European Journal of Cardiovascular Medicine

Trauma represents the foremost cause of mortality and disability in people under the age of 40 across the globe, contributing to almost six million deaths each year and imposing a significant burden of chronic morbidity. (1,2) Although emergency care has progressed considerably, gaps in trauma management persist, as shown by differences in outcomes even between countries with similar healthcare systems. (1) Survival and long-term recovery largely depend on prompt identification and rapid treatment of life-threatening problems, especially those affecting airway, breathing, and circulation. (1,3)

Emergency Medicine (EM) has emerged as a comparatively recent specialty in India. With the National Medical Commission mandating dedicated EM departments in medical colleges, emergency physicians are increasingly assuming frontline roles in trauma care, particularly in rural and district hospitals where continuous radiology services are limited. (4) In these settings, EM physicians not only direct resuscitative efforts but also often interpret point-of-care imaging, such as extended focused assessment with sonography for trauma (eFAST), chest X-rays, and computed tomography (CT) of the brain; to guide urgent decision-making. (5,6)

Imaging at the bedside is integral to trauma workup. According to the ATLS® protocol, chest X-ray, pelvic X-ray, and FAST/eFAST are performed during the primary survey, with CT scans added during the secondary survey. (3,7) Chest radiography is the most widely used tool, especially for identifying pneumothorax, hemothorax, and rib fractures. (8)

Likewise, eFAST is established as a rapid, non-invasive modality for identifying hemoperitoneum, pneumothorax, and pericardial effusion, and is endorsed as the standard of care by both the Eastern Association for the Surgery of Trauma (EAST) and the American College of Emergency Physicians (ACEP). (9–11) In cases of acute head trauma, non-contrast CT of the brain remains the gold standard, enabling prompt detection of intracranial hemorrhage, contusions, and midline shift that may necessitate neurosurgical intervention. (12,13)

Multiple studies have reported that emergency physicians achieve acceptable accuracy in trauma imaging interpretation. For chest radiographs, agreement with radiologist’s ranges from 83% to 99% for key findings such as pneumothorax and hemothorax. (8,14) Meta-analyses of eFAST indicate sensitivities between 74% and 93%, with specificities consistently above 95%. (9,10,15) Likewise, emergency physicians demonstrate substantial diagnostic concordance in CT brain interpretation, particularly for subdural and extradural hematomas. (6,16,17) Nonetheless, interpretation errors remain, often influenced by the physician’s training, the clinical setting, and the quality of the images obtained.

With the growing involvement of emergency physicians in frontline trauma management in India, it is imperative to assess the accuracy of their imaging interpretations relative to radiologists. The present study was designed to compare the diagnostic performance of emergency physicians and radiologists in interpreting chest radiographs, eFAST, and CT brain in trauma cases. In addition, secondary analyses evaluated their performance in HRCT thorax and CECT abdomen.

Study design and setting

This was a cross-sectional analytical study conducted at a tertiary care teaching hospital in Ahmedabad, India. The study was carried out between January 2019 and June 2021 in the Emergency Department (ED) of Sardar Vallabhbhai Patel Institute of Medical Sciences and Research (SVPIMSR). During this period, the hospital also served as a designated COVID-19 center, which affected trauma case volumes and overall sample size.

Study population

All trauma patients presenting to the emergency department during the study period and undergoing bedside imaging, namely chest radiography, extended focused assessment with sonography in trauma (eFAST), and/or computed tomography (CT) brain were eligible for inclusion. Patients with minor traumatic injuries not requiring imaging were excluded from the study.

Imaging modalities

·         Chest X-ray: All chest radiographs were obtained in the supine anteroposterior (AP) view, as patients with trauma were immobilized and unable to sit upright.

·         eFAST: Extended Focused Assessment with Sonography for Trauma was performed using a curvilinear probe for abdominal and pelvic views, and a high-frequency linear probe for pleural evaluation.

·         CT Brain: Non-contrast CT (NCCT) brain was performed using standard trauma protocols.

·         Secondary modalities: High-resolution CT (HRCT) thorax and contrast-enhanced CT (CECT) abdomen were performed when clinically indicated and analyzed as secondary outcomes.

Interpretation and blinding

Initial imaging interpretations were performed by emergency department physicians at the bedside and recorded on a predesigned proforma. The radiologists’ reports served as the reference standard. Emergency physicians were blinded to the radiologists’ interpretations at the time of initial assessment. Patient management and treatment decisions were guided by the radiology-confirmed reports.

