Aspiration occurs commonly in severe trauma patients, and risk factors are known to include loss of consciousness, facial injury, and elevated intra-abdominal pressure [1,2]. Although it is not possible to ascertain the exact incidence of aspiration in severe trauma, depending on the research methods, the incidence is reported to be 6–54% [3-6]. Aspiration is known to cause lung disease from mild bronchiolitis to aspiration pneumonia and aspiration pneumonitis [7,8], and the development of these lung diseases increases the rate of progression acute respiratory distress syndrome (ARDS) [9,10]. Therefore, prevention and awareness of aspiration is an important issue in the management of severe trauma patients.

Previous studies on aspiration in severe trauma patients have investigated the presence of foreign bodies in the oropharynx, or in the larynx during endotracheal intubation [9,10]. These methods are limited by the fact that oropharyngeal foreign bodies cannot be considered clear evidence of aspiration, and that laryngeal foreign bodies cannot be checked without attempting intubation. It is expected that the diagnostic accuracy of testing for aspiration could be improved by using chest computed tomography (CT). However, there have been few previous studies on severe trauma patients showing findings of aspiration on chest CT. Hence, the purpose of the present study was to identify risk factors for aspiration in severe trauma patients by comparing patients who showed signs of aspiration lung disease on chest CT and those who did not. In addition, clinical differences were compared between the two patient groups, including the rate of endotracheal intubation in the emergency room (ER), the intensive care unit (ICU) admission rate, and the in-hospital mortality rate.

There were 3,159 patients registered in the KTDB between January 1st 2014 and December 31st 2019. Of these, 1,333 patients without chest CT scans were excluded, and 1,326 patients with AIS-chest ≥1 were excluded. A further 75 patients were excluded due to mis-recorded or duplicate chart numbers, transfer from the ER to another hospital, being under 18 years of age, or due to trauma affecting the interpretation of CT scans, and the remaining 425 patients were included in the final analysis. Of these 425 patients, there were 48 patients showing aspiration on chest CT (11.2%) and 377 patients showing no aspiration (88.7%) (Fig. 1).

When general characteristics were compared between the aspiration group and the non-aspiration group, endotracheal intubation was performed in the ER for 239 patients in total, with 42 patients from the aspiration group (87.5%) and 197 patients from the non-aspiration group (52.3%), which was a significantly higher rate of intubation in the aspiration group (p=0.000). The mean age was 53.5 years (median value, 40.0–64.7 years) in the aspiration group and 54.0 years (median value, 40.0–67.0 years) in the non-aspiration group, which was not a statistically significant difference (p=0.613). Likewise, there were no significant differences in sex, mechanism of accident, time of accident or past medical history, or status at discharge from the ER (Table 1).

When patient status upon arrival at the ER was compared between the two groups, in terms of mental status, the aspiration group showed 11 mild patients (22.9%), 6 moderate patients (12.5%), and 31 severe patients (64.6%), while the non-aspiration group showed 217 mild patients (57.6%), 37 moderate patients (9.8%), and 123 severe patients (32.6%), meaning that there was a significantly higher proportion of severe patients in the aspiration group (p=0.000). In AIS as well, the median AIS head score was 4.0 (4.0–5.0) in the aspiration group and 4.0 (3.0–5.0) in the non-aspiration group, which was higher in the aspiration group (p=0.046), and the median AIS abdomen score was 0.0 (0.0–0.0) in the aspiration group and 0.0 (0.0–2.0) in the non-aspiration group, which was significantly lower in the aspiration group (p=0.027). Median oxygen saturation was significant lower in the aspiration group, at 96.0 (89.2–98.0), than in the non-aspiration group, at 98.0 (95.0–99.0) (p=0.002). The vital signs of blood pressure, heart rate, respiratory rate, and body temperature, and the ISS showed not significant differences between the groups on arrival in the ER (Table 2).

In a logistic regression analysis including age, sex, time of accident, and GCS on admission to the ER, relative to the GCS mild (13–15) group, the moderate (9–12) group showed an odds ratio (OR) for aspiration of 2.976 (CI, 1.024–8.647), and the severe (3–8)group showed an OR of 5.073 (CI, 2.442–10.539). Consistent with the chi-square tests above, there were no statistically significant differences in aspiration rate according to age, sex, or time of accident (Table 3).

In a study by Benjamin et al. [9], when aspiration was judged by the presence or absence of an aspiration event, the incidence of aspiration among severe trauma patients was 25.4%. Aspiration pneumonitis develops within minutes when a patient aspirates ≥0.3 cc per 1 kg body weight of highly acidic gastric contents [12]. Aspiration lung diseases are characterized by bilateral, ill-defined, alveolar consolidations, multifocal patch infiltrates, posterior dominant distribution of lung opacities, and segmental or lobar consolidation on chest CT scans [13,14]. In the present study, based on chest CT scans taken soon after trauma, the incidence of aspiration among severe trauma patients was found to be 11.2%. The fact that the incidence was lower than the previous study is thought to be due to differences in how aspiration was identified, and identifying aspiration based on chest CT findings consistent with aspiration lung disease is expected to be the more diagnostically accurate method.

