As patients live longer and anesthetic safety improves, the number of intermediate- to high-risk operations continues to grow annually. This is significant, because the World Health Organization has estimated that there are 230 million major surgical procedures undertaken annually.1 Despite the improvement in patient management, however, noncardiac surgery is still associated with significant morbidity and mortality. For example, in the decade preceding 2004, the 30-day postoperative mortality rate for intermediate- to high-risk noncardiac surgery was 1.3% in the Canadian province of Ontario, a jurisdiction with universal health care.2 Cardiac events remain the most common cause of serious postoperative morbidity and mortality. If these results are applicable to the remainder of the world, >2 million persons are at risk of sustaining a perioperative myocardial infarction, cardiac arrest, cerebral vascular accident, or death annually. This serious public health issue requires continued research focused on ameliorating these risks.
In this respect, the recent developments related to nitrous oxide are particularly relevant. Nitrous oxide inhibits methionine synthetase, which mediates an increase in homocysteine. In nonsurgical settings, it is well recognized that long-term increases of plasma levels of homocysteine are an independent risk factor for coronary artery and cerebral vascular disease.3 Pathophysiologically, increases in homocysteine have been found to cause endothelial dysfunction.4 Hyperhomocysteinemia has also been shown to impair myocardial substrate utilization,5 and enhance platelet aggregation.6 Nitrous oxide–induced inactivation of methionine synthetase increases plasma homocysteine after surgery.7 Nitrous oxide has been implicated in the genesis of myocardial ischemia, presumably through this effect on homocysteine levels. Cardiovascular events can plausibly be predicted to be increased with acutely elevated homocystinemia; such increases have been reported by Badner et al.,8 in moderate-risk patients having carotid endarterectomy. Lastly, preoperative administration of folate and B vitamins can inhibit the nitrous oxide–induced increase in homocysteine.9 Similarly, genetic variations have been shown to alter the homocysteinemic response to nitrous oxide as well.10 These findings in part led to the conduct of the ENIGMA-I trial.11
This trial (ENIGMA-I) was conducted in an unselected group of low-cardiac-risk surgical patients. However, it was a large randomized, controlled trial that found that nitrous oxide increased the risk of wound infection, severe vomiting, atelectasis, and pneumonia. Additionally, patients receiving nitrous oxide had longer intensive care unit stays, possibly indicating an increased incidence of more serious complications. The ENIGMA-I trial, however, was not designed to detect acute, short-term differences in myocardial infarction or death. Nevertheless, the number of confirmed myocardial infarctions was greater in the nitrous oxide group compared with the control group (1.3% vs 0.7%; adjusted P = 0.19). Furthermore, when all reports (confirmed and unconfirmed) of myocardial infarction were included, there was a marked increase in the incidence of myocardial infarction in the nitrous oxide group (30 vs 10 cases; P = 0.002). There were also 9 postoperative deaths in the nitrous oxide group and 3 in the control group (P = 0.10). The findings suggested that nitrous oxide may be particularly detrimental in those patients at risk of cardiac events.12 In a substudy of ENIGMA-I, 394 unselected patients had homocysteine measured preoperatively and on the first postoperative day.13 This study showed that, in addition to nitrous oxide, a deficiency of folate or B vitamins increased the incidence of hyperhomocysteinemia 4-fold. Hyperhomocysteinemia was associated with a doubling of the rate of major complications. In a further substudy of 59 patients, nitrous oxide was associated with a decrease in flow-mediated vasodilatation, which is a form of endothelial dysfunction. The results of these investigations lend further support to the theory that nitrous oxide increases adverse cardiac events.14
In this issue of Anesthesia & Analgesia, Leslie et al.15 have published the longer-term results of the original ENIGMA-I trial. They found that patients exposed to nitrous oxide had an increased incidence of myocardial infarction, in a mean follow-up period of 3.5 years after a nitrous oxide–based general anesthetic. However, despite nitrous oxide acutely increasing the incidence of both infection and pneumonia, as noted above, and now in this study, evidence of more long-term myocardial infarctions, there was no difference in mortality. It is interesting to note that this investigation found a significant interaction between abdominal surgery and nitrous oxide, in that there was a 36% reduction in the death rate in patients having nitrous oxide–free anesthesia in nonabdominal surgery. This interaction, however, was not seen in abdominal surgery, which constituted almost 60% of the study population. This finding may be attributable to the large number of low-risk procedures such as laparoscopic cholecystectomies that were included in the trial. Finally, high homocysteine levels were not associated with increased mortality; paradoxically, however, hyperhomocysteinemia was associated with an increased incidence of myocardial infarction. Restated, nitrous oxide anesthesia is associated with increased myocardial infarction, but not mortality.
