Prognostic prediction of long-term survival in patients with type A aortic dissection undergoing surgical repair: development of a novel prognostic index

Abstract

Background

This study developed and investigated the prognostic significance of a comprehensive biomarker for postoperative type A aortic dissection (TAAD) patients.

Methods

A retrospective cohort of 175 TAAD patients who underwent open surgery at the First Affiliated Hospital of Chongqing Medical University, China, between September 2017 to December 2020, was included in the analysis. The least absolute shrinkage and selection operator (LASSO) method was employed to select indicators, thereby constructing a comprehensive biomarker, termed the comprehensive physiological response indicator (CPRI). The prognostic significance of the CPRI was assessed employing the Kaplan-Meier method and log-rank test. Univariate and multivariate Cox regression model were applied to identify independent prognostic factors for TAAD. A prognostic nomogram was constructed based on the CPRI and other nine blood and clinical indicators. The predictive performance of prognostic models and individual indicators was evaluated by determining the area under the curve (AUC) of the receiver operating characteristic (ROC) analysis.

Results

A comprehensive prognostic biomarker (CPRI) was developed, incorporating cardiac troponin T (cTnT), red blood cell count (RBC), white blood cell count (WBC), absolute neutrophil count (ANC), and absolute lymphocyte count (ALC). In the cohort of type A aortic dissection (TAAD) patients who underwent open surgery, those with lower preoperative CPRI values exhibited significantly poorer overall survival, with a HR of 2.325 (95% CI: 1.126–4.802) in a multivariate Cox regression analysis. The CPRI was identified as an independent prognostic factor for TAAD patients. Additionally, a nomogram constructed based on the CPRI demonstrated superior predictive accuracy compared to other models, with an area under the curve (AUC) of 0.874 versus 0.592, 0.514, and 0.577 for the respective models.

Conclusion

Our study suggested that CPRI may be a useful comprehensive prognostic biomarker for predicting the long-term survival of TAAD patients. The nomogram based on CPRI can be considered a valuable tool to identify high risk TAAD patients with poor prognosis.

Introduction

Aortic dissection (AD) is a life-threatening vascular emergency characterized by a tear in the intimal layer of the aorta, allowing blood to enter the middle layer (media) and create a new cavity, known as a false lumen, alongside the original aortic lumen (true lumen) [3]. Based on the Stanford classification, AD is categorized into two types: Type A, which primarily involves the ascending aorta, and Type B, which is confined to the descending aorta or extends distally to the abdominal aorta [6].

The mortality rates associated with AD underscore its severity. Pre-hospital mortality for Type A aortic dissection (TAAD) can reach 48.6%, with a 30-day mortality rate of 47.4% among those who survive to hospital admission [1, 2]. While type B aortic dissection (TBAD) generally has a more favorable prognosis, 25–40% of TBAD patients present with complex conditions, leading to an in-hospital survival rate of only 50% [4]. A 13-year retrospective analysis revealed significantly poorer long-term outcomes for patients with complex TBAD [5]. These high mortality rates necessitate a comprehensive evaluation of AD prognosis and the development of robust prognostic indicators.

The histological features of AD are primarily characterized by degeneration of the aortic media, which is marked by the severe degradation of the extracellular matrix. This degradation is closely linked to abnormal phenotypic transformation and apoptosis of smooth muscle cells, fragmentation of elastic fibers, and breakdown of collagen [7–8]. During the subclinical stage of AD, the pathological process involves the recruitment of various immune cells, including macrophages and lymphocytes, along with the release of cytokines such as tumor necrosis factor (TNF)-α, monocyte chemotactic protein (MCP)-1, and vascular endothelial growth factor (VEGF), which collectively contribute to the progression of the disease [9].

