N‐terminal pro‐B‐type natriuretic peptide and D‐dimer combined with left atrial diameter to predict the risk of ischemic stroke in nonvalvular atrial fibrillation

Abstract Objectives We aimed to explore the potential role of N‐terminal pro‐B‐type natriuretic peptide (NT‐proBNP), d‐dimer, and the echocardiographic parameter left atrial diameter (LAD) in identifying and predicting the occurrence of ischemic stroke (IS) in patients with nonvalvular atrial fibrillation (NVAF). Methods We conducted a retrospective study of 445 patients with NVAF in the First Affiliated Hospital of Nanchang University. They were divided into the NVAF (309 cases) and NVAF with stroke (136 cases) groups according to whether acute ischemic stroke (AIS) occurred at admission. Multivariate logistic regression was used to analyze the odds ratio (OR) of NT‐proBNP, d‐dimer, and LAD for IS. The predictive value of NT‐proBNP, d‐dimer, and LAD in identifying the occurrence of IS in NVAF was determined by plotting the receiver operating characteristic (ROC) curves. Results NT‐proBNP, d‐dimer, and LAD levels were significantly higher in the NVAF with stroke group than in the NVAF group (p < .05). NT‐ProBNP, d‐dimer, and LAD were independently associated with IS in NVAF patients (odds ratio [OR] = 1.12, 95% confidence interval [CI]: 1.08–1.16; OR = 1.87, 95% CI: 1.37–2.55; OR = 1.21, 95% CI: 1.13–1.28, p < .01). The optimal cutoff points for NT‐ProBNP, d‐dimer, and LAD levels to distinguish the NVAF group from the NVAF with stroke group were 715.0 pg/ml, 0.515 ng/ml, and 38.5 mm, respectively, with the area under the curve (AUC) being [0.801 (95% CI: 0.76–0.84); 0.770 (95% CI: 0.72–0.85); 0.752 (95% CI: 0.71–0.80), p < .01]. The combined score of NT‐proBNP, d‐dimer, and LAD improved the predictive efficacy of the single index, with an AUC of 0.846 (95% CI: 0.81–0.88, p < .01), sensitivity of 77.2%, and specificity of 76.4%. Conclusion NT‐proBNP, d‐dimer, and the echocardiographic parameter LAD have outstanding value in predicting the risk of IS in patients with NVAF.


| INTRODUCTION
Atrial fibrillation (AF) is currently the most common tachyarrhythmia in clinical practice, and its incidence is increasing every year. 1 By the end of 2019, approximately 59.7 million patients had AF (including atrial flutter) worldwide. 2 AF can lead to stroke, myocardial infarction, and heart failure, which greatly increase the risk of death in patients. [3][4][5] Nonvalvular atrial fibrillation (NVAF) can be regarded as the main independent risk factor of ischemic stroke (IS). 6 Compared with other causes of stroke, acute ischemic stroke (AIS) associated with NVAF not only has an acute onset, rapid progression, and serious neurological loss but also has a higher recurrence rate and mortality, which imposes a huge medical burden on society. [7][8][9] Several AF-related IS risk stratification tools have been established internationally, such as the CHADS2, CHA2DS2-VASc, and ABC stroke scores. 10 The CHA2DS2-VASc score is currently the most widely used risk stratification tool in clinics and is used for the early identification of medium-and high-risk patients. 11 Previous studies [12][13][14][15][16] have found that N-terminal pro B-type natriuretic peptide (NT-proBNP), D-dimer, and the cardiac structural index left atrial diameter (LAD) are strongly associated with the development of AF and could help refine the risk assessment of IS related to NVAF. PAULIN et al. 17 combined three serum biomarkers, high-sensitivity troponin (HsT), brain natriuretic peptide (BNP), and D-dimer, with the general clinical characteristics of patients, and identified the potential value of hematological parameters in predicting the occurrence of IS in AF. Due to the complex process of thromboembolism in patients with NVAF, current research focuses on different markers. However, in view of their extensive clinical use, easy access, and promotion, they are still highly beneficial in guiding clinical work.
This study aimed to investigate the risk factors for NVAF-related IS and analyze the potential value of NT-proBNP, D-dimer, and LAD in the early prediction of NVAF-related IS. This provides a reliable basis for early prevention and treatment of NVAF-related IS.

