Normoxic versus Hyperoxic Reperfusion: A Retrospective Study of Ischemia–Reperfusion Injury in Patients Undergoing On-Pump Coronary Artery Bypass Grafting

This article examines a clinically important question in cardiac surgery: whether the oxygenation strategy used during cardiopulmonary bypass can influence ischemia–reperfusion injury in patients undergoing elective on-pump coronary artery bypass grafting. The authors compared two approaches used during surgery: a normoxic reperfusion strategy and a more conventional hyperoxic strategy. Their central hypothesis was that avoiding excess oxygen exposure during reperfusion might limit oxidative stress and reduce myocardial injury after CABG. Based on the data presented, the study argues that a more physiologic oxygen target may be preferable to routine hyperoxia in this setting. 

The study was a retrospective observational analysis conducted at a single center and included 50 consecutive adult patients undergoing isolated elective primary CABG with cardiopulmonary bypass. The patients were divided equally into two groups of 25. The normoxia group was managed with lower arterial oxygen targets during bypass, while the hyperoxia group received a more traditional oxygen-rich strategy. The investigators report that baseline characteristics and procedural variables were generally comparable between groups, which helps reduce, though not eliminate, concerns about major imbalance between the cohorts. 

The paper gives substantial attention to perfusion practice. During bypass, the normoxia group was managed with arterial oxygen tensions closer to physiologic levels, whereas the hyperoxia group maintained much higher PaO2 values. Importantly, indexed oxygen delivery remained comparable between groups over time, suggesting that the normoxic approach did not compromise systemic oxygen transport. This is one of the article’s most useful practical points: lower oxygen tension did not appear to reduce overall oxygen delivery when perfusion variables were otherwise properly managed. The charts on pages 6 and 7 reinforce this distinction, showing similar indexed oxygen delivery but clear separation in arterial oxygen tension and mixed venous oxygen saturation patterns between the two strategies. 

The most clinically interesting findings concern myocardial protection. All patients developed some degree of postoperative myocardial injury, but the extent of injury differed between groups. The normoxia group had significantly lower postoperative CK-MB levels and smaller reductions in left ventricular ejection fraction. They also demonstrated a higher postoperative cardiac index, suggesting better preserved cardiac performance after surgery. These results support the idea that hyperoxic reperfusion may worsen myocardial stunning or oxidative damage, while a normoxic strategy may better protect the heart during the vulnerable reperfusion period. On page 9, the figures visually summarize these differences, showing better preservation of ventricular function and lower inotropic requirements in the normoxia cohort. 

Another major theme of the article is oxidative stress. The authors measured multiple biomarkers related to oxidative injury and antioxidant capacity. The normoxia group had lower malondialdehyde levels, indicating less lipid peroxidation, and the paper reports more favorable oxidative stress profiles overall. This is biologically plausible because reperfusion in the presence of very high oxygen concentrations can amplify reactive oxygen species production. The discussion connects these biochemical findings to known mechanisms of myocardial ischemia–reperfusion injury, including endothelial dysfunction, mitochondrial injury, microvascular impairment, and inflammatory activation. In that framework, hyperoxia may be easy to deliver but not necessarily benign. 

The article also reports additional physiologic measures, including oxygen extraction ratio, mixed venous saturation, lactate, and cerebral oximetry. These data are relevant because they help answer a practical concern often raised by surgical and perfusion teams: whether reducing oxygen tension could create hidden systemic hypoxia. In this study, that concern was not strongly supported. Indexed oxygen delivery remained similar, and the normoxic strategy was not associated with obvious worsening in global perfusion markers. Instead, the normoxia group appeared to achieve a balance between adequate oxygen transport and avoidance of excessive oxidative burden. 

That said, the paper’s conclusions should be interpreted with caution. This was not a randomized controlled trial. It was retrospective, single-center, and involved only 50 patients. Those design features introduce risks of selection bias, confounding, and limited generalizability. Even though the groups looked similar at baseline, unmeasured differences in practice patterns, patient biology, operative details, or postoperative management could still have influenced the results. The study also focused mainly on short-term surrogate and early clinical outcomes rather than long-term endpoints such as mortality, major adverse cardiac events, or durable functional recovery. 

Even with those limitations, the study is valuable because it addresses a common but underexamined habit in cardiac surgery: permissive hyperoxia during bypass and reperfusion. The results challenge the assumption that more oxygen is automatically better. Instead, the paper supports a physiology-guided approach in which oxygenation is tailored to maintain adequate delivery while avoiding unnecessary hyperoxia. For perfusionists, anesthesiologists, and cardiac surgeons, this work adds to the growing discussion about precision oxygen therapy in the operating room. For researchers, it provides a rationale for larger prospective randomized studies designed to test whether normoxic reperfusion can improve hard clinical outcomes after CABG. Overall, the article suggests that oxygen strategy is not just a technical detail of bypass management, but a potentially modifiable determinant of myocardial injury and postoperative cardiac performance. 

3
This is a useful clinical study with a focused question, biologic plausibility, and meaningful perioperative endpoints, but it is limited by its retrospective design, single-center setting, and small sample size of 50 patients. Because it is not randomized or blinded and relies largely on short-term physiologic and biomarker outcomes, it provides supportive but not definitive evidence.