The acute onset and rapid progression of AD in the clinical stage disrupt internal homeostasis through multiple mechanisms, affecting various organ systems. For instance, damage to the aortic wall and the invasion of blood into the media can provoke a localized inflammatory response. This inflammation activates the immune system, leading to the release of inflammatory mediators, as evidenced by elevated levels of markers such as C-reactive protein (CRP), interleukins (ILs), and TNF [10–11]. Moreover, blood infiltration into the aortic media may result in localized thrombosis and damage to the endocardium or vascular endothelium, which in turn can cause abnormalities in coagulation and fibrinolytic systems [12]. AD can also cause significant changes in cardiac workload, triggering a stress response in the myocardium. This may compromise coronary artery blood flow, leading to myocardial ischemia or even infarction, with the potential release of biomarkers such as troponin [13]. Consequently, patients with AD exhibit various abnormalities in biological indicators. These abnormalities not only reveal the underlying pathophysiological processes but also highlight the systemic impact of this condition.

The identification of specific biomarkers associated with these pathophysiological events is crucial for developing new therapeutic strategies and improving patient outcomes. Previous studies have explored the prognostic value of various indicators, including troponin [14], D-dimer [19, 20], CRP [16], interleukin-6 (IL-6) [15], neutrophil-to-lymphocyte ratio (NLR) [18], and eosinophil percentage [17]. Abnormal levels of these indicators in the preoperative phase have been associated with survival in AD patients. However, most studies have focused on only one or two factors related to AD prognosis, failing to comprehensively investigate their combined effects.

Therefore, the aim of this study was to thoroughly examine the effects of various blood indicators and demographic characteristics on the long-term survival of TAAD patients. Additionally, we sought to construct novel prognostic indicators using the LASSO procedure to provide evidence for better prognostic management and individualized decision-making in the context of TAAD.

Material and method

Patients

This study is a retrospective analysis. From September 2017 to December 2020, consecutive patients with type A aortic dissection (TAAD) admitted to the Department of Cardiothoracic Surgery at the first affiliated hospital of Chongqing medical university were enrolled. Patients were eligible for inclusion if they met the following criteria: (1) diagnosis of Stanford TAAD confirmed by aortic computed tomography angiography; (2) patients who underwent surgical repair for TAAD. Exclusion criteria were as follows: (1) absence of clinical or laboratory data; (2) patients with hematological disorders, infections, or treatments that might affect biomarkers; (3) patients who died directly or indirectly from causes other than TAAD. Ultimately, based on criteria above, 175 patients were included in the study (Fig. 1A). The study was approved by the independent ethics committee at the First Affiliated Hospital of Chongqing Medical University (2024380-01) and was conducted in accordance with the ethical standards of the World Medical Association Declaration of Helsinki.

Fig. 1

Workflow for Patient Inclusion (A), Data Selection (B) and Data Analysis (C). Abbreviations: VIF Variance Inflation Factor, C-index Concordance Index, TAAD Type A Aortic Dissection

Data collection

As shown in Fig. 1B, the baseline characteristics of the patients were retrospectively reviewed from the electronic medical record system, including gender, age, smoking history, history of hypertension, diabetes mellitus, cardiovascular disease, blood pressure at admission, and heart rate. Blood indicators were measured within 24 h of hospital admission and prior to surgery. The measured biomarkers included absolute neutrophil count (ANC), white blood cell (WBC) count, red blood cell count (RBC), hemoglobin (Hb), platelet count, absolute lymphocyte count (ALC), monocyte count, serum albumin, uric acid, urea nitrogen, alanine aminotransferase (ALT), aspartate aminotransferase (AST), cardiac troponin T (cTnT), myoglobin (Mb), D-dimer, fibrinogen and D-dimer.

Follow-up and treatments

All patients included in the study underwent surgical repair. The follow-up period for each patient started three months after discharge and concluded on August 20, 2024. Specifically, trained interviewers conducted telephone follow-ups with each discharged TAAD patient at 3, 6, and 12 months post-discharge, and every six months thereafter, to support post-discharge care. Following initial enrollment, additional follow-ups were conducted to confirm the survival status of the included patients. Overall survival was defined as the period from diagnosis confirmed by CTA to death due to aortic dissection or the most recent follow-up. The primary endpoint event of this study is defined as the presence of death in the patient cohort under investigation. The maximum overall survival time was 2,445 days, with a median overall survival time of 1,190 days.

