The Narcotrend (MT MonitorTechnik, Bad Bramstedt, Germany), an electroencephalographic monitor designed to measure the effects of anaesthetics on the brain in terms of ‘depth of anaesthesia’, has recently been marketed in Europe.
Like the Bispectral Index Monitor (BIS™; Aspect Medical Systems, Natick, MA, USA), the Narcotrend performs a computerized analysis of the electroencephalogram (EEG). Narcotrend EEG processing results in an index indicating the hypnotic component of anaesthesia, the so-called Narcotrend Index (NI), ranging from 100 (fully awake) to 0 (very deep hypnosis). Alternatively the monitor provides a categorical scale (Narcotrend Stages A-F), which is more popular than the NI, but less suitable for scientific purposes (see Table 1).
In adult patients Narcotrend monitoring is associated with reduced drug consumption and faster emergence in propofol-remifentanil anaesthesia when compared to a standard clinical practice , whereas paediatric data on the impact of the NI on anaesthetic drug requirement are currently not available.
The purpose of this study in paediatric patients undergoing minor urological surgery under total intravenous anaesthesia (TIVA) with propofol and remifentanil was to evaluate the impact of NI guidance on propofol consumption as the primary end-point and on emergence times as the secondary outcome measure. We hypothesized that an NI guided TIVA technique in children would result in decreased propofol consumption and faster emergence compared to a TIVA protocol based on conventional clinical practice.
Following approval from the institutional Ethics Committee of the University of Regensburg, Germany, and in accordance with the Helsinki Declaration, informed parental consent from 30 paediatric patients, scheduled for minor urological surgery at the Children's Hospital St. Hedwig, Regensburg, Germany, was obtained. Children were considered eligible for enrolment in our study, if they were ASA physical status I or II, aged from 1 to 11 yr and without contraindications for TIVA with propofol and remifentanil. Patients were excluded from the study if they suffered from significant cardiovascular, respiratory or neurological disease or if they were taking chronic medication known to affect the central nervous system. According to hospital standards all children had a topical local anaesthetic cream (EMLA®; AstraZeneca, Wedel, Germany) applied to the skin over a visible vein on the dorsum of a hand or forearm approximately 1 h before induction of anaesthesia and 30 min later oral pre-medication (midazolam 0.5 mg kg−1; 0.5 mL kg−1 of a syrup preparation, up to a maximum dose of 15 mg) was given.
Vital signs were monitored using a three-lead electrocardiogram (ECG), pulse oximetry, and non-invasive automatic blood pressure (BP), together with measurements of inspired and end-tidal concentrations of oxygen and carbon dioxide and the NI. Immediately before the start of the induction patients were randomized by a computer generated assignment to one of two different TIVA protocols (conventional practice: Group C, n = 15; NI-monitoring guided practice (target NI 60 ± 5): Group NI, n = 15).
The induction was performed using the same technique for both study groups: after establishment of venous access a slow bolus injection (over 15 s) of remifentanil 1 μg kg−1 was given, followed by a constant infusion of 0.3 μg kg−1 min−1. One minute later, anaesthesia was induced with propofol 2 mg kg−1. After loss of consciousness (LOC) manual ventilation via face mask was established, and a laryngeal mask was inserted. During anaesthesia patients were connected to a semiclosed anaesthetic circuit (Cicero®; Draeger, Lübeck, Germany) and mechanically ventilated to normocapnia (end-tidal carbon dioxide 35-40 mmHg) with 30% oxygen in air.
From the moment of laryngeal mask insertion the conduct of the anaesthetic was different for the two study groups: in Group C the anaesthetist was unrestricted concerning decisions on propofol infusion rates and blinded to the screen of the Narcotrend monitor. The remifentanil infusion rate was set to 0.3 μg kg−1 min−1 throughout the surgical procedure until the start of skin closure or (in case of circumcision) the last 5 min of surgery in both study groups.
In the NI monitoring guided group (Group NI) propofol infusion rates were adjusted to keep the NI within a range of 60 ± 5 from laryngeal mask insertion until the beginning of skin closure.
