Introduction

Inhalation sedation is a conscious sedation technique used to manage anxiety that an individual dental patient can feel about dental treatment. During inhalation sedation a gaseous sedative is mixed with oxygen to provide relief of that anxiety. The commonest inhalational sedative agent used in the UK is nitrous oxide. Conscious sedation in the UK is defined as; 'a technique in which the use of a drug or drugs produces a state of depression of the central nervous system enabling treatment to be carried out, but during which verbal contact with the patient is maintained throughout the period of sedation. The drugs and techniques used to provide conscious sedation for dental treatment should carry a margin of safety wide enough to render loss of consciousness unlikely'.1,2,3 At present the inhalation sedation agent of choice is nitrous oxide.1,2,3 During inhalation sedation, nitrous oxide is incrementally introduced to the individual patient's response to produce a feeling of relaxation and anxiolysis.4

The success in managing dental anxiety using nitrous oxide inhalation sedation has been confirmed by many clinical studies.5,6,7,8 These studies examined the success of inhalation sedation in the management of paediatric dental cases. Successful completion of treatment was found to occur in around 84% of cases.

The use of nitrous oxide, however, is not without local and global environmental effects. Chronic exposure of medical and dental staff to nitrous oxide can lead to adverse effects.9,10 These adverse effects can be haematological,11,12 neurological,13 hepato-renal,9 reproductive14 and ophthalmological.15 Nitrous oxide is also a drug of misuse by dental professionals16 and in wider society. Globally nitrous oxide is a 'greenhouse gas'17 and in the atmosphere can combine with oxygen to produce the ozone depleting compound nitric oxide.18 It is good clinical governance to prevent unnecessary exposure of staff to nitrous oxide. In the UK, limits of exposure to staff form part of the Control of Substances Hazardous to Health Regulations (COSHH) in particular the Occupational exposure standards (OES) introduced in January 1996.19 Exposure of staff can be decreased by: efficient scavenging, suitable nosepieces, good technique to reduce mouth breathing and using well-ventilated rooms.

The increased appreciation of the difficulties associated with using nitrous oxide has led to the examination of alternative inhalation sedation agents (Table 1). The agents that have been examined are:

Table 1 A comparison of the ideal inhalation sedation agent with nitrous oxide and sevoflurane
Full size table

From these studies it might be assumed that any of the above agents might be a suitable alternative to nitrous oxide. However, isoflurane31 is too irritant and pungent for patients to tolerate; xenon is extremely expensive23 and must be used via a rebreathing circuit; and sevoflurane, while expensive, lacks equipment dedicated for its use an inhalation sedation agent.

Objective

Clinical trials are used to test the effectiveness of either a new drug, a new formulation of an existing drug or a new means of delivering a drug. Trials can be designed to establish if the effect on a particular outcome measure, perhaps a reduction in the level of cholesterol, is superior to the effect of a placebo. This is referred to as a 'placebo controlled trial'. However, it might be considered unethical to test a new drug against a placebo if an effective treatment for the particular outcome measure already exists. In a 'superiority trial' a new drug might be tested to see if it performs better than the existing therapy. In such trials a 'null hypothesis', that there is no difference between the drugs, is tested using an appropriate statistical test. The trial attempts to demonstrate that the hypothesis can be rejected, by showing a statistically significant difference. Failure to demonstrate a difference does not mean that the drugs are equivalent. Researchers may hope to demonstrate equivalence between a new drug and the standard therapy. The new drug may be cheaper or provide a second line treatment if, or when, the first line treatment has proved ineffective or poorly tolerated. It is not possible to demonstrate exact equivalence between two drugs but the aim of an equivalence trial is to demonstrate that any differences in a particular outcome lie within an accepted range known as the equivalence margin. It is crucial that equivalence trials follow exact methods to prevent false results being produced.32

In this equivalence trial, the aim was to explore if sevoflurane could be considered equivalent to nitrous oxide as an inhalation sedation agent for dental treatment in adults in terms of patient and user acceptability.

