This Week in Surgery — Jun 11, 2026

Welcome to This Week in Surgery. This week we're covering 9 notable papers spanning perioperative optimization and quality improvement, challenges in critical care, and the evolving landscape of surgical technology and diagnostics. Let's dive in.

First, we look at strategies to improve outcomes across the surgical journey. A major paper in The New England Journal of Medicine provides robust evidence for a simple intervention to reduce bleeding [1]. The TRACTION trial was a large, multicenter, cluster-randomized trial across 10 Canadian hospitals, evaluating a hospital-wide policy of giving tranexamic acid during major noncardiac surgery. Among over 8,200 patients, most undergoing cancer surgery, the policy significantly reduced the need for red-cell transfusions. The transfusion rate dropped from 9.8% in the placebo group to 7.4% in the tranexamic acid group. Crucially, this benefit came without an increased risk of venous thromboembolism, which occurred in 2.1% of patients in both groups, meeting the pre-specified criterion for noninferiority. The takeaway is clear: a hospital policy of routine tranexamic acid in high-risk noncardiac surgery is safe and effective at conserving blood products. Staying with perioperative care, a large retrospective cohort study in the World Journal of Surgery examined the real-world impact of prehabilitation using a major United States claims database [8]. Looking at over 136,000 patients undergoing major procedures like CABG, colectomy, and pancreatectomy, researchers found that only 6.6% received services defined as prehabilitation, such as physical therapy or nutrition counseling, in the months before surgery. While prehabilitation was associated with a modest reduction in any major postoperative complication and a more notable decrease in the risk of myocardial infarction, it was also linked to significantly higher perioperative costs. The benefit for reducing MI was most pronounced in patients undergoing CABG. This study suggests a potential cardiac benefit to prehabilitation but raises important questions about cost-effectiveness and patient selection. From individual interventions to system-wide change, a report in the Journal of the American College of Surgeons details the impact of a province-wide Enhanced Recovery After Surgery, or ERAS, program in Quebec for colorectal surgery [4]. Comparing outcomes before and after implementation in over 3,200 patients, the program led to improved adherence to ERAS protocols and a reduction in median hospital length of stay from 4 days to 3 days, without an increase in complications. Importantly, the study found that the degree of benefit was directly tied to how well hospitals stuck to the program; each 10% increase in adherence was associated with a further reduction in length of stay and a 5.5% drop in complications. This underscores that for large-scale quality initiatives to succeed, sustained local adherence is the critical ingredient.

Now we turn to the critical care setting, where two major trials this week challenge existing or proposed therapies. First, in The New England Journal of Medicine, a large, open-label factorial trial investigated two commonly used mucoactive agents for patients with acute respiratory failure on mechanical ventilation [2]. Nearly 2,000 patients were randomized to receive carbocisteine, nebulized hypertonic saline, both, or usual care alone. The results were definitively negative. Neither agent, alone or in combination, reduced the duration of mechanical ventilation. Worse, both were associated with harm. Carbocisteine was linked to a more than six-fold increase in clinically important upper GI bleeding, while hypertonic saline was associated with a nearly six-fold increase in bronchoconstriction and a thirteen-fold increase in hypoxemia during nebulization. This trial provides strong evidence to abandon the routine use of these agents in this patient population. In a similar vein, the VICTORY randomized trial published in JAMA brings a decisive end to the debate over high-dose intravenous vitamin C for patients with severe burns [3]. The phase 3 trial, conducted across 24 burn centers, randomized adults with burns covering 20% or more of their body surface area to either high-dose vitamin C or placebo. The trial was stopped early at its first interim analysis because of futility and a clear signal of harm. The primary composite outcome of 28-day mortality or persistent organ dysfunction occurred more frequently in the vitamin C group, at 40.8% versus 29.7% in the placebo group. Looking at mortality alone, the results were stark: 28-day mortality was significantly higher in the vitamin C group, at 15% versus 7.6% for placebo—roughly double the risk. Hospital mortality was also significantly increased. This trial definitively shows that high-dose intravenous vitamin C is not beneficial and is likely harmful in patients with severe burn injury.

