Hospital Mortality In Patients With Acute Aortic Dissection And Candidate Predictors Of Diagnostic Biomarkers

Objective:Type A acute aortic dissection (AAD) confers higher in-hospital mortality. It is controversial that plasma D-dimer can predict in-hospital mortality of AAD. Our study aims to investigate the correlation of plasma D-dimer with in-hospital mortality and analyze risk factors of in-hospital mortality in patients with type A AAD.Methods:The study designed as a prospectively observational study. From2012-2to2013-5, a series of consecutive patients with suspect AAD presented to our emergency department was enrolled, and plasma D-dimer level was measured (stago-evolution, France) immediately following admission. The diagnosis of type A AAD was confirmed by aorta angiography with multi-detector computed tomography for each patient. To define risk factors of in-hospital mortality, patient’s clinical characteristics were collected. All patients were followed up during hospitalization. Patients were divided into two groups:the deceased group, who died during hospitalization, and the survival group. The risk factors for in-hospital mortality were determined by using univariate and multivariate Cox proportional hazards analyses.Results:A total of194patients with type A AAD were included,134males, average age (51.7±11.1years). During a mean of13days hospitalization, death occurred in27(13.9%) patients. The plasma D-dimer level of the deceased group was significantly higher than that of the survival group (13.9±7.6μg/ml vs.8.5±6.8μg/ml, P=0.000). The platelet counts of the deceased group was significantly lower than that of the survival group (125±68×109/L vs.187±74×109/L, P=0.000). The plasma Scr level of the deceased group was higher. Admission blood pressure levels of the deceased group were lower than the survial group (118/64mmHg vs.140/75mmHg, P<0.01). More patients in the survival group undergone surgery(72.5%). On univariate Cox regression analysis, admission D-dimer level≥20μg/ml was associated with increased risk of in-hospital death(HR,3.964;95%CI,1.830～8.588; P=0.000). The in-hospital mortality was significantly higher in patients with plasma D-dimer level≥20μg/ml than in those with plasma D-dimer level<20μg/ml (34.2%vs.9.0%, log rank P<0.001). In model1, after adjustment for age, systolic blood pressure, platelet counts, Scr level, a high admission D-dimer level (≥20μg/ml) was still a powerful independent predictor of in-hospital mortality (HR:3.709,95%CI (1.558-8.827)). Given that surgical intervention is key treatment measure to improve the prognosis, so we constructed Cox regression model2which include model1and surgical intervention. The results showed that high admission D-dimer level did not independently correlate with in-hospital mortality any more. Meanwhile admission platelet counts (HR:0.993;95%CI (0.987～0.999)), systolic blood pressure (HR:0.979,95%CI (0.965-0.992)) and surgery (HR:0.033;95%CI(0.008-0.146)) independently correlated with in-hospital mortality. Correlation analysis indicated a negative correlation between admission D-dimer level and platelet counts(r=-0.442, P<0.001). Type A AAD patients who did not undergone surgery with a higher admission D-dimer≥20ug/ml confers higher in-hospital mortality. However when these patients undergone surgery, their in-hopital mortality were similar.Conclusions:Our results suggest admission lower systolic blood pressure, lower platelet counts and conservative treatment are independent predictors for in-hospital mortality of AAD. A high admission D-dimer level might be a powerful predictor for increased in-hospital mortality in patients with type A AAD, these patients may benefit more from surgery. Objective:The excessive mortality associated with acute aortic dissection (AAD) makes the need for accurate diagnosis critical. Biomarker-assisted diagnosis of acute aortic dissection is important for initiation of treatment and improved survival. However, identification of biomarkers for AAD in blood is a challenging task. The present study aims to discover the potential AAD biomarkers using a transcriptomic strategy.Methods:Ascending aortic tissue was collected from eight male patients with Stanford Type A aortic dissection and eight healthy male organ donors. Array based genome-wide gene expression profiling were performed on a panel of aortic tissues from4AAD patients (mean age,49.1±4.9years) and4controls (mean age,47.9±6.7years). The differentially expressed genes were validated using quantitative reverse transcriptase PCR (qRT-PCR) and westen blot in the whole tissue panel. The plasma levels of potential biomarker were determined in two independent case-control cohorts by enzyme linked immunosorbent assay. Twenty type A AAD patients and eighteen normal controls were enrolled in the pilot cohort. Sixty five type A AAD patients, thirty two type B AAD patients and forty four normal controls were included in the second validation cohort. Difference between groups was evaluated using Student’s t test or Mann-Whitney test. Sensitivity and specificity were calculated in relation to the final diagnosis. Receiver operating characteristics (ROC) curves were constructed by plotting sensitivity (true-positive fraction) vs.1-specificity (false-positive fraction) for discrimination between controls and patients with AAD. The area under the curve was calculated.Results:Transcriptome data demonstrated that a total of18genes were significantly up-regulated and28genes were significantly down-regulated among AAD tissues (FDR-adjusted p<0.01, foldchange>3.0). The differentially expressed gene set was highly enriched among functional categories involved in Focal adhesion and Regulation of actin cytoskeleton pathways (p<0.001). By using literature data (Gene Ontology, UniProt, and Human Plasma Proteome), we identified three proteins (angiopoietin2(ANGPT2), chemokine (C-C motif) ligand3(CCL3) and FN1), as candidate biomarkers for blood-based detection of AAD. The qRT-PCR assay demonstrated that ANGPT2and CCL3increased2.4-fold (P<0.01) and3.6-fold (P=0.02) in the entire patient group, compared with control specimens, respectively. The western-blot analysis indicated that ANGPT2in AAD group increased4.2-fold (P=0.033). In the pilot study, although the plasma concentrations of CCL3were unchanged(P=0.51), ANGPT2levels were significantly1.8-fold higher in AAD patients than in controls(P=0.002). In the validation study, ANGPT2was also significantly elevated in AAD patients as compared with controls (medians4.1ng/mL vs.1.4ng/mL, p=6.9E-9). ROC curve analysis showed that ANGPT2was highly predictive of a diagnosis of acute aortic dissection (area under ROC curve0.81, p<0.0001). Sensitivity and specificity were72%and80%, respectively, at the cutoff value of2.46ng/mL.Conclusions:Our study indicate that Focal adhesion and Regulation of actin cytoskeleton pathways are probably invoved in pathogenesis of AAD. Moreover, our preliminary study offers the first evidence that ANGPT2is associated with AAD and could be a promising biomarker for early detection of AAD. Future studies with a larger sample size of normal and AAD patient samples should be pursued.