Incidence, mortality, and predictive factors associated with acute respiratory distress syndrome in multiple trauma patients living in high-altitude areas: a retrospective study in Shigatse

Introduction

The acute respiratory distress syndrome (ARDS) was first reported in 1967, and the diagnostic criteria were created in 1992 (Ashbaugh et al., 1967; Bernard et al., 1994). Subsequently, the new Berlin Definition became widely utilized by clinical physicians for the evaluation and diagnosis of ARDS (Ranieri et al., 2012). As a prevalent complication, ARDS seriously affects patients’ health, and leads to poor prognosis or even mortality (Beitler et al., 2022). A prospective analysis (Rubenfeld et al., 2005) of ARDS incidence reported an annual occurrence of 190,000 cases in USA, while the hospital mortality rate reached 38.5%. In the LUNG-SAFE study, researchers evaluated the incidence and outcomes of ARDS in over 29,000 patients from 50 countries (Bellani et al., 2016). Results showed the prevalence of ARDS reached 10% among ICU patients and 23% among those receiving ventilation. Additionally, the hospital mortality rates surprisingly reached 46.1% for those with severe ARDS. Moreover, ARDS treatments bring heavy financial burden, with data indicating mean total hospitalization costs of 12,336 USD (Adhikari et al., 2010; Oh & Song, 2022). In summary, the high morbidity and mortality of ARDS not only bring heavy burden on global health but also threaten patient prognosis.

Numerous risk factors have been identified in association with the development of ARDS, including major trauma (both surgical and accidental), pneumonia, sepsis, and others (Matthay et al., 2019; Pham & Rubenfeld, 2017). Despite significant advancements have achieved in ARDS research, further investigation is needed to establish effective preventive measures for this condition (Meyer, Gattinoni & Calfee, 2021; Ranieri et al., 2012). Previous studies have successfully developed specific scoring models for predicting the risk of postoperative complications including ARDS (Arozullah et al., 2001; Gajic et al., 2011; Kor et al., 2014). These models help clinical physicians quickly identify patients at high risk of developing ARDS, thereby improving patient prognosis.

According to reports, ARDS has emerged as a significant contributor to morbidity and mortality in trauma cases, with mortality rates reaching up to 40% (Fan, Brodie & Slutsky, 2018). In another retrospective study, researchers reported that over 51% of multiple trauma patients developed ARDS (Haider et al., 2017). However, treatment options for ARDS remain limited, primarily relying on supportive mechanical ventilation (Fan, Brodie & Slutsky, 2018; Rizzo et al., 2023). Consequently, early identification of patients at risk for ARDS development is crucial, and extensive research studies have focused on the identification of risk factors.

Studies have indicated that patients residing at altitudes above 1,500 m exhibit lower partial pressure of oxygen (PaO₂) and arterial oxygen saturation (SaO₂) values compared to those living at sea level (Ortiz et al., 2022). This can be attributed to the gradual decrease in atmospheric pressure and oxygen content in the air with increasing altitude. Consequently, individuals in high-altitude areas have undergone physiological adaptations such as enhanced oxygen delivery, and adjustments in oxygen utilization and metabolism (Guo et al., 2023). These adaptations result in significant differences in respiratory physiology and pathology between patients residing in high-altitude and low-altitude regions (Guo et al., 2023; Ortiz et al., 2022). However, there is a lack of reports focusing on the incidence and mortality of ARDS, as well as potential risk factors for early prediction of ARDS in multiple trauma patients living in high-altitude areas.

The Qinghai-Tibet Plateau, known for its average elevation exceeding 4,000 m, holds the distinction of being the world’s highest plateau and is home to over nine million people in China. In an effort to investigate the incidence and mortality of ARDS among multiple trauma patients residing in high-altitude regions, we conducted a retrospective study at Shigatse People’s Hospital, situated at an average altitude exceeding 4,000 m. Furthermore, we meticulously screened several risk factors and identified two significant factors, enabling us to develop a predictive model for early detection of ARDS.

Methods

Results

In the retrospective study, 186 patients were screened and 18 patients were excluded. Finally, a total of 168 multiple trauma patients were ultimately included (Fig. 1), and their clinical characteristics are presented in Table 1. The majority of these patients sustained injuries as a result of traffic accidents or altercations. Among multiple trauma patients residing in high-altitude areas, the incidence of ARDS was found to be 39.3% (66 out of 168), while the hospital mortality rate was 16.1% (27 out of 168). Additionally, 65.5% (110 out of 168) of patients were diagnosed with craniocerebral trauma, and 53.6% (90 out of 168) were diagnosed with thoracic injuries.

ROC curves and AUC analysis

To evaluate the predictive capacity of the selected risk factors and the risk score model, ROC curves and AUC analysis were conducted. As depicted in Fig. 2, both the ISS score (AUC = 0.75, 95% CI [0.676–0.824]) and thoracic injuries (AUC = 0.75, 95% CI [0.669–0.821]) exhibited promising predictive value for ARDS.

Furthermore, in an effort to enhance the predictive accuracy for ARDS, we utilized the risk score model, which amalgamated the ISS score and thoracic injuries, to construct the ROC curve. As illustrated in Fig. 3, the AUC of the risk score reached 0.82 (95% CI [0.757–0.881]), indicating a substantial effectiveness in early prediction of the incidence of ARDS.