Clinical assessment of the hypnotic level of the patient was performed according to the University of Michigan Sedation Scale for Children (UMSS)  (see Table 2) by a member of the anaesthesia team who was otherwise not involved in the conduct of this study. Under anaesthetized conditions (i.e. the pre-surgery and surgery period) patients were not touched or verbally disturbed in addition to surgical stimulation. In case of no visible reaction to surgery they were assigned to UMSS level 4 (unrousable). Patient movement in response to surgical stimuli, indicating an inadequate low anaesthetic level, as well as wake up reactions during surgery, were specifically noted.
According to departmental standards the laryngeal mask was removed ideally at the moment of return of consciousness (ROC), defined as transition from UMSS Stages 2-4 to UMSS Stage 1 or 0. Presuming adequate spontaneous ventilation and returned eyelash reflex, the laryngeal mask was removed before ROC in case of coughing or head shaking. This decision was made to reduce the risk for development of laryngospasm caused by the laryngeal mask in the throat during the critical time of the emergence period, when the child is highly irritable somewhere between unconsciousness and wakefulness. Postoperative analgesia was provided by rectal diclofenac 1 mg kg−1 and regional block as appropriate (penile nerve block, caudal block, iliohypogastric nerve block), both applied after the induction.
The Narcotrend was attached to each patient according to the manufacturers instructions. Following skin preparation with alcohol and gauze, two silver-silver chloride electrodes (Medicotest A/S, Olstykke, Denmark) were positioned on the left and right lateral parts of the patients forehead with the maximum achievable distance and a third one on the mid-forehead, serving as a referential electrode. Electrode placement and skin preparation were performed until the electrodes' impedance was less than 6000 Ω. The EEG of each patient was recorded continuously with 128 samples per second, a 0.5-Hz high pass filter, a 45-Hz low pass filter and a 50-Hz notch filter, using the Narcotrend® EEG monitor (MT, MonitorTechnik, Bad Bramstedt, Germany). Narcotrend EEG processing leads to a variable called the NI, a dimensionless scale from 0 to 100 (see Table 1). Detailed information about the development of the Narcotrend algorithm has been given by Schultz and colleagues [3,4].
A priori power calculation showed that a sample size of 14 patients in each study group would be required to detect a 20% difference in propofol consumption as the primary outcome measure at a significance level of 5% with 90% power. To compensate for possible non-evaluable dropouts, 15 patients in each group were studied. Data were analysed using SigmaStat, version 3.0 (Systat Software GmbH, Erkrath, Germany). Data were tested for normality using the Kolmogorov-Smirnov method. Groups were then compared by means of a t-test or a Mann-Whitney Rank Sum Test as appropriate. Intragroup comparison was performed using One Way Repeated Measures analysis of variance (ANOVA) or Friedman Repeated Measure Analysis on Ranks. In case of significant differences post hoc all pairwise multiple comparison procedures (Holm-Sidak and Dunn's) were performed. Times to ROC after discontinuation of anaesthetics were compared by means of a Kaplan-Meier survival analysis. Results were considered significant when P < 0.05. Data are presented as mean ± SD or median (inter-quartile range [IQR]) as appropriate.
All children completed the study. In one patient in Group C intraoperative NI data could not be recorded due to a problem with the electrodes' impedance. Study groups were not different with respect to age (Group NI: 4.2 ± 2.5; Group C: 5.7 ± 2.8 yr), weight (Group NI: 17.1 ± 5.1; Group C: 20.4 ± 8.1 kg), gender distribution (female/male ratio: Group NI: 3/12; Group C: 2/13), and duration of propofol infusion (median [IQR] Group NI: 38 [28-54]; Group C: 38 [33-68] min).
