ORIGINAL RESEARCH published: 04 July 2022 doi: 10.3389/fcvm.2022.908053 Blood-Based Biomarkers to Search for Atrial Fibrillation in High-Risk Asymptomatic Individuals and Cryptogenic Stroke Patients Elena Palà 1 , Alejandro Bustamante 1,2 , Jorge Pagola 3 , Jesus Juega 3 , Jaume Francisco-Pascual 4,5 , Anna Penalba 1 , Maite Rodriguez 3 , Mercedes De Lera Alfonso 6 , Juan F. Arenillas 6 , Juan Antonio Cabezas 7 , Soledad Pérez-Sánchez 8 , Francisco Moniche 7 , Reyes de Torres 8 , Edited by: Teresa González-Alujas 5,9 , Josep Lluís Clúa-Espuny 10,11 , Juan Ballesta-Ors 11 , Charles Guenancia, Domingo Ribas 12 , Juan Acosta 13 , Alonso Pedrote 13 , Felipe Gonzalez-Loyola 14,15 , University Hospital Dijon, France Delicia Gentile Lorente 11,16 , Miguel Ángel Muñoz 14,15 , Carlos A. Molina 2 and Reviewed by: Joan Montaner 1* Roberto Latini, 1 Mario Negri Pharmacological Neurovascular Research Laboratory, Vall d’Hebron Institute of Research, Hospital Vall d’Hebron, Universitat Autònoma Research Institute (IRCCS), Italy de Barcelona, Barcelona, Spain, 2 Stroke Unit, Hospital Universitari Germans Trias i Pujol, Badalona, Spain, 3 Stroke Unit, Winnie Chua, Medicine Department, Vall d’Hebrón Hospital and Autonomous University of Barcelona, Barcelona, Spain, 4 Arrhythmia University of Birmingham, Unit-Cardiology Department, Vall d’Hebrón Hospital, Barcelona, Spain, 5 Centro de Investigación Biomédica en Red de United Kingdom Enfermedades Cardiovasculares (CIBER-CV), Madrid, Spain, 6 Stroke Unit, University Hospital of Valladolid, Valladolid, Spain, 7 Stroke Unit, University Hospital Virgen del Rocio, Seville, Spain, 8 Stroke Unit, University Hospital Virgen Macarena, Seville, *Correspondence: Spain, 9 Echocardiography Lab Cardiology Department, Vall d’Hebrón Hospital, Barcelona, Spain, 10 Equip d’Atenció Joan Montaner Primària Tortosa Est, SAP Terres de l’Ebre, Institut Català de la Salut, Tortosa, Spain, 11 Institut d’Investigació en Atenció joan.montaner@vhir.org Primària IDIAP Jordi Gol, Ebrictus Group, Barcelona, Spain, 12 EAP Sant Pere i Sant Pau, DAP Camp de Tarragona, Institut Català de la Salut, Tarragona, Spain, 13 Department of Cardiology, Hospital Universitario Virgen del Rocio, Seville, Spain, Specialty section: 14 Gerència Atenció Primària de Barcelona, Institut Català de la Salut, Barcelona, Spain, 15 Institut d’Investigació en Atenció This article was submitted to Primària IDIAP Jordi Gol, Unitat Suport Recerca Barcelona, Barcelona, Spain, 16 Cardiology Department, Hospital Verge Cardiac Rhythmology, de la Cinta, Institut Català de la Salut, Tortosa, Spain a section of the journal Frontiers in Cardiovascular Medicine Received: 30 March 2022 Background: Atrial fibrillation (AF) increases the risk of ischemic stroke in asymptomatic Accepted: 15 June 2022 individuals and may be the underlying cause of many cryptogenic strokes. We aimed to Published: 04 July 2022 test the usefulness of candidate blood-biomarkers related to AF pathophysiology in two Citation: Palà E, Bustamante A, Pagola J, prospective cohorts representative of those populations. Juega J, Francisco-Pascual J, Penalba A, Rodriguez M, Methods: Two hundred seventy-four subjects aged 65–75 years with hypertension De Lera Alfonso M, Arenillas JF, and diabetes from the AFRICAT cohort, and 218 cryptogenic stroke patients aged Cabezas JA, Pérez-Sánchez S, >55 years from the CRYPTO-AF cohort were analyzed. AF was assessed by 4 weeks Moniche F, de Torres R, González-Alujas T, Clúa-Espuny JL, of monitoring with a wearable Holter device (NuuboTMTM ). Blood was collected Ballesta-Ors J, Ribas D, Acosta J, immediately before monitoring started. 10 candidate biomarkers were measured by Pedrote A, Gonzalez-Loyola F, Gentile Lorente D, Ángel Muñoz M, automated immunoassays (Roche, Penzberg) in the plasma of all patients. Univariate Molina CA and Montaner J (2022) and logistic regression analyses were performed in each cohort separately. Blood-Based Biomarkers to Search for Atrial Fibrillation in High-Risk Results: Atrial fibrillation detection rate was 12.4% (AFRICAT cohort) and 22.9% Asymptomatic Individuals (CRYPTO-AF cohort). 4 biomarkers were significantly increased in asymptomatic and Cryptogenic Stroke Patients. Front. Cardiovasc. Med. 9:908053. individuals with AF [Troponin-T, Angiopoietin-2 (Ang-2), Endocan, and total N-terminal doi: 10.3389/fcvm.2022.908053 pro-B type natriuretic peptide (NT-proBNP)] and 7 biomarkers showed significantly Frontiers in Cardiovascular Medicine | www.frontiersin.org 1 July 2022 | Volume 9 | Article 908053 Palà et al. Biomarkers to Search Atrial Fibrillation higher concentrations in cryptogenic stroke patients with AF detection [growth differentiation factor 15, interleukin 6, Troponin-T, Ang-2, Bone morphogenic protein 10, Dickkopf-related protein 3 (DKK-3), and total NT-proBNP]. The models including Ang-2 and total NT-proBNP [AUC 0.764 (0.665–0.863)], and Ang-2 and DKK-3 [AUC = 0.733 (0.654–0.813)], together with age and sex, showed the best performance to detect AF in high-risk asymptomatic individuals, and in cryptogenic stroke patients, respectively. Conclusion: Blood-biomarkers, in particular, total NT-proBNP, DKK-3, and Ang-2, were associated with AF reflecting two mechanistically different pathways involved in AF pathophysiology (AF stretch and vascular changes). The combination of these biomarkers could be useful in AF screening strategies in the primary care setting and also for searching AF after cryptogenic stroke. Keywords: atrial fibrillation, stroke, biomarkers, screening, cryptogenic stroke INTRODUCTION and in circulation (15, 16). Finally, insulin-like growth factor- binding protein 7 (IGFBP-7) is a marker of myocardial damage Atrial fibrillation (AF) increases the risk of ischemic stroke in that has been independently associated with chronic heart failure asymptomatic individuals and may be the underlying cause of in AF patients (17). many cryptogenic strokes (1). Anticoagulant therapy is the most In the present study, we aimed to test the performance of effective treatment to reduce stroke risk in the presence of AF candidate blood-biomarkers to detect AF in two prospective (2), but this arrhythmia is usually underdiagnosed and, therefore, cohorts: one with asymptomatic high-risk subjects and the other undertreated. Pre- and post-stroke AF searching would increase with cryptogenic stroke patients (Figure 1). AF detection, and subsequently the number of patients that would benefit from primary and secondary stroke prevention treatments (3, 4). MATERIALS AND METHODS Some proteins increase in AF individuals and may be useful as AF biomarkers. Until now, natriuretic peptides [B-type Subjects natriuretic peptide (BNP) and N-terminal pro-B type natriuretic AFRICAT Study peptide (NT-proBNP)] (5, 6) are some of the most promising Asymptomatic individuals between 67 and 75 years old with candidates, given their high