Ceasing routine use of nitrous oxide—a follow up

M. Enlund, L. Edmark and B. Revenäs

Department of Anaesthesia and Intensive Care, Central Hospital, SE-721 89 Västerås, Sweden

{dagger}Conflict of Interest: Dr Enlund is currently by part a paid consultant for Aneo AB, Marsta, Sweden, and is also a clinical investigator for Hudson RCI AB, Upplands Vasby, Sweden. There are no conflicts of interest for the two co-authors.

Accepted for publication: January 10, 2003


    Abstract
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 Appendix 1
 Appendix 2
 References
 
Background. The role of nitrous oxide in modern anaesthesia is questioned. The routine use of nitrous oxide was almost completely stopped in our department after November 1, 2000, and we now report some consequences.

Methods. Staff completed a questionnaire after 6 months, and we analysed the use of hypnotics and opioids after 12 months. The cost of drugs for the year after stopping nitrous oxide was compared with the cost 2 yr before.

Results. Less than half of the 55 staff members who answered the questionnaire used nitrous oxide in the 6 months after the stop, and they did so on only a few occasions. Half of the staff members thought the benefit of nitrous oxide was small. Most supported the change. The use of opioids was stable during the study period, and there was an annual increase of 12–14% in the use of hypnotics during the 3 yr.

Conclusions. The staff questionnaire showed a strong acceptance of the new policy, and the use of other anaesthetic agents did not increase as expected. Has the value of nitrous oxide been overestimated?

Br J Anaesth 2003; 90: 686–8

Keywords: anaesthetics gases, nitrous oxide; anaesthetics volatile, sevoflurane; anaesthetics i.v., propofol; anaesthesia, audit


    Introduction
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 Appendix 1
 Appendix 2
 References
 
Nitrous oxide is a weak anaesthetic (MAC 105%) with a number of side effects (Table 1). Its role in modern anaesthesia is disputed, as there are alternatives available. In our department we stopped routine use of nitrous oxide after November 1, 2000. This discontinuation was not total. Occasionally, nitrous oxide can be useful, and Caesarean sections, and some procedures in Ear-Nose-and-Throat, especially in the case of an adherent tympanic membrane, are excluded from the general discontinuance.


View this table:
[in this window]
[in a new window]
 
Table 1 Potential side effects of nitrous oxide
 
We assumed that when the routine use of nitrous oxide was stopped, the use of opioids and/or hypnotics in anaesthesia would increase to compensate. We had already stopped using nitrous oxide for intra-abdominal surgery in the mid-1980s. This was done to avoid distension of the bowel, because of the property of nitrous oxide to expand gas-containing cavities in the body. We found that stopping the use of nitrous oxide for these procedures was easier than expected. The quality of anaesthesia was unchanged (unpublished observations), and the surgeons appreciated the improved surgical conditions. We also omitted nitrous oxide at some minor anaesthetizing locations away from our Central Surgical Department, which had a limited anaesthesia service (Eye Surgery and Dentistry, including Maxillofacial Surgery), with good results. However, it was not evident if a general stop would be successful, involving all staff members rather than a few enthusiasts. Therefore, we decided to conduct an anonymous staff questionnaire. We also planned to measure the changes in the use of other anaesthetics. No other major changes in anaesthesia practice, or in the types of patient treated were observed over the 3 yr.


    Methods and results
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 Appendix 1
 Appendix 2
 References
 
Anonymous questionnaire
Staff completed an anonymous questionnaire in May 2001, 6 months after stopping the routine use of nitrous oxide (Appendix 1). Ninety-two per cent (55/60) of questionnaires were completed.

More than half of the staff did not use nitrous oxide at all during the 6-month period, and another third used the drug three times or less. Of those who used nitrous oxide for other purposes than the mentioned exceptions, 48% judged the benefit of nitrous oxide use was minor. A majority, 93%, thought that the new strategy should continue (Appendix 1).

