The Predictive Role of Inflammatory Biomarkers in Atrial Fibrillation as Seen through Neutrophil-Lymphocyte Ratio Mirror

Atrial fibrillation (AF) is the most common arrhythmia and is responsible for significant disease burden worldwide. Current evidence has suggested that systemic inflammatory response plays a crucial role in the initiation, maintenance, and progression of AF. So, recent efforts have been directed in search of measurable inflammatory biomarkers as additional tools in severity and prognosis assessment of AF. A simple, and easily obtainable, inflammatory marker is the neutrophil-lymphocyte ratio (NLR), which has shown good performance in preliminary studies as a potential prognostic biomarker in patients with AF. In this work, we performed a thorough review of clinical studies that evaluated the role of C-reactive protein (CRP), interleukin-6 (IL-6), and NLR as predictors of outcomes in AF. We gave a particular emphasis on the NLR because it is a simpler, widely available, and inexpensive biomarker.


Introduction
Atrial fibrillation (AF) is the most common cardiac arrhythmia, which in 2010 was estimated to affect about 33.5 million individuals in the world [1], with a prevalence around 2.3-3.4% among adults [2], and may reach 9% in those aged over 80 years [3]. Studies point to an increase in both incidence and prevalence [1,4,5], as well as the attributable mortality [1,5], so that between 1990 and 2013 AF was the factor with the greatest relative increase in the burden of cardiovascular diseases (CVDs) [5]. Future projections predict a 2-fold increase in the number of cases of AF in 2050 [3]. The AF is associated with significant morbidity and mortality, increasing the risk of stroke and death from all causes [6,7].
Recent investigations have registered significant advances in the understanding of the pathogenic mechanisms underlying AF [8,9]. One of the most explored mechanisms and that has gained more and more space in recent years is the inflammatory response [10,11]. The literature has emphasized the role of inflammation in the initiation, maintenance, and progression of AF [8,12,13]. Several inflammatory biomarkers, as C-reactive protein (CRP) and interleukins, have been associated with the occurrence of AF and its prognosis, including vascular events [10,[14][15][16]. In fact, the incidence of AF is increased in other situations that share significant systemic inflammatory response such as nonalcoholic fatty liver disease [17] and metabolic syndrome [18,19], suggesting a role of inflammation as a mediator between atrial fibrillation and these situations.
Among the inflammatory biomarkers, the neutrophillymphocyte ratio (NLR), defined as the ratio of absolute counts of neutrophils and lymphocytes, has emerged recently as effective outcomes predictor in atrial fibrillation [20,21], a role also demonstrated for ischemic heart disease and stroke [22][23][24]. NLR is a simple, inexpensive, and widely available biomarker and has been shown to be a good predictor of atrial arrhythmias that reflects the role of unbalanced white cells (with the predominance of activated neutrophils) in arrhythmogenesis [8,9,25].

Journal of Biomarkers
Multiple inflammatory markers have been studied as predictors of outcomes in AF, from those with potential direct involvement in the pathogenesis, such as IL-6 and NLR [11,26], and others only as a reflection of underlying immune responses, but apparently without direct participation as CRP [27]. There are those that are being more used in research contexts than in clinical practice. In this paper, we performed a thorough review of clinical studies focusing on CRP, IL-6, and NLR, as they are more reliable in clinical practice, with particular emphasis on the NLR because it is simpler, more widely available, and inexpensive.

