS were managed without advanced airways. There were four endotracheal intubations, three supraglottic airways, and two failed intubations. Those with advanced airways had lower GCS (3 vs. 15, p<0.01) and systolic blood pressure (94 vs. 144, p<0.01). After hospital arrival, 22 SI-GSWs were endotracheally intubated in the trauma bay and two in the OR. No cricothyroidotomies were performed.

Conclusions

A majority of SI-GSWs to the face can be managed by EMS using noninvasive support. Supraglottic airways may be an option for those requiring invasive support.

Introduction

Gunshot wounds to the head or face account for 10–11% of all firearm injuries [1]. Suicide remains a leading cause of death in the United States and has been increasing over the past 20 years. In 2019, over half of all suicides were from firearms, killing over 47,000 people [2]. Homicide-related wounds are typically multiple while self-inflicted gunshot wounds (SI-GSW) to the face are usually single and account for approximately 0.1% of all trauma admissions [3]. Transoral or submental trajectories damage the mandible and midface resulting in airway compromise.

The published frequency of urgent airway management in facial GSWs ranges from 17% to 60% [4]. Often, the airway must be addressed by pre-hospital personnel or secured soon after hospital arrival. Several studies report a 5% rate of cricothyroidotomy and 36–50% will require a tracheostomy for long-term airway management [5], [6], [7]. Multiple studies report on airway outcomes, but less information is available on airway presentations. The objective of this study is to characterize the prehospital and trauma bay airway management of self-inflicted gunshot wounds (SI-GSW) to the face based on the patient's neurologic, hemodynamic, and respiratory status. We hypothesize that most patients with SI-GSWs to the face will require invasive airway management, frequently placed in the pre-hospital setting.

Methods

This study was approved by the Institutional Review Boards at the University Medical Center of Southern Nevada (Las Vegas, NV, USA) and the Kirk Kerkorian School of Medicine at UNLV. A retrospective chart review of self-inflicted GSWs to the face was conducted with a waiver of informed consent from January 2017 through December 2021 at a single, American College of Surgeons-verified Level 1 trauma center with over 4000 admissions per year. Exclusion criteria were age less than 18 years old, pregnant and isolated temporal GSWs without mandible or maxillary involvement.

Emergency Medical Services (EMS) and trauma bay data were collected including initial vitals, Glasgow coma scale (GCS), oxygen delivery device (nasal cannula, non-rebreather mask, or bag valve mask (BVM)), time on scene and in transport, details of advanced airway placement (type, size, attempts, and medications, when available), and need for cricothyroidotomy. Times were obtained from EMS run sheets. Time on scene was calculated by subtracting patient contact from departure time. Time in transport was calculated as the interval from scene departure to hospital arrival. Airways were defined as supraglottic if there was not a cuffed tube below the vocal cords (e.g. King, Combitube, or laryngeal mask airway). Demographic variables were age, gender, ethnicity, and Cormack-Lehane grade. Computed tomography (CT) scans were reviewed for bone and soft tissue injuries. Timing and type of intervention were documented including tracheostomies and reconstructive operations. Hospital data included length of stay, discharge disposition, mortality rate, and hospital charges. To calculate hospital, ICU, and ventilator-free days, we used the methodology described by Yehya et al., with 30 days as a benchmark. If a patient stayed in hospital, ICU, or on a ventilator for more than 30 days, the value was set to zero. For stays under 30 days (x days), the free days were derived as 30-x days [8]. Failure-free days, rather than length of stay proper, was selected to limit the potential confounding of outliers with prolonged length of hospitalization.

Successful advanced airway placement was the primary endpoint. The rates of tracheostomy and cricothyroidotomy were secondary endpoints. Statistical analysis was performed with Stata Version 14 (College Station, TX). Categorical clinical variables are reported as counts and percentages of the cohort. Continuous variables are reported as mean and standard deviation, with ranges when appropriate. Non-normally distributed data are presented as median with interquartile ranges (IQR). Continuous variables were compared using unpaired t-tests, with Chi-square or Fisher exact test for categorical comparisons. Significance was set at a p-value of less than 0.05.

Results

Over the five years, 38 patients presented with an intra-oral or submental SI-GSW. There were 31 males (81.5%) and 7 females, with a mean age of 49.9 years (range, 20–92). The most common race and ethnicity were White, Non-Hispanic which made up 71.1% (n = 27) of the cohort. Eight patients (21.1%) were transferred from outside hospitals with seven arriving endotracheally intubated. One (2.6%) patient presented as a walk-in to the trauma center. The other 29 (76.3%) patients were transported from the field by EMS.