Discussion

ARDS is a highly detrimental complication that significantly impacts the prognosis of patients. Furthermore, the challenging plateau environment further complicates the management of ARDS in high-altitude areas, making early prediction and treatment of utmost importance for improving patient survival in clinical practice. However, limited reports exist on ARDS in multiple trauma patients specifically in high-altitude areas. In this retrospective study, we aimed to investigate the incidence and mortality of ARDS in patients with multiple trauma in the Shigatse area, characterized by an average altitude exceeding 4,000 m. Additionally, we identified several risk factors and established a risk score model for the early prediction of ARDS in these patients. The risk score model demonstrated robust predictive capabilities, with an AUC exceeding 0.8. These findings provide valuable insights for clinical physicians working in high-altitude areas, serving as a reference for their practice.

Rodriguez Lima et al. (2023) developed a predictive model to estimate in-hospital mortality for patients with ARDS due to COVID-19 in high-altitude regions. Their findings identified age, weight, creatinine levels, and the presence of ARDS as risk factors associated with increased patient mortality. In another study, Liu et al. (2022) retrospectively analyzed clinical data from patients with ARDS residing in the Qinghai area (altitude: 2,261 m). They observed a significant increase in ICU mortality corresponding to the severity of ARDS: mild (17.76%), moderate (21.43%), and severe (47.37%). According to Liu’s report, the overall mortality among patients with ARDS who required mechanical ventilation reached 24.0% (55/229).

In our retrospective study, we reported the incidence of ARDS reached 39.3% in multiple trauma patients, and the hospital mortality reached 16.1%. These rates were significantly higher than those reported in studies conducted in low-altitude areas (Kor et al., 2014; Milot et al., 2001; Sanfilippo et al., 2022). Notably, relevant studies Milot et al. (2001), Sanfilippo et al. (2022) investigating ARDS following cardiac surgery reported incidence rates ranging from 0.4% to 8.1%. Additionally, a previous study (Wang et al., 2022) conducted by our team found an ARDS incidence rate of 7.2% (74/1,032) in patients undergoing hepatectomy. In conclusion, high-altitude environments may exert significant effects on the development of ARDS.

Several studies have attempted to identify effective predictors or develop prediction models for early detection of ARDS in patients with specific diseases or injuries (Afshar et al., 2019; Ding et al., 2022; Huang et al., 2021). For instance, Ding et al. (2022) conducted a study on patients with acute pancreatitis and developed a logistic model to predict the in-hospital incidence of ARDS. They found that several independent factors, including white blood cell count, prothrombin time, ALB, serum creatinine, and triglycerides, were associated with the occurrence of ARDS during hospitalization. In a retrospective study by O’Leary et al. (2016) involving 305 trauma patients, clinical predictors of early ARDS were investigated. The study reported an incidence rate of 19.3% (59/305) for ARDS among trauma patients. Additionally, it highlighted that patients with blunt trauma, those who received fresh frozen plasma, or underwent thoracotomy, were at a higher risk of developing early ARDS. These studies contribute to our understanding of potential risk factors and prediction models for the early identification of ARDS in specific patient populations.

In our study, we identified the ISS score and thoracic injuries as independent risk factors for the occurrence of ARDS. Both factors exhibited a good predictive value for ARDS, with an area under the curve (AUC) of 0.75. Furthermore, when combining the ISS score and thoracic injuries into a new risk score, we observed improved effectiveness in predicting ARDS occurrence, with an AUC of 0.82. These findings are consistent with a retrospective analysis conducted by Haider et al. (2017), which also reported a significant association between the ISS score, thoracic injuries, and a higher incidence of ARDS. Specifically, thoracic trauma was identified as an important risk factor for the development of ARDS and its earlier onset in polytraumatized patients.

In summary, our results suggest that both the ISS score and thoracic injuries serve as effective indicators for predicting the occurrence of ARDS in high-altitude areas. Patients with multiple traumas, especially those with higher ISS scores or thoracic injuries, require close monitoring. Prompt measures to prevent ARDS should be implemented, such as oxygen therapy and mechanical ventilation, while fluid therapy should be judiciously administered. However, it is important to acknowledge certain limitations of our study. Firstly, due to its retrospective design, potential biases and confounding factors cannot be completely ruled out. Also, exact causality relationships between ARDS and risk factors cannot be established and further prospective studies are needed. Secondly, our study was conducted at a single center and the prediction model was only validated using internal data. Therefore, further studies are warranted to validate these results across multiple centers. Additionally, prospective studies with larger sample sizes would be necessary in the future to strengthen the evidence base.

Conclusion

In conclusion, we reported the incidence of ARDS in multiple trauma patients lived in Tibetan area, and identified two critical predictive factors (ISS score and thoracic injuries) for early predicting ARDS occurrence. In addition, we constructed a risk score model for early predicting ARDS using the logistic regression, and the AUC reached 0.82. These tools may improve clinicians’ ability to early estimate the risk of ARDS and timely prevent its emergence in high-altitude areas.

Supplemental Information