Acute respiratory distress syndrome (ARDS) is a clinical syndrome characterized by refractory hypoxemia, radiographic severity, and respiratory system compliance. This syndrome is usually associated with sepsis, shock, aspiration, and trauma. Cardiac surgery involving cardiopulmonary bypass (CPB) has been reported to result in ARDS in 0.4–20% of patients, with mortality rates reaching as high as 70–80%.1–3 Beyond mortality, compromised long-term health status and increased medical expenditures could not be ignored contributing to this respiratory failure. Thus, every attempt to decrease the incidence of ARDS should be embraced.

Type A aortic dissection, which involves the ascending aorta irrespective of the site of the intimal tear according to the Stanford system, remains the most common aortic catastrophe despite recent advances in diagnosis and treatment. Prompt surgical repair is required because of an extremely high mortality 22.7% within the first six hours, 33.3% within 12 hours, 50% within 24 hours, and 68.2% within the first two days if left untreated.4 Acute lung injury (ALI) and its more severe form, ARDS, are lethal complications during perioperative management; aortic surgery accounts for 16% of ARDS cases.2 ALI and hypoxemia were observed in more than half of type A acute aortic dissection patients, with ARDS occurring in 10–20% of the population.5,6 Additionally, emergent surgery predisposed type A acute aortic dissection patients to higher risk of ARDS.7 Thus, a better understanding of the independent risk factors for ARDS in type A acute aortic dissection is essential to improve clinical strategies and outcomes.

However, drawing conclusions is challenging because studies on postoperative ARDS have primarily included patients undergoing heterogeneous cardiac surgeries, such as coronary artery bypass grafting (CABG) and multiple valve surgery. Only one recent study reported an occurrence of 15.9% in ARDS following aortic dissection surgery.6 More research is needed to identify pre- and perioperative risk factors for ARDS. Therefore, we firstly established and validated a predictive nomogram to calculate the probability of ARDS for individuals with type A acute aortic dissection, which might enhance the clinical decision-making.

The study aimed to identify independent risk factors and establish a numerical model for postoperative acute respiratory distress syndrome in type A aortic dissection patients.

This study is the first to develop a nomogram of postoperative ARDS in type A acute aortic dissection patients, which illustrates the regression model in an accessible way and thus simplifies the risk assessment. We analyzed clinical characteristics, laboratory values and postoperative outcomes to identify the most relevant variables for multivariate regression analysis. Four independent predictors for postoperative ARDS were identified: BMI, albumin, PCT, and CPB time.

Postoperative ARDS carries a high mortality rate, ranging from 10% to 70%, resulting from the differences in ARDS definitions, types of surgeries, and study populations.1,6,9–11 Previous definitions of ARDS, such as the Murray's Lung Injury Score from 1988, the American–European Consensus Conference definition (AECC) from 1994, and the Berlin definition from 2012, offer varying criteria for diagnosis and intervention.8,12,13 The Berlin definition, which is the most recent one, provided practical guidelines for early recognition and intervention of ARDS. However, the application of the Berlin definition to type A acute aortic dissection patients specifically has not been thoroughly investigated. Studies on ARDS following cardiac surgery often included a mix of procedures involving CPB, such as valvular surgery and CABG. These varied combinations can lead to different incidence and mortality rates. For instance, a study reported that 8.1% of 457 patients undergoing isolated valvular heart surgery developed postoperative ARDS, with a mortality rate of 29.7%.10 In contrast, a broader study including CABG, valve, and combined procedures found mortality of 68.4%.11 Our study observed an ARDS incidence of 28.873% in type A acute aortic dissection patients, which was slightly higher than reported in some studies, possibly due to the high proportion of patients classified as NYHA class IV.6

The pathophysiology of ARDS following CPB has not been totally understood. The prevailing hypothesis involves activation of the inflammatory cascade, including polymorphonuclear leukocyte and cytokine activation, which increases pulmonary permeability and vascular resistance, leading to impaired gas exchange.14 Our findings indicate that CPB time is a significant risk factor in ARDS. Besides, protamine-heparine complexes may trigger immune reactions by releasing anaphylatoxins such as C4a and C3a, potentially increasing ARDS risk.1 Patients with ARDS were reported to have lower preoperative albumin, which was similar in our study, and higher plasminogen activator inhibitor-1, interleukin-8 and surfactant protein-D.15 Up-regulation of tumor necrosis factor-alpha and elevation of serum endocan are also believed to get involved in the early phase of ARDS.16,17 Albumin in the extracorporeal prime can preserve platelet count and reduce chest tube drainage by passivating the synthetic surfaces of the extracorporeal circuit.18 Elevated PCT levels were also associated with ARDS in our study, suggesting it could serve as a potential diagnostic biomarker.19 Timely infection control and interruption of inflammatory processes are crucial for ARDS prevention. Corticosteroids have been shown to improve cardiopulmonary function within three to seven days and alter the ARDS course, with early treatment improving ICU and ventilator outcomes with minimal safety concerns.20

Transfusion-related complications, including transfusion-related acute lung injury and transfusion-associated circulatory overload, are significant concerns. Multiple transfusions have been linked to ARDS risk, with packed red blood cell transfusions exhibiting a dose-dependent effect on pulmonary capillary permeability.21 In a large observational cohort of CABG patients (11963), red blood cell transfusion was associated with higher mortality, renal failure, prolonged ventilatory support.22 A retrospective nested case–control study demonstrated extended durations of CPB, deep hypothermic circulatory arrest, ICU stay, and larger perioperative transfusions volumes (red blood cell, platelet and fresh frozen plasma) in ARDS group of type A acute aortic dissection population.23 Although our study did not identify blood product transfusion as an independent predictor of ARDS, possibly due to sample size limitations and the urgent nature of aortic surgery, meticulous hemostasis and restrictive transfusion strategies could mitigate ARDS risk. A restrictive transfusion strategy targeting a hematocrit of 24% has been shown to be as safe as a liberal strategy in elective cardiac surgery, though caution is needed for elderly patients due to cardiovascular risks.24,25

Obesity's role in ARDS remains debated and is known as the “obesity paradox”. While excess body weight has not consistently been linked to increased mortality in acute lung injury cases,26,27 it is associated with longer hospital stays.28,29 A study of 5023 patients undergoing heart surgery with CPB found that obese patients were at higher risk for postoperative hypoxia, but were somewhat protected against adverse effects from hemodilution and transfusions.30 Our study confirmed that BMI is an independent risk factor of ARDS after aortic surgery. A case–control study identified a BMI cutoff of 24.78 kg/m2, with ARDS likelihood increasing above this threshold.31 While BMI is challenging to modify in emergency settings, weight control strategies could benefit postoperative outcomes.

This study has limitations, including the inability to split the data into training and validation sets due to a limited sample size. Consequently, model validation was performed using bootstrap methods. Additionally, among data from a monocenter there may be some degree of bias. Further research with multicenter data and larger sample sizes is needed to enhance the validity and generalizability of these findings.

We established a straightforward predictive nomogram to estimate the probability of ARDS in patients with type A acute aortic dissection by incorporating four perioperative variables, including BMI, albumin, procalcitonin and CPB time. The accuracy and stability of the model were validated internally using the AUC, calibration plot, and DCA. This nomogram has the potential to serve as a valuable prognostic tool for assessing the risk of ARDS in type A acute aortic dissection patients.

None declared.

None declared.