Statistical analysis

The optimal cutoff values for age, D-dimer, creatinine, age, heart rate (HR), postoperative hospital stay, intensive care unit hospital stay and fibrinogen were determined using X-tiles software (Yale University, New Haven, Connecticut). Pearson’s chi-squared test was applied to assess correlations between variables, and a heatmap was generated to visually depict the degree of correlation among the blood indicators. Survival outcomes for patients with TAAD were evaluated and visualized using the Kaplan-Meier method. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated through univariate and multivariate Cox regression analyses. To assess the predictive capability of the indicators and models, receiver operating characteristic (ROC) curves were generated, and the area under the curve (AUC) was used for comparison. Variance Inflation Factor (VIF) calculation and Pearson’s chi-squared test was applied to assess potential risks of multicollinearity. K-fold cross-validation method and the Concordance Index (C-index) was used to assess the model’s predictive performance (Fig. 1C). In this study, a two-tailed p-value of less than 0.05 was considered statistically significant. All statistical analyses were conducted using SPSS version 27.0 (IBM, Chicago, IL, USA), GraphPad Prism version 9 (GraphPad Software, San Diego, CA, USA), R version 4.4.1 (R Foundation for Statistical Computing, Vienna, Austria), and Python 3.1 (Python Software Foundation, Beaverton, OR, USA).

Results

Patients’ characteristics

A total of 175 consecutive patients with TAAD were enrolled in the study, based on the predefined inclusion criteria. The baseline characteristics of the cohort were summarized in Table 1. This analysis provided a detailed overview of patient demographics and their corresponding mortality outcomes. The column labeled “No. of outcomes (%)” indicated the frequency of deaths within each subgroup, alongside the proportion of these deaths relative to the total number of deaths in the respective category.

Among the enrolled patients, 139 (79.4%) were male, and a higher proportion of deaths occurred in males (40, 74.1%) compared to females (14, 25.9%). The majority of patients were under 52 years of age (122, 69.7%), and this group also accounted for most of the deaths (31, 57.4%). Hypertension (HTN) was prevalent in the cohort, with 120 patients (68.6%) diagnosed with the condition. Patients with HTN exhibited a significantly higher mortality rate (39, 72.2%) compared to those without HTN (15, 27.8%).

The distribution of systolic blood pressure (SBP) in the study population was relatively balanced. However, a higher mortality rate was observed among patients with an SBP of less than 140 mmHg, accounting for 28 deaths (51.9%). Similarly, patients with diastolic blood pressure below 90 mmHg experienced 39 deaths (72.2%). The majority of the study cohort had a pulse pressure exceeding 40 mmHg, and this group also represented the majority of deaths. Furthermore, 130 patients (74.3%) had a body mass index (BMI) outside the normal range, either above 24 or below 18.5. Notably, deaths within this group accounted for approximately 80% of all endpoint events, in contrast to those whose BMI fell within the normal range (18.5–24).

In delayed-admission TAAD patients (≥ 15 h), the survival rate was higher than in early-admitted patients (81.4% vs. 61.0%), likely due to the inclusion of a small proportion of subacute and chronic cases (15, 8.6%). Despite delayed admission, these patients’ relatively stable condition contributed to more favorable survival outcomes. Among the early-admitted patients (< 15 h, 105 cases), 41 (39.0%) died, which reflects a significantly higher mortality rate compared to the delayed-admission group (18.6%). This observation may, on one hand, be due to the inclusion of some non-acute TAAD patients, which resulted in a higher number of survivors and a lower proportion of mortality. On the other hand, it likely reflects a more severe clinical condition in early-admitted patients. These patients, presenting with more pronounced symptoms, may have sought medical attention earlier. However, despite receiving early surgical intervention, the severity of TAAD likely contributed to the higher mortality rate.

Furthermore, using “time from onset to admission” to assess the early treatment and its impact on survival may not provide an accurate reflection. While admission time may appear similar, the actual timing of treatment initiation can vary due to factors such as the completion of essential diagnostic tests, misdiagnosis-related delays, and surgical timing. Although a statistically significant difference in the variable was observed between outcome and survivor groups, this measure may not truly capture the timing of treatment initiation. To improve methodological rigor and ensure more clinically relevant conclusions, we excluded admission time from subsequent analyses.

