Three hundred and seventy four patients were enrolled, including 167 patients with AF (105 undergoing curative AF ablation, 50 undergoing cardioversion, and 12 undergoing AV nodal ablation) and 207 controls without a history of AF or atrial flutter (151 SVT ablation patients and 56 controls with no arrhythmias). The baseline characteristics are shown in . AF patients were older, more likely to be male, white, have hypertension, a history of congestive heart failure, a larger body mass index (BMI), be on a statin, and be on a angiotensin converting enzyme (ACE) inhibitor or angiotensin receptor blocker (ARB). Based on the criteria described in the Statistical Analysis section, the covariates selected as potential confounders for inclusion in multivariate models are listed in .
Baseline characteristics of those with and without atrial fibrillation
Covariates included in the multivariate models
There were no differences in peripheral venous (peripheral or femoral vein) CRP and IL-6 levels between those with and without a history of AF (). However, those with persistent AF had significantly elevated CRP and IL-6 levels compared to those with paroxysmal AF (). In addition, those in AF at the time of the blood draw had significantly higher CRP and IL-6 levels than those in sinus rhythm, whether analyzing all participants, only participants with a history of AF, or only those participants with paroxysmal AF (). In addition, there were no meaningful differences in CRP or IL-6 when comparing patients with a history of AF in sinus rhythm to the control group with no history of arrhythmias.
Figure 1 Median levels of C-Reactive Protein (CRP) in mg/dL and interleukin-6 (IL-6) in ng/ml compared between those with and without an AF history, paroxysmal versus chronic atrial fibrillation, in atrial fibrillation versus in sinus rhythm at the time of the (more ...)
Although those with persistent AF continued to have both higher CRP and IL-6 levels than those with paroxysmal AF after adjusting for age, gender, race, hypertension, congestive heart failure, statin and ACE inhibitor or ARB use, neither CRP nor IL-6 remained higher in the persistent versus the paroxysmal group after adjusting for the rhythm present during the blood draw. In contrast, after adjusting for the same potential confounders as well as persistent versus paroxysmal AF, the presence of AF at the time of the blood draw remained significantly associated with both higher CRP levels (odds ratio [OR] for log CRP 1.79, 95% confidence interval [CI] 1.20–2.66, p=0.004) and higher IL-6 levels (OR for IL-6 1.69, 95% CI 1.11–2.58, p=0.015). Given the log transformation, this means that a doubling of CRP after adjustment (including adjusting for persistent versus paroxysmal AF) was associated with an approximate 50% greater odds of being in AF and that a doubling of IL-6 was associated with an approximate 40% greater odds of being in AF.
Left atrial volume was measured in all curative AF ablation patients (n=105) a median 1 day (IQR 1–3) prior to the blood draw. Those in AF at the time of the blood draw (n=46) had a larger left atrial volume than those in sinus rhythm at the time of the blood draw (n=59): 130 ± 38 ml versus 111 ± 34 ml, respectively (p=0.005). These two groups otherwise did not differ by demographics or medical histories. After adjusting for left atrial volume and AF type (paroxysmal versus persistent), both CRP and IL-6 remained independently associated with the presence of atrial fibrillation versus sinus rhythm at the time of the blood draw ().
Figure 2 Adjusted odds ratios of logCRP and logIL-6 each adjusted for left atrial volume and a history of paroxysmal versus persistent atrial fibrillation as predictors of the presence of atrial fibrillation versus sinus rhythm at the time of the blood draw among (more ...)
Of the 46 AF ablation patients who had cytokine measurements performed in the femoral vein, femoral artery, coronary sinus, and left atrium, 20 were in AF at the time of the blood draw and the remainder were in sinus rhythm. The AF ablation demographics and past medical histories did not differ between those in AF or sinus rhythm at the time of the blood draw; specifically, there were no significant differences regarding age, gender, race, BMI, or proportions with hypertension, heart failure (n=0), coronary artery disease, statin use, and ACE inhibitor or ARB use (all p values>0.1). Due to difficulty withdrawing blood from the luminal coronary sinus catheter in some patients, coronary sinus blood was available in 10 of the patients in AF and 12 of the patients in normal sinus rhythm.
Inflammatory biomarker concentration differences between blood pools are shown in . CRP differences in arterial levels (either femoral artery or left atrial) between those in and out of AF were more pronounced than differences in venous levels.
Figure 3 Median levels of C-Reactive Protein (CRP) in mg/dL and interleukin-6 (IL-6) in ng/ml compared between atrial fibrillation ablation patients in atrial fibrillation (n=20) versus in sinus rhythm (n=26) at the time of the blood draw assessed from the femoral (more ...)
In order to determine the trans-cardiac gradient, the coronary sinus blood levels were subtracted from the left atrial blood levels; the trans-femoral gradient (femoral vein levels subtracted from the femoral artery levels) was used as a control. The median trans-cardiac CRP gradient was positive in those in AF (more CRP in the LA than in the coronary sinus), whereas it was negative in those in sinus rhythm (less CRP in the LA than the coronary sinus), a significant difference (p=0.01, ). The difference in trans-femoral gradients between those in AF versus sinus rhythm was small and did not reach statistical significance (p=0.50, ).
Figure 4 The trans-cardiac (left atrial minus coronary sinus) and trans-femoral (femoral artery minus femoral vein) gradients in median C-Reactive Protein (CRP) and interleukin-6 (IL-6) levels in those in atrial fibrillation (n=20) versus sinus rhythm (n=26) at (more ...)
Although none of the IL-6 comparisons were statistically significant, the trends were all consistent with the findings involving CRP. Of interest, whereas the median trans-cardiac IL-6 gradient was positive in those with AF (more IL-6 in the left atrium than the coronary sinus) and negative in those with sinus rhythm, the median trans-femoral gradient was negative (more IL-6 in the vein than the artery) regardless of AF status ().