Nitrous oxide (N2O) has an unusual toxic side effect, unrelated to its anesthetic action: it inactivates vitamin B12 (cobalamin).1,2 The inactivation of cobalamin and cobalamin-dependent enzymes, such as methionine synthase, is chemically irreversible and can last clinically up to 1 week until new enzyme is synthesized.3 In patients, the inactivating effects of N2O on cobalamin can be observed by an increase in plasma total homocysteine (tHcy) because methionine synthase, the enzyme responsible for the conversion of homocysteine to methionine, requires cobalamin as coenzyme.4,5
Clinically, cobalamin inactivation by N2O and the subsequent increase in plasma tHcy and methionine deficiency can lead to megaloblastic bone marrow changes,6 neuropathy,7 severe spinal cord degeneration,8 and even death, particularly in patients with preexisting inborn or acquired disorders of folate metabolism. In 2003, the death of a 4-month-old infant with a previously unknown severe deficiency in a core enzyme of the folate cycle was attributed to the use of N2O.9
The safety of N2O has been questioned, both in adult10 and pediatric patients.11 However, the inactivating effects of N2O on cobalamin and subsequent increase in plasma tHcy have never been systematically investigated in children. To our knowledge, this is the first study to test the hypothesis that N2O anesthesia causes a significant increase in plasma tHcy in children.
This was an ancillary study, performed on residual plasma samples, from a pharmacokinetic study of approved anesthetic drugs in children, which was approved by the Washington University IRB. Written informed consent was obtained from parents and assent from children. The ancillary study also received IRB approval.
The parent study included pediatric patients scheduled for elective major spine surgery. Most patients had idiopathic scoliosis. Patients could be included if they met the following inclusion criteria: age 5 to 18 years; undergoing general anesthesia and surgery with anticipated postoperative inpatient stay of ≥3 days; and signed, written informed consent from legal guardians and assent from the patient. Patients were excluded if they had a history of liver or kidney disease, or were pregnant or nursing females. An additional inclusion criterion for the ancillary study was that sufficient residual plasma samples were available.
The patients received standard general anesthesia and monitoring. The protocol did not specify the choice of anesthetic drugs, including the use of N2O, which were left at the sole discretion of the anesthesia team.
Plasma samples obtained at baseline and 1, 2, 4, 6, 8, 12, 24, 48, 72, and 96 hours after induction of anesthesia were analyzed. Baseline samples were obtained immediately after an IV line was placed, usually after inhaled induction of anesthesia. Blood samples were immediately put on ice and spun down within 2 hours of collection. Samples were then frozen at −80°C until assayed.
Plasma tHcy was measured on a Roche Hitachi 917 analyzer using the Diazyme homocysteine enzymatic assay (Diazyme Laboratories, Poway, CA) with reagents from Genzyme (Cambridge, MA). This assay has excellent correlation with high-performance liquid chromatography and immunochemical methods with a linear range of 3 to 50 μmol/L and inter %CV (coefficient of variation) values of <5%.12
Peak plasma tHcy concentrations were compared with baseline tHcy and absolute and relative changes were calculated. Cumulative N2O exposure was calculated as the product of average N2O concentration used (as fraction of 1) and the duration of N2O exposure (N2O·min). As an example, if a patient had a N2O anesthesia duration of 300 minutes at 50% concentration (=0.5), the resultant product would be 150 N2O·min. This calculation was done by hand from paper anesthesia records with measurements every 5-minute interval.
To model the effects of N2O anesthesia on plasma tHcy in this longitudinal dataset, we used a linear mixed regression model (random slopes and random intercepts) with log-transformed tHcy as a dependent variable and a first-order autoregressive covariance structure.13 Several models were compared and the most parsimonious chosen based on −2LL (negative log-likelihood) or Akaike Information Criterion. A linear correlation was examined between N2O·min and peak tHcy using the Pearson correlation coefficient and a simple linear regression was performed; results include the 95% confidence intervals. All reported tests are 2-sided and a P value <0.05 was considered statistically significant.
In this study, 27 pediatric patients from the parent study cohort (n = 30) were included (Table 1). Most patients underwent idiopathic scoliosis repair.
At baseline, the median plasma tHcy concentration was 5.1 μmol/L (0.9–9.8 μmol/L; min–max) and several patients (5 of 18 females, 28%; 0 of 9 males, 0%) met the diagnostic criteria for mild hyperhomocysteinemia (normal range for plasma tHcy at 12–19 years of age: female, 3.3–7.2 μmol/L; male, 4.3–9.9 μmol/L) (Table 1).
