Published online August 31, 2007
PEDIATRICS Vol. 120 No. 3 September 2007, pp. e461-e470 (doi:10.1542/10.1542/peds.2006-2577)
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Malviya, S.
Right arrow Articles by Friesen, R. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Malviya, S.
Right arrow Articles by Friesen, R. H.
Related Collections
Right arrow Surgery
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Facebook   Add to Reddit   Add to Technorati   Add to Twitter  
What's this?

ARTICLE

Effect of Age and Sedative Agent on the Accuracy of Bispectral Index in Detecting Depth of Sedation in Children

Shobha Malviya, MDa, Terri Voepel-Lewis, MSN, RNa, Alan R. Tait, PhDa, Mehernoor F. Watcha, MDb, Senthilkumar Sadhasivam, MDc and Robert H. Friesen, MDd

a Section of Pediatrics, Department of Anesthesiology, University of Michigan Health Systems, Ann Arbor, Michigan
b Department of Anesthesiology, Texas Children's Hospital, Houston, Texas
c Department of Anesthesiology, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio
d Department of Anesthesiology, University of Colorado School of Medicine, Children's Hospital, Denver, Colorado


   ABSTRACT
TOP
 ABSTRACT
PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
OBJECTIVE. This study evaluated age- and sedative agent–related differences in bispectral index across observed sedation levels in a large sample of children <18 years of age.

PATIENTS AND METHODS. With institutional review board approval and waiver of consent, data from 4 independently conducted studies were combined in a secondary analysis of 3373 observations from 248 children aged 1 month to 18 years. In these studies, bispectral index values of sedated children were recorded in a blinded fashion, and sedation depth was scored using the University of Michigan Sedation Scale (UMSS). Bispectral index was evaluated across UMSS scores for several age groups and during use of each sedative agent (with/without opioids).

RESULTS. There was a moderate inverse correlation between bispectral index and UMSS for all age groups. There were significant differences in bispectral index across UMSS and between each sedation level except UMSS 3 to 4 in all the age groups and UMSS 0 to 1 in infants. The mean bispectral index and the cutoff values on the receiver-operating-characteristic curve for mild, moderate, and deep sedation were significantly lower in infants ≤6 months compared with older children at each sedation level. Bispectral index was reasonably sensitive and specific in differentiating mild (UMSS 0–1) from deeper (UMSS 3–4) levels of sedation but poorly differentiated between moderate and deep levels of sedation in all age groups. There was a moderate correlation between bispectral index and UMSS during the use of chloral hydrate, pentobarbital, propofol, and midazolam but poor correlation during ketamine or opioid use. Bispectral index values were significantly lower during deep sedation with propofol and pentobarbital compared with midazolam and chloral hydrate.

CONCLUSIONS. Our findings suggest that, although bispectral index may differentiate light from deep sedation in most children, bispectral index must be interpreted cautiously in sedated children, with particular consideration given to patient age and use of sedative agents.

Key Words: bispectral index • sedation depth • children • sedation • University of Michigan Sedation Scale

Abbreviations: BIS—bispectral index • EEG—electroencephalograph • UM—University of Michigan • UC—University of Colorado • CHOP—Children's Hospital of Philadelphia • UMSS—University of Michigan Sedation Scale • CI—confidence interval • ROC—receiver operating characteristic

Investigators have recently suggested that bispectral index (BIS), which correlates inversely with observed sedation depth across populations,1–18 may provide an objective, clinically useful tool to assess sedation depth in children.13,18,19 Health care professionals have expressed interest in adopting this technology into practice outside of the operating room, including the ICU and during sedation for procedures. However, pediatric studies have suggested that BIS values in infants and young children may be different compared with older children and adults at similar anesthetic concentrations or depth of sedation.11,20–23 Furthermore, several studies suggest that BIS may be less reliable in detecting sedation depth during the use of certain sedative and anesthetic agents, suggesting that BIS may be drug dependent.15,17,24–26

Because BIS represents a complex parameter based on electroencephalograph (EEG) signal processing algorithms derived from adult data, maturational differences in the EEGs of infants and young children27 may indeed affect BIS interpretation. It is important to establish the validity of BIS in the assessment of the depth of sedation in various pediatric age groups to ensure appropriate use of this technology in clinical practice.28 Although some studies have demonstrated age-related differences in BIS in anesthetized children, variations in the classification of age groups (eg, <3 years, <1 year, and < 6 months), differences in anesthetic techniques and conflicting results confound the interpretation of the results.11,21,22 These studies have also focused primarily on differences in BIS at similar concentrations of inhaled anesthetics.20,25,26 There are limited data regarding the relationship between BIS and age in sedated children when the level of responsiveness is used to assess the depth of sedation.