expression in the atrium under hypertension and diabetes were included in the AFRICAT pressure (7), and its correlation with atrial enlargement observed cohort, an observational, prospective, multicenter, population- in AF (8). However, the behavior of natriuretic peptides as based screening study for AF (6). Individuals with chronic biomarkers is not perfect and could be complemented by others inflammatory diseases, cancer, or dementia were excluded. AF representative of pathways involved in the AF pathophysiology. diagnosis was assessed by a baseline 12-lead electrocardiogram AF pathophysiology has been described to include atrial and (ECG), 28 days monitoring with a wearable Holter device vascular changes, such as inflammation, myocyte injury, collagen, (NuuboTM ), and/or medical history (MH). From the 359 patients and lipid infiltration. Therefore, interleukin 6 (IL-6), Troponin- included in the cohort, those without blood samples (n = 2), T, growth differentiation factor 15 (GDF-15), and fibroblast and/or short/bad quality Holter registers from the no AF group growth factor 23 (FGF-23), markers of inflammation, myocardial (n = 83) were excluded. In total, 274 individuals were analyzed damage, oxidative stress, and atrial fibrosis, respectively, have in the present study. AF burden was calculated from the Holter been extensively associated with AF occurrence, recurrence, records as minutes being in AF divided by the total minutes and/or AF prognosis (9). of readable records and was expressed as a percentage. Plasma There are other proteins, that althought the literature NT-proBNP measurements were available from all the patients concerning AF is scarcer, are interesting candidates to be further included (6). tested. Endocan (ESM-1) is a marker of inflammation and The AFRICAT study protocol was approved by the clinical endothelial dysfunction proposed as a new biomarker for the research ethics committees of IDIAP Jordi Gol (P15/047) and prediction of stroke risk among patients with AF (10). Bone Hospital Universitari Vall d’Hebron [PR (AG) 133-2015]. All morphogenic protein 10 (BMP-10) is an atrial-specific protein participants signed informed consent before inclusion. The that plays an important role in heart development and predicts study protocol conformed to the ethical guidelines of the 1975 AF recurrence after ablation (11). Angiopoietin-2 (Ang-2) is an Declaration of Helsinki. endothelial growth factor reported to be increased in AF (12– 14). Dickkopf-related protein 3 (DKK-3) is involved in heart CRYPTO-AF Study development and cardiac hypertrophy protection and, in AF Cryptogenic ischemic stroke patients (<72 h after stroke) patients, elevated levels have been found in the atrial appendages over 55 years of age were included in the CRYPTO-AF Frontiers in Cardiovascular Medicine | www.frontiersin.org 2 July 2022 | Volume 9 | Article 908053 Palà et al. Biomarkers to Search Atrial Fibrillation FIGURE 1 | Study design. Patients from two cohorts were included in the present study. In both cohorts, patients were monitored for 28 days to identify paroxysmal AF patients. Biomarkers displayed were measured in all patients. Those underlined had significantly higher concentration in AF patients from the AFRICAT cohort, while the ones in bold had significantly higher concentration in AF patients from the CRYPTOFA cohort. GDF-15 indicates growth differentiation factor 15; IL-6, interleukin 6; Ang-2, angiopoietin 2; BMP-10, bone morphogenic protein 10; DKK-3, dickkopf-related protein 3; ESM-1, endocan or endothelial cell specific molecule-1; FGF-23, fibroblast growth factor 23; IGFBP-7, insulin-like growth factor-binding protein 7; and NT-proBNP, N-terminal pro-brain natriuretic peptide. cohort (18). All patients underwent complete etiological workup After centrifugation at 1,500 g and 4◦ C for 15 min, plasma before the cryptogenic stroke classification (ECG, Doppler aliquots were frozen at –80◦ C until biomarker determination. ultrasound study of the extracranial and intracranial arteries, Biomarkers were centrally measured and blinded to clinical data routine transthoracic echocardiography, and in-hospital ECG (Roche Diagnostics, Penzberg, Germany). GDF-15, IL-6, and continuous automatic monitoring). Patients with known causes Troponin-T were measured with commercial in vitro diagnostic of stroke, previous AF, pacemaker carriers, moderate or severe immunoassays (Roche Diagnostics). Angiopoietin-2, BMP-10, disability after stroke (modified Rankin Scale score > 3), DKK-3, ESM-1, FGF-23, IGFBP-7, and total NT-proBNP and those with clinical worsening were excluded. Patients (including both glycosylated and non-glycosylated forms) were were monitored for 28 days to detect AF with a wearable quantified using a research grade prototype assay on a Cobas Holter device (NuuboTM ) starting within the first 72 h from Elecsys e601 platform (Roche Diagnostics GmbH, Penzberg, stroke symptoms. From the 296 patients included in the Germany) employing Elecsys electrochemiluminescence cohort, 264 completed the Holter monitoring period and technology. More details on these biomarkers are provided in the 218 had available blood samples which were analyzed in the Supplementary Table 1. present study. The number of AF episodes and the longest AF episode in the Holter register were used as AF burden Statistical Analysis measures in this cohort. Statistical analysis was conducted with SPSS version 21 and R Left atrial volume adjusted to body surface index (LAVI, version 3.6.3. Graphs were elaborated with GraphPad Prism 5. ml/m2 ) was measured by biplane transthoracic echocardiography Kolmogorov–Smirnov test was used to assess the normality of in a subset of 152 patients and peak atrial longitudinal strain the data. Data were expressed as number (%) for categorical was evaluated by speckle tracking software (GE EchoPAC ) in R variables and as mean ± SD or median (interquartile range) 131 patients. Plasma NT-proBNP and BNP measurements were for continuous variables, depending on the data distribution. available from all the patients included (5). Association between biomarkers and AF diagnosis was analyzed Written informed consent was obtained from all participants first by univariate analysis. Mann–Whitney U-test or Student’s and the study was approved [PR (AG)49/2014] by the Ethical t-test were used for continuous variables, and the χ2 test was Committee of Vall d’Hebrón Hospital, Valladolid Hospital, used for categorical variables. ANOVA or the Kruskal–Wallis test Virgen Macarena Hospital and Virgen del Rocio Hospital, in line were used to compare > 2 variables depending on the variable with Helsinki guidelines. distribution. The Spearman test was used for correlations. Samples below the limit of detection were substituted by Biomarker Quantification the lowest value detected minus 0.5 for the analysis. Similarly, Blood was collected