All except 1 pediatric patient received N2O intraoperatively. Four children were only briefly (<30 minutes) exposed to N2O during induction of or emergence from general anesthesia. The remaining 23 patients received N2O (average concentration 55%) for the full duration of the spine surgery, which in some patients lasted up to 10 hours.
All children who were exposed to N2O (n = 26) had a subsequent increase in their plasma tHcy concentration by an average of +9.4 μmol/L (geometric mean; 95% confidence interval [CI], 7.1–12.5 μmol/L) or +228% (mean; 95% CI, 178%–279%) (Fig. 1, Table 2). The increase in plasma tHcy was highly significant and peaked between 6 and 8 hours (Fig. 2), usually immediately after the cessation of general anesthesia. Most children experienced a severalfold increase of their plasma tHcy concentration (maximum: +567%) and the maximum observed peak concentration was 38.6 μmol/L. The highest tHcy levels were observed in children with high baseline plasma tHcy levels. No difference was observed between female and male patients. A linear mixed model was used to model the effects of N2O exposure on plasma tHcy within each individual patient and among all patients. The results of the mixed model show a statistically highly significant difference among all patients (P < 0.001), the time points (P < 0.001), and also a significant effect of the cumulative N2O dose (N2O·min, P = 0.02) and time point. Twenty-four hours after anesthesia start time, plasma tHcy concentrations largely reverted to their baseline levels.
The magnitude of the increase in plasma tHcy caused by N2O was strongly correlated with the product of duration of N2O anesthesia and average N2O concentration used (N2O·min) (Pearson r = 0.80; 95% CI, 0.61–0.91; P < 0.0001) (Fig. 3). More than 64% of the observed variation in plasma tHcy increase could be explained by the N2O·min variable, making it a strong and significant predictor.
This study showed that pediatric patients develop a significant increase in plasma tHcy when receiving N2O during prolonged general anesthesia. Some children in our study developed a 400% to 600% increase from their baseline plasma tHcy levels, signifying a magnitude of an inhibitory effect of N2O on folate and methionine metabolism that is larger than frequently observed in adult patients. Even though our patients had very long N2O anesthesia durations, comparable studies in adults reported increases in plasma tHcy of only between 50% and 80%. Badner et al.4 reported a 74% increase; Myles et al.14 only 50% even after >8 hours of anesthesia; and our own previous study showed an increase of 80% for >4 hours of N2O anesthesia.15 Only Ermens et al.3 showed comparable results, although only reported as correlation. Our findings suggest that children experience an at least similar, if not greater, inhibition of methionine synthase and folate metabolism by N2O, but the lack of comparable pediatric studies makes it difficult to generalize the results from a single study.
Baseline plasma tHcy levels varied widely within our study population, and several children had mild hyperhomocysteinemia (normal range for plasma tHcy at 12–19 years of age: female, 3.3–7.2 μmol/L; male, 4.3–9.9 μmol/L). Baseline plasma tHcy levels vary within the population and are influenced by environmental (nutrition, folic acid and vitamin B12 intake) and genetic factors, most importantly the MTHFR C677T polymorphism, considered the most important genetic predictor.16
A surprising finding of this study was that nearly all plasma tHcy levels reverted into the normal range after 24 hours, irrespective of the magnitude of the increase caused by N2O. Until now, it was commonly believed and reported from adult patients that N2O-induced hyperhomocysteinemia extends well beyond the immediate postoperative period and can last up to 1 week after exposure.3 Whether children have a better ability to recover from N2O-induced inhibition of methionine synthase compared with adults or this observation was specific for our study population is unclear. A possible explanation may be that our patients experienced large shifts in their fluid balance including intravascular volume resuscitation and blood transfusions, which may have decreased plasma homocysteine concentrations.
What is the clinical relevance of these findings? Two related but separate issues need to be addressed. First is the question of adverse drug reactions caused by N2O's inhibition of methionine synthase. Over the last 60 years, a multitude of case reports and series have been published showing the following adverse clinical outcomes unequivocally and directly attributed to N2O's inactivation of vitamin B12 and subsequent inhibition of methionine synthase: fatal9 and nonfatal17 neurologic degeneration, myelopathy,8,18 – 21 peripheral neuropathy,7,22 – 25 bone marrow depression and megaloblastic bone marrow changes,6,26 – 31 and an increased risk for infection.32,33 Mostly these toxic side effects were observed only after prolonged or repeated exposure to N2O (>24 hours) or among N2O drug abusers. Although these effects are apparently fairly uncommon, N2O clearly can cause hematologic and neurologic side effects that may be irreversible and/or fatal. The true incidence of the adverse neurologic and hematologic effects of N2O when used in a clinically appropriate manner and dose has never been systematically investigated, neither in adults nor children.