| Data collection
Patients who met the inclusion and exclusion criteria during hospitalization were analyzed by reviewing their medical records and collecting corresponding materials, including (1) basic demographic characteristics, such as sex, age, and body mass index (BMI); (2) medical or drug history, such as smoking, hypertension, diabetes mellitus, coronary heart disease (CHD), previous IS, antiplatelets, and anticoagulants; (3) serological indicators, such as NT-proBNP, D-dimer, fibrinogen, HsT, lipid quadruple, creatinine, uric acid, homocysteine, C-reactive protein (CRP), and all test results obtained from the first venous blood test after admission and before treatment;

| Definition
NVAF was defined as AF without mechanical valves or moderate-tosevere mitral stenosis (usually resulting from rheumatism). Patients with NVAF included all three types of AF (paroxysmal AF, persistent AF, and permanent AF) with arbitrary durations. 18,19 IS was defined as neurological deficits caused by focal brain, spinal cord, or retinal infarction. 20 BMI was defined as weight (kg) divided by the square of height (m) (kg/m 2 ). Smoking is described as a total of ≥100 cigarettes in their lifetime, regardless of whether they have quit smoking now. 21 History of hypertension refers to previous systolic blood pressure ≥ 140 mmHg and/or diastolic blood pressure ≥ 90 mmHg or use of antihypertensive drugs. A history of diabetes was defined as treatment with oral hypoglycemic agents or insulin, fasting glucose level ≥ 7.0 mmol/L, glycosylated hemoglobin ≥ 6.5%, or an oral glucose tolerance test showing 2-h blood glucose ≥ 11.1 mmol/L.
History of CHD was defined by previous episodes of angina pectoris or myocardial infarction, any positive cardiac stress test results, or coronary angiographic pathology signs. 22 History of previous IS was defined as any neurological dysfunction event with or without sequelae. 22

| Risk stratification schemes for ischemic stroke
The CHA2DS2-VASc score was calculated as congestive heart failure, hypertension, diabetes mellitus, vascular disease, age 65-74 years, and female sex; each of the above was scored as 1 point. A history of stroke, transient ischemic stroke, or age ≥75 years was scored as 2 points 23 (Table S1).

| Statistical analyses
Continuous variables with a normal distribution are described as mean ± standard deviation, while continuous variables with a non-

| Predictive value of risk factors
ROC analysis was used to evaluate the predictive value of the risk factors for the occurrence of IS events in patients with NVAF. As shown in Figure 2 and Table 3 Abbreviations: BMI, body mass index; CHD, coronary heart disease; CI, confidence interval; LAD, left atrial diameter; NT-proBNP, N-terminal pro B-type natriuretic peptide; NVAF, nonvalvular atrial fibrillation; OR, odds ratio. a Adjusted for demographics (age, sex, and BMI), clinical risk factors (smoking, hypertension, diabetes mellitus, CHD, and history of stroke), medication history (antiplatelet and anticoagulant drugs), and admission CHA2DS2-VASc score.
F I G U R E 2 Receiver-operator characteristic curves for NT-proBNP, D-dimer, and LAD to predict NVAF with Stroke. Combined score, NT-proBNP combined with D-dimer and LAD. D2, D-dimer; LAD, left atrial diameter; NT-proBNP, N-terminal pro B-type natriuretic peptide.

