Prevalence of Serious Injuries in Low Risk Trauma Patients

Abstract

Objectives:

Computed Tomography (CT) utilization is widespread in contemporary Emergency Departments (EDs). CT overuse leads to radiation exposure, contrast toxicity, overdiagnosis, and incidental findings. This study explores the prevalence of clinically significant injuries in patients identified as low-risk trauma patients (LRTPs) using newly created criteria that account for the patient’s age, trauma mechanism, assessability (which relies on level of consciousness, intoxication, and neurologic deficits), vital signs and other evidence of hypoperfusion, bleeding risk, and past medical history.

Methods:

This was a 6-month retrospective chart review of all LRTPs presenting to a level 1 trauma center in Queens, New York. Data abstraction was performed independently by two abstractors and discrepancies adjudicated by the senior author. Patients were identified using the hospital trauma registry and two reports, created by the researchers, identifying selected chief complaints and discharge diagnoses.

Results:

750 patients were identified of which 352 (46.93%) received one or more CT scans. There were a total of 790 CT scans ordered, of which 731 (92.53%) were negative for acute injury. There were 13 clinically significant injuries of which only one (0.13%) required immediate intervention. There were no mortalities in this LRTP group.

Conclusion:

The incidence of clinically significant injuries in this population is very low and injuries requiring immediate intervention is even lower. CT utilization in LRTPs should be guided by an explicit consideration of benefit and harm for each patient.

Introduction

Use of computed tomography (CT) in trauma has escalated dramatically in American Emergency Departments (EDs) and CT overutilization has been identified as an important care concern. [1–4] In 1989, Kassirer described causes of over-testing, including pressure from patients and administration, fear of malpractice, and the accessibility of diagnostic tests. [5] Between 1998 and 2007, the use of CT in the ED for the evaluation of trauma patients increased three-fold without an increase in the diagnosis of life-threatening conditions [6] and in an observational study by Asha et al., the adjusted odds of receiving a radiation dose of 20 mSv was 2.2 times higher after the introduction of the panscan protocol compared to odds before, with little evidence of reduction in missed injuries. [7]

Some trauma centers are now using pan-CT as standard care for all trauma activations; proponents of this practice suggest that CT scans are highly sensitive in detecting injuries in blunt trauma patients [8–15] and, when performed early in the patient’s ED course, decrease throughput time while expediting the identification of injuries requiring emergent intervention. [13, 15–17] Critics of a “pan-CT standard” cite concerns around overuse leading to a variety of harms including overdiagnosis, radiation exposure, contrast-related kidney injury, allergy, and extravasation, incidental findings, and long-term risk of malignancy. [18–23] Radiation exposure has been estimated to cause 12.5 deaths per 10,000 scans [6, 18, 24] and as many as 1.5% to 2.0% of all cancers in the United States may be attributable to radiation from CT studies. [18] Many studies that correlate CT use with mortality benefit include severe trauma presentations in older patients; [25] these patients are more likely to benefit from CT given their injury burden, and less likely to be harmed by radiation given their age. In developing programs to optimize CT utilization in trauma, it is therefore prudent to focus initial quality improvement efforts on younger patients, presenting with minor trauma, who are more susceptible to the harms of radiation exposure, with the risk of cancer mortality after a similar radiation dose increasing dramatically as age decreases. [26] Our study evaluates newly created criteria that account for the patient’s age, trauma mechanism, assessability (which relies on level of consciousness, intoxication, and neurologic deficits), vital signs and other evidence of hypoperfusion, bleeding risk, and past medical history, to identify a subset of trauma patients who are at low risk for significant injury. The aim of this study is to explore the prevalence of clinically significant injuries and emergent interventions in patients identified as low-risk trauma patients (LRTPs) by these criteria.

Methods

We conducted a retrospective chart review of all LRTPs that presented to Elmhurst Hospital Center (EHC), an urban academic level 1 trauma center in Queens, New York (annual ED census 156,482, trauma activation census of 1003), for 6 months in 2017. Permission for the study was granted by the Institutional Review Board (IRB) at Icahn School of Medicine at Mount Sinai (consent was waived by the IRB).

Data abstraction was performed independently by two abstractors who were aware of the purpose of the study; discrepancies were adjudicated by consensus with input from the senior author. The following data elements were abstracted for each LRTP: age, gender, mechanism of injury, number of CTs ordered, types of CTs ordered, injuries identified on CTs, number of negative CTs, disposition, any missed injuries identified during hospital stay and/or later visits to the ED, 30-day return visit, prisoner or not, incidental findings on CT, time to disposition, and significant intervention.