into EDTA tubes at the time of inclusion samples above the limit of detection were substituted by the of each study In the CRYPTO-FA study, blood was drawn post- highest value detected plus 0.5. Outliers were defined as samples stroke (<72 h). In the AFRICAT study blood was drawn in the with Z-scores of ±3. Outliers were included in the analysis but baseline visit together with the performance of the baseline 12- sensitivity analyses were performed excluding them. lead ECG. In both studies blood was collected before the AF The rms package was used to fit penalized logistic regression monitoring with the Holter device started. models using penalized maximum likelihood estimation. The Frontiers in Cardiovascular Medicine | www.frontiersin.org 3 July 2022 | Volume 9 | Article 908053 Palà et al. Biomarkers to Search Atrial Fibrillation penalty factor applied in each model was selected to minimize them showed significant differences when comparing paroxysmal the Hurvich and Tsai’s corrected AIC of the resulting model. AF cases only detected during the monitoring period (MH-ECG- Backward stepwise using the AIC criteria, and forcing the Holter+) vs no AF cases. Several proteins showed increased levels inclusion of sex and age as covariates, was used to select the in the AF groups with “established AF” (MH + ECG + or MH- biomarkers to include in the model from all the biomarkers ECG + Holter+; Supplementary Figure 1). In this cohort, AF analyzed. Multicollinearity among the predictor variables has burden information was available from 27 patients. The median been checked with the variance inflation factor. In order to assess AF burden was 3% (0.6–14%) and it correlated with FGF-23 the stability of the variable selection process, backward stepwise (r = 0.394, p = 0.042), and Total NT-proBNP (r = 0.525, p = 0.005; was applied to 1,000 bootstrap iterations and the most frequent Table 2). combinations were shown. In the CRYPTO-AF cohort, AF was detected in 50 subjects. The area under the receiving operator curve (AUC) of the No patient had AF when blood was collected. Cryptogenic stroke constructed models was calculated and compared with a basic patients with AF had higher levels of GDF-15, IL-6, Troponin- clinical model including age and sex using the DeLong test. The T, Ang-2, BMP-10, DKK-3, and total NT-proBNP, in comparison integrated discrimination improvement index (IDI) was used to to those without AF (Table 1 and Figure 2). In this cohort, compare their predictive capacity. Also, other cardiovascular risk information about AF duration was available for 38 patients. factors available in each cohort were included in the models in Median number of AF episodes was 10 (3–34.25) and median order to avoid influence of comorbidities. longest AF episode was 626.12 min (130.24–1,409.07 min). The The PanelomiX algorithm (19) was used to select thresholds number of episodes correlated with the level of Ang-2, while for the biomarkers identified. PanelomiX uses the iterative the duration of the longest episode also correlated with GDF- combination of biomarkers and thresholds to obtain 15, Troponin-T, IL-6, and total NT-proBNP (Table 2). The the combination that provides the optimal classification correlations between biomarkers and left atrial function were performance. Two optimization options were used: optimizing reported in Table 2. global accuracy and optimizing the specificity at ≥90% Total NT-proBNP, Ang-2, and Troponin-T were increased in sensitivity. Again, logistic regression models were created both cohorts (Table 1 and Figure 2). When comparing the two including the panels selected by PanelomiX adjusted by age cohorts, some biomarkers presented differential concentration: and sex (including those variables into the model together IL-6, DKK-3, ESM-1, Troponin-T, and total NT-proBNP were with the panel). higher in the CRYPTO-AF cohort compared to AFRICAT The analyses for the selection of the best biomarker cohort, while GDF-15, FGF-23, and IGFBP-7 presented lower combinations were carried out separately and independently in concentrations. These differences were only significant between each of the cohorts. the AF group for DKK-3 and between the no AF group for The performance of the best biomarker combination was GDF-15, Troponin-T, and total NT-proBNP (Figure 2). As the compared to the performance of biomarkers previously measured main difference between the two cohorts, apart from the stroke in the tested cohorts: NT-proBNP (AFRICAT and CRYPTO-AF) itsef was the presence of hypertension and diabetes in all the and BNP [which had higher accuracy than NT-proBNP in the patients included in AFRICAT cohort, the association of these CRYPTO-AF cohort (5)]. pathologies with all the biomarkers was tested in the CRYPTO- AF cohort. GDF-15 (p < 0.001), Troponin-T (p = 0.002), FGF-23 (p = 0.006), and IGFBP-7 (p = 0.018) abundance was higher in RESULTS patients with hypertension. Moreover, patients with diabetes had higher concentrations of GDF-15 (p = 0.017) and Troponin-T Baseline characteristics of each cohort and biomarker results were (p = 0.013), and lower concentrations of ESM-1 (p = 0.048). reported in Table 1. Some biomarkers presented significant correlations with other In the AFRICAT cohort, AF was detected in 34 subjects. biomarkers and age (Supplementary Figure 2). AF cases were classified according to how AF diagnosis was Sensitivity analyses were conducted excluding outliers and performed in 4 groups, defined by medical history (MH) for results were similar (Supplementary Table 1). The only AF, ECG findings, and Holter AF detection. According to this remarkable differences were in the univariate analysis of the classification, 17 individuals were only diagnosed during the CRYPTO-AF cohort, in which IGFBP7 was significant (p = 0.030) Holter monitoring period (MH-ECG-Holter+), 7 individuals had and IL-6 was no longer significant (p = 0.063). a new-onset AF on baseline ECG confirmed during the Holter register (MH-ECG + Holter+), 7 individuals had previous MH of AF, confirmed by ECG in the baseline visit (MH + ECG+), Biomarker Combinations and finally 3 patients had an external diagnosis of AF, previous Asymptomatic High-Risk Subjects to the study or during the study, not detected by the ECG Angiopoietin-2 and Total NT-proBNP were selected or the Holter in the present study (MH + ECG-Holter-). to be included as explanatory variables in the logistic According to this classification, 14 patients had ongoing AF when regression model by backward selection. The model blood was collected. Among the analyzed molecules, 4 proteins including both biomarkers together with age and sex had showed significant differences when comparing asymptomatic an AUC of 0.764 (0.665–0.863), which was better than the individuals with and without AF diagnosis (Troponin-T, Ang-2, performance of Ang-2 alone [AUC = 0.741 (0.640–0.843)] ESM-1, and