Pre-hospital management

Of the 29 patients brought directly from the field, 4 (13.8%) were endotracheally intubated and 3 (10.3%) had a supraglottic airway placed. Two patients (6.9%) had an unsuccessful attempt at intubation and were managed by BVM ventilation until a definitive airway was obtained at the hospital (Fig. 1). Of the 9 patients who had invasive airway management by EMS, 6 (66.6%) had a GCS less than 8 (5 with a GCS of 3) while 2 (22.2%) had a GCS of 15. This cohort had a lower median GCS (3 vs. 15, p = 0.009) and mean SBP (94 vs. 144, p = 0.006) when compared to those without advanced airways. Of those with an advanced airway attempted or successfully placed, 66.6% (n = 6) had rapid sequence intubation (RSI) medications administered. In 3 cases, this was ketamine plus a paralytic. The remaining 3 used etomidate (n = 1), versed (n = 1) or both (n = 1). Three patients did not receive RSI medications due to active cardiopulmonary resuscitation (CPR) while in route to the hospital.

Fig. 1.

Prehospital airway management and mental status

(a) Distribution of airways in patients transported from the field by local EMS. (b) Glasgow Coma Scale (GCS) reported by EMS.

For the remaining 20 patients (69.0%), supplemental oxygen was provided in 10 patients (50%), with 5 nasal cannulas, 3 nonrebreather masks, 1 BVM, and 1 unspecified device. GCS was documented by EMS in 19 cases, with 12 (63.1%) having a GCS of 15. Two patients (10.5%) with a GCS of 8 or less were managed without an advanced airway. Total EMS time from patient contact to hospital arrival averaged 23.9 min. No difference was observed in the scene (6.0 vs. 7.7 min) and transport times (19.8 vs. 16.7 min) by airway management Table 1.

Table 1.

Pre-hospital data by type of airway management.

	Advanced Airway (n = 9)	No Advanced Airway (n = 20)	p
Age, mean ± SD, years	49.6 ± 22.9	50.4 ± 21.4	0.92
Male gender, n (%)	9 (100)	16 (80)	0.15
GCS, median [IQR]	3 [3–13]	15 [11–15]	0.009
Heart Rate, mean ± SD, bpm	76 ± 49	99 ± 19	0.07
Systolic Blood Pressure, mean ± SD, mmHg	94 ± 61	144 ± 26	0.006
Shock Index, mean ± SD	0.88 ± 0.38	0.70 ± 0.15	0.09
Scene Time, mean ± SD, minutes	6.0 ± 4.3	7.7 ± 4.3	0.42
Transport Time, mean ± SD, minutes	19.8 ± 9.8	16.7 ± 8.3	0.39
Total EMS Care Time, mean ± SD, minutes	23.8 ± 3.3	23.9 ± 2.6	0.97

SD, standard deviation; n, number; IQR, interquartile range; bpm, beats per minute; EMS, emergency medical services. Shock Index calculated as heart rate divided by systolic blood pressure.

Trauma bay management

All 11 patients that arrived with an endotracheal tube (7 transfers and 4 EMS) had appropriate positioning confirmed in the trauma bay (Fig. 2). One endotracheal tube from an outside hospital had a significant air leak (damaged balloon) and was replaced over a gum elastic bougie using videolaryngoscopy. Twenty-two (57.9% of cohort) endotracheal intubations were done in the trauma bay: 3 patients with supraglottic airways, 2 unsuccessful EMS intubations, 16 arriving without an advanced airway, and 1 patient who walked into the trauma center. Of these 22 patients, direct laryngoscopy was used in 8 (36.4%) while videolaryngoscopy was used in 10 patients (45.5%). Ten patients (45.5%) had a Cormack-Lehane grade 1 view, with 4 grade 2a (18.2%) and 1 grade 2b (4.5%). Type of laryngoscopy and Cormack-Lehane grade were not documented in the remaining cases. Two patients (5.2%) were taken directly to the OR for airway and hemorrhage management with endotracheal intubation successful in both cases using videolaryngoscopy and having a Cormack-Lehane grade 1. Three patients (7.9%) did not undergo intubation in the trauma bay. There were no cricothyroidotomies performed. Two patients receiving CPR in route did not regain spontaneous circulation and were pronounced deceased before imaging or admission.

Fig. 2.

Airway management at hospital arrival.

Imaging

Twenty-eight (73.7%) GSWs were submental and 10 (26.3%) were intraoral. Thirty-six patients underwent CT imaging with the most common injuries being maxillary (n = 32, 88.9%) and mandible fractures (28, 77.8%). Tongue (24, 66.7%) and dental (18, 50.0%) injuries were also common (Fig. 3). Ten (27.8%) patients had intracranial hemorrhages. Four (11.1%) patients had extravasation from the internal maxillary artery and underwent angioembolization.