Main outcome measures

The outcome measures used to demonstrate equivalence were:

  • Effectiveness of the sedative and anxiolytic effect for patients. This was measured using visual analogue scales and a post-operative questionnaire completed by the patient

  • Apparent effectiveness of the sedative and anxiolytic effect on the patient recorded by an observer experienced in the management of patients undergoing inhalation sedation and who was blind to the type of agent used. This was recorded using visual analogue scales

  • Apparent effectiveness of the sedative and hypnotic effect on the patient and working conditions for an operator experienced in treating patients under inhalation sedation. This was recorded via sedation scores and behavioural characteristics scores

  • Physiological measurement of blood pressure, oxygen saturation, pulse, respiratory rate and bispectral index at 5 minute intervals.

On statistical advice from colleagues at the Department of Anaesthetics, Intensive Care and Pain Medicine, the 95% confidence interval was calculated for the difference in scores between the two groups of patients (one group receiving sevoflurane and the second group nitrous oxide). The primary outcome measure was the response to the question 'Did you find this form of sedation effective?', recorded on a 100 mm visual analogue scale. Assuming that a difference of less than 10 mm between the groups was unimportant, further calculations with a sample size of 20 patients in each group provided a confidence interval of -6.4 to 6.4 mm for no difference, much less than the acceptable 10 mm.

Materials and methods

Ethical approval for the study was gained from The Cardiff and Vale NHS Trust Research and Development Office and Local Research Ethics Committee.

Patient selection

The study involved the provision of conscious sedation to anxious dental patients who had been referred by either a general dental practitioner (GDP), general medical practitioner (GMP) or community dental service practitioner to the sedation suite at Cardiff University School of Dentistry. The management strategies utilised in the sedation suite include the use of behavioural management techniques, hypnosis, inhalation sedation using nitrous oxide and oxygen, intravenous sedation using midazolam, oral sedation and intranasal sedation using midazolam. Intravenous sedation using propofol administered by a consultant anaesthetist is used in a separate area as part of an MSc module in conscious sedation. Patients referred to the sedation suite vary widely in their medical status. For safety and ethical reasons only patients who met the American Society of Anaesthesiologists (ASA) rating I were included in the study. Exclusion criteria were:

Unwillingness to participate in the study

  • ASA II or greater

  • Patients taking psychoactive medication

  • Pregnancy

  • Intellectual impairment that may have prevented the patient understanding the aim of the study and therefore not being able to give informed consent

  • Inability to arrange an escorting adult

  • Participating in other studies

  • Anatomical anomalies such a nasal polyps preventing inhalation of gases

  • Allergy to sevoflurane or other volatile gases

  • Familial or personal history of malignant hyperthermia

  • Patients who had previous experience of inhalation sedation.

Although the presence of an adult escort is not mandatory for inhalation sedation using nitrous oxide, it is a recommendation of the sedation suite that patients who have not experienced inhalation sedation should attend with an adult escort. To avoid the introduction of bias this recommendation was followed for both patient groups.

Additional considerations

Malignant hyperthermia is a fulminant, life-threatening condition33 affecting between 1 in 5,000 and 1 in 200,000 of the population and is familial with a dominant pattern of inheritance. It is triggered by exposure to any of the volatile anaesthetic agents such as sevoflurane and all depolarising muscle relaxants. It is characterised by hypermetabolism, muscle rigidity, muscle injury and increased sympathetic nervous system activity. Because of this risk it was necessary to review the protocol regarding management of malignant hyperthermia34 with anaesthetic colleagues and to have a clear procedure in place should such an event occur.

Study design

The study comprised a stratified, randomised equivalence study of 40 adult patients referred to the sedation suite for dental treatment because of their dental anxiety. At their initial presentation for assessment patients were asked to complete a modified dental anxiety score (MDAS),35 a comprehensive medical history questionnaire and to answer questions about their social and dental history. Patients who required a plastic restoration in a maxillary tooth under infiltration local analgesia and inhalation sedation were selected for inclusion in the study. The rationale was that analgesia of maxillary teeth is less operator sensitive and more reproducible. The patient's blood pressure was recorded. Patients were given a full explanation of what the trial would involve. Patients were also given an information sheet about the trial to take home. This included the reassurance that should they choose to withdraw from the trial then their treatment would proceed without delay anyway. No patient opted to withdraw. Written informed consent was gained from the patient using the standard dental hospital consent form.

