In the prehospital setting, it is imperative to identify potentially life-threatening injuries fast and adequately. These injuries may warrant immediate and tailored actions that potentially determine patient’s outcome.

The current study found that severe head trauma was rarely overlooked prehospital, whereas severe chest injuries were frequently unrecognized.

Other high prevalent injuries that remained frequently unrecognized in the prehospital setting were spine, abdominal, and pelvic injuries ranging from 48% for pelvic injuries up to 57% for spine injuries. This is in the nature of things, as the clinical examination in the field is very demanding and often has to take place under adverse conditions, with limited technical equipment and manpower. Despite these challenges in the field, all study patients in the present study were correctly transferred to a level 1 trauma center.

The high sensitivity for suspected head trauma in the prehospital setting of 96.3% is at the expense of a low specificity of only 47%. These findings are in line with a previous publication by Hasler et al. more than 10 years ago [7]. In this retrospective study, HEMS data of 433 patients who were admitted to a single trauma center were evaluated. Consistent with the current study, all patients were prehospitally examined by a physician. The sensitivity for suspected head trauma in the prehospital setting was reported to be 92.9%, similar to our results. These findings are supported by the fact that the level of consciousness and a brief neurologic examination were adequately assessable in the prehospital setting: patients with decreased level of consciousness or any signs of neurologic deficit are considered as potentially brain injured.

Severe chest injuries were present in almost every second patient included here, but remained unrecognized in almost 40% in the prehospital environment. It is well known that it is very challenging to recognize these conditions on scene. Of note, more than 40% of all missions were carried out during late or night shifts with limited light conditions and more than 30% during winter (November to March). Cold or wet weather and additional clothing may further hinder a proper prehospital clinical examination.

The high number of unrecognized prehospital chest injuries in HEMS missions is in line with previous studies [7, 14]. Interestingly, the current study showed that only 7 of 95 patients (7.4%) with pneumothorax received a chest decompression in the prehospital setting. However, it is also important to emphasize that not every pneumothorax requires a prehospital chest decompression. The four times higher chest decompression rate of pneumothoraces in patients who were preclinically intubated compared to those who were not intubated clearly shows the lower threshold for a chest decompression in intubated patients. The more difficult clinical assessment of intubated patients and the fear that high positive pressure during mechanical ventilation could convert a pneumothorax into a tension physiology may explain these findings [15].

However, the overall very low chest decompression rate of patients with a pneumothorax is of concern because any traumatic pneumothorax can progress and lead to tension physiology that has immediate life-threatening consequences, whereas relatively simple prehospital measures such as chest decompression have the potential to resolve this life-threatening situation. Consequently, it appears that patients may have benefitted from more responsive pneumothorax decompression in prehospital HEMS missions. These findings are also supported by a recently published study of our group which compared severely injured patients with those who died prehospitally [13]. In this study, chest trauma was identified in 45% of fatal missions and in almost 30% of severely injured patients. However, pneumothorax decompression was performed in only 17.2% of the fatal cohort and 3.7% in patients who were severely injured.

The lack of diagnostic tools is a major reason for the diagnostic uncertainty in the prehospital setting. A previous study evaluating 255 trauma patients reported that the use of pulse oximetry in addition to physical examination increased the accuracy of the prehospital diagnosis of lung contusion and early detection of tension pneumothorax [16]. Since 2022, prehospital point-of-care ultrasound has been gradually introduced by Rega HEMS. This tool however, although promising, requires careful evaluation. It is as of yet unclear if prehospital ultrasound leads to improved overall patients’ outcomes, in particular with regard to optimizing prehospital pneumothorax decompression rates [17, 18].

According to the current study, more than 50% of severe abdominal injuries remained unrecognized in prehospital HEMS missions. Similarly, Helm et al. evaluated 479 road traffic accident victims in order to determine the prehospital accuracy [19]. In contrast to the current study, overlooked severe abdominal injuries (AIS ≥ 3) in the prehospital setting were significantly less identified in hypotensive patients with SBP < 90 mmHg (28.6% versus 52.5%, p = 0.025). It is possible that existing hypotension in this cohort raised physician’s awareness for the presence of a possible abdominal trauma. In conclusion, measures such as blood pressure monitoring are helpful to increase diagnostic accuracy, and in the absence of other source of bleeding, abdominal hemorrhage should be strongly considered in the presence of hypotension. The increasing implementation of prehospital point-of-care ultrasound may be a promising tool for early identification of abdominal hemorrhage, however, warrants careful evaluation on different outcome levels [17]. Furthermore, it is apparent that hoist rescue (which is usually done in exposed and difficult terrain) hinders preclinical abdominal examination, but should trigger immediate transportation to a level I trauma center.

Another important finding was that almost 40% of pelvic injuries were unrecognized in the prehospital setting. This finding is in line with previous studies [20].

In German-speaking European countries, a pelvic binder is usually placed based on trauma mechanism or clinical findings at the earliest opportunity in order to reduce the risk of serious pelvic hemorrhage [21, 22]. The low detection rate of pelvic injuries may therefore be a main explanation of the low pelvic binder application rate with less than 60% for patients with unstable pelvic fractures. Hasler et al. [7] reported an even worse prehospital detection rate of pelvic injuries with 52%. This lower number compared to our study may be explained by the fact that Hasler et al. included a different patient population with consecutive trauma patients of any severity admitted to a single trauma center. This is reflected by the low mean ISS of 13 compared to the mean ISS of 22 in the current study.

Knowledge of factors associated with overlooked injuries is of paramount importance in order to help triage and trigger fast transportation to a higher level of trauma care. In the study of Wohlgemut et al., unrecognized injury was more common in patients with polytrauma, shock, and uncertainty of the physician [14]. The level of certainty of prehospital diagnoses was classified as certain or uncertain and was based on the Central Intelligence Agency on how humans describe levels of probability [23]. Uncertainty was discussed to be the result of evolving physiology, reduced patient responsiveness (e.g., from head injury or intoxication), or lack of availability of diagnostic adjuncts.

To our knowledge, this is the first multicenter study evaluating diagnostic accuracy in HEMS missions. Our findings identify particular problems with diagnostic accuracy of potentially life-threatening injuries during HEMS missions and may therefore be used for an improvement of prehospital practice. Furthermore, these data help the clinician in the receiving hospital to understand the difficult prehospital conditions and to better interpret and classify the data from the prehospital phase.

It is important to note that the non-HEMS trauma population and the patients attended to by HEMS differ substantially in Switzerland. As demonstrated, patients transported by HEMS are more severely injured compared to those admitted by GEMS. Furthermore, all HEMS come with a dedicated emergency physician, while such a prehospital physician is rare in non-HEMS missions. In addition, the range of operations differs due to the alpine topography of Switzerland. Many severe trauma cases are caused by recreational sports (mountaineering, climbing, paragliding, etc.) and are not accessible by GEMS. The findings presented in the present study are therefore not directly transferable to modes of transportation other than HEMS.

This study has several other limitations. Patients who died prehospitally and those who were not transported to one of the twelve level 1 trauma centers in Switzerland were not included in the present study. A possible selection bias might therefore be present. However, this is likely to be of limited relevance, because only level 1 trauma centers offer the entire spectrum of specialized polytrauma care and therefore have the national mandate to treat these patients. Moreover, many HEMS missions in this analysis involved individuals participating in recreational activities in the mountains during summer and winter (e.g., skiing, hiking, or climbing). Therefore, the findings may not be directly transferable to other countries.