Training of emergency physicians

The primary investigator completed a mandatory two-week rotation in the radiology department to gain familiarity with imaging interpretation techniques, supplemented by ongoing supervised training during the course of residency.

Outcome measures

The accuracy of ED physicians’ interpretations was compared with radiologists’ reports. For each imaging modality, true positives (TP), true negatives (TN), false positives (FP), and false negatives (FN) were recorded.

•              Sensitivity = TP / (TP + FN)

•              Specificity = TN / (TN + FP)

•              Accuracy = (TP + TN) / (TP + TN + FP + FN)

Statistical analysis

Data were entered into Microsoft Excel 2019 and analyzed using SPSS version 25.0 (IBM Corp., Chicago, IL, USA). Descriptive statistics were calculated, with continuous variables expressed as mean ± standard deviation (SD) or median with interquartile range (IQR), and categorical variables summarized as frequencies and percentages. Diagnostic performance of emergency department physicians compared with radiologists was assessed by calculating sensitivity, specificity, and overall accuracy, along with 95% confidence intervals. A two-tailed p-value of <0.05 was considered statistically significant.

Ethical Considerations

Prior to initiation, the study protocol was reviewed and approved by the Institutional Ethics Committee (IEC).

Cohort characteristics (Table 1)

We included 137 trauma patients, of whom 117 (85.4%) were male. The mean age was 40 years (median 36). Two-wheeler crashes were the leading mechanism with 90 cases (65.7%), followed by pedestrian injuries in 27 patients (19.7%), four-wheeler collisions in 14 patients (10.2%), and other mechanisms in 6 patients (4.4%). Single-system injuries were more common, most frequently brain only in 92 patients (67.2%). Thorax only and abdomen only were seen in 5.1% and 5.8% of patients, while 30 patients (21.9%) had multi-system trauma. Considering overall counts that were not mutually exclusive, brain injuries totaled 124, thoracic injuries 29, and abdominal injuries 27.

Chest radiography (Table 2 and 7)

Agreement between emergency physicians (EPs) and radiologists (Rads) was strong. For pneumothorax, EPs achieved a sensitivity of 66.7%, with 100% specificity and an overall accuracy of 97.8% (6 detected by EPs compared to 9 by Rads, with 3 missed). Hemothorax was most frequently overlooked when minimal on the left side, with 2 detected compared to 7 by Rads. This resulted in 50.0% sensitivity, 100% specificity, and 95.6% accuracy overall. Detection of rib fractures, both single and multiple, showed complete agreement with 100% sensitivity, specificity, and accuracy.

eFAST (Table 3 and 7)

Using severity-graded categories, detection of intra-abdominal free fluid showed a sensitivity of 72.7%, specificity of 99.1%, and overall accuracy of 94.9%. EPs identified 17 cases compared to 22 by Rads, with a single false positive in the moderate group. For thoracic eFAST hemothorax (right and left combined across all severities), sensitivity was 80.8% with 100% specificity, yielding an accuracy of 96.4%. Visceral contusions were recognized with complete concordance, achieving 100% sensitivity and specificity.

CT brain (Table 4 and 7)

Concordance between EPs and radiologists was excellent for intracranial findings. Hemorrhagic contusions were detected with 91.5% sensitivity, 100% specificity, and 94.9% accuracy. Intraventricular hemorrhage and midline shift were identified with 100% accuracy. Subarachnoid hemorrhage was slightly under-recognized, with 68 cases detected by EPs compared to 79 by Rads, giving 86.1% sensitivity, 100% specificity, and 92.7% accuracy. Both subdural and epidural hemorrhages demonstrated strong performance, with sensitivities of 90.9% and 93.0%, and accuracies of 96.4% and 97.8%, respectively. Interpretation of cerebral edema was also highly reliable, with 96.2% sensitivity and 98.5% accuracy.

Subsets: HRCT thorax and CECT abdomen-pelvis (Table 5, 6 and 7)

On advanced thoracic imaging, both pneumothorax and pneumomediastinum were identified with 100% accuracy. Detection of hemothorax, aggregated across sides and severities, demonstrated 88.5% sensitivity, 100% specificity, and 97.1% accuracy. Lung contusions were less sensitive at 76.9% but still showed 100% specificity, maintaining an overall accuracy of 97.1%.

