Role of inflammation in initiation and maintenance of atrial fibrillation in rheumatic mitral stenosis – An analytical cross‐sectional study

Abstract Background Inflammation has been implicated in the initiation and perpetuation of non‐valvular atrial fibrillation (AF). However, there is a lack of similar data on AF in rheumatic heart disease (RHD). The objective of this study was to analyze the association of inflammation as measured by serum inflammatory biomarkers with AF in rheumatic mitral stenosis (Rh‐MS). Methods A comparative cross‐sectional analytical study was conducted on 181 Rh‐MS patients in normal sinus rhythm (NSR; n = 69), subclinical transient AF (SCAF; detected by 24‐hours Holter monitoring; n = 30) and chronic AF (n = 82). Serum hs‐CRP, IL‐6, and sCD‐40L were assessed using ELISA immunoassay and compared in all groups of Rh‐MS with or without AF. Results We found significantly higher serum hs‐CRP and sCD‐40L levels in the overall AF (Chronic AF + SCAF) group (hs‐CRP: 4.5 ± 3.4 vs 2.3 ± 2.9 mg/L, P < .01; sCD‐40L: 6.4 ± 4.8 vs 3.1 ± 3.4 ng/mL, P < .01) and chronic AF subgroup (hs‐CRP: 4.9 ± 3.4 vs 2.3 ± 2.9 mg/L, P < .01; sCD‐40L: 6.9 ± 5.1 vs 3.1 ± 3.4 ng/mL, P < .01) compared to patients with sinus rhythm. There was a statistically significant graded increase of serum IL‐6 level from the NSR to the SCAF (vs NSR: 6.8 ± 3.9 vs 4.0 ± 2.2 pg/mL, P = .03), and chronic AF subgroups (vs NSR: 9.3 ± 6.5 vs 4.0 ± 2.2 pg/mL, P < .01; vs SCAF: 9.3 ± 6.5 vs 6.8 ± 3.9, P = .05) of atrial fibrillation. Conclusions Elevated levels of serum hs‐CRP, IL‐6, and sCD‐40L were strongly associated with overall AF and also with SCAF and chronic AF in Rh‐MS patients indicating a potential role of inflammation in the pathophysiology of rheumatic AF.


| INTRODUC TI ON
Pathophysiology of atrial fibrillation (AF) is highly complex and involves both structural and electrical remodelling of the left atrium. The role of inflammation in initiating and propagating non-valvular atrial fibrillation has been suggested in several studies using inflammatory biomarkers [like high-sensitivity C-reactive protein (hs-CRP), Interleukin-6 (IL-6), Soluble CD-40 Ligand (sCD-40L)] as a measure of systemic inflammation. [1][2][3][4][5][6][7][8][9] However, it is dilemmatic whether chronic inflammation is a risk factor or a consequence of AF. 10 Bruins et al, 11 for the first time, showed a link between inflammation and postoperative non-valvular AF in patients undergoing coronary artery bypass surgery. However, such an association has not been explored widely in patients with rheumatic heart disease (RHD), which is commonly encountered in low-and middle-income group countries. Atrial fibrillation in rheumatic heart disease affects a younger population, is influenced by factors different from those affecting non-valvular atrial fibrillation and has a higher risk of thromboembolism. Previous histopathological studies have shown degenerative remodelling, extensive fibrosis, and evidence of ongoing inflammation in RHD patients with AF, thereby suggesting an association between myocardial inflammation and AF. 12 A previous serum biomarker study also showed statistically higher levels of hs-CRP and IL-6 in RHD patients with paroxysmal and permanent AF. 13 However, all these studies have a small sample size, and their findings have not been validated in larger population. Furthermore, there is a need for non-invasive tests which may allow detection of low-grade inflammation in RHD patients and thus help in identifying patients predisposed to the development of atrial fibrillation.
In the present research, we sought to study the role of inflammation by comparing the levels of inflammatory biomarkers (hs-CRP, IL-6, and sCD-40L) in rheumatic mitral stenosis (Rh-MS) patients with and without AF. Additionally, the association of various echocardiographic and demographic parameters as risk factors for atrial fibrillation was also studied.

| Study design
The comparative cross-sectional analytical study was conducted in a tertiary care referral hospital in New Delhi, India. The clinical study was conducted in accordance with the ethical standards of the institutional committee on human experimentation and with the Helsinki Declaration of 1975, as revised in 2000. The study was approved by the institutional ethics committee. The study duration was three years, and written informed consent was obtained from all eligible patients.  The severity of atrial SEC was scored as, 0 = no SEC, grade 1+ = mild SEC at some part of LA, grade 2 + = severe swirling SEC that appeared throughout the atrium. 15 All the patients who were in sinus rhythm underwent a 24hour ECG recording using 10-lead Holter system (Schiller, Belmont, Australia) to detect subclinical transient atrial fibrillation (SCAF).

| Patient recruitment and assessment
Subclinical transient AF is defined as bursts of irregular tachycardia consisting of more than three consecutive QRS complexes with a rate of >100 beats per minute with no discernible P waves. 16,17 Based on the Holter analysis (Table S1), the patients were categorized in the NSR group (n = 69) and the SCAF group (n = 30). The third group comprised patients with chronic AF (n = 82) ( Figure 1).

