Article Text
PDF
Abstract
Objectives Acute type A aortic dissection is a life-threatening cardiovascular disease commonly seen in emergency department, resulting in substantial mortality and morbidity. We aimed to investigate the prognostic value of N-terminal pro-B type natriuretic peptide (NT-proBNP) among this critically ill population.
Design The design of this study was a retrospective cohort study.
Setting The study population was recruited in the Emergency Department of Fuwai hospital in China from 2018 to 2020.
Participants We consecutively enrolled 829 patients with acute type A aortic dissection and measurable baseline NT-proBNP.
Primary outcome The primary endpoint was 1-year all-cause death.
Results Based on tertiles of NT-proBNP (pg/mL), patients were stratified into low (≤150.3, n=276), intermediate (150.3–667.6, n=277) and high (>667.6, n=276) NT-proBNP groups. Compared with patients with low NT-proBNP, the Kaplan-Meier estimates for primary 1-year mortality were higher in intermediate (32.5% vs 18.1%; HR 1.91, 95% CI 1.35 to 2.69) and high (42.0% vs 18.1%; HR 2.56, 95% CI 1.84 to 3.57) NT-proBNP groups, respectively. After multivariable regression adjusted for confounders, NT-proBNP tertiles were independent predictors for 1-year mortality (adjusted HR for intermediate group 1.52, 95% CI 1.02 to 2.27; adjusted HR for high group 2.17, 95% CI 1.41 to 3.32). Notably, the predictive performance of NT-proBNP for 1-year mortality was greater in patients receiving surgery than conservative treatment (between-cohorts difference in area under the curve 0.13, Delong’s test p=0.04).
Conclusion NT-proBNP provides incremental prognostic information for mortality in patients with acute type A aortic dissection who underwent surgical repairment, which could aid in risk stratification as a pragmatic and versatile biomarker in this critically ill population while having limited prognostic value for those receiving conservative treatment.
- Mortality
- Emergency Departments
- Cardiovascular Disease
Data availability statement
Data are available upon reasonable request. The data that support the findings of this study are available from the corresponding author upon reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
This study consecutively enrolled patients with type A aortic dissection incorporated acute-phase and long-term prognosis representing a well-phenotyped group in China.
Not all participants had an N-terminal pro-B-type natriuretic peptide measurement.
This study was conducted in a single centre which may affect the external validity of the present results.
Introduction
Despite the improvement of diagnostic and therapeutic techniques in recent decades, acute aortic dissection is still a life-threatening cardiovascular disease commonly seen in emergency department, resulting in over half of mortality in patients without proper treatment.1–4 In addition, acute aortic dissection is also a rapid-progressive disorder with the risk of death increased by 1% per hour in the early stage.3 Compared with those with acute type B aortic dissection, patients with acute type A aortic dissection acquire substantially worse in-hospital and long-term prognosis as the ascending aorta is involved.5 6 Therefore, it is of great clinical implication to timely and effectively identify type A aortic dissection patients at higher risk, which would assist clinicians in developing the proper treatment and management strategy to improve the prognosis at the earliest possible stage. However, although there is increasing interest in the use of circulating biomarkers for risk stratification of patients with aortopathy, biomarker expression has not been clearly associated with relevant aortic clinical events.1
Natriuretic peptides, including B-type natriuretic peptides (BNP) and the N-terminal fragment of its prohormone (NT-proBNP), are endogenous cardiac hormones mainly secreted by cardiomyocytes in response to increased stress of cardiac chamber wall.7 As an established biomarker for heart failure,7 8 natriuretic peptides have been proven useful for the diagnosis and risk stratification in several other cardiovascular diseases, including coronary artery disease and valvular heart disease.9–12 Previous studies have demonstrated the prognostic value of NT-proBNP in patients with acute aortic dissection.13–17 However, these small-scale studies were generally conducted in earlier years and mainly focused on acute-phase prognosis with type B aortic dissection. In patients with acute type A aortic dissection, the association between NT-proBNP and long-term prognosis has not been fully clarified, and its clinical implication needs further validation. The present study was designed to investigate the prognostic value of NT-proBNP tertiles for 1-year mortality, and whether the prognostic value differed between patients with conservative and surgery treatment in patients with acute type A aortic dissection in a relatively large cohort.
