Evaluation of a new effect-site controlled, patient-maintained sedation system in dental patients*


  • *

    This study was presented in part to the 2004 Annual Scientific Meeting of the Society for Intravenous Anaesthesia (UK) in Birmingham.

Dr R. M. Chapman
E-mail: roddy.chapman@virgin.net


We have designed a new effect-site controlled, patient-maintained sedation system for delivering propofol. In the previous systems we developed, the patients retained the use of the handset throughout the procedure and were able to increase the level of sedation. However, it was found that this could potentially lead to oversedation. In the present system, the patients were able to increase their level of sedation until a level was reached that was judged by the patients as being adequate to allow them to tolerate the injection of dental local anaesthetic. The handset was then taken from the patients and the effect site concentration of propofol was maintained at that level for the remainder of the procedure. To assess its safety and efficacy, the system was used to sedate 40 patients presenting for dental procedures under sedation. The system was used successfully and treatment was completed in 39 patients. The system was found to be safe. Both surgeon and patient approval scores were high. Although this study demonstrates the efficacy of effect-site controlled, patient-maintained propofol sedation in this group of patients, further work is required to confirm its safety.

At present, patients receiving conscious sedation in the UK for dental procedures usually receive incremental doses of midazolam administered by an operator-sedationist. The safety of this regimen has been questioned in a variety of clinical situations [1–4], and an alternative propofol-based sedation system has been developed recently [5–8]. The majority of target-controlled propofol infusion systems in clinical use are plasma level controlled. It is known that the clinical effect of propofol lags behind changes in its plasma concentration, which reflects the time taken for propofol to move from the plasma compartment to its site of action for sedation or hypnosis in the brain. It is assumed that when the clinical effect of propofol has reached a plateau, the plasma and brain concentrations have equilibrated. The effect-site concentration is a mathematically derived term that corresponds with the clinical effect of propofol [9]. Patient-maintained sedation (PMS) involves the patient controlling a target-controlled infusion of propofol by operating a patient-controlled handset, and has been assessed in a number of surgical models [5–8]. This has recently been changed from controlling the target plasma concentration of propofol, to controlling the effect-site concentration (Ce) [10].

Ultimately, it is hoped that the system will be safe for use in the absence of an anaesthetist. However, in the previous pilot studies the patients retained the use of the handset throughout the procedure and were able to increase the effect site concentration of propofol throughout. It was found that in some instances this could lead to oversedation [10]. Failure to produce over sedation is a prerequisite of a system that can be used safely without the presence of an anaesthetist. In an attempt to avoid this, the system tested in this study differed from previous systems in that the patient did not have use of the handset throughout the whole of the procedure. The study investigated the effect of removing patient control of sedation during the period of the procedure in which the surgeon would no longer be focusing on the patient's sedation. The aim of this study was to determine how effective this new effect-site controlled system would be at providing sedation for dental patients.


After Local Research Ethics Committee approval and written, informed consent, 40 patients presenting for oral surgery under conscious sedation were recruited. Inclusion criteria comprised any ASA physical status 1 or 2 patients referred to the sedation unit at Glasgow University Dental School for oral surgery with conscious intravenous sedation. Exclusion criteria were significant systemic disease (ASA physical status 3–5), patients with epilepsy, and those with severe respiratory disease or an inability to use the handset. Sedation was carried out in the presence of an anaesthetist with access to full resuscitation drugs and equipment. After intravenous access was achieved and routine monitoring (ECG, automated non-invasive blood pressure and pulse oximetry) was attached, sedation was commenced. Supplementary oxygen was not given routinely; oxygen was only administered if Spo2 decreased to < 90%.

The effect-site patient-controlled sedation system used was based on the Marsh pharmacokinetic model [11] with a t½ keo of 2.6 min. The initial target effect-site concentration was set at 1.0 μg.ml−1. The system could increase the plasma concentration of propofol by up to 100% over the target effect-site concentration (‘overshoot’) and, as the target effect-site concentration was approached, the predicted plasma concentration was decreased automatically by the infusion system. The patient's handset button was enabled when the plasma concentration had fallen to within 10% of the target effect-site concentration. Once enabled, a successful double press of the button within 1 s increased the target effect-site concentration by a step of 0.2 μg.ml−1. Patients were instructed to press the button (as in previous studies) until a level of sedation was achieved in which they felt ready to tolerate the injection of dental local anaesthetic – the patients were asked: ‘Do you feel ready for the local anaesthetic injection or do you want to feel a bit more relaxed first?’ When they indicated that they felt ready for the injection, the handset was then taken from them and disconnected from the infusion pump. After disconnection of the handset, the system automatically maintained the effect-site concentration at that level for the remainder of the procedure.