total NT-proBNP; Table 1 and Figure 2). None of (DeLong = 0.044), but similar to the one of total NT-proBNP Frontiers in Cardiovascular Medicine | www.frontiersin.org 4 July 2022 | Volume 9 | Article 908053 Palà et al. Biomarkers to Search Atrial Fibrillation TABLE 1 | AF univariate analysis of the two included cohorts. AFRICAT CRYPTO-AF AF No AF P-value AF No AF P-value (n = 34) (n = 240) (n = 50) (n = 168) Age 70 (67.5–74) 71 (68.25–73) 0.351 78 (73–83) 73 (67–81) 0.005 Sex (%female) 14 (41.2%) 117 (48.8%) 0.408 25 (50%) 83 (49.7%) 0.970 Hypertension 34 (100%) 240 (100%) 1.00 39 (78%) 128 (77.1%) 0.895 Diabetes 34 (100%) 240 (100%) 1.00 11 (24.4%) 37 (23.4%) 0.886 Ischemic cardiopathy 9 (26.5%) 43 (17.9%) 0.234 3 (6%) 11 (6.5%) 1.000 Heart failure* 3 (8.8%) 13 (5.4%) 0.431 1 (2.1%) 3 (2.5%) 1.000 LAVI – – – 31 (28–39) 27 (23–32.8) 0.002 LAS – – – 21.33 ± 19.69 28.64 ± 9.83 <0.001 BNP (pg/ml) – – – 90.70 33.65 <0.001 (49.87–162.30) (14.85–81.87) NT-proBNP (pg/ml) 380.60 112.70 (61.10–206.15) <0.001 425.95 215.90 <0.001 (123.97–1,285.50) (226–816.87) (99.09–463.85) GDF-15 (pg/ml) 2,367.00 2,491 0.375 1,896.5 1,645.5 0.019 (1,329.25–3,198.75) (1,633.5–3,583.00) (1,521.5–2,659.5) (1,169.5–2,356.5) IL-6 (pg/ml) 3.64 (2.07–4.91) 3.15 (1.87–4.89) 0.334 11.19 7.28 0.042 (4.64–20.77) (2.84–15.28) TroponinT (pg/ml) 16.89 (11.77–23.88) 12.88 (9.73–18.85) 0.011 17.19 15.11 0.002 (14.27–28.09) (10.26–20.95) Ang-2 (ng/ml) 2.46 1.67 <0.001 2.43 1.68 <0.001 (1.76–4.11) (1.37–2.14) (1.60–3.68) (1.32–2.31) BMP-10 (ng/ml) 2.24 2.06 0.079 2.26 2.05 0.013 (1.84–2.68) (1.86–2.31) (1.92–2.53) (1.79–2.33) DKK3 (ng/ml) 55.84 54.33 0.413 62.78 55.61 0.001 (51.05–66.29) (47.00–66.11) (55.52–79.56) (49.37–65.10) ESM-1 (pg/ml) 2,087.75 1,815.55 0.012 2,750.00 2,268.80 0.101 (1,831.47–2,460.95) (1,482.70–2,297.15) (2,025.17–3,280.00) (1,846.80–3,226.90) FGF-23 (pg/ml) 167.73 148.80 0.071 134.04 134.74 0.583 (129.75–257.84) (118.39–206.60) (110.00–204.46) (102.11–185.65) IGFBP-7 (ng/ml) 115.66 (99.83–132.21) 107.53 0.088 104.42 94.68 0.052 (94.91–128.29) (85.50–116.84) (82.11–109.56) Total NT-proBNP (pg/ml) 1,821.35 650.93 <0.001 2,123.50 906.18 <0.001 (749.58–4,434.00) (376.97–1,159.77) (1,241.72–3,555.57) (419.09–1,887.40) *Heart failure is considered when LVEF < 40% in th CRYPTO-AF study. LAVI indicates left atrial volume index; LAS, left atrial strain. alone [AUC = 0.771 (0.683–0.860), p < 0.001] (DeLong = 0.731; a sensitivity of 91.2% and a specificity of 47.5% (AUC = 0.7539). Figure 3). The inclusion of the biomarkers as panels to the logistic The model including the biomarkers showed a higher accuracy regression models resulted in an AUC of 0.778 (0.696– in terms of area under the curve than the clinical model [AUC 0.858) and 0.740 (0.657–0.823), respectively, (Figure 3 and 0.559 (0.445–0.673)] (Figure 3 and Table 3). Similar results Supplementary Table 3). Multicollinearity was not detected in were obtained when we adjusted by other cardiovascular factors any of the models. (Supplementary Table 3). After 1,000 boostrap iterations the combination of Total NT- Comparison to Biomarkers Previously Measured in the proBNP and Ang-2 was selected in 20.5% of the times. Other Cohort (NT-ProBNP) frequent combinations included are shown in Figure 4. In the AFRICAT cohort, NT-proBNP (Roche Diagnostics) had A panel including Ang-2 > 1.73 ng/ml and Total NT- been previously measured (6). The pre-commercial total NT- proBNP > 665.86 pg/mL was selected by the PanelomiX proBNP assay, measuring all, glycosylated and non-glycosylated, software as the most accurate (AUC = 0.7331) for the detection forms of NT-proBNP had similar AUC [0.775 (0.69–0.86)] of AF in asymptomatic patients. The panel was positive in comparison to conventional NT-proBNP [0.771 (0.69– when the two markers were above the cut-off (sensitivity, 86)] in the present cohort (De Long = 0.807), although 76.5%; specificity, 70.8%). Alternatively, a panel including Ang- conventional NT-proBNP had a better discriminatory capacity 2 > 2.613 ng/ml and Total NT-proBNP > 632.65 pg/mL, [IDI = 4.78%[1.40–8.17%)]. Both biomarkers presented a strong positive when one of the biomarkers was above the cut-off had correlation (r = 0.944, p < 0.001). Frontiers in Cardiovascular Medicine | www.frontiersin.org 5 July 2022 | Volume 9 | Article 908053 Palà et al. Biomarkers to Search Atrial Fibrillation FIGURE 2 | Boxplot distribution of the biomarker circulating levels in the CRYPTO-AF and the AFRICAT cohort. Boxes extend from the 25th to 75th percentiles. The line in the middle is plotted as the median. Whiskers are drawn according to Tukey methodology (±1.5 IQR) and larger values are plotted as individual points. Significant comparisons are indicated with an asterisk. TABLE 2 | Correlations between biomarkers and measures of AF burden and left atrial function. AFRICAT CRYPTO-AF AF burden (%) Number of AF episodes Longest AF episode LAVI LAS (n = 27) (n = 38) (minutes; (left atrial volumen index; (left atrial strain; n = 38) n = 152) n = 131) GDF-15(pg/ml) r = 0.310, p = 0.115 r = 0.137, p = 0.412 r = 0.327, p = 0.045 r = 0.185, p = 0.023 r = –0.186, p = 0.034 IL-6(pg/ml) r = –0.136, p = 0.498 r = 0.105, p = 0.529 r = 0.379, p = 0.019 r = 0.145, p = 0.074 r = –0.362, p < 0.001 TroponinT (pg/ml) r = 0.359, p = 0.072 r = 0.054, p = 0.430 r = 0.344, p = 0.034 r = 0.245, p = 0.002 r = –0.322, p < 0.001 Ang-2(ng/ml) r = 0.244, p = 0.220 r = 0.395, p = 0.014 r = 0.518, p = 0.001 r = 0.159, p = 0.051 r = –0.302, p < 0.001 BMP-10(ng/ml) r = 0.362, p = 0.064 r = 0.096, p = 0.565 r = 0.100, p = 0.550 r = 0.174, p = 0.033 r = –0.275, p = 0.002 DKK3(ng/ml) r = 0.271, p = 0.171 r = 0.289, p = 0.078 r = 0.309, p = 0.059 r = 0.253, p = 0.002 r = –0.295, p = 0.001 ESM-1(pg/ml) r = –0.081, p = 0.691 r = 0.185, p = 0.267 r = 0.181, p = 0.276 r = 0.227, p = 0.005 r = –0.221, p = 0.012 FGF-23(pg/ml) r = 0.394, p = 0.042 r = 0.088, p = 0.597 r = 0.146, p = 0.381 r = –0.003, p = 0.967 r = –0.061, p = 0.490 IGFBP-7(ng/ml) r = 0.315, p = 0.110 r = 0.122, p = 0.464 r = 0.071, p = 0.674 r = 0.139, p = 0.089 r = –0.134, p = 0.127 Total NT-proBNP(pg/ml) r = 0.525, p = 0.005 r = 0.149, p = 0.373 r = 0.376, p = 0.020 r = 0.417, p < 0.001 r = –0.415, p < 0.001 Significant correlations are represented in bold. If conventional NT-proBNP was also included in the backward of 0.733 (0.654–0.813), which is better than the performance of stepwise model, the combination of Ang-2 and conventional NT- DKK-3 alone [AUC = 0.677 (0.595–0.757)] (DeLong = 0.0.029), proBNP would be selected as the best combination. In fact, the but similar to the performance of Ang-2 alone [AUC = 0.719 accuracy of the model including these biomarkers together with (0.6374–0.802)] (DeLong = 0.462; Figure 3). age and sex was similar [AUC = 0.757 (0.655–0.859)] to the one The model including the biomarkers showed a higher accuracy obtained with the combination of Total