Altered mental status is related to aspiration, and more severely altered mental status is known to be associated with higher risk of aspiration [15,16]. Studies on severe trauma patients have also shown worse mental status in the aspiration group compared to the non-aspiration group [9,10]. Patients with impaired consciousness show decreased or loss of the gag and cough reflexes. These reflexes help to protect against aspiration, and so patients with impaired consciousness are more susceptible to aspiration [17,18]. The present study also showed lower GCS on admission to the ER among severe trauma patients, and multivariate logistic regression analysis showed that, compared to patients with mild (13–15) impairment of mental status, the OR for aspiration for moderate (9–12) patients was 2.976 (CI, 1.024–8.647) and for severe (3–8) patients was 5.073 (CI, 2.442–10.539), meaning that worse mental status was associated with higher risk of aspiration, which was consistent with previous studies.

Fawcett et al. [10] reported that, among severe trauma patients, the aspiration group show higher max head AIS scores than the non-aspiration group, and Benjamin et al. [9] also found that head AIS scores were higher in severe trauma patients with aspiration compared to those without aspiration. Traumatic head injury causes impairment or loss of consciousness [19]. Head injury in severe trauma also often causes vomiting, to the extent that vomiting is considered evidence of a potential intracranial injury, and this could also increase the risk of aspiration [20]. In the present study of severe trauma patients, there was a higher distribution of AIS-head scores in the aspiration group compared to the non-aspiration, which is consistent with previous research.

In two previous prospective studies, aspiration patients showed statistically higher ISS than non-aspiration patients (aspiration 26.0 vs. non-aspiration 17.0, p=0.027; aspiration 25.0±1.7 vs. non-aspiration 21.9±1.5, p-value=0.04) [8,9]. In the present study, there was no significant difference in ISS between the aspiration group and the non-aspiration group, with 24 severe patients (score 16–24, 50%) and 24 critical patients (score >24, 50%) in the aspiration group and 214 severe patients (score 16–24, 56.8%) and 163 critical patients (score >24, 43.2%) in the non-aspiration group (p=0.441). This is thought to be because patients who did not undergo chest CT were not included in the study. Given that head injury and impaired consciousness are risk factors for aspiration, as seen above, isolated brain injury patients may not have undergone chest CT scanning, resulting in selection bias affecting the results. Since recent studies recommend whole-body CT for severe trauma patients [21-24], it would also be beneficial to consider chest CT to check for aspiration in patients with isolated brain injury.

Aspiration pneumonia causes hypoxia and increased infiltration of the lung field [1]. In the present study, the median oxygen saturation was significantly lower in the aspiration group, at 96.0% (interquartile range [IQR], 89.2–98.0%), than in the non-aspiration group, at 98.0% (IQR, 95.0–99.0%) (p-value=0.002). The frequency of endotracheal intubation in the ER was also significantly higher in the aspiration group, at 42 patients (87.5%), than in the non-aspiration group, at 197 patients (52.3%) (p-value=0.000). This is thought to be related to the poorer mental state and hypoxia shown by the aspiration group in this study.

This study had several limitations. First, this was a retrospective study conducted at a single regional trauma center. The number of patients enrolled in the study was insufficient for a more thorough statistical comparison of relevant factors between the aspiration group and the non-aspiration group. Second, because it is difficult to differentiate between aspiration lung disease and traumatic lung injury from radiologic findings, patients with an AIS-chest score of ≥1 point were excluded from this study. Consequently, patients with traumatic lung injury accompanied by head injury were omitted, which could have affected the results by selection bias. Third, regarding the significantly lower oxygen saturation upon arrival at the ER, patients who had been transferred from another hospital or brought by the emergency services (119) are usually intubated or administered oxygen before arrival. Since there are no precise records of oxygen administration before arrival at the hospital, this makes it difficult to precisely analyze differences in oxygen saturation. Fourth, for patients whose initial GCS was not recorded but who had GCS recorded only after intubation, the verbal response was scored as 1 point to proceed with the analysis. This is because there are cases where intubation is performed immediately upon arrival at the ER since the patients subject to this study were severe trauma patients with ISS >15. Although patients who were intubated without the previous record of GCS should be excluded from the study since their GCS cannot be referred to, because the intubated patients before GCS recording tend to be intubated due to severely impaired consciousness, they were scored as 1 point. It is thought that this may have affected the accuracy of the statistical analysis.

In severe trauma patients, poor mental state and head injury increase the risk of aspiration. In order to test for aspiration, it would be useful to perform chest CT for severe trauma patients with head injury to check for accompanying aspiration lung disease. Therefore, in severe trauma patients with head injury or poor mental state, various measures should be considered to reduce the risk of aspiration, and this will need to be investigated further in studies with a larger sample size of patients.