Article, Radiology

Aggressive imaging protocol for hanging patients yields no significant findings

a b s t r a c t

Background: Despite rising rates of hanging injuries, few high-quality data and no national trauma guidelines are available to standardize the evaluation of these patients. We sought to identify the yield of imaging hanging pa- tients at our institution.

Methods: This is a retrospective study at an urban Level I Trauma Center. Charts for patients diagnosed with “Hanging, Strangulation, or Asphyxiation” (ICD-9 E93.0) from February 2008 to March 2014 were reviewed. Fre- quency of imaging orders and their results were recorded. Logistic regression analyses were done to determine factors associated with increased rates of imaging.

Results: 78 patients met inclusion criteria. The average age was 34 years, 86% were male, and 65% were witnessed hangings. In total, 195 CT scans and 67 X-rays were done. Frequency of imaging in our cohort: 77% received CT of the head (CTH); 88% CT of the neck; 85% CT angiography (CTA) of the neck; 86% chest X-ray. Highest-level Trauma activation occurred in 76% of patients and was associated with an increased rate of imaging, with the likelihood of CTH being increased by 31% (p b 0.01), CT of the neck without contrast by 19% (p b 0.01), CTA of the neck by 25% (p b 0.01), and chest X-ray by 25% (p b 0.01). Of the 78 patients, none had significant findings that required in- tervention.

Conclusions: In this study, the highest-level trauma activation was associated with increased rates of imaging, but did not alter patient care. A more selective approach in the evaluation of hanging injuries should be considered.

(C) 2019

Introduction

As of 2013, the Department of Health and Human Services estimates that 900,000 patients present to the emergency department (ED) each year for suicidal ideation. Of these patients, 60,000 of them will also have concurrent self-inflicted injury [1]. Although a significant amount of resources that have been spent to destigmatize mental health and support programs that focus on suicide prevention, suicide still remains the second leading cause of death amongst people ages 10-34 [2]. While the discharge of a firearm is the primary method of self-inflicted injury in completed suicides (48%), strangulation by hanging has steadily in- creased over the last decade and now accounts for 30% of all suicides [3].

? Abstract presented at ACEP Scientific Assembly 2017.

* Corresponding author at: Department of Emergency Medicine, Icahn School of Medicine at Mount Sinai, 1 Gustave L. Levy Place, Box 1149, New York, NY 10029, United States of America.

E-mail addresses: [email protected] (S. Schuberg), [email protected] (K. Shah).

At Elmhurst Hospital Center, a level-1 trauma center in Queens, New York with an annual ED census of 130,000 patient visits per year, histor- ically (and at the time of this study), highest level of trauma team acti- vation (“Red Trauma”) was initiated for patients that present with a hanging mechanism of injury. By protocol, these patients receive exten- sive radiological imaging regardless of Symptom severity. This radio- logic evaluation includes CT head non-contrast and CT angiography of the head and neck. We sought to identify the yield of imaging and to characterize injury patterns based on radiographic findings.

Methods

This is a retrospective, observational single center study conducted at a Level I Trauma Center in an urban setting. All charts for adult pa- tients (age N 18) diagnosed with a Clinical impression of “Hanging, Strangulation, or Asphyxiation” (ICD-9 E93.0) from February 2008 to March 2014 were identified from the electronic medical record system. Specific data elements that were determined a priori by the study inves- tigators were extracted and recorded on a data spreadsheet for review and analysis. Isolated strangulation injuries in the absence of suspension

https://doi.org/10.1016/j.ajem.2019.01.006

0735-6757/(C) 2019

738 S. Schuberg et al. / American Journal of Emergency Medicine 37 (2019) 737739

were excluded due to a fundamental difference in mechanism. Patient data was gathered by personnel trained in the use of both the emer- gency department electronic medical record system (Allscripts) and the hospital electronic medical record system (Quadramed). Extracted data included demographic information, the presence and level of trauma activation, types and results of imaging, procedures, presence of Otolaryngologist consultation, final diagnosis, final disposition, and patient outcomes. Logistic regression analyses were done to determine which factors were associated with increased rates of imaging. Statisti- cal analysis was done using the R statistical computing software. Statis- tical significance was set at P value b0.05. This study was reviewed and approved by our local Institutional Review Board.