In the outcome group, the proportion of patients without shock was notably higher than those with shock (87.0% vs. 13.0%), likely due to the the small number of TAAD patients with shock (11 cases), resulting in lower representation in both the outcome and survivor groups. However, Among patients with shock (11 cases), 7 deaths were observed compared to only 4 survivors, indicating a higher mortality rate (63.6%) in this subgroup. In contrast, among patients without shock (164 cases), the number of survivors (117 cases) far exceeded the number of deaths (47 cases), resulting in mortality rate of 28.7%. These findings suggest that while shock is relatively uncommon in TAAD patients, it is associated with a poorer survival compared to patients without shock.

Table 1 General demographic and clinical characteristics of the study subjects at admission

Correlation between preoperative blood indicators

The heatmap presented Pearson’s correlation coefficients between various clinical variables, offering a visual representation of the strength and direction of these correlations. The color gradient, ranging from light pink to dark red, represented the spectrum from weak to strong correlations. Each cell in the heatmap was labeled with a Pearson’s correlation coefficient, quantifying the linear relationship between two variables. A coefficient of 1 indicated a perfect positive correlation, -1 signified a perfect negative correlation, and 0 reflects no correlation. Asterisks were used to denote statistical significance, with one asterisk (*) representing a p-value of less than 0.05 and two asterisks (**) indicating a p-value of less than 0.01 (Fig. 2).

Fig. 2

Correlation heat map of ten blood indicators. Abbreviations: cTnT Cardiac Troponin T, PLT Platelet Count, RBC Red Blood Cell Count, Hb Hemoglobin, WBC White Blood Cell Count, ANC Absolute Neutrophil Count, ALC Absolute Lymphocyte Count, PCT Procalcitonin Notes: ‘*’ denotes p < 0.05, ‘**’ denotes p < 0.01

In the initial screening of indicators, indicators closely related to patients’ physiological states and immune responses were prioritized, based on expert consultation and a thorough review of the literature, to ensure both statistical significance and clinical relevance. Correlation analysis was subsequently performed to refine the selection process, focusing on variables with lower inter-correlations and higher clinical value. The variables included in the final correlation analysis were cardiac troponin T (cTnT), Fibrinogen, D-dimer, platelet count (PLT), red blood cell count (RBC), hemoglobin (Hb), white blood cell count (WBC), absolute neutrophil count (ANC), absolute lymphocyte count (ALC), and procalcitonin (PCT). Despite high correlations between variables such as WBC and ANC or RBC and Hb, they were retained as they reflect important physiological characteristics in TAAD patients, including inflammation, immune function, overall health, and acute blood loss or fluid redistribution.

Pearson’s chi-squared test revealed that preoperative WBC was significantly correlated with Fibrinogen, D-dimer, RBC, Hb, and ANC (p = 0.002, p = 0.002, p = 0.003, p < 0.001, p < 0.001, respectively), while no significant correlations were observed with cTnT, PLT, ALC, and PCT. Strong correlations were identified between certain indicators, with the most prominent being between WBC and ANC (correlation coefficient: 0.97), followed by the correlation between RBC and Hb (correlation coefficient: 0.67). The high correlation between WBC and ANC suggested a parallel increase in these values, which, in patients with AD, may reflect an intense inflammatory response. The elevated neutrophil count, as indicated by ANC, likely represented a reaction to tissue damage and the inflammatory processes associated with dissection formation. However, aside from these strong correlations, most other correlations were relatively weak.

Of particular interest was the negative correlation between fibrinogen and D-dimer (correlation coefficient: -0.27), indicating that in TAAD patients, lower fibrinogen levels were associated with higher D-dimer levels. This inverse relationship may reflect the complex interplay between thrombus formation and fibrinolysis, processes that are critical in the pathogenesis of TAAD.

Establishment of a comprehensive blood indicator

Given the observed correlations among preoperative hematological indicators and the limitations of conventional Cox proportional hazards modeling, we sought to improve predictive accuracy by employing the LASSO Cox regression model. This approach was used to assess the prognostic significance of ten clinical indicators in relation to the primary endpoint, mortality. The LASSO method, known for its ability to assign varying coefficients to predictors and reduce the impact of less significant variables, allowed us to refine the model. As a result, the comprehensive physiological response indicator (CPRI) was formulated as follows: 779.3 - (167.82 × cTnT) + (91.72 × RBC) - (144.54 × WBC) + (120.56 × ANC) + (312.15 × ALC). Based on this formula, the 175 patients with TAAD were divided into two groups using the optimal cut-point derived from X-tile software (Yale University, New Haven, Connecticut): the high CPRI group (≥ 966) and the low CPRI group (< 966).