These patients were identified using our hospital trauma registry and two reports, created by the researchers, identifying selected chief complaints: “fall,” “trauma,” “motor vehicle crash,” “motorcycle crash,” “head injury,” “assault victim,” “head laceration” and discharge diagnoses: “motor vehicle accident,” “fall,” “trauma,” “head injury,” “assault,” “pedestrian bicycle accident,” “head trauma, initial encounter,” “blunt head trauma, initial encounter,” “blunt trauma”, “fall (on) (from) other stairs and steps,” “contusion,” “closed head injury,” “head injuries, initial encounter”.

Patients had to meet all six of the following LRTP inclusion criteria:

1. Blunt trauma mechanism in patients aged 18-40 years

2. Persistently normal level of consciousness, no neurologic deficit, and GCS 15 in the ED

3. Without > 1 reading of heart rate >100 bpm or respiratory rate >20 breaths per minute in ED

4. Never hypotensive (systolic blood pressure <90 mmHg) or demonstrating evidence of hypoperfusion

5. Not anticoagulated

6. Without significant baseline disease (ASA class I or II)

Patients were excluded if intoxicated (clinically or serum alcohol level > 80 mg/dL), Emergency Severity Index 4 or 5, transferred from another institution, eloped or left against medical advice, baseline paraplegia/quadriplegia, isolated trauma to the extremities, hanging mechanism, or a medical cause precipitating the trauma.

Injuries were classified as significant according to criteria based on Smith et al [27] (Table 1). Clinically significant facial injuries are not defined by Smith et al [27] and we used the following criteria to classify these: all mandible fractures and any other facial fractures requiring admission for a surgical procedure. Significant intervention was defined as surgery, thoracostomy, pericardial drainage, or blood transfusion within the first 24 hours of patient’s admission.

Table 1:

CT Criteria for Clinically Significant Injury

Head

Substantial epidural or subdural hematoma (1.0 cm in width or with mass effect)

Cerebral contusion 1.0 cm in diameter or at >1 site

Extensive subarachnoid hemorrhage—mass effect or sulcal effacement

Signs of herniation

Basal cistern compression or midline shift

Hemorrhage to posterior fossa

Intraventricular hemorrhage

Bilateral hemorrhage

Depressed or diastatic skull fracture

Pneumocephalus

Diffuse cerebral edema

Diffuse axonal injury

Cervical spine

Any fracture (except those deemed not clinically significant per NEXUS criteria)

Ligamentous injury

Any other injury requiring surgical intervention, causing a neurological deficit, or causing death Chest

Contusion requiring oxygen

Any pneumothorax

Hemothorax

Traumatic pericardial effusion

Pneumomediastinum

Great vessel injury

Esophageal injury

Multiple rib fractures (only if admitted for IV pain medication or epidural)

Flail segment of ribs

Any other injury requiring surgical intervention or causing death

Abdomen and pelvis

Any injury requiring admission or surgical intervention or causing death

Thoracic/lumbar spine

Any fracture (except isolated transverse process fracture or those not requiring a brace)

Any other injury requiring surgical intervention, requiring an orthotic brace, causing a neurological deficit, or causing death

Results

A 6-month chart review demonstrated 750 total patients that met LRTP criteria, of which 352 (46.93%) received one or more CT scans. The male to female ratio was 2.38 and motor vehicle accident (MVA), assault, and mechanical fall contributed to 86.93% of all traumas. Characteristics of included patients are shown in Table 2.

Table 2:

Patient characteristics

Variable	Total Number of Patients (%)	Number of Patients who received ≥ 1 CT (%)	Number of Patients who did not receive a CT (%)
Gender
Male	528 (70.4)	263 (74.7)	265 (66.6)
Female	222 (29.6)	89 (25.3)	133 (33.4)
Mechanism of injury
MVA	250 (33.3)	98 (27.8)	152 (38.2)
Assault	210 (28.0)	128 (36.4)	82 (20.6)
Mechanical fall	192 (25.6)	83 (23.6)	109 (27.4)
Pedestrian struck	32 (4.3)	22 (6.3)	10 (2.5)
Other	66 (8.8)	21 (5.9)	45 (11.3)
Trauma color
Non-activation	715 (95.3)	320 (90.9)	395 (99.2)
Yellow (moderate severity)	27 (3.6)	26 (7.4)	1 (0.3)
Green (stable patients)	8 (1.1)	6 (1.7)	2 (0.5)
Disposition
Discharge	703 (93.7)	326 (92.6)	377 (94.7)
Admit	37 (4.9)	22 (6.3)	15 (3.8)
Psychiatric ED	8 (1.1)	4 (1.1)	4 (1.0)
Observation unit	1 (0.1)	0 (0)	1 (0.2)
Labor and delivery	1 (0.1)	0 (0)	1 (0.2)
Prisoner
Yes	45 (6.0)	29 (8.2)	16 (4.0)
No	705 (94.0)	323 (91.8)	382 (96.0)

A total of 790 CT scans were ordered for 750 patients of which 731 (92.53%) were negative for acute injury. CT imaging was categorized into body regions: head, face, mandible, cervical spine, thoracic spine, lumbar spine, chest, abdomen/pelvis. The most common CT obtained throughout the entire study period was head followed by cervical spine (Table 3).

