Left atrial (LA) size has been associated with cardiovascular outcomes and the success of different therapy strategies in patients with atrial fibrillation (AF) [1,2]. The assessment of LA is therefore recommended in the clinical routine in all AF patients [3]. There are several imaging modalities to assess the LA, though echocardiography is widely available and thus the most frequently used. However, the echocardio-graphic LA diameter (LAD) does not reliably reflect the true size of LA anatomy, as pathological LA is often enlarged asymmetrically during AF progression [4–6]. Computer tomography (CT) is a modality that has been increasingly used to obtain three-dimensional (3D) images prior to AF catheter abla-tion. This provided new insights on the LA shape and volume, which have been proven to be better predictors of AF recurrences after LA ab-lation in comparison to the commonly used anterior–posterior LA diameter [4–6]. Recently we demonstrated that LA dilatation is more pronounced on the coronal plane, as represented by the transversal LA diameter (LA-TV) [7]. Furthermore, LA-TV was associated with AF recurrences and remained stronger predictor for rhythm outcomes compared with the echocardiographic LA diameter. We also demonstrated strong association between pro-inflammatory plasma markers and AF recurrences [8,9]. However, whether LA size parameters are associated with peripheral plasma markers of inflammation is unknown. Consequently, the present study (approved by the ethics committee) recruited 51 consecutive patients presenting for their first AF cath-eter ablation at Heart Center Leipzig. All patients gave informed consent according to institutional guidelines and the Declaration of Helsinki. Echocardiography and cardiac-CT with a multidetector 64-row helical system (Brilliance 64, Philips, Best, Netherlands) were performed (2 ± 1 days) before the procedure. CT data were reviewed using 3D reconstruction (EnSite Verismo, SJM, MN) and LA volume (LAV) was determined after exclusion of the atrial appendage (LAA) and the pulmonary veins (PV). LA was then centered on all three cutting planes and the superior–inferior (SI), transversal (TV) and anterior–posterior (AP) diameters were measured. Measurements were performed offline by an experienced observer and were repeated 4 weeks later by the same investigator and a second blinded reviewer. High sensitive inter-leukin 6 (hsIL-6) was analyzed from pre-procedural blood samples using a commercially available assay. Catheter ablation was performed as previously described [10], with circumferential ablation of the ipsilateral pulmonary veins, verified with a multipolar circular catheter. In patients with persistent AF, additional linear lesions were added at the mitral isthmus and the posterior LA wall to create a " box " lesion. Follow-up was performed with repeated 7-day-Holter ECG recordings at 6, 12, 24 and 36 months. Recurrence was defined as any documented atrial tachycardia or fibrillation episodes of ≥30 s (after a 3 month blanking period). Statistical analyses were performed with SPSS 17 (SPSS Inc., Chicago, USA). Parameters with a p-value b 0.1 in the univariable analysis (UV), were introduced in multivariable analyses (MV) in order to identify with hsIL-6 levels independently associated parameters. A two-tailed p value b 0.05 was considered significant. The clinical characteristics of the study population are presented in Table 1. The intra-and inter-observer correlation coefficients were ≥ 0.88. We found a significant correlation between peripheral hsIL-6 and LA-TV (r 2 = 0.34, p = 0.017) but not with LAV (p = N.S., Fig. 1). On univariable analysis, advanced age, higher BMI and lower eGFR aswell as with left atrial dimensions – e.g. LA-AP, LA-TV and LAD (but not LAV) – were significantly associated with hsIL-6 levels. On mul-tivariable analysis, the levels of hsIL-6 remained associated with age (r 2 = 0.28, p = 0.029), BMI (r 2 = 0.33, p = 0.011), renal dysfunction left atrial; hsIL-6, high sensitive interleukin 6 (hsIL-6).