Validation of post-operative atrial fibrillation in the Western Denmark Heart Registry

Introduction: Post-operative new-onset atrial fibrillation and flutter (POAF) is associated with increased morbidity and mortality following cardiac surgery. Registers and databases are important data sources for observational studies in this research area; hence, the aim was to assess the data validity of the POAF diagnosis in the Western Denmark Heart Registry (WDHR). Methods: We studied a 25% random sample (n = 1,381) from a cohort of 5,532 patients who underwent coronary artery bypass grafting, valve surgery or combinations between 1 January 2011 and 31 December 2013. Registrations of POAF diagnoses in the WDHR were compared with the actual clinical course as documented in the medical records. The positive predictive value for new-onset POAF in the WDHR was calculated. Results: A positive predictive value of 82.5% (95% confidence interval: 78.8-85.7) was found with a sensitivity and specificity of 75.2% and 90.9%, respectively. Conclusions: A diagnosis of new-onset POAF in the WDHR is relatively valid and may be used for contemporary epidemiological studies. Improvements may optimise the regis-try's validity even further, emphasising the importance of continuous validation and maintenance of the registry.