In addition, it is imperative to understand the influence of commonly used sedative and analgesic agents on the relationship between BIS and observed sedation depth if BIS is to be appropriately interpreted during the care of sedated patients. Because BIS primarily reflects cerebral activity, only those agents that depress the level of consciousness by actions on cortical activity should affect BIS values.29 As Sleigh and Barnard have suggested, "knowledge of its boundaries and constraints is essential to the safe use of any monitor."29 Previous studies have provided preliminary data alluding to potential constraints of BIS monitoring during the use of certain sedatives or in certain age groups. Yet, these studies were limited by insufficient power to carefully examine such limitations. Therefore, data from 4 previous studies conducted in a similar manner were combined to yield a sample size sufficient to evaluate the relationship among BIS, age, and sedative agent. The overall objective of this secondary analysis was to evaluate age- and sedative agent–related differences in BIS across observed sedation levels in a large sample of children. This study was designed to test the hypotheses that, at observed depths of sedation, the BIS values in infants and older children will differ significantly, and changes in BIS values across observed sedation depths will depend on the sedative agent used.


   PATIENTS AND METHODS
TOP
 ABSTRACT
 PATIENTS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
With exemption or approval and waiver of consent from the institutional review boards at University of Michigan (UM), University of Colorado (UC), and the Children's Hospital of Philadelphia (CHOP), respectively, raw data from 4 independently conducted studies15,16,22,30 were combined to increase the power of this secondary analysis. These initial studies were performed with written informed consent of the parents of children aged 1 month to 18 years who received a variety of sedative agents for diagnostic or therapeutic procedures in accordance with institutional protocols and guidelines. Disposable pediatric or adult BIS sensors were applied to the foreheads of all of the children in accordance with the manufacturer's instructions, and data were recorded using the A-2000 BIS monitor (software 3.12–3.21, BIS algorithm 3.4, Aspect Medical Systems, Inc, Newton, MA). In each of these studies, children were similarly observed at designated time intervals and the depth of sedation evaluated using the 5-point University of Michigan Sedation Scale (UMSS; where 0 = awake/alert; 1 = minimally sedated; 2 = moderately sedated; 3 = deeply sedated; and 4 = unarousable).31 This observational tool has been shown to possess excellent criterion validity, correlating well with the Observer's Assessment of Alertness/Sedation Scale,30,31 as well as construct validity in children from birth to 18 years old.16,22,30,31 Furthermore, interobserver reliability has been well established in several studies, including 3 of the 4 used in these analyses.16,17,22,30,31 BIS values corresponding with each observation time were independently recorded in a blinded fashion. Only BIS data with a signal quality index >50 were included in the analysis. The following data were also collected: demographics including age, type of procedure, and types of sedative and analgesic agents administered.

Statistical Analysis
Data were analyzed with SPSS 13.0 (SPSS, Inc, Chicago, IL). Nonparametric tests were used for these analyses, because Kolmogorov-Smirnov tests indicated that the BIS variable was not normally distributed at each sedation depth. Spearman's {rho} coefficients were used to evaluate the correlations between BIS and UMSS. Mann-Whitney U tests were used to compare BIS values between age and sedative/analgesic agent groups. Kruskal-Wallis 1-way analyses of variance with repeated measures were used to compare BIS values between observed sedation levels (ie, UMSS score), followed by posthoc Mann-Whitney U tests with Bonferroni corrections for multiple comparisons. These tests were performed for each age and sedative agent group.

A linear mixed model was used to investigate the possibility that within-subject correlation in the repeated BIS measurements recorded for each child may have affected the statistical analyses in some way.32 This model can accommodate missing data in a repeated-measures design (as long as data are missing at random), a lack of balance in the data set (eg, not all the children were measured at the same levels of UMSS the same number of times), and within-subject correlation of the repeated measures on the dependent variable (ie, via the inclusion of random child effects in the model). Specifically, a linear mixed model was fitted to the data, including random child effects (to model within-subject correlation of the dependent BIS measures) and fixed effects associated with UMSS, age group, and sedative agent, in addition to fixed effects associated with the 2-way interactions between UMSS and age group (to investigate the possibility that age group differences are not fixed across levels of sedation) and between age group and sedative agent. The interaction between age group and sedative agent was not found to be significant and was dropped from the final model reported. Assumptions of normality and constant variance for the residuals based on the linear mixed model were investigated and confirmed after fitting the final model.

