The feasibility and safety of a simple method for coronary sinus blood sampling during catheter ablation of arrhythmias

BACKGROUND: Coronary sinus (CS) blood sampling gives an insight into the localized pathophysiology of heart diseases. However, additional specifically-designed or modified catheters were needed in most previous studies, making the convenience of clinical application unsatisfactory. This study aimed to introduce a simple method for CS blood sampling without using additional catheters during catheter ablation (CA) of arrhythmias, and to investigate its feasibility and safety. METHODS: A total of 119 patients undergoing CA were prospectively enrolled, including 25 with paroxysmal supraventricular tachycardia (PSVT), 30 with premature ventricular complexes (PVC), and 64 with atrial fibrillation (AF). Cannulation and sampling of CS was performed via the femoral vein using a conventional 8F SR0(TM) or 8.5F SL1(TM) introducer sheath (St. Jude Medical) guided by a 6F steerable diagnostic catheter (St. Jude Medical). The success rate and any suspicious complications were recorded. Untargeted liquid chromatograph-mass spectrometer (LC-MS)-based metabonomics of CS samples versus peripheral venous samples were also performed. RESULTS: CS blood samples were successfully collected from 114 patients, with an overall success rate of 95.8%. Among patients with different arrhythmias, the success rates were similar, with 96.0% in PSVT, 96.7% in PVC, and 95.3% in AF (P=0.223). Adverse events occurred in four (3.4%) patients, including two patients with occasional atrial ectopic beats without causing any discomfort, and two patients with new-onset paroxysmal AF lasting for about 2 min. No serious complications were noted. Metabonomics analysis showed that CS samples provided a number of different metabolites (93 in PVST, 217 in PVC, and 109 in AF) versus peripheral samples. CONCLUSIONS: Our method for CS blood sampling during CA is feasible and safe, and can provide useful cardiometabolic information that is significantly different from a peripheral sample.