Sevoflurane cannot be delivered via the same dedicated machinery as nitrous oxide so a Boyle's anaesthetic machine was modified by the Clinical Engineering Department, Cardiff and Vale NHS Trust to deliver the sevoflurane. A possible complication of this was that the Boyle's machine could also deliver nitrous oxide mixed with sevoflurane. The machine was further modified to prevent the delivery of nitrous oxide in this way. The Boyle's machine was fitted with a Blease Datum anaesthetic vaporiser (Blease Datum, Beech House, Chiltern Court, Ashridge Rd, Chesham, HP5 2PX, UK). The machine was calibrated using a gas analyser and the concentration dial on the vaporiser clearly marked at 0.3% sevoflurane.

MAC is the minimum alveolar concentration (in volumes per cent) of an anaesthetic gas at 1 atmosphere absolute that prevents movement of 50% of the population to a standard stimulus.36 The effectiveness of anaesthesia and sedation is related to the partial pressure of an inhalation agent in the brain, as opposed to its alveolar concentration. MAC, because it is measurable, is widely regarded as an index of anaesthetic potency. Sevoflurane is a potent agent with a MAC of 2.0% whereas nitrous oxide is a weak agent with a MAC of 110%. Most inhalation sedation for dental treatment occurs at a concentration of nitrous oxide between 25 and 40%.37 This would equate to a concentration of sevoflurane of 0.25% and 0.4%. In order to maintain a wide margin of safety it was decided, with advice from anaesthetic colleagues, to limit the concentration of sevoflurane to 0.3% during the study.

Use of bispectral index monitoring:

Bispectral analysis (BIS) is a measure of the pharmacodynamic effects of anaesthesia and sedation on the central nervous system (CNS).38 The basis for BIS technology is the surface encephalogram that shows total brain activity as a waveform. This waveform varies during the use of sedatives and anaesthetics. The BIS index is derived from utilising a composite of multiple electroencephalography (EEG) signal processing techniques, including bispectral analysis and time domain analysis. Analysis by multivariate statistical models of EEG data from over 5,000 subjects who had received one or more of the most common hypnotic agents provided the database from which the BIS algorithm could be derived. From this a numerical figure was extrapolated and has been used as a parameter to measure the effects of sedation and anaesthesia.39,40 BIS values vary from 100 (fully awake) to 0 (flat line EEG), hence a lower BIS number would indicate reduced brain activity. From the way the BIS value is derived, it can be considered as a statistical function and as an indicator of very recent sedation.41 BIS monitoring was used during this trial to acquire additional data on each patient's response to the sedative administered.

Before each treatment session the study equipment would be set up and tested. The equipment comprised: a MDM Quantiflex nitrous oxide inhalation sedation apparatus; a modified Boyle's machine with a sample line to a gas analyser attached to ensure the correct dosage of sevoflurane; an ECG monitor; a pulse oximeter; an automatic sphygmomanometer and a BIS machine.

Screens were placed to obscure the machinery from the observer. The same observer was used for every patient seen, and the same dedicated sedation nurse assisted with each patient. When the reception staff had informed the operator of the patient's arrival a National Examining Board for Dental Nurses (NEBDN) sedation trained dental nurse would give the patient a MDAS questionnaire to complete and also a Gestalt test to perform. Gestalt is a psychological term used to describe an organised 'whole' as being greater than the sum of its parts. Gestalt tests can be used to measure recovery from anaesthesia.42

On attending the appointment the patient was asked if they were experiencing any discomfort away from the planned filling site and also if there had been any change in medical history. The planned procedure was confirmed, preoperative blood pressure taken and ECG leads attached. The patient was then reclined and the BIS and pulse oximeter leads attached. As the gas analyser made a very distinctive sound, it was turned on for every case to try and avoid bias. The type of agent used was allocated from a master list produced by a random number generator.