Table 3:

Number of CTs by body region

Body region	Number of CTs
Head	274
Facial	113
Mandible	14
Cervical spine	204
Thoracic spine	57
Lumbar spine	62
Chest	31
Abdomen/pelvis	40

Thirteen clinically significant injuries were identified in this cohort, including a patient with multiple rib fractures and a hemopneumothorax, another patient with 2 rib fractures requiring admission for pain control, a patient with a rib fracture and pneumothorax, a patient with multiple intracranial hemorrhagic contusions and subarachnoid hemorrhage, two patients with pelvic fractures requiring admission, another patient with fracture of the occipital calvarium, a patient with spinal compression fracture, four patients with facial fractures, and a patient with rib fractures and splenic laceration. There were 37 admissions and 16 (43.24%) of these were to orthopedics for extremity injuries.

Out of the 13 clinically significant injuries, one (0.13%) patient had an injury requiring immediate intervention (splenic laceration requiring IR embolization within 24 hours). There were no mortalities in this LRTP group and none of the patients returned to the ED with a missed injury.

Among the 715 LRTPs not activated for trauma, the mean number of CT scans per patient was 0.91; 8 “green” activation patients who met LRTP criteria received a mean of 2.38 scans per patient and 27 “yellow” activation patients received a mean of 4.48 CT scans (see discussion).

Discussion

CT rapidly and accurately identifies serious injuries and “clears” patients of such injuries, facilitating rapid disposition. [28] Unfortunately, CT use carries a variety of important harms, mandating that physicians, charged with acting in their patients’ best interest, use discretion when deciding on whether to use CT in the evaluation of a given patient.

The primary aim of this observational study was to explore the prevalence of clinically significant injuries and subsequent emergent interventions in a subset of trauma patients who are designated low risk for serious injury on their initial evaluation by a set of newly created clinical criteria. LRTP criteria were developed de novo through consensus, by a committee that included designees from the hospital Trauma Surgery and Emergency Medicine departments.

Trauma activation at our facility, which involves a protocoled personnel response from both the emergency and surgery services, occurs according to three categories of severity: Red, Yellow, and Green, which correspond to critical severity (Level I), moderate severity (Level II) and stable patients, respectively (see supplementary material). However, 95.33% of LRTPs did not involve a trauma activation and were exclusively managed in the ED by emergency physicians. Level II (Yellow) trauma activation patients in this cohort were 4.89 times more likely to receive a CT scan compared to a non-activation trauma. Though trauma activation implies higher injury severity, all included patients had to meet LRTP criteria and therefore had similar baseline characteristics.

The majority of the injuries identified in our study were facial fractures that did not require any intervention (Table 4). Though some authors have associated the use of CT in trauma with decreased ED length of stay (LOS); [13, 15–17] in the present study, the average LOS when the patient did not receive a CT was 356 minutes versus 516 minutes when the patient received one or more CT scans. The use of CT was associated with a statistically significant increase in average LOS (p<0.0001). 93.73% of patients in this cohort were discharged and none of the patients returned to the ED with a missed injury. 43% of LRTP admissions were to orthopedics for extremity injuries that were identified on X-rays.

Table 4:

Injuries identified on CTs (number of patients)

Head

Frontal lobe hemorrhagic contusion (2), Occipital fracture, Subarachnoid hemorrhage (4), Subdural hematoma, Temporal bone fracture (2)

Face/mandible

Lamina papyracea fracture, Lateral buttress fracture, Mandible fracture (3), Maxillary fracture (5), Nasal bone fracture (21), Orbital fracture (13), Zygomatic arch fracture, Zygomaticofrontal fracture, Zygomaticomaxillary complex fracture (2)

Spine

Lumbar transverse process fracture, Thoracic compression fracture, Thoracic endplate fracture, Thoracic spinous process fracture

Chest

Hemopneumothorax, Pneumothorax, Rib fractures (4)

Abdomen/Pelvis

Acetabular fracture, Ilium fracture, Pubic rami fracture, Sacral wing fracture, Splenic laceration

Limitations

This is a retrospective observational study that does not account for provider medical decision making; that a CT scan is negative does not imply that it was inappropriately ordered. The study utilizes the trauma registry and two reports created by the researchers; patients that would have otherwise met criteria but were not captured by these reports could have skewed our results. The present study was conducted at a single center and therefore may not be applicable to other settings.

Conclusion

Our study demonstrates that the incidence of clinically significant injuries in LRTPs is very low and injuries requiring immediate intervention is even lower. CT utilization in this population should be guided by an explicit consideration of benefit and harm for each patient.