NT-proBNP and Ang- in terms of area under the curve than the clinical model [AUC 2 (DeLong = 0.444), but the discrimination capacity was better 0.631 (0.550–0.712)] (Figure 3 and Table 3). Similar results [IDI = 1.96% (0.15–3.77%)]. were obtained when we adjusted by other cardiovascular factors (Supplementary Table 3). Cryptogenic Stroke Patients After 1,000 boostrap iterations the combination of Ang-2 and DKK3 were selected by backward selection to be Total NT-proBNP and Ang-2 was selected in 39.1% of included as explanatory variables of post-stroke AF. The model the times. Other frequent combinations included Total including both biomarkers together with age and sex had an AUC Nt-proBNP (Figure 4). Frontiers in Cardiovascular Medicine | www.frontiersin.org 6 July 2022 | Volume 9 | Article 908053 Palà et al. Biomarkers to Search Atrial Fibrillation FIGURE 3 | Receiver operating characteristic (ROC) curves of the constructed models for the two cohorts. (A, B) Performance of the models constructed for the AFRICAT cohort. (C, D) Performance of the models constructed for the CRYPTO-AF cohort. (A, C) ROC curves including Ang-2,Total NT-proBNP, or DKK-3 alone, and combined as continuous variables for each cohort. (B, D) ROC curves for the clinical model alone (Age + Sex) and in combination with the biomarkers as continuous variables or dichotomized according to the panel with the best accuracy. TABLE 3 | Logistic regression analyses and additional predictive value of blood biomarkers as continuous variables in the AFRICAT and the CRYPTO-AF cohort. AFRICAT CRYPTO-AF Clinical model (Age + Sex) Clinical model (Age + Sex) Clinical model (Age + Sex) Clinical model (Age + Sex) +Biomarkers_continuous +Biomarkers_continuous Logistic regression, OR (95% CI) Age 0.982 (0.929–1.037), p = 0.517 0.933 (0.837–1.041), p = 0.221 1.044 (1.010–1.080), p = 0.010 1.036 (0.999–1.073), p = 0.056 Sex 0.962 (0.739–1.252), p = 0.773 0.667 (0.342–1.301), p = 0.235 0.955 (0.574–1.590), p = 0.861 0.977 (0.573–1.666), p = 0.931 Ang2 (ng/ml) – 1.635 (1.216–2.198), p = 0.001 – 1.451 (1.160–1.814), p = 0.001 TotalNTproBNP (pg/ml) – 1.000 (1.000–1.000), p = 0.030 – – DKK-3 (ng/ml) – – – 1.016 (0.997–1.036), p = 0.087 IDI statistics Total IDI (95% CI) 16.8% (8.32–25.4%) 9.87% (4.95–14.79%) P-value 0.0001 0.000084 ROC curve AUC 0.559 (0.445–0.673) 0.764 (0.665–0.863) 0.631 (0.550–0.712) 0.733 (0.654–0.813) DeLongTest p = 0.0012 p = 0.0163 De Long test compared the performance of the clinical model with the biomarkers, and the clinical model alone in each cohort. AUC indicates area under the curve; OR, odds ratio; and ROC, receiver operator characteristic. Frontiers in Cardiovascular Medicine | www.frontiersin.org 7 July 2022 | Volume 9 | Article 908053 Palà et al. Biomarkers to Search Atrial Fibrillation FIGURE 4 | Five most frequent biomarker combinations after 1,000 stepwise logistic regression boostrap iterations. X axis represents de percentage of boostrap iterations in which the combination was selected. In total, 150 combinations were selected in the AFRICAT cohort, and 72 in the CRYPTO-AF cohort. A panel including Ang-2 > 1.52 ng/ml and DKK- have been measured in blood samples from 492 patients of two 3 > 54.07 ng/mL was selected by the PanelomiX software cohorts. The two selected study populations included (high-risk as the most accurate (AUC = 0.695) for the diagnosis of AF after asymptomatic individuals and cryptogenic stroke patients) would stroke. The panel was positive when the two markers were above benefit from AF diagnosis markers. the cut-off (sensitivity, 76%; specificity, 64.1%). Alternatively, a As a result, four proteins (Troponin-T, Ang-2, ESM-1, and panel including Ang-2 > 1.41 ng/ml and DKK-3 > 43.52 ng/mL, Total NT-proBNP) showed increased levels in the asymptomatic also positive when both biomarkers were above the cut-off, had population with AF. Also, we must highlight that the tendency a sensitivity of 90% and specificity of 40.1% (AUC = 0.647). The for most of the markers measured was to increase in those addition of the biomarker panels to the logistic regression models groups where “AF burden” was higher (i.e., AF diagnosed by resulted in an AUC of 0.738 (0.662–0.813) and 0.717 (0.643– several methodologies), while no biomarker was able to detect 0.792), respectively, (Figure 3 and Supplementary Table 3). paroxysmal AF in our cohort. Complementary to the biomarker Multicollinearity was not detected in any of the models. abundance distribution among groups, correlations between AF burden and biomarkers were calculated, but only two biomarkers Comparison to Biomarkers Previously Measured in the showed significant correlations (total NT-proBNP and FGF-23). Cohort (BNP and NT-ProBNP) However, we should highlight that the lack of registers of some In the CRYPTO-AF cohort, NT-proBNP (Roche Diagnostics) of the patients of the MH + ECG + group, which would have and BNP (Siemens Healthcare) had been previously probably presented the highest AF burden measures, and in measured (5). which monitoring was optional, reduced the statistical power of The pre-commercial total NT-proBNP assay had higher AUC these correlations. These results are supported by other studies [0.718 (0.64–0.79)] than conventional NT-proBNP [0.668 (0.589– in which biomarkers like NT-proBNP presented higher values 0.747)] (DeLong = 0.002; IDI = 1.67% [0.509–2.831%]) but according to the AF burden (20). We may hypothesize that the similar to BNP [0.722 (0.645–0.799); DeLong = 0.82; IDI = – biomarker profiles observed reflected a more “advanced” stage 0.044% (–1.64-1.55%)] in the present cohort. Conventional NT- of the disease, with more pathophysiological changes, such as proBNP (r = 0.939, p < 0.001) and BNP (r = 0.839, p < 0.001) atrial stretch and dilatation (associated with NT-proBNP release), presented a strong correlation with the pre-commercial assay. taking place in patients with established AF, in comparison to If BNP and conventional NT-proBNP were also included in the patients with paroxysmal AF. According to a published meta- backward stepwise model, the combination of DKK-3 and Ang-2 analysis, stroke risk is lower in paroxysmal AF, but still higher would be selected as the best combination. than in patients with sinus rhythm (9). Therefore, the difficulty of the proteins tested to identify paroxysmal cases reduces its DISCUSSION potential as screening biomarkers in asymptomatic populations. Conversely, almost all the proteins tested presented significantly In the present study, 10 candidate AF blood-biomarkers of elevated levels in patients with AF detected after a cryptogenic various biological pathways described for AF pathophysiology stroke, underlining the usefulness of biomarkers in this setting. Frontiers in Cardiovascular Medicine | www.frontiersin.org 8 July 2022 | Volume 9 | Article 908053 Palà et al. Biomarkers to Search Atrial Fibrillation