The Novel CHA 2 DS 2-VASC-FSH Score is Predictive of Severe Coronary Artery Disease on Coronary Angiography in Patients with Atrial Fibrillation and Unstable Symptoms

Objective: AF may create confusion about the presence of severe or unstable coronary artery disease in cases with unstable symptoms. Novel scores and markers are needed to determine severe coronary artery disease in such patients. We aimed to test the newly developed CHA2DS2-VASc-FSH score, developed by adding family history for coronary artery disease, hyperlipidemia, and smoking to the original CHA2DS2-VASc score, in the prediction of severe CAD in patients with AF and unstable symptoms.


Original Article
Introduction Atrial fibrillation (AF) is a common cause of emergency department admissions, particularly when associated with a rapid ventricular response [1].AF may create myocardial ischemia or injury in the case of underlying severe coronary artery disease (CAD), but it may also mimic cardiac ischemia or infarction through elevated cardiac biomarkers and electrocardiogram (ECG) changes indicative of ischemia in the absence of underlying severe or unstable CAD [2].Thus, there is a need for reliable predictors of severe or unstable CAD as the cause of signs and symptoms of patients with AF and unstable symptoms.It has been reported that the CHADS 2 and CHA 2 DS 2 -VASc scores, which were originally developed for the risk prediction of stroke and death [3][4][5][6], could predict severe CAD in various clinical settings [7][8][9].Recently, the CHA 2 DS 2 -VASc-FSH score was formulated, which includes hyperlipidaemia, smoking, and the family history of CAD, in addition to the original CHA 2 DS 2 -VASc score, and substitutes the male gender for female gender.Hence, by incorporating the majority of risk factors for CAD, this newly developed score provides a comprehensive risk assessment for CAD [9].The aim of this study was to determine the prevalence of severe CAD and the predictive ability of the newly developed CHA 2 DS 2 -VASc-FSH score for the prediction of severe CAD in comparison with conventional risk scores developed for AF, as well as other demographic, clinical, electrocardiographic, and biochemical parameters in patients with AF and unstable symptoms.
tive of cardiac ischemia, including chest pain or one or more of its equivalents, including resting dyspnea, palpitations, altered consciousness, syncope, diaphoresis, agitation, decompensated heart failure, or pulmonary edema, and all of them underwent invasive coronary angiography in the index hospitalization.
AF was classified into the categories of newonset, paroxysmal, persistent, and permanent AF categories that were determined according to the 2014 American College of Cardiology/ American Heart Association/Heart Rhythm Society task force on practice guidelines, and the 2010 European Society of Cardiology Guidelines [10,11].Rapid ventricular response was defined as the resting heart rate greater than 100/min during AF at rest.Severe chronic renal disease referred to a severely (GFR ≤30 mL/min) or very severely (GFR ≤15 ml/min and/or receiving renal replacement therapy) reduced renal function (Stage 4 and 5 chronic kidney disease, respectively).Previous CAD was defined as invasively or non-invasively detected coronary luminal irregularities, plaques, or insignificant narrowings (≤50%) previously treated medically, and also significant (≥50%) lesions previously treated with either medical therapy, percutaneous coronary intervention, or coronary bypass surgery.The family history of CAD was defined as having a first degree relative diagnosed with CAD before the age of 65 for women and 55 for men.The history of smoking included past or current smoking.The history of heart failure included past or current signs and symptoms of heart failure with both low (<40%) and preserved ejection fraction (≥40%) and/or other objective evidence of cardiac dysfunction.Hypertension was defined as ongoing therapy for hypertension, systolic blood pressure of ≥140 mmHg or diastolic blood pressure of ≥90 mmHg; diabetes mellitus (DM) was defined as a fasting blood glucose level >126 mg/dL or blood glucose ≥200 mg/dL, or using antidiabetic drugs; hyperlipidemia was defined as hypercholesterolemia (serum TC >5.72 mmol/L), high levels of LDL-c (>3.1 mmol/L), low levels of HDL-c (<0.9 mmol/L), hypertriglyceridemia (serum TG >1.70 mmol/L), or taking antihyperlipidemic medications.The history of ischemic cerebrovascular accident was defined as having a history of ischemic transient ischemic attack or ischemic cerebral stroke.Previous non-cerebral embolic events were defined as thromboembolic events in any organ other than the brain.Cardiac biomarkers of injury, namely mass CK-MB and troponin I (cTnI), were measured both at admission and at 3 hours of followup.The cTnI testing was performed with an Architect STAT troponin I analyzer (Abbott Diagnostics, Illinois, USA) with a 99 th percentile cut-off value of 0.01 μg/L.Numerical troponin I values and binary (present/absent) troponin I positivity were recorded at the admission and at follow-up.Troponin positivity referred to a troponin I measurement above the upper reference limit (URL).A delta troponin I change between the admission and follow-up troponin I values was recorded for every patient and defined by either as a 50% increase in the follow-up troponin I value compared to the admission value when the latter was below the URL, provided that either of the measurements was above the URL; or a 20% increase in the followup troponin I value when the admission value was above URL [12].Renal function tests, electrolytes, and complete blood count parameters, including leucocyte count, neutrophil count, lymphocyte count, the red cell distribution width (RDW), hemoglobin level, platelet count, platelet-to-lymphocyte ratio, and neutrophil-tolymphocyte ratio, were also measured.
Electrocardiographic changes suggestive of myocardial ischemia or injury were analyzed from 12-lead ECG recordings made at admission and 3-6 hours later, and they were defined as ≥1.0 mm horizontal or downsloping ST segment depression 0.08 seconds after the J point, symmetrical T inversion in at least two contagious precordial leads excluding V1 and V2, newly developed pathological Q waves defined as initial negative waves at least 0.04 sec in duration and 0.1 mv or >¼ of the proceeding R wave in size, or newly developed loss of the R wave progression across the precordial leads.Echocardiographic examinations were performed using a Vivid 7 Vingmed echocardiography device (Vingmed, GE, Horten, Norway).Echocardiographic left ventricular ejection fraction (LVEF) was determined by the modified Simpson' s method, and systolic wall motion abnormality for each left ventricular segment was assessed by 2D echocardiography.Left ventricular hypertrophy was defined as interventricular septal and/or posterior left ventricular wall thicknesses exceeding 11 mm.
All patients in this study underwent invasive coronary angiography at consulting physician' s discretion, on the basis of signs of symptoms suggesting acute cardiac ischemia or infarction.A significant coronary lesion was defined as luminal narrowing of ≥50% in any major epicardial vessel or one of its branches.In patients with a previous percutaneous coronary intervention, significant coronary lesions were defined as significant stent restenosis (≥50%) in previously stented vessels, or de novo sig-nificant luminal narrowing (≥50%) in previously untreated vessels.In patients who underwent a prior coronary bypass grafting operation, severe CAD was defined as significant de novo lesions in bypass grafts supplying one or more epicardial vessels, in non-bypassed coronary vessels, or in native vessels at sites distal to the insertion sites of bypass grafts.A non-significant coronary lesion was defined as luminal irregularities or plaques, or non-severe coronary lesions not exceeding 50%.The study population was classified into two groups depending on coronary angiography on invasive coronary angiography, and the two groups were compared with respect to the CHADS 2 , CHA 2 DS 2 -VASc, and CHA 2 DS 2 -VASc-FSH scores, as well as demographic, clinical, electrocardiographic, echocardiographic, and laboratory results.
The CHADS 2 score was the sum of 1 point each for heart failure, hypertension (HT), age ≥75 years, and DM and 2 points for prior stroke or transient ischemic attack.The CHA 2 DS 2 -VASc score was calculated by assigning 1 point each for Congestive heart failure (CHF), HT, age 65-74 years, DM, vascular disease, and female gender and 2 points each for prior stroke or TIA and age ≥75 years.The newly formulated CHA 2 DS 2 -VASc-FSH score consists of CHF, HT, age ≥75 years (double score), DM, previous stroke/TIA (double score), vascular disease, age 65-74 years, gender (male), hyperlipidaemia, smoking, and a family history of CAD.
The exclusion criteria included asymptomatic AF cases, cardiovascular arrest or asystole, any wide complex tachyarrhythmia presumed or suspected to be of ventricular origin, presistent (>20 minutes) ST segment elevation; electrical cardioversion (pharmacological and spontaneous cardioversion were permissible) or chest compressions causing troponin I elevation; and recent history of myocardial infarction in the last 15 days.All study data were accessed via Başkent University Hospital' s automation system.