Methods
Study population
A total of 847 consecutive patients were recruited with acute type A aortic dissection diagnosed by aortic CT angioplasty in the emergency department of Fuwai hospital from January 2018 to December 2020. Acute aortic dissection was diagnosed by CT and classified according to the Stanford system: (1) type A involves the ascending aorta, regardless of the site of the primary intimal tear; and (2) type B involves only the descending aorta. Adult patients were eligible for inclusion if they were diagnosed with acute type A aortic dissection with onset time ≤14 days from symptom to diagnosis. Recurrent aortic dissection was excluded in the present study. All participants provided written informed consent.
Data collection and follow-up
All data were obtained from the electronic health records. Demographic characteristics, cardiovascular risk factors, comorbidities, in-hospital assessment, laboratory biomarkers and treatment strategy were recorded in real time by medical personnel. For NT-proBNP, blood samples were collected into EDTA-anticoagulant tubes by venipuncture in emergency department, and the sample would be sent to the laboratory immediately for analysis. Plasma NT-proBNP concentration was measured using an Elecsys proBNP, Cobas E analyser (Roche Diagnostics GmbH, Mannheim, Germany) within a measurable range between 5 and 35 000 pg mL/L. Risk classification of patients was performed according to tertiles of NT-proBNP: (1) low NT-proBNP, ≤150.3 pg/mL; (2) intermediate NT-proBNP, 150.3–667.6 pg/mL; and (3) high NT-proBNP, >667.6 pg/mL. In subgroup analysis, patients were further stratified according to the treatment strategy into conservative group or surgery (open repair) group.
The primary endpoint for the present study was the all-cause death within 1 year from emergency contact (ie, date of emergency admittance). The secondary endpoint was the 30-day rate of all-cause death. Considering that the visualisation of the relationship between NT-proBNP and prognosis is more intuitive when it is used as a categorical variable (eg, KM curve, mortality increase with increasing tertiles), the results with NT-proBNP tertiles were used as the primary outcome.
Statistical analysis
Continuous variables are expressed as mean±SD or median (IQR) and categorical variables are presented as counts (%). Restricted cubic splines were applied to delineate the curve of associations between baseline NT-proBNP level and the risk of all-cause death. The receiver-operating characteristic (ROC) curve analysis with AUC was used to compare the prediction capability for 1-year mortality between cohorts with conservative or surgery treatments using DeLong’s test.18 For 1-year outcome, considering ROC curve analysis could not adjust impact of potential confounders, cox proportional hazards model was used to estimate HR and 95% CI, while logistic regression model was used to estimate OR and 95% CI for 30-day outcome. Multivariable adjusted analysis was used to identify independent predictors. The candidate variables for multivariable analysis were identified using historical confounder definition based on clinical knowledge and previous literature reports.19 The included covariates were age, admission SBP, smoking, syncope, coma, time from onset to admission, left ventricular diameter, left ventricular ejection fraction, pericardial effusion, troponin I, creatinine, C reactive protein, and artery affected—coronary artery. Subgroup analysis was performed according to the treatment strategy (ie, conservative treatment and surgery), and the p value for interaction was calculated from a multivariable Cox proportional hazards model. Unless otherwise specified, a two-sided p value <0.05 was considered to indicate statistical significance. All statistical analyses were performed using R software, V.4.2.0 (R Foundation for Statistical Computing, Vienna, Austria).
Patient and public involvement
None.
Results
A total of 847 consecutive patients with acute type A aortic dissection were enrolled between January 2018 and December 2020, among which 18 patients without available baseline NT-proBNP data (n=7) or completed 1-year follow-up (n=11) were excluded (figure 1). Therefore, 829 patients were included in the present study. The median baseline NT-preBNP was 308.0 pg/mL (IQR 104.8–974.5).
Flow chart. NT-proBNP, N-terminal pro-brain natriuretic peptide.
Risk classification of patients was performed according to tertiles of baseline NT-proBNP level (pg/mL): (1) ≤150.3 (low NT-proBNP group, n=276); (2) 150.3–667.6 (intermediate NT-proBNP group, n=277); (3) >667.6 (high NT-proBNP group, n=276).
