Significant pain or discomfort associated with the intravenous (i.v.) injection of propofol  is of concern despite a number of therapeutic stratagems. The incidence rate of pain on injection of propofol ranges from 25% to 100% [2-5] if a vein on the dorsum of the hand is used, and significantly less (3-36%) if it is injected into larger proximal veins [6,7]. Several mechanisms have been proposed to explain the pain following propofol injection [6,8]. This pain may occur immediately or be delayed within 10-20 s of injection . Immediate pain probably results from a direct irritant effect whereas the delayed pain may result from an indirect effect via the kinin cascade .
A variety of measures have been employed to reduce this pain , the most common being the administration of lidocaine as a pretreatment or as an admixture to the propofol solution itself [11-13]. The methods of action of lidocaine in this setting are possibly due to its local anaesthetic effect on the vein and the stabilization of the kinin cascade . In a systematic review, lidocaine (0.5 mg kg−1 i.v.) given with a rubber tourniquet on the forearm 30-120 s before the injection of propofol was the most efficacious treatment. Of the patients, 80-90% will not have any pain using this stratagem . Others have suggested the use of a combination of different analgesic techniques .
Nitrous oxide is a well-established analgesic agent and the inhalation of nitrous oxide - as an adjunct to i.v. induction of general anaesthesia with propofol - has been shown to be safe [15-18]. Its analgesic effects are believed to be due to its opioid nature  and ability to stimulate descending inhibitory neurons . The present study was designed to examine the analgesic effect of nitrous oxide on preventing pain during propofol injection.
Our hospital's Ethics Committee approved the study and written informed consent was obtained from all patients. The study was a prospective, randomized, observer-blinded trial and planned with a power of 0.9 to detect a 30% difference in the incidence of pain following propofol injection, with a significance level (P) of 0.05. This required the recruitment of 132 patients. All were ASA I-II undergoing general anaesthesia for elective surgery. Patients in ASA III-V and those who suffered from cardiac conduction defects, epilepsy, those taking antidysrrhythmic drugs or receiving analgesic drugs in the previous 24 h were excluded.
A member of the anaesthesia team took responsibility for anaesthesia and another to record pain following propofol injection. Patients were randomly allocated to one of three groups. Randomization was conducted by a computer with the code sealed until arrival of the patient in operating room. Patients in the control group were preoxygenated with 100% oxygen (120 s). Patients in the nitrous oxide group were preoxygenated with 50% nitrous oxide in oxygen (120 s). In both groups, anaesthesia was induced with propofol with no added lidocaine. Patients in the lidocaine group were preoxygenated with 100% oxygen and anaesthesia was induced with propofol to which lidocaine had been added . This mixture, prepared by adding 2 mL lidocaine 1% (20 mg) to 18 mL propofol (10 mg mL−1) resulted in a lidocaine concentration of 1 mg mL−1. The propofol and propofol-lidocaine mixtures were used within 15 min of preparation. The propofol administered to all groups was at room temperature. The investigator recording the pain scores was blinded to the drugs given and the gas mixture administered to the patients (flowmeters were covered by cardboard). The patients who were not premedicated had a Luer-Lock® 20-G cannula inserted into a vein on the dorsum of the hand. The cannula was flushed with 0.9% saline. No other injectate was used before induction of anaesthesia with propofol.
Routine monitoring was established before preoxygenation and arterial pressure, heart rate and SPO2 were recorded; these measurements were repeated at 3 and 5 min after the start of the propofol injection, and then every 5 min. The speed of injection was carefully controlled by hand. Propofol was injected at 1 mL s−1, the injection was stopped at 5 s and then continued after pain scores were assessed at 10 s. On initial injection (after 5 s), the degree of pain experienced by the patient was scored by observation of any verbal response and behavioural signs such as facial grimacing or arm withdrawal. At 10 s, if there was no observed pain response, the patient was asked a standard question about comfort at the injection site (Table 1). This scoring system was described by McCrirrick and Hunter . After scoring of pain intensity, induction of anaesthesia proceeded as determined by the anaesthetist in charge of the patient's care. All patients received 100% oxygen after assessment of the injection pain. Anaesthesia was subsequently maintained with an inhalational technique of isoflurane, oxygen and an air mixture. After recovery from anaesthesia, recall of pain following injection was assessed by the nurses in the recovery room.
The physical characteristics and haemodynamic variables of the three groups were assessed using the Kruskal-Wallis ANOVA. Pearson's χ2-test and Fisher's exact test assessed differences in the incidence and severity of pain on injection. A significance level (P) of 0.05 was chosen.
We recruited 140 patients. Five were excluded due to difficulty with venous cannulation. The three groups were comparable with regard to patients' age, weight and gender ratio (Table 2).
In the control group, 26 of 45 (58%) patients had pain on propofol injection compared with 11 of 45 (24%) patients in the nitrous group (95% CI: 14-52%, P = 0.001) and 11 of 45 patients (24%) in the lidocaine group (95% CI: 14-52%, P = 0.001). The distribution of pain scores in the study groups is shown in Table 3. Fifteen of 45 (33%) patients in the control group had a pain score of 2 or 3 (moderate, severe) compared with six of 45 (13%) (P = 0.04) and two of 45 (4%) (P = 0.0007) patients in the nitrous oxide and lidocaine groups, respectively. There was no significant difference between the groups (P = 0.2). The proportion of patients with recall of injection pain is shown in Figure 1.
In all groups, SPO2 increased from baseline values after 1 min of inhalation of the gaseous mixture and maintained at a median of 100% throughout the induction of anaesthesia. Haemodynamic and SPO2 changes did not differ between the three groups (Table 4).
