Abstract / Summary
Abstract New-onset atrial fibrillation is the most frequent arrhythmia after cardiac surgery and is widely attributed to the inflammatory response to cardiopulmonary bypass, yet whether measured inflammation or constitutional predisposition governs individual risk is unsettled. In the prospective INFLACOR cohort, designed to relate perioperative inflammation to postoperative complications, we contrasted first-day postoperative inflammatory markers with variants at atrial-fibrillation loci. Of 525 adults undergoing first-time cardiac surgery on cardiopulmonary bypass, 364 without pre-existing atrial fibrillation formed the analytical sample. Four variants were genotyped: rs67249485 at the 4q25 locus upstream of PITX2 , rs2359171 in ZFHX3 , rs883079 in TBX5 and rs11773845 in CAV1 . A treated first episode from the second postoperative day onwards occurred in 115 patients (31.6%). Only rs67249485 was associated with the arrhythmia: risk-allele frequency was 23.5% in patients who fibrillated and 15.7% in those who did not (allelic risk ratio 1.38, 95% confidence interval 1.08 to 1.76; heterozygote risk ratio 1.55). No postoperative inflammatory marker, and none of ten inflammation-related variants genotyped under the original protocol, distinguished the groups. In multivariable analysis only age (odds ratio 1.71 per decade, 1.36 to 2.16) and the heterozygous PITX2 genotype (odds ratio 1.91, 1.19 to 3.06) remained independent predictors. The arrhythmia was associated with early respiratory failure, acute kidney injury and delirium but not with mortality at 30 days, one year or five years. Susceptibility appears to be set by the patient’s constitution rather than by the inflammatory response to the operation. ClinicalTrials.gov NCT01020409, registered 24 November 2009, retrospectively registered.