Fig. 3.

Injury locations in SI-GSWs to the face

Thirty-six patients underwent CT imaging to characterize their craniofacial injuries. Maxillary fractures were most common (89%), with only 28% having an intracranial hemorrhage.

Hospital course

Thirty-six of 38 patients having SI-GSW to the face were admitted to the hospital, with 30 (79.0%) surviving to discharge. Twenty-five (69.4%) had a tracheostomy placed at a median of 2 days [IQR 1–4] post-admission. Tracheostomies were evenly divided between percutaneous and open procedures. Twenty-six patients underwent facial reconstruction with a tracheostomy placed in 24. The median number of operative procedures was 2 [IQR 0–3]. The initial reconstructive procedure occurred within 48 h in 9 (34.6%) patients and after in 17 (65.4%). Nine patients underwent a tracheostomy during the same anesthetic event as a reconstructive procedure (Table 2). Overall, the average hospital-free days were 6.3 ± 8.2, with ICU-free days of 16.1 ± 10.4 and ventilator-free days of 17.3 ± 11.4 days. Only eleven patients (29.0%) were discharged home. Nineteen patients (50.0%) were discharged to other facilities: long term acute care (n = 6), skilled nursing or rehab (n = 6), inpatient psychiatric facility (n = 3), acute care hospital (n = 2), jail (n = 1) and hospice (n = 1). Eight patients died for an overall mortality of 21.1%. Median hospital charges were $340 K (IQR $159K-615 K).

Table 2.

Comparison of Patients Undergoing Early Versus Late Reconstruction.

	Early Reconstruction (n = 9)	Late Reconstruction (n = 17)	p value
Procedures, median [IQR]	1 [1–2]	2 [2–3]	0.02
Tracheostomy, n (%)	9 (100)	15 (88)	0.28
Enucleation, n (%)	2 (22)	4 (24)	0.94
Mandibulomaxillary Fixation, n (%)	6 (67)	7 (41)	0.21
Tooth Extraction, n (%)	3 (33)	5 (29)	0.83
Flap reconstruction, n (%)	0 (0)	6 (35)	0.04
Hospital-free days, mean±SD	7.2 ± 7.1	4.1 ± 1.4	0.23
ICU-free days, mean±SD	20.7 ± 6.8	16.6 ± 8.2	0.22
Ventilator-free days, mean±SD	23.3 ± 5.3	17.5 ± 10.1	0.12
Hospital Charges, mean±SD, $K	478 ± 171	743 ± 472	0.12

IQR, interquartile range; n, number; SD, standard deviation; ICU, intensive care unit. Failure-free days calculated from a 30-day baseline. Hospital charges are reported in thousands of dollars ($K).

Discussion

Self-inflicted gunshot wounds to the face, although uncommon, are highly morbid and sometimes fatal injuries. A trajectory through the lower face has the potential to compromise the patient's airway and mental status. Most patients transported from the field were managed without an advanced airway. Surprisingly, nearly half of the advanced airways were supraglottic devices. In the trauma bay, airways were secured with a combination of direct and videolaryngoscopy without the need for cricothyroidotomy.

The initial management of patients with facial trauma is based on Prehospital Trauma Life Support and Advanced Trauma Life Support principles starting with ensuring airway patency [9,10]. Over 75% of this cohort were transported from the field by local EMS with two-thirds of those managed without an advanced airway. Extended prehospital time is associated with a risk of mortality. Brown et al. found an increased odds of mortality in patients with prolonged scene time, which was partially mediated by pre-hospital intubation [11]. Hypotension, GCS less than 8, and non-extremity firearm injuries are indicators of time-sensitive injuries. Patients displaying these signs have increased mortality with increased transport time [12]. No difference was observed in EMS times by airway type. Over half (68%) of the patients who sustained a SI-GSW had a GCS >8, though patients receiving an advanced airway had significantly lower GCS and SBP. Prehospital intubation in patients with penetrating trauma results in higher odds of mortality when adjusting for multiple variables [13]. This was especially true for hypotensive patients who had an increased risk of cardiac arrest after emergent intubation [14]. Cervical spine injuries in this population are rare therefore spinal motion restriction should be avoided [15]. As found in this cohort, most SI-GSWs to the face with normal GCS and hemodynamically stable may be managed with upright positioning, suctioning, and supplemental oxygen.