Drug expenditure
We were supplied with statistics from the Hospital Pharmacy, the Hospital Service Division, and our own records of surgical and anaesthetic activity for the period from 2 yr before the stop and 1 yr after. We gave general anaesthesia (GA) to 8744, 8750, and 8509 patients annually during the study period with an annual total of 13 895, 14 724, and 14 750 h, respectively. Drug dosage was related to the number of anaesthesia hours month by month for hypnotics (propofol and sevoflurane) and opioids (alfentanil, fentanyl, and remifentanil). It was not possible to separate when sevoflurane or propofol was used. Therefore, the expenditure of each hypnotic was divided by the total number of hours. The use of alfentanil and remifentanil was measured as fentanyl equivalents to allow comparison (Appendix 2).

We found a steady increase in the annual expenditure on hypnotics from November 1998 to November 2001 (Table 2). Before the discontinuance there was a 14% increase in hypnotic consumption, and the year after there was a 12% increase. For the three opioids, expressed as fentanyl equivalents, use was stable over the years, as was the proportion of intra-operative regional blocks combined with GA (Appendix 2 and Table 2). The obvious decrease in nitrous oxide expenditure is clear (Table 2).


View this table:
[in this window]
[in a new window]
 
Table 2 The proportion of general anaesthetics combined with an intraoperative regional block, and the expenditure on hypnotics (propofol and sevoflurane), opioids (alfentanil, fentanyl and remifentanil, expressed as fentanyl equivalents), and nitrous oxide together with the reported incidence of critical incidents (per thousand anaesthetics) during the 2 yr before and the year after discontinuing the routine use of nitrous oxide
 
Outcome data
There was one case of awareness in 1999, none thereafter. Overall, the incidence of critical incidents decreased over the study period (Table 1). The data for 30-day mortality or morbidity rate was inconclusive because of unreliable data sampling in the different surgical departments during those years. We did not investigate at this time whether our surgeon colleagues observed the change in anaesthetic policy. We knew that when we discontinued nitrous oxide in intra-abdominal surgery, the surgeons appreciated this measure. Recovery Ward staff reported that the incidence of PONV decreased. However, this was not properly investigated, and it might be attributable to more use of propofol compared with sevoflurane (not in Table 2). We did not try to relate the time patients spent at the Recovery Ward to the change in policy. Too many factors besides the anaesthetic drugs influence this measure.


    Comment
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 Appendix 1
 Appendix 2
 References
 
The increase in opioids and hypnotics use necessary to replace nitrous oxide could be considerable according to published data. For every 10% of nitrous oxide administered a reduction of 0.1 MAC of an inhaled hypnotic would be anticipated.1 A 25% reduction of the intravenously administered hypnotic propofol is possible when 67% nitrous oxide is used.2 However, expenditure on anaesthetic agents did not increase as expected. The expenditure of hypnotics increased, but no more than the increase measured the year before, an increase for which we have no obvious explanation. Expenditure on opioids was quite stable over the periods.

We conclude that the staff accepted the withdrawal of the routine use of nitrous oxide, and this had surprisingly little effect on the expenditure on hypnotics and opioids. Has the role of nitrous oxide been overestimated?3 4


    Acknowledgements
 
Pharmacy Manager Anders Wickberg, Hospital Pharmacy, Central Hospital, Västerås, Sweden provided pharmacy statistics. Service manager Gunnar Larsson, Hospital Service, Central Hospital, Västerås, Sweden provided nitrous oxide statistics. Consultant Anaesthetist Hans Rudstam, currently at the National Board of Health and Welfare, Stockholm, Sweden provided anaesthesia statistics.


    Appendix 1
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 Appendix 1
 Appendix 2
 References
 
Questionnaire results given 6 months after stopping the routine use of nitrous oxide. The response rate was 92% (55/60).