A disadvantage of propofol is pain following injection, which can be quite distressing for patients. This kind of pain is a clinical outcome that is both common and important to avoid. When clinical importance and frequency are considered, propofol-induced pain has been ranked seventh of 33 clinical outcomes . The 58% incidence rate of pain following injection of propofol without analgesia in our control group agrees with the findings of others . Similarly, the reduction in injection pain associated with 20 mg lidocaine is similar in magnitude to that reported by King and colleagues . The decreased incidence of injection pain associated with nitrous oxide and oxygen mixture has previously not been reported.
Nitrous oxide is a useful analgesic agent  and has been in use for more than 150 yr. The administration of nitrous oxide has been investigated as an additional inhaled drug for the induction of anaesthesia and has been demonstrated to reduce the induction dose of propofol . It has also been shown not to compromise oxygenation  or alter haemodynamic variables  during the induction of anaesthesia. The short duration of breathing this gas mixture (120 s) in our study may have decreased potential risks of hypoxia, vomiting, excitement and delirium associated with nitrous oxide inhalation.
The acceptability of breathing gas mixtures of oxygen and nitrous oxide during the induction period was not studied; however, none of the study patients complained postoperatively of any unpleasant experiences. A limitation of our study is that it did not compare nitrous oxide to lidocaine given intravenously with a tourniquet in situ (the currently most efficacious treatment ). Though this was an observer-blinded rather than a double-blinded study, observer bias was unlikely to have had an effect on study outcome as observers used a standard question to assess pain.
In conclusion, we demonstrated that the administration of a nitrous oxide/oxygen mixture during the induction of anaesthesia significantly reduces the incidence of pain associated with propofol injection. This agent was equally effective as lidocaine mixed with propofol. As nitrous oxide is administered during induction of anaesthesia and has different analgesic actions to lidocaine in this setting, a combination therapeutic strategy as previously suggested  may be a useful focus in further studies.
1. Sebel PS, Londdon JD. Propofol: a new intravenous anesthetic. Anesthesiology
2. Stark RD, Binks SM, Dutka VN, O'Connor KM, Arnstein MJA, Glen JB. A review of the safety and tolerance of propofol ('Diprivan'). Postgrad Med J
3. Brooker J, Hull CJ, Stafford M. Effect of lignocaine on pain caused by propofol injection. Anaesthesia
4. McCulloch MJ, Lees NW. Assessment and modification of pain on induction with propofol (Diprivan). Anaesthesia
5. Hynynen M, Korttila K, Tammisto T. Pain on i.v. injection of propofol (ICI 35 868) in emulsion formulation [Short communication]. Acta Anesthesiol Scand
6. Scott RP, Saunders DA, Norman J. Propofol: clinical strategies for preventing the pain of injection. Anaesthesia
7. Gehan G, Karoubi P, Quinet F, Leroy A, Rathat C, Pourriat JL. Optimal dose of lignocaine for preventing pain on injection of propofol. Br J Anaesth
8. Klement W, Arndt JO. Pain on intravenous injection of some anaesthetic agents is evoked by unphysiological osmolality or pH of their formulations. Br J Anaesth
9. Briggs LP, Clarke RS, Dundee JW, Moore J, Bahar M, Wright PJ. Use of di-isopropyl phenol as main agent for short procedures. Br J Anaesth
10. Picard P, Tramèr MR. Prevention of pain on injection with propofol: a quantitative systematic review. Anesth Analg
11. Nathanson MH, Gajraj NM, Russell JA. Prevention of pain on injection of propofol: a comparison of lidocaine with alfentanil. Anesth Analg
12. Helbo-Hansen S, Westergaard V, Krogh BL, Svendsen HP. The reduction of pain on injection of propofol: the effect of addition of lignocaine. Acta Anaesthesiol Scand
13. King SY, Davis FM, Wells JE, Murchison DJ, Pryor PJ. Lidocaine for the prevention of pain due to injection of propofol. Anesth Analg
14. Tan CH, Onsiong MK. Pain on injection of propofol. Anaesthesia
15. Johnson GW, St John Gray H. Nitrous oxide inhalation as an adjunct to intravenous induction of general anaesthesia with propofol for day surgery. Eur J Anaesthesiol
16. Khoo ST, Woo M, Kumar A. Preoxygenation techniques: the value of nitrous oxide. Acta Anaesthesiol Scand
17. Carlier S, Van Aken H, Vandermeersch E, Thorniley A, Byttebier G. Does nitrous oxide affect hemodynamic effects of anaesthesia induction with propofol? Anesth Analg
18. Herregods L, Rolly G, Versichelen L, Rosseel MT. Propofol combined with nitrous oxide-oxygen for induction and maintenance of anaesthesia. Anaesthesia
19. Gillman MA, Lichtigfeld FJ. Clinical role and mechanisms of action of analgesic nitrous oxide. Int J Neurosci
20. Komatsu T, Shingu K, Tomemori N, Urabe N, Mori K. Nitrous oxide activates the supraspinal pain inhibition system. Acta Anaesthesiol Scand
21. McCrirrick A, Hunter S. Pain on injection of propofol: the effect of injectate temperature. Anaesthesia
22. Macario A, Weinger M, Truong P, Lee M. Which clinical anesthesia outcomes are both common and important to avoid? The perspective of a panel of expert anesthesiologists. Anesth Analg
23. Nathanson MH, Gajraj NM, Russell JA. Prevention of pain on injection of propofol: a comparison of lidocaine with alfentanil. Anesth Analg
24. Gillman MA, Lichtigfeld FJ. Pharmacology of psychotropic analgesic nitrous oxide as a multipotent opioid agonist. Int J Neurosci
25. Ng JM, Hwang NC. Inhaling nitrous oxide reduces the induction dose requirements of propofol. Anesth Analg