Overview of Inflammatory Biomarkers in Atrial Fibrillation
Several studies have demonstrated the association between inflammatory markers and the incidence, severity, response to treatment, and prognosis in AF [14][15][16]28]. An analysis of a large study with 17,120 participants, without prior history of arrhythmia, high-sensitivity C-reactive protein (hs-CRP), was associated with a 36% increase in the risk of developing AF (hazard ratio [HR]: 1.37, -trend < 0.01) for each increasing tertile above baseline, with persistently high risk, comparing the highest to the lowest hs-CRP tertile, even after adjustment for potential confounders (hazard ratio [HR] 1.96; < 0.01) [29]. In other studies in patients with AF, CRP was a significant predictor of stroke [14,30] and peripheral embolism [31]. In treated patients who underwent electrical cardioversion, a high hs-CRP was an independent predictor of AF recurrence even after adjusting for confounders variables [32,33]. In another study following patients after catheter ablation, a high hs-CRP was an independent predictor of recurrence ( = 0.021) during a median follow-up of 15 months [15]. The contribution of inflammation in atrial activity seems to begin early, as demonstrated in a study where a high CRP was an independent risk factor for spontaneous contrast in transesophageal echocardiography [34]. This reflects that electromechanical impairment begins before any electrocardiographic visible dysrhythmia.
Other inflammatories biomarkers associated with AF and its progression are interleukin-6 (IL-6) [35] and interleukin-18 [36]. In one of these studies, with 3,762 adults with chronic kidney disease, a high plasma IL-6 level was associated with AF at baseline (Odds Ratio [OR], 1.61; = 0.001) and predicted new-onset AF (OR, 1.25; = 0.03) during a mean follow-up of 3.7 years [35]. In a study with patient in oral anticoagulation for AF, a high-sensitivity interleukin-6 (hsIL6) was a predictor of long-term cardiovascular events (HR 1.97, = 0.002) and all-cause mortality (HR 2.48, < 0.001) [37]; and adding hsIL6 to the clinical risk scores (CHADS2 and CHA2DS2-VASc) improved the discrimination index value for prediction of long-term cardiovascular events and death [37]. In another study with rhythm control strategy, IL-6 and CRP were significantly higher in those with AF recurrence than in those maintaining sinus rhythm (mean IL-6: 1. 84

Incidence and Prevalence.
A high NLR is associated with increased incidence of AF, as was evident in a prospective cohort, with 275 patients who underwent nonemergency coronary artery bypass grafting, where the group with postoperative AF had higher preoperative NLR (median 3.0 versus 2.4, = 0.001) [21]. These findings were also evident in a study with patients undergoing coronary angiography, with stent placement, for acute ST-segment elevation myocardial infarction, where those who developed AF had higher postcatheterization NLRs at 48 hours (median 5.23 versus 3.00, = 0.05) and 96 hours (median 4.67 versus 3.56, = 0.03) [59], suggesting an inflammatory contribution to new-onset postprocedural AF. In another study with diabetic patients, NLR was significantly higher in those with AF than in the AFfree group (mean 2.87 ± 1.3 versus 2.2 ± 1.56, = 0.019) and was an independent risk factor for AF (OR 3.486, = 0.004) using 2.38 as cut-off [60].

Severity and Incidence of Stroke.
A high NLR not only predicts higher incidence of AF but also is a predictor of disease severity and risk of stroke [61,62]. In a study with 309 patients with nonvalvular AF, a high NLR (>2.59) was an independent risk factor for the presence of left atrial thrombus on transesophageal echocardiography (TEE) (OR 1.59; < 0.02) [61]. In another study with TEE, a high NLR (>2.92) was a predictor of reduced (<10 cm/sec) left atrial appendage wall velocity (LAAWV) in patients with paroxysmal AF [63]. A large retrospective cohort including 32.912 patients with AF showed that each increase in NLR quartile above the lowest was associated with a significant increase in risk of stroke with HRs of 1.11 (0.91-1.35), 1.25 (1.03-1.51), and 1.56 (1.29-1.88) for the second, third, and highest quartiles, respectively; and adding NLR to CHA2DS2-VASc risk score improved the accuracy for prediction of stroke [62]. Even in those in oral anticoagulation, a high NLR level was a predictor of stroke [20].

Treatment Response and Mortality.
A high NLR is also a predictor of poor response to treatment as shown in a study, where it predicted AF recurrence after successful cardioversion with amiodarone [64]. In another study with 251 patients with symptomatic AF who underwent cryoablation, a high preablation NLR (>3.15) was a predictor of postprocedural disease recurrence (HR 2.15, 95% CI 1.70 to 2.73, < 0.001) [65]. Regarding mortality, no specific study relating NLR and increased mortality in patients with AF was found. However, a high NLR was an independent predictor of short-and longterm mortality in patients with stroke in general (irrespective of being cardioembolic or atherosclerotic) [24,66,67]. Table 2 summarizes the clinical studies that have assessed the role of NLR as a prognostic biomarker in AF.