Data were further analyzed based on the clinically important end points of mild sedation or discharge readiness (ie, UMSS 0–1), moderate sedation (UMSS 2), and deep sedation (ie, UMSS 3–4), in keeping with the definitions of the American Society of Anesthesiologists and the American Academy of Pediatrics.16,22 To evaluate the accuracy of BIS in predicting sedation depth, receiver operating characteristic (ROC) curves were generated for various age groups for dichotomized UMSS scores at each of these sedation depths. ROC curves describe the performance characteristics of the test (ie, BIS) in predicting the outcome (ie, sedation depth). Different points on the ROC curve represent different cutoff points, each balanced between maximizing sensitivity and specificity.33 A larger area under the curve represents a "better" test, with areas under the curve of 0.75 to 0.92 indicating a "good" test. P values <.05 (with appropriate corrections for multiple comparisons) were accepted as statistically significant. Correlations of 0.26 to 0.49 were considered low, 0.5 to 0.69 were considered moderate, 0.7 to 0.89 were considered high, and 0.9 to 1.00 were considered very high.34


   RESULTS
TOP
 ABSTRACT
 PATIENTS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Data from 3373 observations obtained in 248 children from birth to 18 years of age were received and combined for this study. For the purpose of clarity, the results are presented as analyses of age and BIS and as analyses of sedative agents and BIS.

Analyses of Age and BIS
Data analysis for the relationship between age and BIS was based on 983 observations obtained in 128 children after excluding children who received opioids or ketamine, because previous studies5,10,24,35–37 and data from part 2 of these results suggest that these drugs have a potentially confounding effect on BIS values. This sample included 411 observations in 42 children (aged 2.7 ± 4.6 years) from UM; 207 in 17 children (2.8 ± 2.6 years) from CHOP; and 365 in 69 (4.7 ± 4 years) from UC. A total of 440 observations (45%) were recorded during sedation with chloral hydrate, 355 (35%) during sedation with midazolam, 144 (15%) during sedation with pentobarbital, and 44 (5%) during sedation with propofol.

Analyses were performed for several select age groups based on the current knowledge of EEG development in infants,27,38 previously defined groups,22 and exploratory analyses. The sample included 168 observations (17%) from infants ≤6 months of age, 149 (15%) in infants aged >6 to 12 months, and 666 (68%) in children >12 months of age. Table 1 presents the distribution of BIS values across UMSS scores for each age group. There were moderate and significant inverse correlations between BIS and UMSS scores for infants ≤6 months ({rho} = –.63; 95% confidence interval [CI]: –.53 to –.72]), those >6 to 12 months ({rho} = –.6; 95% CI: –.49 to –.70]), and for children >12 months ({rho} = –.69; 95% CI: –.65 to –.73]). BIS values were not significantly different between age groups for UMSS scores of 0 but were lower in children ≤6 months of age compared with older age groups for UMSS scores 1, 2, 3, and 4, respectively. There were no significant differences in BIS values between the 2 older age groups at any level of sedation. Kruskal-Wallis analysis of variance demonstrated significant differences in BIS across all UMSS comparisons in each age group (P < .05). Furthermore, posthoc individual comparisons found significantly different BIS values between each sedation level except between UMSS 3 vs 4 in all the age groups and UMSS 0 vs 1 in infants >6 months to 12 months of age.


View this table:
[in this window]
[in a new window]

  		TABLE 1 BIS Values for Each Sedation Level

 
The mixed-model analyses found a significant interaction between UMSS and age group (F8,948.73 = 4.705; P < .001; Satterthwaite approximation used for degrees of freedom). Investigation of estimated marginal BIS means based on the final model, which included fixed effects of UMSS, age group, sedative agent, and the 2-way interaction of interest, confirmed that the mean BIS values were identical in 3 age groups at UMSS 0 but significantly lower for the younger age group compared with the older 2 age groups at every subsequent level of sedation (even after a conservative Bonferroni adjustment to the significance values for the multiple comparisons of means within each sedation level).

Sensitivity and Specificity of BIS in Predicting Sedation Depth
The sensitivity and specificity (ie, accuracy) of BIS in predicting sedation depth was examined for each age group at each level of sedation and at clinically important end points of discharge readiness (ie, UMSS 0–1)16 and deep sedation (ie, UMSS 3–4).22 These data are described in detail in Table 2, and those for discharge ready and deep sedation cutoffs are depicted in Fig 1. These analyses indicate that BIS is reasonably sensitive and specific in differentiating 0 and 1 from deeper levels of sedation and UMSS 3 and 4 (combined) from lighter levels of sedation in all age groups. However, BIS poorly differentiates between moderate and deep levels of sedation. BIS cutoff values for mild sedation (ie, UMSS 1) were substantially lower for infants ≤6 months (ie, BIS > 68) compared with the older age groups (BIS >83–84). Similarly, BIS cutoff values for deep sedation (UMSS 3–4) were lower (≤62) for infants compared with the older groups (≤77). There were no differences between the 2 older age groups in BIS cutoff values.