On CECT abdomen–pelvis, severity-based assessment of free fluid yielded a sensitivity of 68.2%, with perfect specificity (100%) and an overall accuracy of 94.2%. Most of the discrepancies were attributable to missed cases with only minimal fluid collections. Visceral contusions involving the kidney and liver were detected with high reliability, showing 90.0% sensitivity, 100% specificity, and 99.3% accuracy. For contrast leaks from bowel or vascular structures, sensitivity was 75.0%, while specificity remained 100%, resulting in an overall diagnostic accuracy of 97.8%.

Overall interpretation

Across all imaging modalities, EP interpretations demonstrated consistently high specificity, almost always close to 100%, and accuracy typically greater than 95%. The main sensitivity deficits were confined to subtle or small-volume findings, most notably minimal hemothorax on the left side and minimal free fluid. In contrast, detection of clinically significant lesions such as pneumothorax on HRCT, midline shift, intraventricular hemorrhage, and major subdural or epidural hematomas showed excellent concordance with radiologists.

Table 1. Demographic and clinical profile of trauma patients

Table 2. Diagnostic accuracy of chest X-ray interpretation: emergency physicians vs. radiologists

Table 3. Diagnostic accuracy of eFAST interpretation: emergency physicians vs. radiologists

Table 4. Diagnostic accuracy of CT brain interpretation: emergency physicians vs. radiologists

Table 5. Diagnostic accuracy of HRCT thorax interpretation: emergency physicians vs. radiologists

Table 6: Diagnostic accuracy of CECT abdomen interpretation: emergency physicians vs. radiologists

Table 7. Overall sensitivity, specificity, and accuracy across all imaging modalities

* CXR hemothorax = right + left (minimal + mild).
† “Free fluid” uses severity rows only (to avoid double counting with site-specific).
‡ eFAST hemothorax = right + left, all severities.
§ HRCT hemothorax = right + left, all severities.
‖ CECT visceral contusion = kidney + liver combined.

TP = min (EP, Rad) = Rad − Missed
We can’t have more true positives than the gold-standard positives (Rad) or more than EP positives; therefore, we take the minimum.
If we already have Missed by EP (i.e., FN), then TP = Rad − Missed.

FP = max (EP − Rad, 0)
If EP called more positives than the radiologist, the excess are false positives.
If EP ≤ Rad, FP is 0 (we use max to avoid negative numbers).

TN = 137 − Rad − FP
Out of the 137 patients, we subtract all who were radiologist-positive (Rad) and all EP false positives (FP). The remainder are true negatives.

This study highlights that emergency physicians (EPs) are able to interpret commonly used bedside trauma imaging with a high level of diagnostic accuracy when compared with radiologists. Across the three cornerstone modalities in early trauma evaluation, namely chest radiographs, eFAST, and CT brain, EPs achieved sensitivities and specificities that fall within clinically acceptable limits, particularly for lesions of immediate life-threatening relevance.

Chest radiograph interpretation by EPs showed strong agreement with radiologists. Pneumothorax was detected with an accuracy approaching 98%, a result consistent with earlier studies that reported concordance exceeding 90%. (8,14) In contrast, hemothorax was less consistently recognized, especially in cases of minimal left-sided collections, leading to a sensitivity of only 50% despite perfect specificity. This observation reflects previous reports indicating that small-volume hemothoraces are often overlooked on supine radiographs regardless of reader expertise. (18) By comparison, rib fractures, both single and multiple, were identified with complete agreement, supporting earlier findings that osseous injuries are more readily detected than subtle parenchymal or pleural abnormalities. (19)

Bedside ultrasonography using the eFAST protocol was also interpreted with high reliability. Overall accuracy for intra-abdominal free fluid was around 95%, though sensitivity fell to 73% owing to occasional under-recognition of minimal collections. Importantly, no moderate or large hemoperitoneum was missed. These findings correspond with systematic reviews reporting sensitivities of 74–93% and specificities exceeding 95% for EP-performed eFAST. (9,15,20) In line with earlier studies, false negatives in this cohort were confined to very small-volume fluid, which often has limited acute relevance. (21) Thoracic applications of eFAST were similarly robust, with hemothorax detected at 81% sensitivity and 100% specificity, reinforcing evidence that trained EPs can perform and interpret trauma ultrasound with reliability. (22)

Interpretation of CT brain by EPs demonstrated excellent diagnostic performance, with accuracies surpassing 90% for hemorrhagic contusions, subdural hematomas, and extradural hematomas. All cases of clinically significant midline shift were detected without error. These findings align with prior reports citing sensitivities between 79–92%, depending on lesion type. (13,16,17) While subarachnoid hemorrhage (SAH) was slightly under-recognized, with sensitivity of 86%, overall accuracy remained high at 93%, comparable to existing literature. (23) Given the central role of CT brain in neurosurgical decision-making, these findings emphasize that EPs can reliably detect life-threatening intracranial pathology in the acute setting.