| DISCUSSION
Our research investigated the potential predictive value of NT-proBNP, D-dimer, and the echocardiographic parameter LAD for the development of IS in NVAF patients who were at a high risk for cardiogenic stroke. The results showed that NT-proBNP, D-dimer, and LAD levels were significantly higher in NVAF patients with stroke than in those without stroke. Increased levels of NT-proBNP, D-dimer, and LAD are of great value in confirming the diagnosis of IS and can be used as potential predictors of IS in NVAF patients.
BNP is produced primarily by cardiomyocytes in response to increased end-diastolic pressure and/or volume expansion and then enzymatically cleaved to NT-proBNP. 24,25 Previous research [26][27][28] have found that NT-proBNP is a predictor of AF and thromboembolic events and is independently associated with an increased risk of IS. In addition, a single-center study 29 showed that the inclusion of NT-proBNP in the CHA2DS2-VASc score increased the predictive ability of the risk of IS or systemic thromboembolism by 17% in anticoagulated patients with AF.
NT-proBNP levels were significantly higher in patients with IS than in those without IS, and there was a definite correlation between NT-proBNP and IS (AUC: 0.801, 95% CI: 0.76-0.84, cutoff point: 715.0 pg/ ml). A previous study 30 found that NT-proBNP levels 24 h after acute IS were higher than those in controls, which is consistent with our findings.
Among the coagulation indicators in patients with AF, plasma D-dimer has been extensively studied as an indicator of thrombosis. 31 Levels of D-dimer are elevated compared with matched controls in sinus rhythm and appeared to remain elevated despite successful cardioversion, demonstrating a correlation between D-dimer and NVAF. 15 In an observational study, 32 AF patients with high D-dimer levels also had a higher risk of IS, transient ischemic attack, and arterial thrombotic events. Numerous studies [33][34][35] had reported that D-dimer level is elevated during acute stroke, is associated with stroke subtypes and volume, and is significantly elevated in cardioembolic ischemic stroke. In recent years, prospective studies 36,37 have revealed that D-dimer levels were correlated to acute stroke, and can be a valuable and independent short-term prognostic marker for acute stroke.
Therefore, assessment of D-dimer levels is of great value in the early prevention of IS events in patients with NVAF. The incidence of stroke was also significantly higher in the D-dimer level ≥ 0.515 ng/ml group in our study (82.4% vs. 38.5%, p < .01).
A meta-analysis of 22 clinical studies by Njoku et al. 38 suggested that the prevalence of AF is significantly and positively correlated with increased LAD, leading to myocardial dysfunction, hemodynamic changes, and inflammatory factors that increase the incidence of thrombosis. Left atrial enlargement may also lead to IS and thrombosis by promoting endothelial damage. 39,40 In our study, decreased LVEF and increased LVDd were also independent risk factors for the occurrence of IS in patients with NVAF. Therefore, we speculate that LVEF and LVDd could also predict the occurrence of cardiogenic cerebral embolism to a certain extent.
Despite widespread clinical use, there are limitations in the ability of the CHA2DS2-VASc score to predict the risk of IS in patients with NVAF. Evidence 41 from a recent systematic review shows that this score has not ideal predictive power (c-statistic of 0.6-0.7). Our study confirms to some extent the potential predictive value of the 3 identified risk factors or biomarkers (LAD, NT-proBNP and D-dimer), which will likely contribute to the refinement of the CHA2DS2-VASc score and be useful for the early benefit of NVAF patients. The present study also has some limitations. First, this was a single-center retrospective study with limited sample size; a large multicenter study is required to reduce this bias. Second, considering that this is an observational research, further prospective studies are needed to measure the true value of these indicators.

| CONCLUSION
This study indicates that NT-proBNP, D-dimer, and LAD are reliable biomarkers for detecting the occurrence of IS and can be used as independent predictors of IS in patients with NVAF. NT-proBNP and Abbreviations: AUC, area under the curve; CI, confidence interval; LAD, left atrial diameter; NT-proBNP, N-terminal pro B-type natriuretic peptide. a NT-proBNP combined with D-dimer and LAD.
D-dimer levels combined with LAD had higher sensitivity and specificity in predicting IS in NVAF than the CHA2DS2-VASc score.
The results of this study will help further screen for NVAF patients at a higher risk of IS and help clinicians administer anticoagulation therapy as soon as possible.