Our final theme covers the intersection of surgery with technology, from new robotic platforms to the nuances of artificial intelligence and molecular diagnostics. In Surgical Endoscopy, a prospective study from a single center in Italy provides a head-to-head comparison of three major robotic platforms for hysterectomy: the da Vinci X, the Versius, and the Hugo RAS [9]. In this non-randomized study of 150 patients, the overall performance, defined by a composite outcome of 'optimal hysterectomy', did not differ significantly between the three systems. There were some minor differences in secondary outcomes—for example, the Hugo was associated with less blood loss than the da Vinci X, and both the Hugo and Versius were linked to lower pain scores on postoperative day three. The most significant differences were in cost. When accounting for both platform-related indirect costs and consumables, the Versius and Hugo systems were substantially cheaper per procedure than the da Vinci X. This study suggests that while clinical performance is similar, cost considerations may play a major role in platform selection. Also in Surgical Endoscopy, a SAGES white paper tackles the complex and timely issue of risk and liability in the deployment of artificial intelligence systems in surgery [7]. The authors propose a framework that divides risk into three categories: risks inherent to the AI system itself, risks introduced by the surgeon using it, and risks from how the institution deploys it. They highlight that while surgeons remain the ultimate decision-makers, liability for errors could extend to AI developers for faulty design and to hospitals for negligent implementation. The paper stresses that safe integration requires robust governance, ongoing surveillance of AI performance, and clear credentialing for clinicians, moving the conversation beyond just technical accuracy to the entire ecosystem of AI in the operating room. Shifting to diagnostics, a paper in Annals of Surgery explores a way to improve the workup of indeterminate pulmonary nodules, a common clinical challenge [5]. Researchers evaluated whether adding blood-based biomarkers to the standard Mayo clinical prediction model could improve diagnostic accuracy. They tested a panel of four cancer protein biomarkers and antibodies to histoplasma, a common cause of benign granulomatous nodules. In two distinct patient cohorts, adding the full panel of blood tests to the Mayo model significantly improved its predictive power, increasing the area under the curve from around 0.7 to 0.8. For intermediate-risk nodules, the combined model correctly reclassified over 40% of patients, helping to better separate benign from malignant disease. This suggests a combined biomarker-clinical model could be a valuable tool to refine patient management. Finally, a study in Annals of Surgical Oncology offers a critical caution on the use of circulating tumor DNA, or ctDNA, for cancer surveillance [6]. In a secondary analysis of three prospective studies in stage II-III colon cancer, researchers found that peritoneal metastases represent a significant 'blind spot' for ctDNA. Recurrences that were isolated to the peritoneum were much more likely to be missed by ctDNA tests. This is likely because peritoneal disease sheds less tumor DNA into the systemic circulation. This finding has direct implications for surveillance strategies, reminding clinicians that a negative ctDNA test does not rule out recurrence, particularly in the peritoneum, and traditional imaging remains essential.

If you only have time for one paper this week, make it the TRACTION trial on tranexamic acid in The New England Journal of Medicine [1]. This large, well-designed trial provides definitive, practice-changing evidence that a simple, inexpensive hospital-wide policy can safely reduce blood transfusions in a broad population of patients undergoing major noncardiac surgery.

Here are the key takeaways from this week in Surgery. First, a hospital-wide policy of intraoperative tranexamic acid for major noncardiac surgery reduces the need for red-cell transfusions without increasing the risk of venous thromboembolism [1]. Second, for critically ill patients on mechanical ventilation, neither carbocisteine nor nebulized hypertonic saline shortens ventilation duration and both are associated with significant harm, including GI bleeding and bronchoconstriction [2]. Third, high-dose intravenous vitamin C does not improve outcomes in severe burn injury and is associated with increased mortality; it should not be used [3]. Fourth, while new robotic surgery platforms offer similar clinical performance, their cost profiles can differ substantially, a key factor for hospital adoption [9]. And finally, circulating tumor DNA surveillance for colon cancer has a significant blind spot for detecting isolated peritoneal metastases, meaning imaging remains a critical component of follow-up [6].

That's your roundup for This Week in Surgery. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.