The second question is, What are the clinical consequences of an acute increase in plasma homocysteine concentrations? This question is much more difficult to answer. As the substrate for methionine synthase, homocysteine accumulates because of the inhibition by N2O. It is therefore possible that an acute increase in plasma tHcy in itself causes no additional adverse effects; alternatively, a direct toxic effect of increased plasma tHcy may also be possible. Whereas a chronic plasma homocysteine increase at the level observed in our study (up to 40 μmol/L) is unequivocally associated with a substantially increased risk of atherosclerosis, coronary artery disease,34 and premature death in adults,35 as well as venous thromboembolism in children,36 the effects of an acute increase in plasma tHcy are largely unknown. The direct effects of an acute increase in blood homocysteine levels on the cardiovascular system have mostly been shown experimentally37 – 39 and have not been unequivocally correlated with any clinical outcomes (separate from N2O's inhibition of methionine synthase).
This study has several strengths and limitations. It must be emphasized that this study was exploratory in nature, taking advantage of a unique opportunity to collect serial blood samples in a pediatric population, and therefore was not designed to investigate any clinical outcomes.
The present observational study did not have a formal control group that did not receive N2O. Two independent lines of evidence strongly suggest that, in our study, the exposure to N2O was causally linked to the subsequent increase in plasma tHcy. First, it is generally accepted that in nonexperimental, observational studies, the presence of a dose-response relationship infers, but naturally does not prove, causality. In our study, a very strong relationship between the cumulative N2O dose and subsequent increase in homocysteine was found (r = 0.80). Second, it has been repeatedly shown in >6 human studies (>1000 patients) that in the absence of N2O, patients will not develop an increase in postoperative plasma tHcy.3,4,14,40 – 42 Based on these preexisting data, it is a reasonable prediction that we would have observed a similar result (no increase in plasma tHcy without N2O) if our study had had a formal control group.
The study was limited in its generalizability because it enrolled mostly teenagers undergoing major spine surgery and no patient younger than 10 years of age. No study has investigated the interaction among N2O, folate metabolism, and homocysteine in younger children (0–10 years) thus far, so our findings should not be extrapolated to younger children. Because most of our patients had scoliosis, it is possible that the findings of this study were influenced by scoliosis-associated conditions or spine surgery (prone position, large fluid shifts). Our study did not determine any influence of gene variants, most notably the MTHFR C677T polymorphism, which is considered the most important genetic predictor for baseline blood tHcy concentrations and which also has a significant effect on the increase in plasma tHcy after N2O anesthesia.15
This study showed that children undergoing N2O anesthesia develop significantly increased plasma tHcy concentrations. The clinical relevance of this observation is unknown and would require well-designed randomized controlled trials.
Name: Peter Nagele, MD, MSc.
Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.
Attestation: Peter Nagele has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.
Conflicts of Interest: Peter Nagele received research funding from Roche Diagnostics.
Name: Danielle Tallchief, BS, RN.
Contribution: This author helped conduct the study.
Attestation: Danielle Tallchief has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Conflicts of Interest: Danielle Tallchief reported no conflicts of interest.
Name: Jane Blood, BS, RN.
Contribution: This author helped conduct the study.
Attestation: Jane Blood has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Conflicts of Interest: Jane Blood reported no conflicts of interest.
Name: Anshuman Sharma, MD.
Contribution: This author helped design the study, conduct the study, and write the manuscript.
Attestation: Anshuman Sharma has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Conflicts of Interest: Anshuman Sharma reported no conflicts of interest.
Name: Evan D. Kharasch, MD, PhD.
Contribution: This author helped design the study, conduct the study, and write the manuscript.
Conflicts: Evan D. Kharasch reported no conflicts of interest.