Unexpectedly, one-third of the advanced airways placed in the prehospital setting were supraglottic devices. These airways benefit from a blind insertion technique and potentially being faster to place; however, in penetrating maxillofacial trauma there are additional considerations [16]. Dental injuries can introduce foreign bodies into the oropharynx, while comminuted mandible fractures with posterior displacement can decrease airway patency [17]. Though second-generation supraglottic airways have improved sealing and, in some cases, a channel for gastric decompression, their use in the setting of intraoral hemorrhage or injury has not been well studied [16]. These devices may serve as a temporary rescue adjunct until a definitive airway is obtained [18,19]. In this cohort, supraglottic airways were used in a few patients with depressed GCS and an intraoral trajectory.

Many patients with GSWs to the face will require urgent airway control [5,20]. EMS endotracheal intubation success rate varies between 53 and 90% of cases [16]. Blood in the airway, cervical immobility, and facial trauma are associated with failed prehospital endotracheal intubation [21]. Conversion to a definitive airway after a prehospital supraglottic device may include tracheostomy or endotracheal intubation via direct laryngoscopy or videolaryngoscopy, with craniofacial injuries having an increased risk of complications and need for tracheostomy [22]. Successful intubation in this cohort occurred with both direct and videolaryngoscopy; though Pieters et al. have shown an improved first-attempt success for difficult airways with video-assisted laryngoscopy [23]. Similarly, a Cochrane review reports that videolaryngoscopy reduces the rate of failed intubation and first-attempt success [24]. Though not observed in this cohort, a 4–5% rate of cricothyroidotomy has been observed in other studies of GSWs to the face [5,25]. The rate of tracheostomy in this cohort (69%) was higher than reported in previous studies (36–50%) and may be related to the inclusion criteria of only intraoral or submental GSWs [3,6,7]. Mandible and midface fractures have been shown to have the highest rate of tracheostomy [3]. Most of the tracheostomies were placed for patients undergoing reconstruction, especially those needing mandibulomaxillary fixation.

Patients surviving their SI-GSW frequently require craniofacial reconstruction. Our overall mortality rate (21%) was lower than other studies of intraoral and submental trajectories [3,[25], [26], [27]]. This may reflect a lower frequency of concurrent TBIs which is a known risk for mortality in GSWs to the face [3,25]. Reconstruction has been previously defined as early if occurring within 48–72 h of injury [20,28]. The late reconstruction group in this study had more procedures and flap reconstruction, but no significant difference in length of stay or hospital charges. In comparison to Murphy et al.’s SI-GSWs to the face, our cohort had a similar frequency of tracheostomy (69% vs 60%), more intermaxillary fixation (50% vs. 4%), and discharges to home (29% vs. 8%) and fewer deaths (21% vs. 38%). This may reflect their inclusion of frontal GSWs in the cohort [28]. Another series of facial GSWs reported a 37% tracheostomy rate, 50% intermaxillary fixation rate, and 14% rate of free tissue transfers for reconstruction [6]. Consistently, facial GSW survivors require intensive surgical management of their injuries [20].

This study is limited by the small cohort size and incomplete airway management data. Despite the trauma volume of this Level 1 center, self-inflicted GSWs to the face remain a rare mechanism. As a regional trauma referral center, over 20% of the cohort was transferred from other facilities without EMS data available from their field transport. Despite a review of all available EMS and electronic medical records, airway data was frequently missing, including Cormack-Lehane grade, type of blade used, and the precise number of attempts. No cricothyroidotomies were performed in this small patient set. Given the 5% emergent surgical airway rate reported in the literature, [5,25] the cohort is underpowered to draw a definitive conclusion for the modern SI-GSW to the face population. While the availability of videolaryngoscopy and fiberoptic bronchoscopy may have improved intubation success, the practitioner managing these challenging airways must be prepared to intervene surgically. Given the variability seen in management, additional data is needed to evaluate these trends on a larger scale. Evaluation of airway management in facial GSWs may benefit from a multicenter trial.

Conclusion

Self-inflicted GSWs to the face create a complex injury profile that may compromise the patient's airway and mental status. Despite potentially unstable injuries to the mandible and midface, patients that can protect their airway should be managed with non-invasive support (suctioning, upright positioning, and supplemental oxygen) until arrival at the trauma center. In those with depressed GCS or hypoxia, advanced airways may be required but the potential for a difficult airway cannot be underestimated. Supraglottic airways may be an alternative to endotracheal intubation in select patients.

Presentation

Quickshot presented at the Southwest Surgical Congress 2022 Annual Meeting, April 24–27, in Wigwam, Arizona.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.