Also, similarly to what we found in asymptomatic individuals, both biomarkers should be tested in other studies to ensure NT-proBNP, together with other biomarkers, such as Ang-2, their additive results. Moreover, the presented cut-offs, selected correlated with AF burden measures. automatically by the PanelomiX algorithm might serve as a In the AFRICAT cohort the combination of Ang-2 and total reference for replication purposes in further studies. In the NT-proBNP showed the best results, while in the CRYPTOFA present study we only explored cut-offs optimizing accuracy and cohort the combination of Ang-2 with DKK3 was the one sensitivity. The combination of both biomarkers, if confirmed selected, but also the combination of Ang-2 and Total NT- in future studies, might be used in clinical practice to screen proBNP was between the most frequent combinations after a for AF and a high sensitivity is aimed for screening purposes. bootstrapping strategy. These biomarkers represented different Then, in those cases with elevated markers, intensive monitoring pathways affected by AF. Based on its biology, NT-proBNP and periods, and anticoagulation, in case AF is confirmed, would DKK3 showed atrial changes whileAng-2 is a marker of vascular be recommended as primary or secondary stroke prevention damage, suggesting that both atrial and vascular changes are strategies. Also, we must highlight that all the biomarkers present in AF pathophysiology. On the one hand, total NT- were measured by automatic immunoassays which makes the proBNP is released by the cardiomyocytes in response to pressure translation into the clinical practice easier. or volume overload indicating cardiac dysfunction (21). NT- According to our results, total NT-proBNP correlated with proBNP had been previously presented as one of the most other natriuretic peptides and their performance was similar. promising AF biomarkers (9, 22), and even validated in the two Also, when combined with Ang-2 and basic clinical variables, the cohorts presented here (5, 6). Interestingly, this molecule has natriuretic peptides previously tested presented similar results. nine known O-glycosylation sites, some of them located in the One of the strengths of the current study is that we measured central region, where most of the commercial immunoassays the same proteins in two cohorts electrocardiographically are directed, making the molecule almost “invisible” when monitored during 1 month, with the main difference that one glycosylated. Consequently, most assays only detect a subfraction presented a previous stroke and the other did not. Therefore, of endogenous NT-proBNP (23) but in the present study, the biomarker changes between cohorts may be influenced by the total NT-proBNP assay detected both glycosylated and non- index stroke. In line with our results, IL-6, had been described glycosylated NT-proBNP forms. On the other hand, DKK3, a to increase after stroke as part of the post-stroke inflammatory member of the dickkopf family, is a cardiac-enriched protein response with a maximum increase on day 3 (27). Similarly, the that plays an important role in heart development and presents level of NT-proBNP was known to increase after stroke within cardioprotective functions as a negative regulator of pressure the first 2 days (28). In our study, this biomarker only showed overload-induced cardiac hypertrophy (24). In AF, increased differences between cohorts in the group without AF, and we DKK-3 levels have been previously reported, in atrial and blood may hypothesize that the effect of stroke on NT-proBNP is not so samples (15, 25). important in AF individuals as the biomarker is already increased Next to such atrial changes (reflected by NT-proBNP, or in those patients. Also, previous results showed an increase DKK3), Ang-2, a marker of endothelial dysfunction, suggests that in ESM-1 after stroke, although, in contrast with our study, there is a pronounced vascular remodeling in AF patients, which the biomarker was tested in large-artery atherosclerotic stroke could add a further causality to the increased thromboembolic patients (29). Moreover, differences between cohorts may be due risk in AF next to the blood stasis in the atrium generally to their baseline differences. Essentially, all the AFRICAT patients described as major stroke risk. The contribution of endothelial had hypertension and diabetes whereas the percentage of these injury to the stroke risk is further substantiated by the pathologies was lower in the CRYPTO-AF cohort. Therefore, observation that restoration of the rhythm by ablation or biomarkers with higher levels in the first study may be associated antiarrhythmic drugs is generally not normalizing the stroke with these pathologies. Our results and previous publications risk (26). From the few studies that had explored Ang-2 in AF supported this hypothesis for GDF-15, FGF-23, and IGFBP-7 individuals, Lip et al. found significantly higher levels of this (30–32). Finally, another explanation is that patients with AF protein together with other angiogenic factors, and platelets in detected after stroke may have an increased thromboembolic AF patients compared to healthy controls, hypothesizing the role risk and/or a higher degree of atrial substrate underlying AF, of this protein in the AF prothrombotic state (13, 14). Ang-2 had affecting the expression and release of several proteins. This also been recently associated with the risk of incident AF in the hypothesis could explain the rise of biomarker levels in the Framingham cohort (12). Also, our results complement the ones AF group of the CRYPTO-AF cohort in comparison to the of the study of Staszewsky et al. (25), in which a panel with similar same group of the AFRICAT cohort and could be the reason biomarkers was tested in patients with history of AF and total NT- why more biomarkers were increased in AF in this cohort in proBNP and Ang-2 were significantly associated with ongoing comparison to the general population. In particular, this could AF together with BMP-10. In their study only total NT-proBNP explain why DKK-3 was associated with AF in the CRYPTOFA predicted recurrent AF. cohort, while this biomarker did not show differences in the Although in our study the combination of two biomarkers AFRICAT cohort. Furthermore, and supporting this hypothesis, (Ang-2 and Total NT-proBNP, or Ang-2 and DKK-3) optimized most of the biomarkers tested correlate either with LAS or with the AIC of the model and was selected by a backward stepwise LAVI, measures of enlargement, and deformability of the atria, strategy, when we compared their combined performance with that evaluate the atrial substrate underlying AF (33). All the the DeLong test vs the performance of each biomarker alone, differences explained and their possible influence in biomarker we obtained similar results. Therefore, the combination