Is there an explanation for these seemingly paradoxical results? As noted, one hypothesis being advanced suggests that the adverse postoperative events are mediated through a vascular mechanism. Moreover, nitrous oxide mediates an increase in homocysteine and secondarily causes endothelial dysfunction. Consider that mortality was 50% higher (albeit not statistically significantly) in the patients with hyperhomocysteinemia. However, only 13% of the study population is estimated to have developed this biochemical abnormality, and fully one-third of the subjects that did develop postoperative homocysteine increases were not exposed to nitrous oxide. This underpowering was compounded by the fact that approximately 20% of the study population (342 patients) was lost to follow-up, potentially underestimating the number of deaths. In the original analysis by the authors, patients lost to follow-up were treated as alive and event-free at 30 days. The validity of this assumption is debatable. In 2 sensitivity analyses that were conducted to address this methodologic shortfall, the missing data were imputed, first by using a logistic regression model for death, and second by assuming that the patients lost to follow-up with strokes and myocardial infarctions occurred randomly. These key assumptions, and that the events occurred randomly in the patients lost to follow-up, is open to challenge because loss to follow-up did not occur randomly. Ten percent of patients were not followed because of a resource shortfall at the recruiting center, which is a selection bias and by definition therefore did not occur randomly. Furthermore, because this was a multinational study, it is not safe to assume that the events would be uniform across all sites, because methionine synthetase is understood to vary within populations, based on a genetic profile. This again brings into question the assumption that these lost data can be modeled based on the remainder of the data. Independent of the resource issue, another 5% of the whole population was lost to follow-up. Patients lost to this type of follow-up tend to be from lower socioeconomic backgrounds.16 These patients are likewise associated with increased morbidity, and one should not assume that the death rate in this group was similar to the remainder of the population. Thus, it is quite probable that this investigation has underestimated the true death rate. However, it is safe to assume that these factors influencing the loss to follow-up would also underestimate the number of myocardial infarctions. Underestimation of the true event rates is an important consideration because in any “negative” study, the issue of sample size and power is of paramount importance. ENIGMA was underpowered to answer questions about mortality and these factors further reduce the ability to detect an effect; thus, a much larger study would be required to show the effect conclusively. For example, a trial of 7000 patients would be required to demonstrate a 30% reduction in a population with a baseline event rate of 5%.
Alternatively, the longer-term results of the ENIGMA trial reported in this issue, whereby myocardial infarctions are increased without any change in mortality, may in fact represent the true outcome. In a somewhat analogous manner, the POISE (Perioperative Ischemia Evaluation) trial showed that perioperative metoprolol use decreased the incidence of myocardial infarction, but ironically had a higher mortality rate.17 Why, for instance, would we expect that acute short-term exposure to nitrous oxide would influence longer-term postoperative mortality or myocardial infarction? One of the few trials showing a long-term benefit of perioperative management, the use of atenolol,18 had serious methodologic flaws of withdrawing patients receiving β-blockers and excluding the mortality that occurred within the first week. It is possible that nitrous oxide acts as a “stress test” and that people with increases in homocysteine represent a population at risk for small myocardial vascular events. These factors may be the result of genetic predisposition, lifestyle choices, and/or poor diet. Thus, it is entirely possible that the myocardial results presented herein are merely an association. Nitrous oxide, therefore, causes an increase in patients predisposed to developing hyperhomocysteinemia and this is associated with causing a myocardial infarction, but not death. A recent large randomized, controlled trial in nonsurgical patients in which patients with known hyperhomocysteinemia were randomized to B vitamins and placebo (which also normalizes homocysteine levels) found no difference in the major outcomes of death or vascular events.19 Thus, hyperhomocysteinemia may just be a marker associated with increased vascular outcomes but not death.