Inhalation sedation involves the use of 100% oxygen for an initial 2 mins and then the incremental introduction of the sedative agent titrated to the patient's response. During the trial the actions involved in administering both the real agent and the actions involved in administering the alternative agent were performed so that the observer was unclear as to which agent was being used. Topical anaesthetic gel was applied to the local anaesthetic infiltration site for 1 min and then an infiltration of 2.2 ml 4% articaine and 1:100,000 adrenaline was slowly given. After allowing 5 mins for full analgesia to take place the removal of caries and then restoration of the tooth was completed using conventional dental instruments and materials. On completion of treatment 100% oxygen was administered for 4 mins to prevent possible diffusion hypoxia. The patient was asked to complete another Gestalt test and a post-operative questionnaire. Both the observer and the operator completed questionnaires. To avoid placing undue pressure on the patient the operator would place himself outside of the patient's direct line of vision while being able to still monitor his or her wellbeing. The completed questionnaires were collected and placed in a secured folder and locked in a cabinet.

Results

Forty patients completed the clinical trial. There were 18 male patients and 22 female patients with an average age of 34 years. The age range was 18 to 62 years. On completion of the study, the anonymised raw data was entered onto an Excel spreadsheet and passed for analysis to a statistician in the Department of Anaesthetics, Intensive Care and Pain Medicine of Cardiff University.

Analysis involved chi-squared analysis for all of the question and answer sections except answer seven, 'Do you remember being told the treatment was complete for today', which was analysed using Fischer's exact test because of low numbers. The results from the visual analogue scales were analysed using the Mann-Whitney U-test.

Discussion

Analysis of the results showed that there was equivalence between nitrous oxide and sevoflurane in the domains examined in this trial (Tables 2,3,4,5,6). On reflection there is scope for further study in this area. Sevoflurane was used at a low-dose in this study and this could be increased in further studies, in conjunction with anaesthetic colleagues, while remaining well below the MAC of 2%. Another possible area of future study could be the development of machinery dedicated to the delivery of sevoflurane as a dental inhalation sedation agent.

Table 2 Responses to MDAS, observer and patient VAS questionnaire (median [IQR] range)
Full size table
Table 3 Responses to patient questionnaires. Missing data reflects non-completion of questionnaire by some participants
Full size table
Table 4 Physiological results for patients given sevoflurane
Full size table
Table 5 Physiological results for patients given nitrous oxide
Full size table
Table 6 Results of the physiological measurements
Full size table

Conclusion

Forty patients who were recruited for treatment in the study completed their planned dental procedure, for many this was their first dental experience for some time due to their anxiety. The technique of administering nitrous oxide inhalation sedation was familiar to the operator and the supporting sedation qualified dental nurses. The technique of administering small percentages of sevoflurane was less easy but by no means difficult, although the use of the gas analyser meant an additional machine to monitor for the assisting dental nurse. In a clinical trial setting in a dental teaching hospital there were fewer time pressures at each treatment episode than would be the case in primary care and so the safe delivery of an agent such as sevoflurane, using non-dedicated machinery, was achieved without any complications. The questionnaires were generally well received and only one patient failed to complete them adequately. The operator and observer both felt that the questions they were expected to answer were reasonable and within their sphere of experience. In a future study it would perhaps be better if the operator also had visual analogue scales to complete, as these proved to be a very valuable tool for recording subjective experiences. In a trial such as this the use of the Trieger test to assess recovery from conscious sedation might be regarded as being excessive but because of the lack of experience in using sevoflurane it was considered prudent to be able to assess recovery at the end of the treatment session.

The selection criteria for this type of study had to be rigid and correctly applied to avoid introducing any bias. During screening of patients to try and find a suitable participant it was not uncommon to see 10 or 12 patients before anyone who fitted the criteria was examined. For the operator, who was a part-time student working mostly in primary care, this proved worryingly time consuming. Fortunately, many patients appreciated that they were attending a dental teaching hospital and that participation in a study such as this was to be expected and so once the process was explained only two patients declined to take part.

Patients who took part in the study were accepting of the monitoring and recording equipment used and many were interested in what the equipment was there to do. This distraction provided a useful way of introducing the patient into the treatment session.

In summary, the statistical analysis and experience gained within the limits of the study showed that sevoflurane was equivalent to nitrous oxide when administered at low doses for inhalation sedation in this study.