Propofol consumption (median [IQR]) was significantly lower in Group NI (7.0 [6.4-8.2] mg kg−1 h−1) than in Group C (9.3 [8.3-11.0] mg kg−1 h−1; P < 0.001), whereas the time interval between termination of anaesthetics and ROC as revealed by Log-Rank Test (see Fig. 1) was not different between study groups (Group NI: mean [95% CI] 12.8 [11.2-14.4] min; Group C: 16.4 [12.6-20.2]min). Box and whisker plots (95th, 75th, 50th, 25th, and 5th percentiles) of NI values vs. case milestones are displayed in Figure 2. Inter-group comparison showed no differences in NI values at corresponding case milestones. Individual patient data of median NI values during surgery vs. total propofol consumption and the time interval from termination of anaesthetics to ROC show a much wider distribution with outliers in Group C compared to Group NI (see Fig. 3).
Heart rate (HR) and mean arterial pressure (MAP) were not different between study groups and remained within normal age related limits throughout the study period (see Fig. 4). There were no intraoperative observations of inadequate anaesthesia or wake up reactions in any patient.
In this paediatric study propofol requirement significantly decreased when NI guidance was used compared to a conventional practice based on avoidance of clinical signs of inadequate anaesthetic depth and maintenance of haemodynamic stability. Some of our results came unexpectedly and we are unable to definitely explain them on the basis of the data obtained from this study.
We could not demonstrate a statistically significant effect of NI monitoring on emergence times (Log-Rank Test: P = 0.10). However, as visible from the Kaplan-Meier survival plot (Fig. 1) there is a tendency towards earlier ROC in Group NI.
Figure 2 shows a considerable larger variability of NI values in Group C under anaesthetized conditions compared to Group NI. However, as revealed by the Mann-Whitney Rank Sum Test there were no statistically significant differences in NI values between the study groups.
Probably we were unable to detect differences in times to ROC and in NI values between groups because the study was underpowered with respect to these secondary items - the sample size was determined to provide the study with adequate power to detect a difference in propofol consumption between the study groups as the primary outcome measure.
Haemodynamic parameters, besides visible movement reactions to surgical stimulation the most important parameters for clinical judgement of anaesthetic depth, were not different between study groups. It could be hypothesized, that despite significantly different propofol infusion rates there were no differences in anaesthetic depth between the two study groups. After all, NI values were not different between groups. On the other hand it is well known from recent studies in both adults  and children [6,7], that haemodynamic parameters are poor indicators of anaesthetic depth.
Recart and colleagues  as well as White and colleagues  recently mentioned the possibility of investigator bias if anaesthetists caring for patients according to a conventional clinical practice were previously experienced with depth of anaesthesia monitors. In our study patients in the conventional practice group were therefore anaesthetized by anaesthetists who had no previous experience with the Narcotrend or other depth of anaesthesia monitors. Patients in the NI guided group were anaesthetized by an investigator (F.W.) with longstanding experience in the use of various depth of anaesthesia monitors, among them the Narcotrend monitor. All investigators were previously experienced with propofol/remifentanil TIVA in children. However, investigator bias appears to be a problem that cannot be completely ruled out in studies like this.
Beyond the explanatory power of sophisticated statistical procedures, visual comparison of the individual patient data displayed in Figure 3 led us to the impression, that probably the hypnotic component of anaesthesia in some children of the conventional practice group was very light, as indicated by NI values of 82-87, with a total range from 22 to 87, whereas in the NI guided group there was a rather narrow distribution of NI values from 49 to 62. Despite these high NI values observed in Group C, which are more likely to indicate sedation rather than light anaesthesia (see Table 1), there were no clinical signs of inadequate anaesthesia, such as unusual haemodynamic fluctuation or movement in response to surgery. Maybe perfect analgesia from remifentanil infusion and regional block prevented the anaesthetists caring for the patients in Group C from detecting clinical signs of inadequate anaesthetic depth, whilst conducting a low-hypnotic component anaesthesia.
Furthermore the range of times to ROC was obviously smaller in Group NI (8-19 min) compared to Group C (8-32 min), indicating the usefulness of NI monitoring for an improved prediction of recovery times.