Results

In total, 100 charts were identified for review. 22 charts were ex- cluded, 21 for being isolated strangulations and 1 due to unclear docu- mentation during an electronic medical record downtime. Of the remaining 78 patients: 86% were male and 76% were incarcerated. 65% of patients were found hanging at the scene. Of these 78 patients, 67 (87%) of them had a GCS equal to 15(Table 1). The most common evaluations included highest-level trauma activation (76%), Computed Tomography (CT) of the head without contrast (77%), CT of the neck without contrast (88%), CT angiography (CTA) of the neck (85%), X- ray of the chest (86%), and otolaryngology consultation for visualization of the oropharynx (76%) (Table 2). In total, 195 CT scans and 67 X-rays were performed. Of the patients, 4 presented with unstable vital signs, 4 were intubated in the field, and 7 were intubated in the emergency de- partment. In addition, 7 patients presented with cardiac arrest in the field, of which 6 did not survive, and 1 survived with minimal deficits. Imaging was not performed in 4 of the cardiac arrest patients. The total mortality amongst these patients was 8%. Amongst patients with initial GCS b8, the absolute mortality was 67%. All patients with a GCS N8 were discharged from the hospital.

There were 5 patients with findings identified on imaging. Of these, 2 were deemed not related to the clinical presentation. These 2 patients were discharged home without return visits. The other 3 patients had findings that warranted further observation and follow up. Of these, 1 was found to have a sternum fracture with subcutaneous air. The second had a questionable left carotid dissection on CT angiography. A follow- up magnetic resonance angiography showed no dissection. The third patient was noted to have a mildly dislocated right thyroid cartilage on CT neck. He was otherwise well and without aerodigestive symp- toms. He was evaluated by ENT and discharged with routine follow up. Surgical intervention was not required in any patient.

Logistic regressions were done to determine if there were statisti- cally significant factors leading to increased rates of imaging. Highest- level trauma activation was associated with an increased rate of all im- aging: CT of the head 31% more likely to occur (p b 0.002), CT of the neck 19% more likely to occur (p b 0.005), CTA of the neck 25% more likely to occur (p b 0.001), and X-ray of the chest 25% more likely to occur (p b

Table 1

Patient demographics.

Age, mean (range)

32 (12-84)

Male (%)

71 (92%)

Incident at correction facility (%)

59 (76)

Found hanging (%)

Glasgow Coma Scale (GCS)

51 (65)

Median ED arrival GCS (range) 14 (3-15)

GCS = 15 (%) 67 (87)

GCS b8 (%) 9 (12)

Initial emergency department vitals

Heart rate (range) 80 (22-149)

Systolic blood pressure (range) 136 (86-173)

Discharged alive (%) 71 (92)

LOS, days (range) 2 (1-20)

Table 2

Frequency of imaging and consultation

Highest level of trauma activation (%) 59 (76)

CT head non-contrast (%) 60 (77)

CT neck non-contrast (%) 69 (88)

CT neck angiograph (%) 66 (85)

X-ray of chest (%) 67 (86)

Otolaryngology consult (%) 59 (76)

0.001). Highest-level trauma activation was also associated with a 21% increase in otolaryngology consultation (p b 0.05). In addition, incarcer- ated patients were noted to have a 21% increase in rate of highest-level trauma activation (p b 0.05) (Table 3).

Discussion

Hanging has become an increasingly frequent method for suicide at- tempt in the United States [3]. In the emergency department, patients will rarely present following a classic ‘judicial hanging’ whereby the noosed patient drops at distance equal to their own height. By this mechanism, the neck is forced into hyperextension with resulting cervi- cal spine fracture and spinal cord injury [4,5]. More common are non- judicial hangings, in which a combination of tight ligature around the neck with or without minor fall from height results in the obstruction of jugular Venous outflow [6-8]. This results in a syncopal event that re- laxes the cervical musculature and allows the weight of the patient to be placed on the ligatured neck [6]. This subsequent tension on the neck then allows for the other rarer injuries of the hypopharynx, arteries, and cervical spine to occur [6]. Despite this sequence of events, the ob- served injury patterns are highly variable in both type and severity and depend largely on the duration of hanging and whether the injury was with complete or incomplete suspension from the ground [8-13].

