Risk factors for one-year mortality following discharge in patients with acute aortic dissection: development and validation of a predictive model in a cross-sectional study

AAD is a severe condition affecting the aorta. Although the incidence rate of AAD is relatively low, patients diagnosed with this condition generally have a poor prognosis [13]. The high mortality rate associated with AAD often leads clinical practitioners to prioritize managing patients’ conditions during their hospital stay, potentially neglecting their long-term prognosis after discharge. With advancements in technology, we have the capacity to intervene in the early stages of AAD and slow down its progression at present. However, in China, there is a lack of post-discharge follow-up on patients, leading to neglect of their rehabilitation and subsequent treatment, which can worsen the disease after initial treatment [14]. Therefore, it is crucial to assess the risk of postoperative outcomes for AAD patients based on their hospitalization and establish a comprehensive follow-up system. It could enhance the recovery rate of discharged patients and reduce the medical burden on them and their families. Thus, we identified five potential risk factors and developed a risk prediction model for one-year post-discharge mortality, facilitating early identification of patients at risk.

Traditionally, researchers have observed gender disparities in the incidence of AAD. Males have a higher prevalence of AAD compared to females and tend to present with a wider range of aortic lesions [15, 16]. Additionally, male patients are more likely to be diagnosed with complications such as inadequate blood supply to vital organs [17]. Conversely, although the condition was reported to be less severe in females, they may have a lower attributable mortality rate for AD and its complications, consistent with our findings [18]. In our risk prediction model, both univariate and multivariate logistic regression analyses revealed that gender strongly predicts the mortality risk within one year. Female patients were assigned a higher score in our risk prediction model, indicating the need for healthcare professionals to prioritize monitoring their disease progression after discharge and intervene as necessary. To understand the elevated mortality rate of female patients after discharge, we conducted a literature search. In the clinical diagnosis and treatment of heart disease, patients of different genders also have different tendencies towards treatment methods. Female patients are more inclined to choose transcatheter aortic valve implantation for the treatment of valvular heart disease, while the willingness to choose interventional surgery for ischemic heart disease is significantly lower than that of males [19]. Different options bring different outcomes. Although there is currently no clear research indicating significant gender differences in the selection of treatment options and clinical prognosis for AAD, some studies have provided us with clues. Existing studies showed that male patients experience more frequent chest pain symptoms in the early stages of AAD, while female patients often present with poor peripheral vascular perfusion and neurological dysfunction before receiving treatment. The presence of invisible early symptoms and severe complications contributes to worse conditions at the time of treatment, worsening their prognosis [20]. Moreover, female patients often have smaller arterial diameters, making them more susceptible to the impact of AAD on fine blood vessels, further deteriorating their prognosis [21]. Therefore, it is crucial to focus not only on the in-hospital development of AAD patients but also on improving the survival rates of female patients after discharge.

In our research, we found that the duration of hospitalization could be a potential risk factor for one-year post-discharge mortality. Longer hospital stays were associated with a decreased risk of death. The choice of therapeutic approaches for different types of AAD varies, with surgery being the optimal option to extend patients’ lives and improve their prognosis [3]. Surgery leads to superior treatment outcomes compared to internal medicine treatment, but it also requires longer hospital stays. Patients opting for internal medicine treatment need extended hospital stays to assess drug efficacy and mitigate side effects. Furthermore, AAD often leads to early-onset complications, and an extended hospitalization period allows clinical physicians to manage these complications in a timely and effective manner, which significantly improves patient prognosis [22].

BUN is a crucial indicator reflecting renal function and plays a vital role in the occurrence and progression of cardiovascular diseases such as heart failure [23]. Liu’s research demonstrated that BUN is an independent mortality predictor during hospitalization among AAD patients, with higher BUN levels corresponding to a higher risk of hospital mortality, consistent with our findings [24]. In the predictive model we constructed, a BUN level > 20 mg/dL during hospitalization was assigned a score of 100, indicating that BUN is the most influential factor in nomogram for the risk of death within one year after discharge. Inadequate visceral perfusion is one of the most severe complications following the onset of AAD, and it has a significant impact on renal function [25]. Ischemia-induced kidney damage may impair the elimination of BUN from the body, resulting in increased BUN levels. Therefore, early detection and prediction of potential renal function damage, along with early intervention, are advantageous for improving the prognosis of AAD patients [24]. Additionally, our predictive model showed that the use of adrenaline somewhat increased the risk of death within one year after discharge. Although adrenaline plays a crucial role in overcoming hypotension during the onset of AAD, its side effects deserve attention. Adam et al. reported a case of an electrical storm induced by the use of adrenaline after AAD surgery, as it can cause or exacerbate arrhythmias [26]. Wang et al.‘s research also demonstrated that high concentrations of catecholamines can promote the expansion of the false lumen, thereby increasing the risk of rupture [27]. Furthermore, adrenaline directly stimulates glomerular contraction, leading to reduced renal blood supply in AAD patients, ultimately increasing BUN levels [28]. Hence, comprehensive assessment of visceral perfusion and function before administering vasopressors, particularly adrenaline, is essential.

Arrhythmias are the most common cardiac rhythm changes observed after AAD and are also a frequent cause of hemodynamic disorders associated with AAD [29]. Currently, there has been no literature directly indicating the relationship between arrhythmia and short-term mortality of AAD patients. However, in nomogram, the use of amiodarone was found to be beneficial for improving the one-year survival rate of patients after discharge. Unfortunately, due to unavailability of medical records from the MIMIC database, we couldn’t determine the specific indications for administering amiodarone to these patients. Further research is required to explore the relationship between amiodarone and short-term mortality in AAD patients.

Our results indicated that gender, length of hospital stay, BUN_Max, use of adrenaline, and use of amiodarone were potential risk factors for death within one year after discharge in AAD patients. These results emphasize some important factors that need to be noted in the management of AAD patients. For patients of different genders, clinical doctors need to assess the progression of the disease in more detail. Meanwhile, paying attention to the control and optimization of hospitalization days can help improve the prognosis of patients. In addition, monitoring renal function indicators and careful selection of medication use are also key to reducing the risk of short-term mortality.

This study had some limitations. Firstly, although we have included data from two hospitals, the study period was relatively short and the number of included patients was relatively small, which inevitably leads to some biases (such as Incident Probability bias, sampling error). The existing sample size may lead to the neglect of certain factors that affect the occurrence and prognosis of AAD, such as age, underlying diseases, etc. Due to racial differences, we had to convert continuous variables into binary variables, which, although beneficial for the applicability of nomograms, also led to the occurrence of some biases. Secondly, due to limitations in sample size, we had to exclude indicators with high missing values, potentially resulting in an incomplete assessment of short-term mortality risk using only the five mentioned risk factors. Additionally, the lack of access to patients’ medical records limited our understanding of their treatment process and reduced the reliability of our constructed nomogram. In the future, by increasing the sample size and improving the follow-up mechanism, more meaningful conclusions may be obtained.