Prognostic impact of preoperative CPRI in patients with TAAD

In a comparison between patients with TAAD who had lower versus elevated preoperative CPRI values, a significant improvement in survival was observed in those with higher CPRI levels (p = 0.001, Fig. 3). This association was confirmed in both univariate and multivariate Cox regression models (Table 2), where CPRI was identified as a significant predictor of survival outcomes. In the univariate analysis, patients with a CPRI score below 966 exhibited a higher hazard ratio (HR) of 2.617 (95% CI: 1.422–4.815, p = 0.002), and this remained significant in the multivariate analysis with an HR of 2.325 (95% CI: 1.126–4.802, p = 0.023). These findings from the multivariate Cox regression analysis underscore the robustness of CPRI as an independent prognostic factor for survival, with a p-value of 0.023.

Fig. 3

Kaplan-Meier survival curves of overall survival stratified by preoperative CPRI in 175 TAAD patients (with log-rank test). Abbreviations: CPRI comprehensive physiological response indicator

Table 2 Univariate and multivariate Cox regression analysis for overall survival

In addition to CPRI, other independent prognostic factors for overall survival identified in the multivariate analysis included age (< 52 vs. ≥52, HR: 0.336, 95% CI: 0.179–0.628), postoperative hospital stay (< 13 vs. ≥13, HR: 14.407, 95% CI: 6.274–33.083), ICU hospital stay (< 3 vs. ≥3, HR: 7.449, 95% CI: 2.793–19.867), presence of back pain (no vs. yes, HR: 0.471, 95% CI: 0.248–0.893), abdominal pain (no vs. yes, HR: 0.369, 95% CI: 0.157–0.866), creatinine (< 138 vs. ≥138, HR: 0.280, 95% CI: 0.118–0.662), fibrinogen (< 3.2 vs. ≥3.2, HR: 3.334, 95% CI: 1.499–7.414), ST-segment deviation on ECG (no vs. yes, HR: 2.133, 95% CI: 1.011–4.500), and heart rate (< 77 vs. ≥77, HR: 0.440, 95% CI: 0.197–0.980).

Furthermore, CPRI demonstrated superior predictive ability compared to other blood and clinical indicators at admission, including back pain, abdominal pain, ST-segment deviation on ECG, fibrinogen, creatinine, and age. The area under the curve (AUC) values for these indicators were 0.629 (95% CI: 0.509–0.685) for CPRI, 0.597 (95% CI: 0.509–0.685) for heart rate, 0.506 (95% CI: 0.413–0.600) for back pain, 0.565 (95% CI: 0.469–0.662) for abdominal pain, 0.518 (95% CI: 0.426–0.611) for ST-segment deviation on ECG, 0.576 (95% CI: 0.487–0.666) for fibrinogen, 0.610 (95% CI: 0.514–0.706) for creatinine, and 0.593 (95% CI: 0.499–0.687) for age (Fig. 4).

Fig. 4

Comparison of AUC between blood and clinical indicators at admission for patients with TAAD. Abbreviations: CPRI comprehensive physiological response indicator, AUC area under the curve Note Abnormal ECG means ST-segment deviation on ECG

Next, A prognostic nomogram was developed based on the CPRI and nine additional factors identified through multivariate analysis to quantitatively predict 1-year, 3-year, and 5-year overall survival (Fig. 5).

Fig. 5

Nomogram composed of Age, CPRI, Postoperative hospital stay, ICU hospital stay, Back pain, Abdominal pain, Creatinine, Fibrinogen, Abnormal ECG and Heart rate. Abbreviations: CPRI comprehensive physiological response indicator. Note: Abnormal ECG means ST-segment deviation on ECG

Model validation and performance assessment

This nomogram (Model 1) outperformed previously reported prognostic models for patients with AD, including Model 2 (NLR), Model 3 (PLR), and Model 4 (uric acid + age + D-dimer) [33, 34]. The area under the curve (AUC) values for the models were 0.874 (95% CI: 0.816–0.931) for Model 1, 0.592 (95% CI: 0.503–0.681) for Model 2, 0.514 (95% CI: 0.421–0.606) for Model 3, and 0.577 (95% CI: 0.486–0.668) for Model 4 (Fig. 6).