Baseline characteristics
Baseline characteristics of the NT-proBNP tertiles are summarised and stratified in table 1. Among 829 patients, 587 were men (70.8%), with an average age of 55.1 years. The median NT-preBNP levels were 74.0 (IQR 40.7–105.4), 308.0 (IQR 219.0–444.9) and 1490.5 (IQR 974.3–3108.5) in low, intermediate and high NT-proBNP groups, respectively. Compared with the lowest tertiles group, patients with higher NT-proBNP tertiles tended to have higher level of advanced age, heart rate, previous coronary artery disease, previous aortic disease, time from onset to admission, left ventricular diameter, creatinine, C reactive protein and troponin I, with lower levels of male proportion, admission blood pressure, smoking status, left ventricular ejection fraction and haemoglobin. In addition, the percentage of surgery treatment was decreased along with the increasing NT-proBNP levels (77.5%, 61.7% and 50.4%, respectively, p<0.001).
Patient characteristics according to the tertiles of NT-pro BNP levels
Prognostic value of NT-proBNP among the whole cohort
A total of 256 (30.9%) deaths occurred during 1-year follow-up, and the 30-day death was documented in 233 (28.1%) patients. Comparisons of demographic data and clinical characteristics of patients stratified by 1-year or 30-day outcomes are presented in online supplemental tables S1 and S2. Median NT-proBNP level (pg/mL) in 1-year survivors versus non-survivors was 236.3 (IQR 90.9–794.0) versus 517.2 (IQR 200.2–1449.9; p<0.001; figure 2), and in patients without versus with 30-day death was 248.0 (IQR 91.5–846.5) and 482.0 (IQR 195.7–1489.0), respectively (online supplemental figure S1).
Supplemental material
Baseline NT-proBNP as a predictor of 1-year outcome. NT-proBNP, N-terminal pro-brain natriuretic peptide.
As a categorical variable, Kaplan-Meier curves showed a graded risk for 1-year mortality with higher NT-proBNP levels (log-rank p<0.001) (figure 3A). Compared with patients with low NT-proBNP, the risk of 1-year death was higher in intermediate (32.5% vs 18.1%; HR 1.91, 95% CI 1.35 to 2.69, p<0.001), and high groups (42.0% vs 18.1%; HR 2.56, 95% CI 1.84 to 3.57, p<0.001), respectively (online supplemental table S3).
Death within 1 year from emergency contact according to NT-proBNP levels. Incidence of 1-year all-cause death is presented according to (A) NT-proBNP tertiles and (B) continuous value of ln NT-proBNP among patients with acute type A aortic dissection. NT-proBNP, N-terminal pro-brain natriuretic peptide.
As a continuous variable, restricted spline curve analysis showed that there was a monotonic increase in the risk of 1-year death with increasing NT-proBNP concentrations (P for linearity=0.57; figure 3B). The ln NT-proBNP was significantly associated with 1-year mortality (HR 1.24, 95% CI 1.15 to 1.34, p<0.001; online supplemental table S3).
Multivariable adjustment analysis
In addition, by multivariable analysis, age, admission SBP, smoking, coma, time from onset to admission, left ventricular ejection fraction, creatinine and NT-proBNP tertiles (adjusted HR for intermediate group 1.52, 95% CI 1.02 to 2.27, p=0.04; adjusted HR for high group 2.17, 95% CI 1.41 to 3.32, p<0.001) were independent predictors for 1-year mortality (table 2). Similar results were observed for the secondary endpoint (table 2 and online supplemental figure S2).
Independent predictors of clinical outcomes
Performance of NT-proBNP tertiles in patients with conservative or surgery treatment
The comparison of baseline characteristics and clinical outcomes grouped by the treatment strategy was shown in online supplemental table S4. ROC analysis was performed in surgery and conservative treatment cohort separately to compare the predictive performance of NT-proBNP. As depicted in figure 4, NT-proBNP showed greater predictive power in surgery treatment subgroup (AUC 0.64, 95% CI: 0.54 to 0.74) when compared with conservative treatment subgroup (AUC 0.51, 95% CI: 0.44 to 0.59), with significantly between-cohorts AUC difference (ΔAUC 0.13, 95% CI: 0.01 to 0.25, p=0.04).
The receiver-operating characteristic curve of NT-proBNP for predicting 1-year death in conservative and surgery treatment cohorts. AUC, area under curve; NT-proBNP, N-terminal pro-brain natriuretic peptide.