of levels prevented a common analysis pooling both cohorts. Also, Frontiers in Cardiovascular Medicine | www.frontiersin.org 9 July 2022 | Volume 9 | Article 908053 Palà et al. Biomarkers to Search Atrial Fibrillation the AF detection rate was higher in the CRYPO-AF cohort DATA AVAILABILITY STATEMENT (22.9%) than the AFRICAT cohort (12.4%). This is in line with previous studies in which prevalence rates of AF were higher The raw data supporting the conclusions of this article will be in studies including patients after cryptogenic strokes, than in made available by the authors, without undue reservation. studies including primary care patients (3, 34). Samples in both studies were taken in a single time-point, just before the monitoring period started. However, longitudinal ETHICS STATEMENT changes in these markers may be useful to monitor the degree of atrial substrate and search for AF (34). Therefore, the study of This study was reviewed and approved by Comité de ética serial samples should be explored in future studies as well their de investigación con medicamentos y comisión de proyectos implementation into AI-based personalized prevention strategies de investigación del hospital universitari Vall d’Hebrón and medical decisions. [PR(AG)459/2020]. The patients/participants provided their The present study has several limitations that should be written informed consent to participate in each respective study mentioned. First, patients were only monitored for 28 days. (AFRICAT and CRYPTO-AF). Longer monitoring periods would probably have detected a higher amount of AF patients (35), which were classified in the no AF group adding background noise to the analysis. AUTHOR CONTRIBUTIONS Second, as two different cohorts were included, some registered variables and protocols slightly differed. This is the case of EP: samples handling, statistical analysis of the data, and drafted AF burden measurement, which in the AFRICAT cohort was the manuscript. AB, JP, JJ, MR, MD, JFA, JC, SP-S, FM, RT, calculated as a percentage and in the CRYPTO-AF cohort as JC-E, JB-O, DR, FG-L, MÁ, and CM: patient recruitment. JF-P, the number of episodes and the duration of the longest one, JA, DG, and AlP: Holter-ECG readings. TG-A: Ecocardiographic making the comparison difficult. Another limitation was the lack measurements. AnP: samples handling. JM and CM: obtained of echocardiographic information in the AFRICAT cohort. Also, funding and developed the protocol for the studies included in we did not correct the models with other variables beyond age the present manuscript. All authors have critically reviewed the and sex as these were the only variables measured in both cohorts manuscript and approved the final version. that did not present missing information. As a result, the AUC of the models presented was moderate (between 0.7 and 0.8) and the addition of other clinical variables or biomarkers may improve FUNDING their performance. Another limitation is that biomarkers were selected from the literature and we can not discard that other This work was supported by the Instituto de Salud Carlos III non-tested proteins could perform better. Finally, the number of (PI15/02265, PI18/00804, and RD16/0019/00021); la Fundació AF patients was limited and some associations may have been Marató de TV3 in the research call “La Marató 2014: missed. Validation studies in larger cohorts are needed. malalties del cor” (grant number: 201528-30-31-3); the European Union’s Horizon 2020 research and innovation program (grant agreements N◦ 847770 and N◦ 754517); and the Vall d’Hebrón CONCLUSION Institute of Research (predoctoral grant to EP). Blood-biomarkers, in particular, total NT-proBNP, DKK- 3, and Ang-2, were increased in AF patients and reflect SUPPLEMENTARY MATERIAL two mechanistically different pathways involved in AF pathophysiology (AF stretch and vascular changes). They could The Supplementary Material for this article can be found be useful in AF screening strategies in the primary care setting online at: https://www.frontiersin.org/articles/10.3389/fcvm. and also for searching AF after cryptogenic stroke. 2022.908053/full#supplementary-material REFERENCES 4. Schnabel RB, Haeusler KG, Healey JS, Freedman B, Boriani G, Brachmann J, et al. Searching for atrial fibrillation poststroke: a white paper of the 1. Virani SS, Alonso A, Benjamin EJ, Bittencourt MS, Callaway CW, Carson AF-SCREEN international collaboration. Circulation. (2019) 140:1834–50. AP, et al. Heart disease and stroke statistics—2020 update: a report from the 5. Palà E, Pagola J, Juega J, Francisco-Pascual J, Bustamante A, Penalba A, et al. American heart association. Circulation. (2020) 141:139–596. doi: 10.1161/ BNP over NT-proBNP to predict incident atrial fibrillation after cryptogenic CIR.0000000000000757 stroke. Eur J Neurol. (2020) ene:14579. doi: 10.1111/ene.14579 2. Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic therapy to 6. Palà E, Bustamante A, Clúa-Espuny JL, Acosta J, Gonzalez-Loyola F, prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Ballesta-Ors J, et al. N-terminal pro B-type natriuretic peptide’s usefulness Med. (2007) 146:857–67. doi: 10.7326/0003-4819-146-12-200706190-00007 for paroxysmal atrial fibrillation detection among populations carrying 3. Freedman B. Screening for atrial fibrillation a report of the AF-SCREEN cardiovascular risk factors. Front Neurol. (2019) 10:1126. doi: 10.3389/fneur. international collaboration. Circulation. (2017) 135:1851–67. 2019.01226 Frontiers in Cardiovascular Medicine | www.frontiersin.org 10 July 2022 | Volume 9 | Article 908053 Palà et al. Biomarkers to Search Atrial Fibrillation 7. Doyama K, Fukumoto M, Takemura G, Tanaka M, Oda T, Hasegawa K, 26. Pencina MJ, D’Agostino RB Sr., D’Agostino RB Jr., Vasan RS. Evaluating the et al. Expression and distribution of brain natriuretic peptide in human right added predictive ability of a new marker: from area under the ROC curve atria. J Am Coll Cardiol. (1998) 32:1832–8. doi: 10.1016/S0735-1097(98)0 to reclassification and beyond. Stat Med. (2008) 27:157–72. doi: 10.1002/sim. 0494-X 2929 8. Bossard M, Kreuzmann R, Hochgruber T, Krisai P, Zimmermann AJ, 27. Perini F, Morra M, Alecci M, Galloni E, Marchi M, Toso V. Temporal Aeschbacher S, et al. Determinants of left atrial volume in patients with atrial profile of serum anti-inflammatory and pro-inflammatory interleukins in fibrillation. PLoS One. (2016) 11:e0164145. doi: 10.1371/journal.pone.016 acute ischemic stroke patients. Neurol Sci. (2001) 22:289–96. doi: 10.1007/ 4145 s10072-001-8170-y 9. Noubiap JJ, Sanders P, Nattel S, Lau DH. Biomarkers in atrial fibrillation: 28. Fonseca AC, Matias JS, Melo TPE, Pires C, Geraldes R, Canhão P, et al. Time pathogenesis and clinical implications. Card Electrophysiol Clin. (2021) course of NT-proBNP levels after acute ischemic stroke. Acta Neurol Scand. 13:221–33. (2013) 128:235–40. 