Fig. 6

Comparison of AUC between four models for patients with TAAD. Abbreviations: AUC the Area Under the Curve

To ensure the accuracy and stability of the model, Variance Inflation Factors (VIF) were calculated among the model’s variables to assess potential risks of multicollinearity. Typically, a VIF value exceeding 10 indicates severe multicollinearity. As shown in Fig. S1, all variables in our model had VIF values below 2, suggesting a minimal risk of multicollinearity.Additionally, the correlation matrix revealed generally low correlations among variables, with most Pearson’s correlation coefficients near zero, further supporting the strong independence of the variables (Fig. S2).

We also validated the model’s performance using K-fold cross-validation, employing the Concordance Index (C-index) as the measure of predictive accuracy. On the training set, the model achieved an average C-index of 0.8730 with a small standard deviation of 0.0083, demonstrating strong predictive ability and consistent performance across folds. On the validation set, the average C-index was 0.8546—slightly lower than the training set—indicating robust predictive power on unseen data. The standard deviation on the validation set was 0.0169, reflecting some variability but maintaining overall stability. The small difference of 0.0184 in C-index between the training and validation sets suggests the model generalizes well and has a low risk of overfitting (Table S1).

In summary, the low VIF values and minimal correlations among model variables indicate a negligible risk of multicollinearity. Cross-validation results further support the model’s reliability, with high C-index values on both training and validation sets. While the validation set showed slightly lower performance, this difference is acceptable, highlighting the model’s strong generalization capability. The use of LASSO regression in this study also helps mitigate the risk of overfitting. However, to comprehensively validate the model’s predictive performance, testing on an independent external dataset remains a crucial future direction.

Discussion

To the best of our knowledge, this study is one of the few efforts in devising a comprehensive blood-based index, CPRI, by integrating inflammatory and immune-related parameters via the LASSO method. Our data indicate that the CPRI is a robust independent prognostic factor for TAAD, with higher preoperative CPRI values associated with improved long-term survival. Notably, the CPRI outperformed other blood and clinical indicators, underscoring its superior predictive capacity. The predictive accuracy of our nomogram was significantly enhanced by incorporating a variety of blood and clinical indicators, including CPRI, age, postoperative hospital stay, ICU hospital stay, back pain, abdominal pain, creatinine, fibrinogen, ST-segment deviation on ECG, and heart rate. This comprehensive model demonstrated superior predictive performance compared to other prognostic models previously reported for patients with TAAD [33, 34].

AD is a pathologically complex condition driven by multiple mechanisms, including disruptions in intercellular signaling, phenotypic changes in vascular smooth muscle cells, endothelial cell injury, depletion of smooth muscle cells, degradation of the extracellular matrix, inflammatory responses, and oxidative stress, among other processes [21,22,23,24,25,26,27]. A variety of studies have identified biomarkers associated with the development and progression of AD, offering potential diagnostic and prognostic value. Key biomarkers include inflammatory mediators such as IL−6, IL−10, C-reactive protein, and soluble ST2, as well as D-dimer and cardiac markers like NT-pro BNP and cTnT [28].

Importantly, the intricate interplay of these factors, rather than any single indicator, reflects the prognosis of AD more accurately. Relying on a single biomarker may not provide a sufficient basis for predicting patient outcomes, limiting its utility in clinical settings. Among these biomarkers, D-dimer is widely recognized for its role in indicating hypercoagulability and secondary hyperfibrinolysis. Elevated D-dimer levels have been shown to have prognostic value in several studies [20, 29]. However, in our study, D-dimer did not maintain statistical significance in the multivariate analysis, possibly due to the limited sample size, which may have constrained the study’s power to detect a true association. Alternatively, D-dimer may not be the sole or primary determinant of prognosis in TAAD, with other, more robust biomarkers or clinical variables potentially playing a more significant role.