How often have you used nitrous oxide after stopping the routine use (besides C-section and the ENT-procedures in which nitrous oxide is in routine use)?

Never 1–3 times 3–5 times >5 times

51% 36% 5% 8%

Was the effect from nitrous oxide as you expected it?

Yes, no doubt Yes, reasonable Doubtful No

52% 31% 17% 0%

Will there be situations when nitrous oxide is indispensable?

This question was directed to those who did not use nitrous oxide during the study period.

Yes, no doubt Doubtful No

6% 58% 33%

Shall we continue with the new policy, that is, limit the use of nitrous oxide to a few strict indications and at occasional attempts?

Yes No Do not know Blank

93% 0% 5% 2%


    Appendix 2
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 Appendix 1
 Appendix 2
 References
 
Opioid potencies (converted from plasma concentration to dose)
Plasma concentrations of fentanyl 3.3 ng ml–1 and alfentanil 279 ng ml–1 provide equivalent analgesia (EC50 for skin incision in combination with nitrous oxide),5 6 while plasma concentrations of fentanyl 2.1 ng ml–1 and remifentanil 4.8 ng ml–1 are equivalent (in combination with TCI propofol 2.5 µg ml–1).7 The pharmacokinetic program TIVA trainer©8 was used to simulate the above-mentioned concentrations of opioids for a 60 min infusion to a 70 kg patient. The calculated drug given was then used to calculate the dose equivalents of fentanyl and alfentanil, and fentanyl and remifentanil, respectively. A factor of 1:24.7 and 1:2.5 was found when calculating the fentanyl equivalent for alfentanil and remifentanil, respectively. The actual hourly expenditure of each opioid is shown down.

Fentanyl Alfentanil Remifentanil

mg h–1 mg h–1 mg h–1 x10

Two years before 0.100 0.68 0.73

The year before 0.094 0.58 0.30

The year after 0.110 0.60 0.83


    References
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 Appendix 1
 Appendix 2
 References
 
1 Nakahara T, Akazawa T, Nozaki J, Sasaki Y. Effect of nitrous oxide at sub-MAC concentrations on sevoflurane MAC in adults. Masui 1997; 46: 607–12[Medline]

2 Davidson JA, Macleod AD, Howie JC, White M, Kenny GN. Effective concentration 50 for propofol with and without 67% nitrous oxide. Acta Anaesthesiol Scand 1993; 37: 458–64[ISI][Medline]

3 Schwilden H, Schuttler J. 200 years of nitrous oxide (laughing gas) and the end of an era? Anasthesiol Intensivmed Notfallmed Schmerzther 2001; 36: 640

4 Ropcke H, Wirz S, Bouillon T, Bruhn J, Hoeft A. Pharmaco dynamic interaction of nitrous oxide with sevoflurane, desflurane, isoflurane and enflurane in surgical patients: measurements by effects on EEG median power frequency. Eur J Anaesthesiol 2001; 18: 440–9[CrossRef][ISI][Medline]

5 Glass PS, Doherty M, Jacobs JR, Goodman D, Smith LR. Plasma concentration of fentanyl, with 70% nitrous oxide, to prevent movement at skin incision. Anesthesiology 1993; 78: 842–7[ISI][Medline]

6 Ausems ME, Hug CC Jr, Stanski DR, Burm AG. Plasma con centrations of alfentanil required to supplement nitrous oxide anesthesia for general surgery. Anesthesiology 1986; 65: 362–73[ISI][Medline]

7 Vuyk J, Mertens MJ, Olofsen E, Burm AG, Bovill JG. Propofol anesthesia and rational opioid selection: determination of optimal EC50-EC95 propofol-opioid concentrations that assure adequate anesthesia and a rapid return of consciousness. Anesthesiology 1997; 87: 1549–62[CrossRef][ISI][Medline]

8 Engbers F, Kenny G, Sutcliffe N and the European Society for Intravenous Anaesthesia. www.eurosiva.org