View this table:
[in this window]
[in a new window]

  		TABLE 2 Optimal BIS Cutoff Values, Sensitivity, and Specificity for Each Level of Sedation

 

Figure 1
View larger version (23K):
[in this window]
[in a new window]

  		FIGURE 1 ROC curves for BIS at discharge: readiness and deep sedation. A, Mild sedation in infants ≤6 months of age; B, deep sedation in infants ≤6 months of age; C, mild sedation in infants >6 to 12 months of age; D, deep sedation in infants >6 to 12 months of age; E, mild sedation in children >12 months of age; F, deep sedation in children >12 months of age.

 
Analyses of Sedative Agents and BIS
In the analysis of the effects of sedative agents on BIS, we excluded data from children ≤6 months of age, because young age was a confounding factor (see "Analysis of Age and BIS"). Data are, therefore, presented for 2136 observations obtained in 209 children. A total of 362 of these observations were made in 35 children (aged 4.8 ± 6 years) from UM, 1358 from 94 (5.6 ± 3.7 years) at CHOP, and 416 in 80 (5 ± 3.6 years) from UC. A total of 358 observations (17%) were made in children who received chloral hydrate, 1037 (49%) in those who had received midazolam, 676 (32%) in those who had received pentobarbital, 44 (2%) in those who had received propofol, and 21 (1%) in those who had received ketamine. A total of 1300 observations (61%) were made in children who had received opioids as adjuvants.

There were significant inverse correlations between BIS and UMSS during the use of all sedative agents, with the exception of ketamine (Table 3). However, the correlations between BIS and UMSS were poor to low in the presence of opioids. Indeed, BIS values were significantly higher in children who received an opioid as a secondary agent with chloral hydrate at all levels of sedation and with midazolam at deep (ie, UMSS 3–4) levels of sedation (P ≤ .001). The distribution of BIS across sedation levels for each group was, therefore, evaluated for observations obtained in the absence of opioids. These data are presented in Table 4. Kruskal-Wallis analysis of variance demonstrated significant differences in BIS across UMSS scores (P ≤ .001) in all the sedative agent groups with the exception of ketamine. However, when corrected for multiple comparisons, BIS was not significantly different between UMSS 0 and 1 in any of the sedative agent groups or between 1 and 2 in the propofol or pentobarbital groups. Although BIS was different between UMSS 2 and 3 in the pentobarbital group, there were too few data at other levels of sedation in this group for meaningful comparisons. BIS values were not significantly different between sedative agent groups for the clinically important end point of discharge readiness (ie, UMSS 0–1; Table 5). However, BIS remained significantly higher during moderate sedation with midazolam compared with other agents and higher during deep sedation with chloral hydrate and midazolam compared with propofol and pentobarbital (Table 5).


View this table:
[in this window]
[in a new window]

  		TABLE 3 Correlations (95% CIs) Between BIS and UMSS for Each Sedative Agent Group

 

View this table:
[in this window]
[in a new window]

  		TABLE 4 BIS Values at Each Sedation Level in the Groups

 

View this table:
[in this window]
[in a new window]

  		TABLE 5 BIS Values at Clinically Important Sedation Levels

 
Lastly, mixed-model analyses demonstrated a significant main effect of sedative agent, with significantly higher mean BIS values for chloral hydrate and midazolam compared with pentobarbital (P < .001 after a Bonferroni adjustment for multiple comparisons of the BIS means). Furthermore, results based on the mixed model indicated that the variance of the random child effects was still significant (P < .01), even after assessing the fixed effects of the sedative agent and age group, which suggests that there could be additional child-level factors explaining between-child variance in BIS means.


   DISCUSSION
TOP
 ABSTRACT
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
The purpose of sedation in intensive care and procedural settings is to provide anxiolysis, facilitate patient comfort, and prevent patient movement, thereby promoting compliance with treatment. Assuring appropriate depth of sedation is largely based on observations of patient responsiveness. It has been suggested that BIS monitoring may provide a useful guide for the assessment and titration of sedation depth in children.13,15,18 The limitations of BIS relative to its variability under certain pharmacologic conditions and in certain populations, such as infants, however, must be considered for its appropriate use in any setting. This study found that, compared with older children, BIS values in infants ≤6 months were significantly lower at each observed sedation level, as were the cutoff values for mild and deep sedation. Furthermore, BIS values decreased across sedation levels during the use of chloral hydrate, midazolam, propofol, and pentobarbital in children >6 months of age. During the use of ketamine or when opioids were added to the sedative regimen, BIS values remained high despite diminished patient responsiveness. These data suggest that BIS values should be interpreted with caution in infants and during the use of opioids and ketamine in children.