10. Ceyhun G. The relationship of the serum endocan level with the CHA2DS2- 29. He XW, Ke SF, Bao YY, Hong WJ, Shen YG, Li C, et al. Serum levels VASc score in patients with paroxysmal atrial fibrillation. Egypt Heart J. (2021) of endocan and endoglin are associated with large-artery atherosclerotic 73:4–9. doi: 10.1186/s43044-021-00132-1 stroke. Clin Chim Acta. (2018) 478:157–61. doi: 10.1016/j.cca.2017.1 11. Reyat JS, Chua W, Cardoso VR, Witten A, Kastner PM, Nashitha Kabir S, 2.040 et al. Reduced left atrial cardiomyocyte PITX2 and elevated circulating BMP10 30. Drew DA, Katz R, Kritchevsky S, Ix JH, Shlipak MG, Newman AB, et al. predict atrial fibrillation after ablation. JCI Insight. (2020) 5:1–16. doi: 10.1172/ Fibroblast growth factor 23 and blood pressure in older adults: the health, jci.insight.139179 aging, and body composition study. Hypertension. (2020) 76:236–43. 12. Ko D, Benson MD, Ngo D, Yang Q, Larson MG, Wang TJ, et al. Proteomics 31. Adela R, Mohammed SA, Kanwal A, Vishwakarma G, Reddy PNC, Banerjee profiling and risk of new-onset atrial fibrillation: framingham heart study. J SK. Elevated levels of GDF-15 is associated with increased angiotensin II Am Heart Assoc. (2019) 8:e010976. doi: 10.1161/JAHA.118.010976 in hypertensive patients with Type 2 diabetes. Per Med. (2016) 13:325–36. 13. Choudhury A, Freestone B, Patel J, Lip GYH. Relationship of soluble CD40 doi: 10.2217/pme-2016-0030 ligand to vascular endothelial growth factor, angiopoietins, and tissue factor 32. Ibrahim NE, Afilalo M, Chen-Tournoux A, Christenson RH, Gaggin HK, in atrial fibrillation: a link among platelet activation, angiogenesis, and Hollander JE, et al. Diagnostic and prognostic utilities of insulin-like growth thrombosis? Chest. (2007) 132:1913–9. doi: 10.1378/chest.07-1565 factor binding protein-7 in patients with dyspnea. JACC Hear Fail. (2020) 14. Freestone B, Chong AY, Lim HS, Blann A, Lip GYH. Angiogenic factors 8:415–22. in a trial fibrillation: a possible role in thrombogenesis? Ann Med. (2005) 33. Goldberger JJ, Arora R, Green D, Greenland P, Lee DC, Donald M, et al. 37:365–72. doi: 10.1080/07853890510037392 Evaluating the atrial myopathy underlying atrial fibrillation: identifying 15. Liu B, Li X, Zhao C, Wang Y, Lv M, Shi X, et al. Proteomic analysis of atrial the arrhythmogenic and thrombogenic substrate. Circulation. (2015) appendages revealed the pathophysiological changes of atrial fibrillation. Front 132:278–91. Physiol. (2020) 11:573433. doi: 10.3389/fphys.2020.573433 34. Diederichsen SZ, Haugan KJ, Kronborg C, Graff C, Højberg S, Køber L, et al. 16. Piek A, Suthahar N, Voors AA, de Boer RA, Silljé HHW. A combined Comprehensive evaluation of rhythm monitoring strategies in screening for bioinformatics, experimental and clinical approach to identify novel cardiac- atrial fibrillation: insights from patients at risk monitored long term with specific heart failure biomarkers: is Dickkopf-3 (DKK3) a possible candidate? an implantable loop recorder. Circulation. (2020) 141:1510–22. doi: 10.1161/ Eur J Heart Fail. (2020) 22:2065–74. doi: 10.1002/ejhf.1988 CIRCULATIONAHA.119.044407 17. Blum S, Aeschbacher S, Meyre P, Kühne M, Rodondi N, Beer JH, et al. 35. Garg PK, Norby FL, Wang W, Krishnappa D, Soliman EZ, Lutsey PL, Insulin-like growth factor-binding protein 7 and risk of congestive heart et al. Association of longitudinal changes in cardiac biomarkers with atrial failure hospitalization in patients with atrial fibrillation. Heart Rhythm. (2021) and ventricular arrhythmias (from the atherosclerosis risk in communities 18:512–19. doi: 10.1016/j.hrthm.2020.11.028 [ARIC] study). Am J Cardiol. (2021) 158:45–52. doi: 10.1016/j.amjcard.2021. 18. Pagola J, Juega J, Francisco-Pascual J, Bustamante A, Penalba A, Pala E, 07.043 et al. Large vessel occlusion is independently associated with atrial fibrillation detection. Eur J Neurol. (2020) 27:1618–24. Conflict of Interest: Neurovascular research laboratory, Vall d’Hebron Institute of 19. Robin X, Turck N, Hainard A, Tiberti N, Lisacek F, Sanchez JC, et al. Research (VHIR) received institutional research support from Roche Diagnostics PanelomiX: a threshold-based algorithm to create panels of biomarkers. Transl that funded part of the present study. Proteomics. (2013) 1:57–64. doi: 10.1016/j.trprot.2013.04.003 20. Plitt DC, Chung EH, Mounsey JP, Schwartz JD, Pursell IW, Gehi AK. Relation JM and AB have participated in scientific advisory boards organized by ROCHE on of atrial fibrillation burden and N-terminal pro-brain natriuretic peptide. Am the topic of “New biomarkers in Atrial Fibrillation.” J Cardiol. (2013) 111:1315–8. doi: 10.1016/j.amjcard.2013.01.273 21. Goetze JP, Bruneau BG, Ramos HR, Ogawa T, de Bold MK, de Bold AJ. Cardiac Publisher’s Note: All claims expressed in this article are solely those of the authors natriuretic peptides. Nat Rev Cardiol. (2020) 17:698–717. and do not necessarily represent those of their affiliated organizations, or those of 22. Markus A, Valerie S, Mira K. Promising biomarker candidates for the publisher, the editors and the reviewers. Any product that may be evaluated in cardioembolic stroke etiology. A brief narrative review and current opinion. this article, or claim that may be made by its manufacturer, is not guaranteed or Front Neurol. (2021) 12:1–10. doi: 10.3389/fneur.2021.624930 endorsed by the publisher. 23. Semenov AG, Feygina EE. Standardization of BNP and NT-ProBNP Immunoassays in Light of the Diverse and Complex Nature of Circulating BNP- Copyright © 2022 Palà, Bustamante, Pagola, Juega, Francisco-Pascual, Penalba, Related Peptides. 1st ed. Vol. 85. Amsterdam: Elsevier Inc (2018). doi: 10.1016/ Rodriguez, De Lera Alfonso, Arenillas, Cabezas, Pérez-Sánchez, Moniche, de Torres, bs.acc.2018.02.001 González-Alujas, Clúa-Espuny, Ballesta-Ors, Ribas, Acosta, Pedrote, Gonzalez- 24. Al-Khatib SM, Allen LaPointe NM, Chatterjee R, Crowley MJ, Dupre ME, Loyola, Gentile Lorente, Ángel Muñoz, Molina and Montaner. This is an open-access Kong DF, et al. Rate– and rhythm-control therapies in patients with atrial article distributed under the terms of the Creative Commons Attribution License fibrillation. Ann Intern Med. (2014) 160:760–73. (CC BY). The use, distribution or reproduction in other forums is permitted, provided 25. Staszewsky L, Meessen JMTA, Novelli D, Wienhues-Thelen UH, Disertori M, the original author(s) and the copyright owner(s) are credited and that the original Maggioni AP, et al. Total NT-proBNP, a novel biomarker related to recurrent publication in this journal is cited, in accordance with accepted academic practice. atrial fibrillation. BMC Cardiovasc Disord. (2021) 21:553. doi: 10.1186/s12872- No use, distribution or reproduction is permitted which does not comply with 021-02358-y these terms. Frontiers in Cardiovascular Medicine | www.frontiersin.org 11 July 2022 | Volume 9 | Article 908053