The patients were monitored continuously throughout the procedure. Records were made of heart rate, blood pressure, Spo2, respiratory rate, calculated effect-site concentration and level of sedation (Table 1) at the following times: before sedation, at 5-min intervals during sedation and before discharge from the recovery area. In addition, the lowest and highest values for each parameter were noted. Any interventions performed by the anaesthetist were also noted. Patients completed a post-procedure questionnaire in which they were asked to rate their overall experience using a four-point scale (1 – no recall, 2 – not unpleasant, 3 – unpleasant, 4 – very unpleasant). The surgeon also completed a questionnaire in which both ease of procedure, i.e. the quality of the operating conditions produced by the sedation (1 – easy, 2 – mild difficulty, 3 – difficult, 4 – extremely difficult), and patient co-operation (1 – very good, 2 – good, 3 – average, 4 – minimal) were scored using a four-point scale. The main outcome measures in the study were lowest oxygen saturation without supplementary oxygen, and patient and surgeon scores as assessed by the post-procedure questionnaires.

Table 1.  Sedation Scale adapted from the Observers Assessment of Alertness Sedation Score [17].
5Responds readily to nameNormal
4Lethargic response to nameMild slowing
3Responds to name only if called repeatedlySlurring
2Response only after mild proddingNot recognisable
1No response to mild prodding or shaking 


Ten male and 30 female patients were recruited; their mean (SD) age was 32.4 (12.7) years. All were ASA physical status 1 or 2. The dental procedures being carried out were 37 third molar extractions, two removals of alveolar cysts and one excision biopsy of an oral lesion. Of the 40 patients recruited, 39 completed their treatment satisfactorily and achieved an adequate level of sedation as defined by the term conscious sedation[12]. The mean (SD) [range] final effect-site concentration reached before local anaesthetic injection, and continued throughout the procedure, was 1.5 (0.4) [1.0–2.5]μg.ml−1.

Heart rate, respiratory rate, and systolic and diastolic arterial pressures all remained stable within 20% of baseline measurements. The mean (SD) of the patients' lowest oxygen saturations was 96 (2.4)%. The lowest arterial oxygen saturation for each patient is shown in Fig. 1. No patient required emergency intervention in the form of supplementary oxygen or airway manoeuvres. Times from start of sedation to discharge are shown in Table 2.

Figure 1.

Histogram showing the lowest recorded oxygen saturations.

Table 2.  Times recorded during the study. Values are mean (SD) [range].
Time from start of sedation to start of procedure; min8.9 (4.8) [2–20]
Duration of procedure; min15.6 (7.9) [1–39]
Time from start of sedation to discharge; min39.5 (9.6) [20–63]
Time from end of procedure to discharge; min15.1 (4.8) [3–25]

The group as a whole spent a mean (SD) of 55.7 (37.7)% of the procedure time at sedation level 5 (see Table 1). Twenty-four patients (60%) reached a sedation level of 4, and spent a mean (SD) of 73.5 (11.3)% of their time at this level. Only one patient reached a sedation score of 3, and even then only spent 8% of the procedure at this level.

Thirty-nine of the patients used the system satisfactorily to maintain a level of sedation that was adequate for their comfort and the completion of the surgical procedure. Of the 40 patients who completed a postoperative questionnaire, 39 were happy with the sedation they received and would have the same system again. The patient who did not complete the study had been extremely anxious before the procedure and initially refused to proceed. After counselling, she received sedation and the dental local anaesthetic but refused to proceed with surgery; she was subsequently referred to have the surgery carried out under a general anaesthetic. No patients reported nausea at any stage. Pain on injection of propofol, the patients' recall of, and satisfaction with, the procedure are detailed in Table 3. The surgeons' rating of the ease of procedure and overall patient co-operation are shown in Table 4.