Evaluation of advanced imaging, including HRCT thorax and CECT abdomen, showed a similar trend. Specificity was consistently high, while sensitivity was modestly reduced for subtle abnormalities. On HRCT, pneumothorax and pneumomediastinum were detected with 100% accuracy, hemothorax reached 88% sensitivity, and lung contusions 77% sensitivity. On CECT abdomen, free fluid detection achieved 94% accuracy, though minimal collections were occasionally overlooked. Visceral contusions involving the kidney and liver and vascular contrast leaks were identified with sensitivities of 90% and 75%, respectively, with near-perfect specificity. These results are consistent with international reports indicating that EP interpretations of thoracoabdominal CT achieve accuracy rates within 10% of radiologists. (24,25)

The relevance of these findings is amplified in India and other low- and middle-income countries, where radiologists are not universally available round the clock, particularly in rural and district settings. In such environments, EPs may be the only clinicians available to interpret imaging. Our results strengthen global evidence that, with appropriate training, EPs can interpret trauma imaging reliably and enable timely intervention. (26,27)

Certain limitations merit mention. The sample size was smaller than expected due to reduced trauma volumes during the COVID-19 pandemic. Being a single-center study, the findings may not be generalizable across diverse healthcare environments. Inter-observer variability among EPs was not assessed, which may have influenced consistency. Finally, subtle abnormalities such as minimal hemothorax or trace free fluid were more frequently missed, underscoring the need for structured training and imaging curricula to further improve diagnostic performance.

Emergency physicians demonstrated consistently high diagnostic accuracy across essential trauma imaging modalities, including chest radiographs, eFAST, CT brain, and advanced studies such as HRCT thorax and CECT abdomen. Sensitivity was occasionally lower for subtle or low-volume findings such as minimal hemothorax or small free-fluid collections, but clinically critical conditions such as pneumothorax, midline shift, and major intracranial or thoracoabdominal injuries were detected with accuracy comparable to radiologists. These results emphasize the important role of trained emergency physicians in facilitating timely trauma diagnosis and management, particularly in healthcare settings where radiology support is limited or not continuously available. The integration of structured radiology training within emergency medicine curricula, together with the adoption of emerging technologies such as artificial intelligence, has the potential to further refine diagnostic performance and enhance the overall quality and timeliness of trauma care.

1.       Artigas Martín JM, Martí de Gracia M, Claraco Vega LM, Parrilla Herranz P. Radiología e imagen en el traumatismo grave. Med Intensiva. 2015;39(1):49-59.

2.       Centers for Disease Control and Prevention. Global Road Safety. Available from: https://www.cdc.gov/injury/features/global-road-safety/index.html.

3.       Advanced Trauma Life Support (ATLS®): The 10th edition. Chicago: American College of Surgeons; 2018.

4.       Harrigal CL, Erly WK. On-call radiology: community standards and current trends. Semin Ultrasound CT MR. 2007;28(2):85-93.

5.       Körner M, Krötz MM, Degenhart C, Pfeifer KJ, Reiser MF, Linsenmaier U. Current role of emergency US in patients with major trauma. Radiographics. 2008;28(1):225-42.

6.       Shah H, Jarwani B, Gajjar M. Can ED physician interpret NCCT brain reliably in head injury victims’ emergency management? J Evol Med Dent Sci. 2015;4(8):1306-11.

7.       Kool DR, Blickman JG. Advanced Trauma Life Support. ABCDE from a radiological point of view. Emerg Radiol. 2007;14(3):135-41.

8.       Safari S, Baratloo A, Negida AS, Sanei Taheri M, Hashemi B, Hosseini Selkisari S. Comparing the interpretation of traumatic chest X-ray by emergency medicine specialists and radiologists. Arch Trauma Res. 2014;3(4):e22189.