Subgroup analysis was conducted to investigate the impact of treatment strategy (surgery or conservative treatment) on the association between NT-proBNP tertiles and all-cause mortality. In surgery treatment cohort, the rate of 1-year mortality was significantly increased in intermediate group (7.6% vs 2.8%; HR 2.79, 95% CI 1.06 to 7.33, p=0.04) and high group (7.9% vs 7.6%; HR 2.89, 95% CI 1.07 to 7.81, p=0.04) when compared with low NT-proBNP (≤155.0 pg/mL) group (figure 5 and table 3). In conservative treatment group, compared with low NT-proBNP group, the rate of 1-year mortality was comparable in intermediate group (72.6% vs 71.0%; HR 1.00, 95% CI 0.69 to 1.45, p=0.99) and high group (76.6% vs 71.0%; HR 1.05, 95% CI 0.74 to 1.49, p=0.79). Notably, there was a significant interaction between NT-proBNP tertiles and treatment strategy for 1-year death (P for interaction=0.04) (figure 5 and table 3). Similar results were observed for 30-day mortality, although surgery treatment cohort did not reach statistical significance. However, no significant interaction between NT-proBNP levels and treatment strategy was observed for 30-day death (P for interaction=0.18).
Outcomes in patients stratified by NT-proBNP tertiles in conservative and surgery cohorts. NT-proBNP, N-terminal pro-brain natriuretic peptide
Clinical outcomes in cohorts with conservative or surgery treatment, according to NT-proBNP tertiles
Discussion
The present study was focused on association of baseline NT-proBNP levels and mortality (ie, acute-phase and long-term mortality) in patients with acute type A aortic dissection, and the main findings are as follows: (1) in our primary analysis, baseline NT-proBNP tertiles were independent predictor of acute-phase or 1-year survival after multivariate adjustment; and (2) in our secondary analysis, NT-proBNP was more predictive of long-term outcomes in patients with acute type A aortic dissection undergoing surgery treatment. Therefore, baseline NT-proBNP, as a user-friendly and incremental prognostic factor, could assist in profiling risk among patients with acute type A aortic dissection.
NT-proBNP has been routinely used as a diagnostic tool for heart failure; besides, it has also been proven to be a novel and useful biomarker for the risk stratification of several other cardiac diseases and even non-cardiac conditions.10 20–22 A previous study has reported that the level of NT-proBNP was significantly higher in those with acute aortic dissection.23 In addition, several studies have demonstrated the prognostic value of NT-proBNP in patients with acute aortic dissection.15–17 For the first time, a prospective study of 104 type A aortic dissection patients revealed that higher levels of NT-proBNP predicted the occurrence of 30-day mortality and short-term major adverse events (ie, postoperative heart failure, neurological deficit, lung failure, renal failure or sepsis).15 Another study of 67 patients verified that NT-proBNP was an independent risk factor of in-hospital death in patients with type A aortic dissection.17 However, these studies on type A aortic dissection were limited by the relatively small sample size and the lack of long-term follow-up results. The present study further validated the prognosis value of NT-proBNP in the acute phase or 1 year later with the largest sample size so far (n=829).
Although the development of surgical repairment and intensive care has greatly improved the prognosis of type A aortic dissection, several studies still reported relatively high mortality rates.24 25 Many factors have been identified as predictors for short-term mortality; however, there is currently no established blood biomarker for risk stratification.17 As a non-specific preoperative biomarker, it is not comprehensive to use NT-proBNP alone as a risk predictor despite it being confirmed as an independent predictor in the present study. However, combined with the existing clinical risk factors, NT-proBNP could substantially improve prognosis prediction, which could assist physicians in identifying high-risk patients and enhance perioperative and follow-up management.
In the present study, a total of 305 (36.8%) received conservative treatment, and the reasons are as follows: (1) 133 (16.0%) patients suffered aortic rupture prior to emergency surgery, resulting in death and no opportunity for surgery; (2) some patients with multiple comorbidities are not suitable for surgery due to the contraindications after evaluation, which received conservative management; and (3) a small number of patients refused surgery due to the treatment costs.26 Early surgical repair has been recommended as the gold standard treatment for most patients with acute type A aortic dissection, which can significantly reduce mortality. This is also reflected in the present study, in which the 1-year mortality was 5.7% and 74.1% in the surgical and conservative treatment group, separately. Compared with the surgical group, patients in the conservative group had worse basic conditions and were more likely to have severe complications such as hypotension, shock, pericardial effusion and heart failure, which may be the cause of the elevated NT-proBNP and worse prognosis. Thus, we suggest that it should be cautiously interpreted the prognostic value of NT-proBNP in conservative cohorts. Besides, in the subgroup analysis, mortality risks were significantly higher in patients with higher NT-proBNP tertiles among surgical cohort while they were comparable in conservative cohort, and a significant interaction was observed between NT-proBNP tertiles and treatment strategy for 1-year death, indicating that only a particular population with surgery requirement might benefit of using NT-proBNP in their risk stratification.