Table 3.  Results of the patients' postoperative questionnaires. Values are number (%).
Pain on infusion of propofol; n7 (17.5%)
Recall of injection of local anaesthetic; n33 (82.5%)
Recall of surgery; n33 (82.5%)
Pain with local anaesthetic injection
 No recall8 (20%)
 Not unpleasant18 (45%)
 Unpleasant11 (27.5%)
 Very unpleasant3 (7.5%)
Overall score
 No recall3 (7.5%)
 Not unpleasant35 (87.5%)
 Unpleasant2 (5%)
 Very unpleasant0 (0%)
Table 4.  Results of the surgeons' postoperative questionnaires. Values are number (%).
Ease of procedure
 Easy10 (25%)
 Mild difficulty22 (55%)
 Difficult8 (20%)
 Very difficult0 (0%)
Patient co-operation
 Very good35 (87.5%)
 Good1 (2.5%)
 Average3 (7.5%)
 Minimal1 (2.5%)


Propofol target-controlled infusion systems have been used by anaesthetists for sedation in a variety of clinical situations [13]. Low-dose propofol has been shown to have satisfactory anxiolytic properties when used as a sedative [14], and has been evaluated in a number of intermittent bolus and infusion regimes [13, 15, 16]. The rapid onset and offset of propofol gives it advantages over midazolam as a sedative agent. In direct comparison, it has been shown to allow more accurate titration, quicker discharge, superior anxiolysis, less psychomotor impairment and fewer side-effects such as arterial oxygen desaturation [8].

This study has provided useful practical data on the efficacy of the new effect-site controlled patient-maintained sedation system. The system was used successfully and treatment was completed in 39 patients. In this study, patients were ready for local anaesthetic injection after a mean of 8.9 min, compared to 13 min for those using the plasma-controlled patient-maintained sedation system in a previous study [8]. The effect-site algorithm allows the user to increase sedation more rapidly than in previous models [7, 8] although it uses the same basic pharmacokinetic model.

The system was found to be safe within the context of this study in that no patient suffered significant oxygen desaturation (Spo2 < 90%) requiring supplementary oxygen. All patients were sedated to a level in keeping with the definition of conscious sedation[12]. In particular, no patients were oversedated (sedation score = 2), and all maintained verbal contact throughout the procedure.

In general, patient satisfaction was high and the vast majority would be happy to use the system again if they require further treatment. The one patient who did not complete the study did receive sedation and injection of the dental local anaesthetic. However, the procedure had to be abandoned due to the unwillingness of the patient to proceed beyond the local anaesthetic injection. It was therefore a failure of sedation in general and not a specific failure of this sedation system. This patient was a dental phobic, and required a general anaesthetic rather than conscious sedation to have the procedure performed.

The surgeon rated patient co-operation as very good in 87.5% of cases, which is very encouraging. As in previous studies with patient-maintained sedation in dental patients, it was found that most patients had recall of parts of the procedure [8]. However, as in previous studies, this does not appear to have had a negative effect on the patients' perception of the system, as evidenced by the fact that all 39 patients who completed the study expressed a willingness to repeat the experience if necessary.

The same system has been tested on volunteers without removing the patient control handset, thus allowing the volunteers continually to titrate the effect-site concentration until the end of the study period. In this artificial situation in which volunteers deliberately attempted to oversedate themselves, the system proved to be unsatisfactory, with five of the 20 subjects reaching potentially unsafe levels of sedation [10]. However, when used in the routine clinical setting of this study, with the effect-site concentration fixed after the most stimulating event (local anaesthetic injection), it appears to provide safe and reliable sedation.

In previous clinical studies with patient-maintained sedation, the patients retained the use of the handset and were able to increase the effect-site concentration of propofol throughout the procedure [6–8]. It was found that in some instances at high effect-site levels this could lead to the patient becoming uncooperative, which could impair the utility of the system. The patient-maintained sedation system used in this study allowed the patient to increase the effect-site concentration of propofol until a level was reached that was judged by the patient as adequate to tolerate the injection, which is probably the most anxiety-provoking part of the procedure. After this injection, analgesia should be provided by the local anaesthetic itself, and anxiolysis was continued by propofol which was maintained at the pre-injection effect-site concentration. Previously, when allowed to titrate their plasma propofol level, patients have been shown to increase the level received between local anaesthetic injection and the end of the procedure. This fixing of the effect-site concentration should therefore considerably improve the safety of the system.

The results of this study are encouraging and although it is too early to recommend this system for use without the presence of anaesthetist, it does appear to be a safe system in routine clinical practice.