The propofol-sparing effect of NI guiding paediatric TIVA is comparable the effect of auditory-evoked potential monitoring in children using the AEP-Monitor/2 (Danmeter A/S, Odense, Denmark) with its AAI™ Index of anaesthetic depth. In a recent study, which was performed in our own institution, AAI guidance of propofol/remifentanil TIVA for paediatric strabismus repair led to a 34% reduction of propofol consumption and significantly shorter emergence times compared to a conventional clinical practice . However, caution must be taken whilst trying to compare the results of both studies: the NI and the AAI have completely different target ranges for ‘surgical anaesthesia’. We can furthermore almost be sure, that anaesthetic depth in an NI range of 60 ± 5 in this study is different from anaesthetic depth in an AAI range of 30 ± 5 in the AAI study, as indicated by different propofol consumption (NI: 7.0 [6.4-8.2] (median [IQR]); AAI: 4.2 ± 1.7 (mean ± SD) mg kg−1 h−1).
Besides preventing the patient from accidental intraoperative awareness the ultimate use of anaesthesia depth monitors is to avoid the use of excessive levels of anaesthesia, and therefore promote rapid emergence and recovery. Only in Group C there was one patient with very deep anaesthesia, as visible from a median intraoperative NI value of 22, a propofol infusion rate of 18.4 mg kg−1 h−1 and a time to ROC of 32 min.
Consequently, the results of this study do not address the overall validity of the Narcotrend in paeditric anaesthesia, but should be regarded as an initial indication for the usefulness of Narcotrend guidance of paediatric TIVA with propofol-remifentanil. NI guidance of propofol delivery results in less drug consumption when compared to a standard practice without evidence of intraoperative awareness or excessive deep levels of anaesthesia. Despite reduced propofol requirements under NI guidance we were unable to demonstrate an effect on emergence times.
The authors gratefully acknowledge Professor Wolfgang Roesch, MD, Head, Department of Paediatric Urology, Children's Hospital St. Hedwig, University of Regensburg, Germany, for his friendly cooperation during the conduct of this study.
The Narcotrend monitor was provided on loan by the manufacturer, MT MonitorTechnik, Bad Bramstedt, Germany. MT MonitorTechnik was not involved in the design, conduct or analysis of this study. The study itself was supported solely by departmental funding.
1. Kreuer S, Biedler A, Larsen R et al
. Narcotrend monitoring allows faster emergence and a reduction of drug consumption in propofol-remifentanil anesthesia. Anesthesiology
2. Malviya S, Voepel-Lewis T, Tait AR et al
. Depth of sedation in children undergoing computed tomography: validity and reliability of the University of Michigan Sedation Scale (UMSS). Br J Anaesth
3. Schultz B, Kreuer S, Wilhelm W et al
. The Narcotrend Monitor: development and interpretation algorithms [in German]. Anaesthesist
4. Schultz B, Grouven U, Schultz A. Automatic classification algorithms of the EEG monitor Narcotrend for routinely recorded EEG data from general anaesthesia: a validation study. Biomed Tech
5. Struys MM, Jensen EW, Smith W et al
. Performance of the ARX-derived auditory evoked potential index as an indicator of anesthetic depth: a comparison with bispectral index and hemodynamic measures during propofol administration. Anesthesiology
6. Denman WT, Swanson EL, Rosow D et al
. Pediatric evaluation of the bispectral index (BIS) monitor and correlation of BIS with end-tidal sevoflurane concentration in infants and children. Anesth Analg
7. McCann ME, Bacsik J, Davidson A et al
. The correlation of bispectral index with endtidal sevoflurane concentration and haemodynamic parameters in preschoolers. Paediatr Anaesth
8. Recart A, White PF, Wang A et al
. Effect of auditory evoked potential index monitoring on anesthetic drug requirements and recovery profile after laparoscopic surgery: a clinical utility study. Anesthesiology
9. White PF, Ma H, Tang J et al
. Does the use of electroencephalographic bispectral index or auditory evoked potential index monitoring facilitate recovery after desflurane anesthesia in the ambulatory setting? Anesthesiology
10. Weber F, Seidl M, Bein T. Impact of the AEP-Monitor/2-derived composite auditory-evoked potential index on propofol consumption and emergence times during total intravenous anaesthesia with propofol and remifentanil in children. Acta Anaesthesiol Scand