Statistical Analysis
Continuous study data were presented as the mean±standard deviation or median (minimum-maximum) depending on the normality of their distribution assessed by the Kolmogorov-Smirnov test.Categoric variables were presented as the number and percentage.The difference between the groups with respect to continuous variables was tested with Student' s t-test or the Mann-Whitney U test, with the former being used for normally distributed variables and the latter for non-normally distributed ones.The difference with regard to categoric variables was tested using the chisquared test.Two separate univariate analyses were then performed to determine the correlation between examined study parameters and severe CAD and unstable CAD on coronary angiography.A binary logistic regression analysis was performed to determine the significant independent predictors of severe CAD on coronary angiography.
The comparison of the two groups with and without severe CAD revealed that the significant CAD group had significantly greater rates of hypertension, DM, CAD, CVA, severe renal disease, chest pain as the presenting symptom, and unstable lesions (p<0.05 for all comparisons); significantly greater CHA 2 DS 2 -VASc (p<0.01) and CHA 2 DS 2 -VASc-FSH scores (p<0.05), and significantly greater RDW and creatinine levels (p<0.05 for both).The two groups did not significantly differ with respect to cardiac biomarker levels, any troponin I positivity, ischemic ECG changes, echocardiographic wall motion abnormalities, or left ventricular hypertrophy (Table 3).

