Reply to Ngaage

We thank the author for his comments [1]. It is clear that preoperative atrial fibrillation (AF) has different implications among patients with valve disease and those requiring isolated coronary artery bypass graft surgery (CABG). However, patients with coronary artery disease also share important risk factors with patients with some of the more common valve pathologies, such as senescent calcific aortic stenosis and ischaemic mitral regurgitation. Furthermore, in our study, only 11% of the entire study cohort was isolated valve patients. Therefore, the study’s findings are largely driven by the CABG patients, and the small proportion of isolated valve patients was not expected to introduce unacceptable heterogeneity within the study population nor appreciably change the study’s findings. To check this, we performed subgroup analysis among the CABG and valve patients separately and did not find substantive effect modification. Furthermore, as pointed out by the author, the retrospective nature of the study did not allow us to precisely classify patients according to the currently accepted definitions of paroxysmal, persistent and permanent AF [2]. However, in everyday clinical practice, it is often difficult to get a clear sense of the duration of AF and to reliably classify the patient presenting for cardiac surgery in AF without a previous history of the dysrrthythmia, and we do not feel that the lack of classification necessarily discredits our findings. Given that all of the patients with AF examined in this study had to have AF on preoperative electrocardiogram or a history of treatment for AF documented in the medical record, we probably have a larger proportion of persistent and permanent AF patients represented in this study, and our results must be interpreted with this in mind. Finally, a point was made about the low utilisation of oral anticoagulation as a potential mechanism for the adverse effect of AF on cardiac surgery outcomes. We did not have data on long-term use of medical therapies postoperatively including oral anticoagulants. The paper quoted in the discussion [3] was a population-based assessment (i.e., largely non-surgical patients) of warfarin use in the province where our centre is located, and appropriate treatment with warfarin compared favourably with contemporary literature. As a surrogate for warfarin use, we looked at the death and re-hospitalisation rates of AF patients with mechanical valves and found these to be significantly higher compared with those patients without AF (hazard ratio = 1.57, p = 0.01), suggesting that the negative effects of AF are not entirely obviated by and perhaps independent of the anticoagulation received for the mechanical prosthesis. Bando et al. [4] found similar results in a cohort of fully anticoagulated mechanical valve recipients with AF; however, this hypothesis remains to be evaluated in a future study. A particular strength of our study is the use of administrative re-hospitalisation data to determine outcomes in a large contemporary cohort of cardiac surgical patients. This eliminates the need to rely on patient self-report [5], a methodology we believe makes the telling of a good story far less compelling!

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.