Although the prevalence of suicide by hanging continues to increase, there remains little high quality data and no national trauma guidelines to help standardize the initial evaluation and treatment of these pa- tients. A few retrospective studies over the last two decades have ana- lyzed injury patterns and patient outcomes to guide future management. These data suggest that patients who present with high risk features (altered level of conscious as predicted by Glasgow Coma Scale(GCS), signs and symptoms of neurologic, airway, or vascular com- promise) need a dedicated evaluation that may include non-contrast CT head and C-spine and CT angiography of the head and neck [8-12,14,15]. What remains less clear is the evaluation of patients without such high risk features.

A few prior studies which found an incidence of cervical fracture of 5-7% and Vascular injury of 2% amongst all patients regardless of pre- senting features [8-10]. This lead to a prevailing recommendation in the literature to liberally screen hanging patients with the use of ad- vanced imaging (CT and MRI) regardless of their presenting symptoms [9]. This recommendation led to the adoption of a protocolized screen- ing process in our institution. In our study, the vast majority of patients (75%) received this full “Red” trauma evaluation that included the mo- bilization of scarce Hospital resources including senior surgical attend- ing, anesthesia, and OR staff. Additionally, N75% of patients had prioritized imaging including CT non-contrast head and cervical spine

Table 3

Highest level of trauma activation associated with increased likelihood of imaging and otolaryngology consultation.

CT head non-contrast 31% (p b 0.002)

CT neck non-contrast 19% (p b 0.001)

CT neck angiography 25% (p b 0.001)

X-ray of chest 25% (p b 0.001)

Otolaryngology consult 21% (p b 0.05)

S. Schuberg et al. / American Journal of Emergency Medicine 37 (2019) 737739 739

as well as CTA head and neck. Despite the use of these resources, no sig- nificant pathology was identified. Our findings question the utility of continuing this practice.

Several prior studies have sought to identify risk factors associated with significant injury amongst self-hanging patients in order to reduce over utilization of resources. Initial GCS upon emergency department presentation seems to be an important risk factor to predict both signif- icant injury and mortality [10,11,16]. Consistent with these studies, we found that patients with a GCS b8 had an absolute mortality of 67%. No patient with a GCS N8 died in our study. Martin et al. found similar re- sults in their large retrospective analysis of the National Trauma Data Bank, which found that patients with a GCS of b15 had an inpatient mor- tality of 29% compared to 1.5% amongst those with GCS 15 [9].

Conversely, a GCS of 15 may identify patients at low risk for clinically significant injury. In our analysis of 78 patients, a high proportion had a GCS of 15 (87%). Amongst these patients, we found no radiographic ev- idence of injury. Additionally, these patients had normal nasopharyngoscopies performed by a consulting otolaryngologist. This is consistent with a recent study by Subramanian et al. that found that a patient with a GCS of 15 and no signs or symptoms to suggest cervical or aerodigestive injury effectively ruled out significant injury amongst their study group [17]. This is similar to findings by Salim et al. who found in their cohort of patients a similar low rate of injury (2.6%) amongst patients with GCS N13 [8]. It is unclear from their data whether these injuries lead to significant Changes in management. In the larger National Trauma Data Bank study, patients with GCS 15 at time of dis- charge had no severe functional disability suggesting that they suffered only minor injuries [9]. Given these results, we are in agreement with Subramanian et al. that a screening process which incorporates both GCS with signs of aerodigestive or cervical Spinal injury may be a more efficient use of resources [17].

Our study has several limitations that are inherent in a retrospective review. This is a single center study with a high proportion of patients from a local correctional facility. As such, these results may not be appli- cable to other patient populations. Furthermore, patients presenting from a correctional facility may have underlying secondary gain and fac- tiously report or feign significant hanging injury. This could account for the low level of injury noted in our study, although 59% of these patients were reportedly found hanging at the corrections facility. In addition, while all efforts were made to identify patient charts through use of ICD-9 codes, some cases could have been missed through this process. We anticipate this number is very low given the protocolized trauma screening these patients received in our department.