Attestation: Evan D. Kharasch has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
In 2007, Peter Nagele was awarded a Foundation for Anesthesia Education and Research (FAER)-mentored career development award (mentor: Evan Kharasch, MD, PhD) to investigate the role of pharmacogenetics in the development of adverse outcomes related to nitrous oxide anesthesia, particularly perioperative myocardial infarction. The FAER award was instrumental in creating a strong and rigorous research program in perioperative pharmacogenomics and nitrous oxide research and subsequently led to a successful National Institutes of Health career development grant application. What the FAER grant also created was a strong, productive mentor-mentee relationship that transcended beyond the funding period and subsequently transformed into a collaboration. Therefore, both mentor and mentee are extremely grateful for having received support from FAER. The current article is a logical extension of the research topic presented in the FAER award application and aims to investigate the effects of nitrous oxide in children.
1. Deacon R, Lumb M, Perry J, Chanarin I, Minty B, Halsey MJ, Nunn JF. Selective inactivation of vitamin B12 in rats by nitrous oxide. Lancet 1978;2:1023–4
2. Sanders RD, Weimann J, Maze M. Biologic effects of nitrous oxide: a mechanistic and toxicologic review. Anesthesiology 2008;109:707–22
3. Ermens AA, Refsum H, Rupreht J, Spijkers LJ, Guttormsen AB, Lindemans J, Ueland PM, Abels J. Monitoring cobalamin inactivation during nitrous oxide anesthesia by determination of homocysteine and folate in plasma and urine. Clin Pharmacol Ther 1991;49:385–93
4. Badner NH, Drader K, Freeman D, Spence JD. The use of intraoperative nitrous oxide leads to postoperative increases in plasma homocysteine. Anesth Analg 1998;87:711–3
5. Koblin DD, Waskell L, Watson JE, Stokstad EL, Eger EI II. Nitrous oxide inactivates methionine synthetase in human liver. Anesth Analg 1982;61:75–8
6. Amos RJ, Hinds CJ, Amess JAL, Mollin DL. Incidence and pathogenesis of acute megaloblastic bone-marrow change in patients receiving intensive care. Lancet 1982;320:835–9
7. Layzer RB, Fishman RA, Schafer JA. Neuropathy following abuse of nitrous oxide. Neurology 1978;28:504–6
8. Marie RM, Le Biez E, Busson P, Schaeffer S, Boiteau L, Dupuy B, Viader F. Nitrous oxide anesthesia-associated myelopathy. Arch Neurol 2000;57:380–2
9. Selzer RR, Rosenblatt DS, Laxova R, Hogan K. Adverse effect of nitrous oxide in a child with 5,10-methylenetetrahydrofolate reductase deficiency. N Engl J Med 2003;349:45–50
10. Jahn UR, Berendes E. Nitrous oxide: an outdated anaesthetic. Best Pract Res Clin Anaesthesiol 2005;19:391–7
11. Baum VC. When nitrous oxide is no laughing matter: nitrous oxide and pediatric anesthesia. Pediatr Anesth 2007;17:824–30
12. La'ulu SL, Rawlins ML, Pfeiffer CM, Zhang M, Roberts WL. Performance characteristics of six homocysteine assays. Am J Clin Pathol 2008;130:969–75
13. West BT. Analyzing longitudinal data with the linear mixed models procedure in SPSS. Eval Health Prof 2009;32:207–28
14. Myles PS, Chan MT, 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
15. 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
16. Bathum L, Petersen I, Christiansen L, Konieczna A, Sorensen TI, Kyvik KO. Genetic and environmental influences on plasma homocysteine: results from a Danish twin study. Clin Chem 2007;53:971–9
17. Felmet K, Robins B, Tilford D, Hayflick SJ. Acute neurologic decompensation in an infant with cobalamin deficiency exposed to nitrous oxide. J Pediatr 2000;137:427–8
18. Hadzic A, Glab K, Sanborn KV, Thys DM. Severe neurologic deficit after nitrous oxide anesthesia. Anesthesiology 1995;83:863–6
19. Lacassie HJ, Nazar C, Yonish B, Sandoval P, Muir HA, Mellado P. Reversible nitrous oxide myelopathy and a polymorphism in the gene encoding 5,10-methylenetetrahydrofolate reductase. Br J Anaesth 2006;96:222–5
