We read the article by Amar et al.  with great interest. Their trial addressed whether N-acetylcysteine can prevent atrial fibrillation after thoracic surgery. A secondary analysis examined perioperative systemic markers of inflammation, oxidative stress, and the incidence of atrial fibrillation after discharge. We thank the authors for contributing to our understanding on this important topic. However, we have two questions regarding subgroup analysis.

First, the primary endpoint analysis included surgical procedures such as open thoracotomy, minimally invasive video-assisted thoracoscopic, robotic-assisted approaches, and Ivor Lewis esophagectomy. However, there was not a subgroup analysis by surgical modality. Of note, the incidence of postoperative atrial fibrillation is influenced by the extent of lung resection  and by the operative site (left more than right lobectomy). Moreover, lobectomy removes more lung tissue than segmentectomy, which may induce myocardial hypoxia, thereby triggering atrial fibrillation.  A subgroup analysis of N-acetylcysteine effect according to surgical approach may be informative.

Second, the conclusion that N-acetylcysteine did not result in lower systemic markers of inflammation and oxidative stress should be interpreted with caution given the lack of power to identify these secondary outcomes. Perioperative drugs such as opioids and dexmedetomidine may affect intraoperative stress and inflammatory response. Oxycodone, for instance, has antioxidant and anti-inflammatory properties that protect against surgical stress by inhibiting the production of proinflammatory cytokines and lipid peroxidation  Therefore, it would be of interest to perform a more comprehensive analysis of potential pharmacologic influence on the results.

Postoperative atrial fibrillation is an important clinical outcome in patients undergoing noncardiac thoracic surgery, and there is growing interest in its prevention and treatment. However, multiple factors contribute to postoperative atrial fibrillation and inflammatory response. If there is substantial heterogeneity of these risk factors in the study population, risk-stratified analysis is especially important for enhancing the credibility of clinical trials.