Underlying Mechanisms, Pathways, and Relationship between Biomarkers in AF
In relation to the underlying mechanisms, despite the consistency of the studies regarding the epidemiological association between inflammation and AF, there is still a substantial scarcity of data in basic sciences giving the pathophysiologic background to this link. In the case of the CRP, an acute phase protein, it seems to be more a marker of underlying immune responses than an active participant in the pathogenesis of the disease. This is reinforced by the fact that genetic polymorphisms that are associated with the double increase in CRP showed no significant association with the AF [27].
On the other hand, multiple factors interfere with CRP that would be very difficult to control in primary studies, to evaluate possible pathophysiological nexus. For this reason, we focus our description more on those related to NLR, and IL-6, which are more than simple reflectors, seeming to be actively involved in the pathogenesis of AF. Neutrophil-lymphocyte ratio is a derived marker, expressing an imbalance in leukocytes with the dominance of neutrophils over lymphocytes, which may be only the "tip of the iceberg" of a deeper imbalance in the immunologic response. This seems particularly true from the observation that a high NLR is associated with the excessive activation of interleukin-17 (IL-17) axis in AF [26], which is a cytokine produced mainly by T-helper 17 (Th17) a subset of T-helper cells. In fact, the differentiation of Th17 cells from naïve T cells is mediated largely by IL-6 [73][74][75], a cytokine produced mainly by macrophages, which are neutrophils infiltrating tissues. So, IL-6 would induce, at T-helper cells level, the polarization of the differentiation favoring the effectors Th17 cells over the regulatory T (Treg) cells [76]. Th17 cells produce IL-17 that, among other functions, is responsible for the increase in fibrosis, which is a crucial component in AF [77][78][79][80]. It is interesting that those diseases that have the IL-17 as a cornerstone of its pathophysiology, like psoriasis, or with significant increase of its levels such as nonalcoholic fatty liver disease and metabolic syndrome are associated with increased incidence of AF [17,19,[81][82][83], suggesting a role of this cytokine as a mediator between AF and these clinical conditions.
The IL-17 is also associated with the upregulation of transforming growth factor beta (TGF-) signaling pathways [84], another potent promoter of atrial fibrosis and consequent AF [85][86][87]. In addition, IL-17 stimulates the production of more proinflammatory cytokines such as tumor necrosis factor-(TNF-) and IL-6 [88,89] and regulates tissue infiltration by neutrophils and myocyte apoptosis, which can start and engage various other pathophysiological pathways including oxidative stress and hypercoagulability [90][91][92]. IL-17 induces the production of IL-6, a potent inductor of IL-17 synthesis, occurring in this way, a refeeding on the axis [73,74].
The IL-6, as we described, plays a critical role in the regulation of IL-17 axis [73,75]. IL-6, together with IL-23, induces the differentiation of Th17 cells from naïve T cells, at the same time that inhibits TGF-beta-induced Treg differentiation, favoring, in this way, the T17/Treg imbalance [73,74,76]. There is still a significant gap between the epidemiological evidence of inflammation in AF and the current understanding of underlying physiopathological basis. The clear understanding of this association is still an object of future studies from basic science to clinical practice level.

Concerns and Limitations of the Use of Inflammatory Biomarkers in AF
The main concerns and doubts that arises from the potential use of inflammatory biomarkers in AF is about the additional value of using a panel of two or more biomarkers in predicting AF outcomes than the isolated use, and if there is a superiority of a biomarker in relation to others in AF. No primary study has evaluated the additional value of two or more biomarkers in comparison with the isolated use. Even studies that studied the correlation between biomarkers [11,14,38] did not evaluate the additive effect of them to predict outcomes. Despite this gap, it is very likely that in clinical practice a panel of 2 biomarkers or more may be better than the isolated use of one, for predicting AF-related outcomes. So, the evaluation of the additive effect of combined use should be a subject for future studies. On the other hand, the primary studies are very controversial about the superiority of a biomarker in relation to others. So, in light of the current literature, there are no sufficient data to support such point, highlighting only that NLR is more easily accessible and inexpensive than other biomarkers as hs-CRP and IL-6.

Conclusion and Future Directions
The review of the available evidence shows that inflammatory biomarkers such as IL-6, IL-17, and NLR have a crucial role in the pathogenesis of AF. They represent an additional, noninvasive tool, with good performance, to predict newonset disease, persistence, treatment response, recurrence, the risk of complications, and mortality in AF. The available evidence suggests that NLR, a simpler, widely available, and inexpensive biomarker, is a predictor of incidence, treatment success, recurrence, and thromboembolic complications.
Next studies should be addressed to clarify the underlying mechanisms in AF, to establish the additional value of using a panel of two or more biomarkers in predicting AF outcomes, to evaluate the superiority of a biomarker in relation to others, and to test the value of different biomarkers in different situations in the setting of AF.