20. Layzer RB. Myeloneuropathy after prolonged exposure to nitrous oxide. Lancet 1978;2:1227–30
21. Rosener M, Dichgans J. Severe combined degeneration of the spinal cord after nitrous oxide anaesthesia in a vegetarian. J Neurol Neurosurg Psychiatry 1996;60:354
22. Alarcia R, Ara JR, Serrano M, Garcia M, Latorre AM, Capablo JL. Severe polyneuropathy after using nitrous oxide as an anesthetic: a preventable disease? [in Spanish] Rev Neurol 1999;29:36–8
23. Flippo TS, Holder WD Jr. Neurologic degeneration associated with nitrous oxide anesthesia in patients with vitamin B12 deficiency. Arch Surg 1993;128:1391–5
24. Kinsella LJ, Green R. ‘Anesthesia paresthetica': nitrous oxide-induced cobalamin deficiency. Neurology 1995;45:1608–10
25. Waclawik AJ, Luzzio CC, Juhasz-Pocsine K, Hamilton V. Myeloneuropathy from nitrous oxide abuse: unusually high methylmalonic acid and homocysteine levels. WMJ 2003;102:43–5
26. Amess JA, Burman JF, Rees GM, Nancekievill DG, Mollin DL. Megaloblastic haemopoiesis in patients receiving nitrous oxide. Lancet 1978;2:339–42
27. Amos RJ, Amess JA, Nancekievill DG, Rees GM. Prevention of nitrous oxide-induced megaloblastic changes in bone marrow using folinic acid. Br J Anaesth 1984;56:103–7
28. Nunn JF, Chanarin I, Tanner AG, Owen ER. Megaloblastic bone marrow changes after repeated nitrous oxide anaesthesia: reversal with folinic acid. Br J Anaesth 1986;58:1469–70
29. Nunn JF, Sharer NM, Gorchein A, Jones JA, Wickramasinghe SN. Megaloblastic haemopoiesis after multiple short-term exposure to nitrous oxide. Lancet 1982;1:1379–81
30. O'Sullivan H, Jennings F, Ward K, McCann S, Scott JM, Weir DG. Human bone marrow biochemical function and megaloblastic hematopoiesis after nitrous oxide anesthesia. Anesthesiology 1981;55:645–9
31. Skacel PO, Hewlett AM, Lewis JD, Lumb M, Nunn JF, Chanarin I. Studies on the haemopoietic toxicity of nitrous oxide in man. Br J Haematol 1983;53:189–200
32. Myles PS, Leslie K, Chan MT, Forbes A, Paech MJ, Peyton P, Silbert BS, Pascoe E. Avoidance of nitrous oxide for patients undergoing major surgery: a randomized controlled trial. Anesthesiology 2007;107:221–31
33. Myles PS, Leslie K, Silbert B, Paech MJ, Peyton P. A review of the risks and benefits of nitrous oxide in current anaesthetic practice. Anaesth Intensive Care 2004;32:165–72
34. Clarke R, Lewington S. Homocysteine and coronary heart disease. Semin Vasc Med 2002;2:391–9
35. Nygard O, Nordrehaug JE, Refsum H, Ueland PM, Farstad M, Vollset SE. Plasma homocysteine levels and mortality in patients with coronary artery disease. N Engl J Med 1997;337:230–6
36. Kosch A, Koch HG, Heinecke A, Kurnik K, Heller C, Nowak-Gottl U. Increased fasting total homocysteine plasma levels as a risk factor for thromboembolism in children. Thromb Haemost 2004;91:308–14
37. Abahji TN, Nill L, Ide N, Keller C, Hoffmann U, Weiss N. Acute hyperhomocysteinemia induces microvascular and macrovascular endothelial dysfunction. Arch Med Res 2007;38:411–6
38. Chambers JC, McGregor A, Jean-Marie J, Kooner JS. Acute hyperhomocysteinaemia and endothelial dysfunction. Lancet 1998;351:36–7
39. Myles PS, Chan MT, Kaye DM, McIlroy DR, Lau CW, Symons JA, Chen S. Effect of nitrous oxide anesthesia on plasma homocysteine and endothelial function. Anesthesiology 2008;109:657–63
40. Badner NH, Beattie WS, Freeman D, Spence JD. Nitrous oxide-induced increased homocysteine concentrations are associated with increased postoperative myocardial ischemia in patients undergoing carotid endarterectomy. Anesth Analg 2000;91:1073–9
41. Foschi D, Rizzi A, Zighetti ML, Bissi M, Corsi F, Trabucchi E, Mezzetti M, Cattaneo M. Effects of surgical stress and nitrous oxide anaesthesia on peri-operative plasma levels of total homocysteine: a randomised, controlled study in general surgery. Anaesthesia 2001;56:676–9
42. Rao LK, Francis AM, Wilcox U, Miller JP, Nagele P. Pre-operative vitamin B infusion and prevention of nitrous oxide-induced homocysteine increase. Anaesthesia 2010;65:710–5