Increased Long-term Mortality After a High Perioperative Inspiratory Oxygen Fraction During Abdominal Surgery

Abstract

A lthough a high perioperative inspiratory oxygen fraction (80%) has been recommended to avoid perioperative wound infections, one of the largest recent trials, the PROXI (perioperative oxygen inspired) trial, found no reduction in surgical site infection and a higher 30-day mortality rate in the group receiving 80% oxygen. In this follow-up study of the PROXI trial, the association between long-term mortality and perioperative oxygen fraction was assessed in patients undergoing abdominal surgery. From patients underwent elective or emergency laparotomy and were randomized to receive either 80% or 30% oxygen during for 2 hours after surgery. At follow-up, on February 4, 2010, survival was analyzed using Kaplan-Meier statistics and the Cox proportional hazards model. Vital status was obtained on 1382 of the 1386 patients after a median follow-up of 2.3 years (range, 1.3Y3.4 years). One hundred fifty-nine (23.2%) of 685 patients died in the 80% oxygen group compared with 128 (18.3%) of 701 patients in the 30% oxygen group (hazard ratio [HR], 1.30; 95% confidence interval [CI], 1.03Y1.64). In patients undergoing cancer surgery, the HR was 1.45 (95% CI, 1.10Y1.90), and after noncancer surgery, the HR was 1.06 (95% CI, 0.69Y1.65). Administration of 80% oxygen in the perioperative period was related to markedly increased long-term mortality, which appeared to be statistically significant in patients undergoing cancer surgery but not in noncancer patients. COMMENTOccasionally, studies are published that turn conventional thinking on its head; this is one such study. For a number of years, investigators have been studying the advantages of administering high-concentration oxygen in the operating room and postanesthesia care unit (PACU). This line of inquiry was stimulated by several studies that showed a benefit of decreased abdominal wound infections with administration of 80% oxygen during surgery and for the first 2 hours in the PACU. 1 A large subsequent trial, the PROXI study, however, showed no difference in wound infection with high-concentration oxygen administration. 2 That study also examined 30-day mortality, which was not different between the groups.The current study examines the long-term mortality of those enrolled in the PROXI trial. The trial was conducted in Denmark, which has a comprehensive medical record collection system. Indeed, there were only 4 of 1386 patients who were lost to follow-up. In the original PROXI trial, patients scheduled for elective or emergent laparotomy were randomized to receive 30% or 80% oxygen in the operating room and for 2 hours in the PACU.Surprisingly, there was a greater chance of not surviving the follow-up period in the 80% oxygen group. Looking at subgroups, it was apparent that the increased mortality was seen in patients undergoing laparotomy for cancer surgery. The HR was 1.45 with the 95% CI of 1.10Y1.90, which is a very notable difference. The noncancer laparotomy patients had no different outcome. The authors speculate as to the cause of increased mortality in ...

Intraoperative surgical site infection control and prevention: a position paper and future addendum to WSES intra-abdominal infections guidelines

“…The PROXI trial [125], that is a large, multicenter, randomized trial involving 1400 patients undergoing abdominal surgery, found no evidence of any beneficial effect of supplemental oxygen; in fact, SSI occurred in 131 of 685 patients (19%) receiving 80% oxygen and in 141 of 701 (20%) receiving 30% oxygen [odds ratio 0.94 (95% confidence interval 0.72-1.22), P = 0.64]. Indeed, a long-term follow-up study (median 2.3 years after surgery) found poorer survival in the supplemental oxygen group [126].…”

Cyclic and constant hyperoxia cause inflammation, apoptosis and cell death in human umbilical vein endothelial cells

“…1,2 Only recently, perioperative hyperoxia has been associated with higher patient mortality and morbidity although the detailed injurious mechanisms remain unknown. [3][4][5][6] While the effects of constant hyperoxia on the human body have been investigated to some extent, little is known with regard to cyclic hyperoxia that may occur during O 2 -administration in the presence of altered cardiac or pulmonary function. For example, cyclic hyperoxia can be caused by cyclic recruitment and derecruitment of lung atelectasis (cyclic atelectasis, CA) during high-dose O 2 -administration.…”

“…39 However, recent studies have questioned a net beneficial effect of hyperoxia that seems associated with increased mortality and morbidity. [4][5][6] The mechanisms by which hyperoxia may harm the human body include increase in reactive O 2 -species, reduced organ blood flow, and disturbances in cellular redox balance. 1,2 The present study adds a novel concept to this topic by proposing that extreme constant and cyclic hyperoxia potentially may cause harm to the vascular endothelium by signaling pathways including the oxidative stress, NFjB, Notch, and PPAR pathway.…”

Oxygen supplementation for critically ill patients—A protocol for a systematic review

“…High inspiratory oxygen concentrations have been associated with adverse outcomes in emergency medical conditions, including exacerbation of COPD, 28 resuscitation after cardiac arrest, [29][30][31] myocardial infarction, 32 and traumatic brain injury. 33 Additionally, treating perioperative patients with a high FiO 2 may be associated with increased mortality without reducing surgical site infections in surgical patients 34,35 and it might be linked to increased longterm risk of myocardial infarction. 36 These adverse outcomes may be caused by decreased local blood flow on normal and nondiseased vasculature induced by the vasoconstrictive effect of hyperoxaemia, 37 which have been described in the heart 38 and brain.…”

“…32 Recently two meta-analyses of observational studies found an association between hyperoxaemia and mortality in critically ill patients. 50,51 Some single studies have indicated an association between hyperoxaemia and mortality, 23,29,35 whereas others have not. 18,[52][53][54][55] Finally, a new single centre randomised clinical trial conducted in a general ICU population, showed that a hyperoxaemic target level of PaO 2 resulted in markedly increased mortality as opposed to a more conservative approach with a PaO 2 target within the normal physiological range.…”