In summary, in our study population with overall high GCS scores, we found no radiographic evidence of Significant injuries. Given that no medical management was changed by the imaging performed, we recommend that a higher level trauma activation and CT of the head, CT/CTA of the neck and CXR should be reserved for patients with judicial hangings (drop greater than or equal to the patient’s height), GCS b 15,

cervical spine tenderness, symptoms of neurological, vascular or airway injury and those with Abnormal vital signs. This more selective ap- proach to evaluate the patient with a mechanism of hanging is likely to reduce unnecessary cost, staff utilization, and radiation exposure.

Financial disclosures

None.

Conflict of interest

No conflicts of interest to be declared.

References

  1. Owens PL, Fingar KR, Heslin KC, Mutter R, Booth CL. Emergency department visits related to suicidal ideation, 2006-2013: statistical brief #220. Healthcare Cost and Utilization Project (HCUP) Statistical Briefs. Rockville (MD): Agency for Healthcare Research and Quality (US); 2006http://www.ncbi.nlm.nih.gov/books/NBK442036, Accessed date: 5 August 2018.
  2. National Suicide Statistics. National center for injury prevention and control. Ctr Dis Control Prev 2018https://www.cdc.gov/violenceprevention/suicide/statistics/index. html.
  3. National Violent Death Reporting System 2003-2015. Centers for disease control and prevention. www.cdc.gov/injury/wisqars/nvdrs.html; 2018.
  4. Newton K. Neck. In: Marx J, Hockberger R, Walls R, editors. Rosen’s Emergency Med- icine Concepts and Clinical Practice. 7th ed. Philadelphia: Mosby; 2010.
  5. Hua A, Shah KH, Garg M, Legome E, Ufberg J. A hanging and its complications. J Emerg Med 2016;51(6):691-6. https://doi.org/10.1016/j.jemermed.2016.07.016.
  6. Iserson KV. Strangulation: a review of ligature, manual, and postural neck compres- sion injuries. Ann Emerg Med 1984;13(3):179-85.
  7. Pesola GR, Westfal RE. Hanging-induced status epilepticus. Am J Emerg Med 1999; 17(1):38-40.
  8. Salim A, Martin M, Sangthong B, Brown C, Rhee P, Demetriades D. Near-hanging in- juries: a 10-year experience. Injury 2006;37(5):435-9. https://doi.org/10.1016/j.in- jury.2005.12.013.
  9. Martin MJ, Weng J, Demetriades D, Salim A. Patterns of injury and functional out- come after hanging: analysis of the National Trauma Data Bank. Am J Surg 2005; 190(6):836-40. https://doi.org/10.1016/j.amjsurg.2005.05.051.
  10. Nichols SD, McCarthy MC, Ekeh AP, Woods RJ, Walusimbi MS, Saxe JM. Outcome of cervical near-hanging injuries. J Trauma 2009;66(1):174-8. https://doi.org/10.1097/ TA.0b013e31817f2c57.
  11. Matsuyama T, Okuchi K, Seki T, Murao Y. Prognostic factors in hanging injuries. Am J

    Emerg Med 2004;22(3):207-10.

    Penney DJ, Stewart AHL, Parr MJA. Prognostic outcome indicators following hanging injuries. Resuscitation 2002;54(1):27-9.

  12. Hanna SJ. A study of 13 cases of near-hanging presenting to an Accident and Emer- gency Department. Injury 2004;35(3):253-6.
  13. Aufderheide TP, Aprahamian C, Mateer JR, et al. Emergency airway management in hanging victims. Ann Emerg Med 1994;24(5):879-84.
  14. Kaki A, Crosby ET, Lui AC. Airway and respiratory management following non-lethal hanging. Can J Anaesth 1997;44(4):445-50. https://doi.org/10.1007/BF03014468.
  15. Solhi H, Pazoki S, Mehrpour O, Alfred S. Epidemiology and prognostic factors in cases of near hanging presenting to a referral hospital in Arak, Iran. J Emerg Med 2012;43 (4):599-604. https://doi.org/10.1016/j.jemermed.2011.09.035.
  16. Subramanian M, Hranjec T, Liu L, Hodgman EI, Minshall CT, Minei JP. A case for less workup in near hanging. J Trauma Acute Care Surg 2016;81(5):925-30. https://doi. org/10.1097/TA.0000000000001231.

Leave a Reply

Your email address will not be published. Required fields are marked *