In our study, we identified that long-term survival was significantly worse in TAAD patients older than 52 years, with age serving as an independent prognostic factor. This finding highlights the importance of paying close attention to middle-aged TAAD patients, as the risk of adverse outcomes is not limited to the elderly population. In addition, ST-segment deviation on ECG [31], fibrinogen levels [32], and heart rate [30] were identified as independent prognostic factors for TAAD, consistent with results from previous studies. Notably, aside from these established indicators, we also identified creatinine, back pain, abdominal pain, postoperative hospital stay, and ICU hospital stay as independent prognostic factors for TAAD. These factors have not been extensively investigated in this context.

Expanding on the clinical presentation of TAAD, we observed significant individual variability in symptom expression. The classic manifestation is severe, often described as a tearing chest pain that is intolerable. Intriguingly, our findings suggested that patients presenting with back and abdominal pain at admission tend to have a poorer prognosis. This association may indicate disease progression, with pain spreading and intensifying over time, adversely impacting long-term survival. The recognition of these symptoms as potential prognostic indicators could facilitate the early identification of patients at an elevated risk of poor outcomes in clinical practice, thereby informing tailored management strategies.

We also found that patients with longer postoperative hospitalizations and extended ICU stay had superior long-term prognoses, with these factors emerging as independent predictors of long-term survival in TAAD patients. This association could be partially explained by the observation that patients who were able to tolerate a longer hospitalization may have had better overall health at baseline. In contrast, patients with more severe disease may have experienced death at an earlier stage, prior to the potential benefits of longer-term care being realized. Additionally, extended hospital stays may reflect the benefit of more comprehensive postoperative monitoring and care, which is crucial for the prevention and timely management of complications. Hence, it is imperative to acquire additional research evidence to determine whether the optimization of hospital stay duration should be incorporated as a goal within the perioperative management and monitoring paradigm for patients with TAAD.

Although individual indicators have demonstrated some prognostic value in patients with TAAD, their performance in terms of comprehensiveness and accuracy can be enhanced. In the present study, Pearson chi-square analysis revealed strong correlations between certain indicators, such as a Pearson correlation coefficient of 0.97 between WBC and ANC, and a coefficient of 0.67 between RBC and Hb. These high correlations may reduce the effectiveness of statistical tests and limit the predictive power of the model. Therefore, the development of easily accessible, effective and comprehensive assessment tools is crucial for the prognostic management of AD.

To address this, we developed a novel prognostic indicator, CPRI, using the LASSO regression method. The CPRI integrates cTnT, RBC, WBC, ANC, and ALC. We assessed the contribution of these variables to clinical outcomes and applied coefficients to eliminate variables contributing minimally. Our results demonstrated that in TAAD patients, the prognostic value of CPRI is significantly superior to other blood and clinical indicators, such as back pain, abdominal pain, ST-segment deviation on ECG, fibrinogen, creatinine, and age. The AUC values for these indicators are as follows: CPRI 0.629, back pain 0.506, abdominal pain 0.565, ST-segment deviation on ECG 0.518, fibrinogen 0.576, creatinine 0.610, and age 0.593. Furthermore, the predictive performance of the model incorporating CPRI, postoperative hospital stay, creatinine, fibrinogen, heart rate, back pain, abdominal pain, and ICU hospital stay (Model 1, AUC: 0.874) was significantly superior to other models (Model 2–4, with AUCs of 0.592, 0.514, and 0.577, respectively).

While current study provides valuable preliminary findings, it has certain limitations warrant consideration. First, the relatively small sample size may limit the generalizability and reliability of our conclusions. Therefore, future studies with larger cohorts are necessary to validate the clinical utility. Second, as a single-center study, our research may be prone to selection bias, multi-center studies are still needed. Finally, CPRI and the associated prognostic model proposed in our study are still in their initial stages and require further refinement to ensure applicability and accuracy across diverse populations and healthcare settings.

Conclusion

This study introduced a novel prognostic indicator, CPRI, which has been shown to be strongly associated with long-term survival in patients with TAAD. Furthermore, when contrasted with other established prognostic models, the nomogram model incorporating CPRI and other clinical indicators demonstrated superior predictive performance.