Lastly, although this trial may represent a true outcome, readers should also consider that the results of the present investigation may be analogous to those in the EVAR (EndoVascular Aneurysm Repair) investigations.20 In these prospective studies, endovascular repairs of aortic aneurysms were found to reduce the short-term morbidity and mortality compared with open repair. The benefit was short lived, however, and at the end of the 2-year follow-up, there was no net benefit in terms of mortality. In the ENIGMA follow-up study in this issue, the authors were not able to precisely time myocardial infarction or stroke, or provide survival curves in which it would be possible to discern early differences in survival.15 This would allow identification of an early increased incidence of mortality due to infection, pneumonia, and myocardial infarction, which could diminish over time when deaths due to cancer (75% of the study mortality) predominate. This in itself is not surprising because cancer is frequently the primary reason for having surgery, suggesting that the major cause of death after major surgery is a result of the primary disease, not the associated comorbidities.
The results provided by Leslie et al.15 are a valuable addition to our knowledge about the effects of intraoperative nitrous oxide. The article is undoubtedly intriguing; however, it still does not answer several important safety aspects. There is clearly a need for more information on this subject. We hope that the ENIGMA-II trial,21 a prospective study of intermediate- or high-risk patients randomized to the use of nitrous oxide that is currently enrolling patients, will help provide some additional answers.
Name: W. Scott Beattie, MD, PhD, FRCPC
Conflicts of Interest: Dr. Beattie is a member of the ENIGMA-II Steering Committee and holds a peer-reviewed grant from the Ontario Heart and Stroke Foundation to conduct ENIGMA-II in Canada. Dr. Beattie is supported in part through the R. Fraser Elliot Endowment and the University of Toronto, Department of Anesthesia.
Name: Neal H. Badner, MD, FRCPC
Conflicts of Interest: Dr. Badner is a member of the ENIGMA-II Steering Committee and holds a peer-reviewed grant from the Ontario Heart and Stroke Foundation to conduct ENIGMA-II in Canada.
1. Weiser TG, Regenbogen SG, Thompson KD, Haynes AB, Lipsitz SR, Berry WR, Gawanda AA. An estimation of the global volume of surgery: a modelling strategy based on available data. Lancet 2008;372:139–44
2. Wijeysundera DN, Austin PC, Beattie WS, Hux JE, Laupacis A. Outcomes and processes of care related to preoperative medical consultation. Arch Intern Med 2010:170;1365–74
3. Clarke R, Daly L, Robinson K, Naughten E, Cahalane S, Fowler B, Graham I. Hyperhomocysteinemia: a risk factor for vascular disease. N Engl J Med 1991;324:1149–55
4. Chambers JC, McGregor A, Jean-Marie J, Obeid OA, Kooner JS. Demonstration of rapid onset vascular endothelial dysfunction after hyperhomocysteinemia: an effect reversible with vitamin C therapy. Circulation 1999;99:1156–60
5. Wald DS, Law M, Morris JK. Homocysteine and cardiovascular disease: evidence on causality from a meta-analysis. BMJ 2002;325:1202–8
6. Mayer EL, Jacobsen DW, Robinson K. Homocysteine and coronary atherosclerosis. J Am Coll Cardiol 1996;27:517–27