Discussion
AF may complicate an acute coronary syndrome [13,14], or it may also cause myocardial injury or infarction by creating a myocardial supply-demand mismatch (Type 2 myocardial ischemia), particularly when associated with rapid ventricular response, or by coronary embolism [15,16].Unfortunately, AF may also be associated with electrocardiographic ST-T changes and troponin positivity mimicking myocardial ischemia and injury in the absence of severe CAD [2].It is often difficult to discern AF-induced pseudo-infarction/ischemia patterns from Type 2 myocardial ischemia resulting from underlying severe CAD, or a true acute coronary syndrome plus AF.Among patients with AF and unstable signs and symptoms, clinicians should determine the pretest probability of underlying severe or unstable CAD to make decisions about hospitalization and invasive coronary angiography.
So far, only a few studies have examined the prevalence and predictors of severe CAD in patients with AF and unstable symptoms and found prevalences ranging from 18.2% to 32.5% [17,18].Androulakis et al. [17] reported that there was no significant correlation between ischemic ECG changes and underlying occult CAD in patients with AF paroxysms.Alghamry et al. [18] found that dyspnea, ST depression, and a history of CAD predicted significant CAD   CHA 2 DS 2 -VASc score.Supporting this idea, the CHA 2 DS 2 -VASc score was correlated to severe CAD in the univariate analysis, but it did not independently predict severe CAD in a multivariate analysis whereas the CHA 2 DS 2 -VASc-FSH score did.The CHA 2 DS 2 -VASc-FSH score was comparably predictive of significant CAD in early-stage AF types such as new-onset and paroxysmal AF, and the older AF types such as persistent and permanent AF.This suggests that it is the combination of risk factors and thus a greater CAD burden, but not the age of AF, that plays a role in the development of CAD, and the CHA 2 DS 2 -VASc-FSH score is equally predictive of severe underlying CAD in the overall AF population with unstable symptoms.It should also be noted that the CHA 2 DS 2 -VASc-FSH score helps provide the risk stratification not just in AF, but in all cases of ischemic heart disease, and it can be used in all of ischemic heart disease for prognostication.Therefore, its role as a prognostic sign should be further studied.
This study has some limitations.First, it was a retrospective analysis, and therefore it is limited by the inherent limitations of its design.Hence, it involved only patients with AF who underwent coronary angiography, and therefore the prevalence and the significant predictors of CAD remained unclear in the overall AF population, including the conservatively managed one.Furthermore, we lacked information In conclusion, in patients with AF and unstable symptoms warranting further management, the newly defined CHA 2 DS 2 -VASc-FSH score can successfully predict severe CAD on coronary angiography.In clinical settings where clinical doubt exists whether symptoms are caused by severe CAD, or when severe CAD should be excluded for other purposes, a CHA 2 DS 2 -VASc-FSH score of at least 3 can be used to make a decision to proceed with invasive coronary angiography.In the future, the CHA 2 DS 2 -VASc-FSH score can be incorporated with other biochemical and echocardiographic predictors of atherosclerosis in patients with AF.

Figure 1 .
Figure 1.Receiver operating characteristics (ROC) curves of the thromboebolic risk scores in the prediction of severe coronary artery disease in the overall atrial fibrillation group.
about in-hospital or 30-day mortality and thus could not ascertain the prognostic value of the CHA 2 DS 2 -VASc-FSH score or delta troponin I change beyond coronary lesion characteristics.There is also a need for the validation of the CHA 2 DS 2 -VASc-FSH in other AF settings, such as in outpatients, postoperative period, and ST elevation myocardial infarction.

Table 1 .
Clinical and pathologic characteristics of the patients

Table 2 .
Findings on coronary angiography and coronary intervention results

Table 3 .
Comparison of the study groups with respect to demographic, biochemical, electrocardiographic, and echocardiographic variables CAD: coronary artery disease; DM: diabetes mellitus; HT: hypertension; HL: hyperlipidemia; CRD: chronic renal disease; COPD: chronic obstructive pulmonary disease; CVA: cerebrovascular accident; NLR: neutrophil-to-lymphocyte ratio; PLR: platelet-to-lymphocyte ratio; RDW: red cell distribution width; ECG: electrocardiogram; LVEF: left ventricular ejection fraction; LVHT: left ventricular hypertrophy; WMA: wall motion abnormality on echocardiography in patients hospitalized for AF but concluded that troponin elevation was common among these patients and was not a reliable predictor for underlying severe coronary lesions.Our study revealed that among patients with AF and unstable symptoms the prevalence of severe CAD was 48.6%, and it revealed that neither cardiac biomarkers nor ischemic elec-trocardiographic changes were able to predict underlying severe CAD.On the other hand, CHA 2 DS 2 -VASc-FSH score and severe chronic renal disease emerged as significant independent predictors of severe CAD.Our results suggest that among patients presenting with AF and unstable symptoms, severe CAD can be independently predicted by the CHA 2 DS 2 -2 -VASc score because the former incorporates a greater number of risk factors for CAD and thus provides a greater clue about the underlying CAD than the traditional