There are several possible interpretations for the increased mortality in patients with elevated NT-proBNP levels. First, the increased plasma NT-proBNP levels were proven to be associated with cardiovascular dysfunction in critically ill patients regardless of surgery or not.27–30 And cardiac dysfunction is a common and significant predictor of poor prognosis among critically ill patients.31 32 Second, the occurrence and development of acute aortic syndrome involved activation of inflammatory pathway,3 24 and studies have demonstrated that systemic inflammation state contributed to morbidity and mortality in acute aortic syndrome.33 34 Moreover, severe systemic inflammation could further induce or exacerbate cardiac dysfunction which contributes to the increased plasma levels of NT-proBNP.35 Third, the troponin I levels were gradually increased along with the elevation of NT-proBNP levels, indicating a relatively poor coronary perfusion in patients with high NT-proBNP level, which is also an important predictor of mortality.36 Fourth, NT-proBNP levels are associated with abnormal kidney function,37 which could independently predict acute-phase and long-term prognosis. Finally, increased plasma levels of NT-proBNP may reflect the overall disease severity and the proportion of patients who received surgery was significantly reduced along with elevated NT-proBNP levels.
In addition, it was of great interest to observe that NT-proBNP levels significantly elevated along with the increase of time from onset of symptoms to admission, further indicating the importance of early diagnosis and treatment of acute type A aortic dissection in improving survival.3 4 15
Limitations of the study
The strength of the present study is this large-scale retrospective cohort of type A aortic dissection incorporated acute-phase and long-term prognosis, which reflect the current status of diagnosis and treatment of aortic dissection in China to a certain extent. However, this study has several limitations. First, this study was conducted in a single centre, although the enrolled patients came from multiple provinces in China; the external validity of the present study needs to be further confirmed in future multicentre studies. Second, longer follow-up results are warranted (eg, 3 years or 5 years) to further investigate the prognostic value of NT-proBNP especially for patients who underwent index surgery. Third, the impact of NT-proBNP levels on outcomes other than mortality, such as life quality and ischaemic events, is also worth investigating in future studies. Fourth, although the possible confounders were adjusted by multivariate analysis, we cannot exclude an effect from residual confounding (from measured covariates) and unmeasured confounders due to the observational design (eg, patient management at the emergency department, operating theatre and intensive care unit). Finally, serial measurements of NT-proBNP levels are not available in this study, and the impact of the dynamic change of NT-proBNP on outcomes cannot be evaluated. Therefore, the findings of the present study are hypothesis generating, and the clinical implications of NT-proBNP levels among patients with type A aortic dissection should be evaluated in future massive prospective multicentre studies.
Conclusions
NT-proBNP provides incremental prognostic information for mortality in patients with acute type A aortic dissection who underwent surgical repairment, which could aid in risk stratification as a pragmatic and versatile biomarker in this critically ill population while having limited prognostic value for those receiving conservative treatment. Further large-scale prospective studies are needed to confirm these findings.
Data availability statement
Data are available upon reasonable request. The data that support the findings of this study are available from the corresponding author upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This study was approved by the Ethics Committee of Fuwai Hospital (Approval no: 2016-847 ) and followed the principles of the Declaration of Helsinki. Participants gave informed consent to participate in the study before taking part.
References
Footnotes
SL, XB and QL contributed equally.
Contributors All authors have participated in the work. KD, RF and SL contributed to the conception or design of the work. SL, CS, QL, HW, RF, SY, CZ, JG, XC, SQ, YL and XB contributed to the acquisition of data for the work. RZ, XB, WL and SY contributed to the statistical analysis of the data. SL, XB and QL drafted the manuscript. KD and RF critically revised the manuscript and finalised the manuscript. KD acted as guarantor.
Funding The present study was supported by National High Level Hospital Clinical Research Funding (grant number: 2023-GSP-QN-17, 2023-GSP-GG-02), CAMS Innovation Fund for Medical Sciences (CIFMS) (grant number: 2021-I2M-1-008), and National Natural Science Foundation of China (grant number: 82070369).
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.