7. Badner NH, Drader K, Freeman D, Spence JD. The use of intraoperative nitrous oxide leads to postoperative elevations in plasma homocysteine. Anesth Analg 1998;87:711–3
8. Badner NH, Beattie WS, Freeman D, Spence JD. Nitrous oxide-induced elevated homocysteine concentrations are associated with increased myocardial ischemia in patients undergoing carotid endarterectomy. Anesth Analg 2000;91:1073–9
9. Badner NH, Freeman D, Spence JD. Preoperative oral B vitamins prevent nitrous oxide-induced postoperative plasma homocysteine increases. Anesth Analg 2001;93:1507–10
10. Nagele P, Zeugswetter B, Wiener C, Burger H, Hüpfl M, Mittlböck M, Födinger M. Influence of methylenetetrahydrofolate reductase gene polymorphisms on homocysteine concentrations after nitrous oxide anesthesia. Anesthesiology 2008;109:36–43
11. Myles PS, Leslie K, Chan MTV, Forbes A, Paech MJ, Peyton P, Silbert BS, Pascoe E. Avoidance of nitrous oxide for patients undergoing major surgery. Anesthesiology 2007;107:221–31
12. Hynes MD. Catecholamine mechanisms in the stimulation of mouse locomotor activity by nitrous oxide and morphine. Eur J Pharmacol 1983;90:109–14
13. Myles PS, Chan MTV, Leslie K, Peyton P, Paech M, Forbes A. Effect of nitrous oxide on plasma homocysteine and folate in patients undergoing major surgery. Br J Anaesth 2008;100:780–6
14. Myles PS, Chan MTV, McIlory D, Lau C, Symons J, Chen S. Effect of nitrous oxide anesthesia on plasma homocysteine and endothelial function. Anesthesiology 2008;109:657–63
15. Leslie K, Myles PS, Chan MTV, Forbes A, Paech M, Peyton P, Silbert BS, Williamson E. Nitrous oxide and long-term morbidity and mortality in the ENIGMA trial. Anesth Analg 2011;112:387–93
16. Geng EH, Bangsberg DR, Musinguzi N. Understanding reasons for and outcomes of patients lost to follow-up in antiretroviral therapy programs in Africa through a sampling-based approach. J Acquir Immune Defic Syndr 2010;53:405–11
17. POISE Study Group Devereaux PJ, Yang H, Yusuf S, Guyatt G, Leslie K, Villar JC, Xavier D, Chrolavicius S, Greenspan L, Pogue J, Pais P, Liu L, Xu S, Málaga G, Avezum A, Chan M, Montori VM, Jacka M, Choi P. Effects of extended-release metoprolol succinate in patients undergoing non-cardiac surgery (POISE trial): a randomized controlled trial. Lancet 2008;371:1839–47
18. Mangano D, Layug EL, Wallace A, Tateo I. Effect of atenolol on mortality and cardiovascular morbidity after noncardiac surgery. N Engl J Med 1996;335:1713–20
19. Study of the Effectiveness of Additional Reductions in Cholesterol and Homocysteine (SEARCH) Collaborative Group Armitage JM, Bowman L, Clarke RJ, Wallendszus K, Bulbulia R, Rahimi K, Haynes R, Parish S, Sleight P, Peto R, Collins R. Effects of homocysteine-lowering with folic acid plus vitamin B12 vs placebo on mortality and major morbidity in myocardial infarction survivors. JAMA 2010;303:2486–94
20. EVAR trial participants. Endovascular aneurysm repair vs open repair in patients with abdominal aortic aneurysm (EVAR trial 1): randomised controlled trial. Lancet 2005;365:2179–86
21. Myles PS, Leslie K, Peyton P, Paech M, Forbes A, Chan MTV, Sessler D, Devereaux PJ, Silbert BS, Jamrozik K, Beattie S, Badner N, Tomlinson J, Wallace S; ANZCA Trials Group. Nitrous Oxide and Perioperative Cardiac Morbidity (ENIGMA-II) Trial: rationale and design. Am Heart J 2009;157:488–94