| Measurements of inflammatory biomarkers
Biomarkers of inflammation-that is, high sensitivity C-reactive protein (hs-CRP), Interleukin-6 (IL-6) and soluble CD-40 Ligand Cedex, France). As ELISA kits have high fidelity and results may vary between two same ELISA kits from the same or two different manufacturers, all these samples were run along with the quality control (QC) with already defined analysis range to check the accuracy and authenticity of the tests. Additionally, 20 age-matched healthy subjects were added to assess the performance of the biomarkers.

| Statistical analysis
Continuous variables were expressed as mean ± standard deviation (SD) or median (range). The frequency of categorical variables was expressed as percentage (%). Independent t-test was used to compare the normally distributed continuous variable between two groups. Chi-square test was used to analyze the association between two categorical variables. The diagnostic accuracy of the significant inflammatory biomarkers, demographic, and echocardiographic parameters were assessed using the receiver operating characteristic curve analysis. Univariate binary logistic regression was performed to calculate the unadjusted odds ratio, whereas, multiple logistic regression analysis was performed to determine confounding variables adjusted odds ratios. A two-sided P ≤ .05 was considered significant.

| Study population characteristics
The characteristics of the study population (n = 181) and its groupwise description is shown in Table 1

| Results from univariate analysis
Significantly higher number of patients with severe MS had AF. In addition, Rh-MS patients with AF had significantly higher Wilkins score and lower MVA in comparison to the NSR group. The serum hs-CRP, IL-6, and sCD-40L levels (mean ± SD) in the healthy subjects were 1.3 ± 0.8 mg/L, 3.2 ± 1.9 pg/mL, and 0.1 ± 0.1 ng/mL, respectively. Serum inflammatory biomarker (hs-CRP, IL-6, sCD-40L) levels were significantly higher in the Rh-MS patients with AF in comparison to the NSR group (Table 1). In addition, there was a graded increase in the levels of the inflammatory biomarkers across the healthy subjects, NSR and various AF subgroups

| D ISCUSS I ON
Several risk factors, that is, age, hypertension, obesity, heart failure, alcohol intake, and family history, etc have been established   Our study also showed a significantly higher level of sCD-40L in the overall AF and chronic AF groups in comparison to the NSR group. CD-40/CD-40L system belongs to tumour necrosis factor superfamily and has a strong pro-inflammatory, prothrombotic, and atherogenic role. 7 The soluble CD-40L (sCD-40L) has been reported to be an inflammatory marker for non-valvular atrial fibrillation, especially during the first year of initiation. 8 Several studies have also highlighted it as a predictor of thromboembolic complications in AF patients. 9,31 Our study is the first report that showed a significant involvement of sCD40L in the pathophysiology of atrial fibrillation in the RHD population.
Not all severe Rh-MS patients with significant hemodynamic stress develop atrial fibrillation indicating the influence of an unknown triggering factor in the pathophysiologic cascade. It has been hypothesized that chronic inflammation may be 'the distinctive and decisive factor' that completes the missing link between atrial fibrillation and hemodynamic stress in rheumatic mitral stenosis. While the association of elevated levels of inflammatory biomarkers with AF does not imply causation, the presence of a graded response with increasing AF burden in our study lends merit to the hypothesis that chronic inflammation may act synergistically with hemodynamic stress to initiate and perpetuate atrial fibrillation (Figure 3).
Moreover as the severity of inflammation as measured by the biomarker levels increases, the burden of AF increases (from SCAF to chronic AF) ( Figure 2). The implications of such a finding would be manifold. First, these or similar inflammatory markers could be used as predictors for AF. This could be either using the marker alone or as a risk score incorporating clinical, echocardiographic, and biochemical parameters. Second, such predictors of increased risk of AF may allow upfront anticoagulation therapy in patients at risk, without overt AF, and such a strategy may help lower the burden of thromboembolic complication. Thirdly, there may be a potential role of pharmacologic immune modulators in preventing progression to AF in patients with RHD.
In this study, we have used stringent exclusion criteria to eliminate confounding factors associated with chronic inflammation that may bias the results. Although many factors like smoking, alcohol, lack of sleep, mental stress etc can induce oxidative stress, we did not collect those data which is one of the limitations of the study.
Though the sample size was not ideally calculated, our observed significant results with substantial effect ensure that the study power was sufficient to detect the effect. The limitation of our study includes small sample size and cross-sectional design. A small study group limits statistical inferences, especially in the smaller subset of patients with subclinical transient AF. Another concern is that ambulatory rhythm follow-up of only 24 hours might have missed patients with subclinical transient and paroxysmal AF. Although there was a significant association between atrial fibrillation and serum inflammatory biomarkers (ie, hs-CRP, IL-6, and sCD-40L), our study does not establish a clear cause and effect relationship between inflammation and valvular AF due to the cross-sectional study design and small sample size. However, the potential link between chronic lowgrade inflammation and valvular AF in our study may act as a primer for further research in this area which will benefit a large subset of valvular heart disease patients in the developing countries.

| CON CLUS IONS
Our study has shown a significant association between atrial fibrillation and serum inflammatory biomarkers (ie, hs-CRP, IL-6, and F I G U R E 3 Diagram showing probable pathophysiologic mechanisms for initiation and maintenance of atrial fibrillation in rheumatic mitral stenosis. SCAF: subclinical transient atrial fibrillation, AF: atrial fibrillation, hs-CRP: high sensitivity C-reactive protein, IL-6: interleukin-6, sCD-40L: soluble CD-40 Ligand