9.       Netherton S, Milenkovic V, Taylor M, Davis PJ. Diagnostic accuracy of eFAST in the trauma patient: a systematic review and meta-analysis. CJEM. 2019;21(6):727-38.

10.    Brooks A, Davies B, Smethhurst M, Connolly J. Prospective evaluation of non-radiologist performed emergency abdominal ultrasound for haemoperitoneum. Emerg Med J. 2004;21(5):e5.

11.    American College of Emergency Physicians. Emergency ultrasound guidelines. Ann Emerg Med. 2009;53(4):550-70.

12.    Schweitzer AD, Niogi SN, Whitlow CT, Tsiouris AJ. Traumatic brain injury: imaging patterns and complications. Radiographics. 2019;39(6):1571-95.

13.    Khan A, Qashqari S, Al-Ali AA. Accuracy of non-contrast CT brain interpretation by emergency physicians: a cohort study. Pak J Med Sci. 2013;29(2):549-53.

14.    Hetela SA, Aseri ZA, Delaney JS. Comparison of observer performance between emergency physicians at different levels of experience using digital imaging for detecting pneumothorax in the emergency department. Emerg Med. 2012;2:122.

15.    Soyuncu S, Cete Y, Bozan H, Kartal M, Akyol AJ. Accuracy of physical and ultrasonographic examinations by emergency physicians for the early diagnosis of intra-abdominal haemorrhage in blunt abdominal trauma. Injury. 2007;38(5):564-9.

16.    Talebian MT, Kavandi E, Farahmand S, Shahlafar N, Arbab M, Seyedhosseini-Davarani S, et al. Comparing the brain CT scan interpretation of emergency medicine team with radiologists’ report and its impact on patients’ outcome. Emerg Radiol. 2015;22(3):261-8.

17.    Arendts G, Manovel A, Chai A. Cranial CT interpretation by senior emergency department staff. Australas Radiol. 2003;47(4):368-74.

18.    Exadaktylos AK, Sclabas G, Schmid SW, Schaller B, Zimmermann H. Do we really need routine computed tomographic scanning in the primary evaluation of blunt chest trauma in patients with “normal” chest radiograph? J Trauma. 2001;51(6):1173-6.

19.    Ziegler K, Feeney JM, Desai C, Shapiro D, Marshall WT, Twohig M. Retrospective review of the use and costs of routine chest X-rays in a trauma setting. J Trauma Manag Outcomes. 2013;9:2.

20.    Bhoi S, Sinha TP, Ramchandani R, Kurrey L, Galwankar S. Accuracy of FAST performed by non-radiologists compared with radiologists in a level 1 trauma center in India. J Emerg Trauma Shock. 2013;6(1):42-6.

21.    Wongwaisayawan S, Suwannanon R, Prachanukool T, Sricharoen P, Saksobhavivat N, Kaewlai R. Trauma ultrasound. Ultrasound Med Biol. 2015;41(9):2543-61.

22.    Basnet S, Shrestha SK, Pradhan A, et al. Diagnostic performance of the extended focused assessment with sonography for trauma (EFAST) in a tertiary care hospital of Nepal. Trauma Surg Acute Care Open. 2020;5(1):e000438.

23.    Ro YS, Shin SD, Holmes JF, Song KJ, Park JO, Cho JS, et al. Comparison of clinical performance of cranial computed tomography rules in patients with minor head injury: a multicenter prospective study. Acad Emerg Med. 2011;18(6):597-604.

24.    Kang MJ, Sim MS, Shin TG, Jo IJ, Song HG, Song KJ, et al. Accuracy of emergency medicine resident interpretations of abdominal CTs in patients with non-traumatic abdominal pain. J Korean Med Sci. 2012;27(10):1255-60.

25.    DeCorato DR, Kagetsu NJ, Ablow RC. Off-hours interpretation of radiologic images in the emergency department: efficacy of teleradiology. AJR Am J Roentgenol. 1995;165(6):1293-6.

26.    Nathens AB, Jurkovich GJ, Maier RV, Grossman DC, MacKenzie EJ, Moore M, et al. Relationship between trauma center volume and outcomes. JAMA. 2001;285(9):1164-71.

27.    Lyon M, Sturgis L, Lendermon D, Kuchinski AM, Mueller T, Loeffler P, et al. Rural ED transfers due to lack of radiology services. Am J Emerg Med. 2015;33(11):1630-4.