To date, several studies have reported age-related differences in BIS; however, these have been conducted during the use of volatile anesthetics.10–12,20,21 Davidson et al21 found significant differences in prearousal to postarousal BIS values in all children, yet they could not demonstrate differences in BIS across anesthetic concentrations in children <12 months of age, as were found in older children. In addition, compared with older children in that study, infants had significantly lower BIS values at each concentration of anesthetic and at prearousal. Another study evaluating BIS-guided practice reported lower-than-targeted BIS values in infants ≤6 months of age despite extremely low concentrations of inhaled anesthetics.11 Conversely, recent studies found inverse correlations between BIS and age during the use of volatile anesthetics.20,25,26 Together, these studies demonstrate significant variability in BIS in young children; yet, differences in age classification, anesthetic techniques, and use of analgesics confound the interpretation of age-related differences. The present study evaluated BIS relative to patient responsiveness during sedation in the absence of agents that potentially confound the interpretation of BIS and found lower BIS values at similar sedation levels in infants ≤6 months of age compared with older children.

The BIS monitor yields a single dimensionless number that correlates with the patient's sedation depth by comparing EEG signals obtained from that patient with EEG traces derived from adults that are saved in its database.39 BIS differences in infants, therefore, are not surprising, given the rapid maturational development that occurs over the first few months of life. Eeg-Olofsson27 described 4 maturational stages based on evoked potential data (fetal, neonatal, transient, and mature) and suggested that the transition from transient to mature-evoked potential activity occurs between 1 and 2 years of age. Marshall et al38 found that dominant EEG frequencies steadily increased from 5 to 24 months of age, likely explained by concurrent neuronal maturation and the development of myelination across the cortex. Our findings of lower BIS values in younger children are consistent with these described patterns in EEG development. In addition, Eeg-Olofsson27 described interindividual EEG variability in young children, which we similarly observed in our BIS data that are derived from EEG signal processing. In fact, our mixed-model analyses demonstrated that 33% of the variance in BIS between children could not be explained by sedative agent or age group factors but rather by some random child-level factors.

It is well known that different sedative/hypnotic agents do not necessarily produce similar EEG patterns.40 These agents do, however, produce clinically observable sedation levels, although such levels may vary from patient to patient because of variability in responses to drugs and dosages. Because BIS is an indicator of cerebral activity, effects of agents that produce or enhance loss of consciousness without cortical depression are not likely to be reflected in BIS values.29 Although BIS is said to be "insensitive" to specific anesthetics or sedative agents,39 data from several studies suggest that the BIS number may incorrectly reflect depth of hypnosis during the use of certain agents or under certain conditions.25,26,35,37,41 The commonly used sedatives, pentobarbital, chloral hydrate, midazolam, and propofol, all have {gamma}-aminobutyric acid-producing effects, depressing the central nervous system in a dose-dependent fashion. As such, it is not surprising that, in the absence of other agents, the present study found significant correlations between BIS and observed sedation depth with these agents. Studies by McDermott et al15 and by Shields et al,17 however, found poor associations between BIS and UMSS scores in children sedated with chloral hydrate. The results of these studies may have been confounded by the concomitant use of opioids15 and inclusion of infants <6 months of age.17

Several studies have suggested that the addition of opioids altered the patient's level of responsiveness to verbal or noxious stimuli without lowering BIS values.36,42,43 In addition, reductions or escalations in remifentanil concentrations did not change BIS values in other studies.44,45 We found similar discrepancies in BIS values in the presence of opioids. For instance, correlations between BIS and observed responsiveness (ie, UMSS scores) were poor to low for children who received opioids in combination with sedative agents. In addition, BIS values remained higher for these children at moderate-to-deep levels of sedation compared with those who received sedatives alone. Although the exact doses of sedatives and opioids were not available for this analysis, it is likely that children who received opioids required smaller doses of sedatives to facilitate the diagnostic procedure. Indeed, several studies demonstrated that adjuvant opioids effectively lower the anesthetic dosages of propofol needed to produce loss of consciousness,46 showing that agonist opioids enhance the hypnotic affect of anesthetics without altering BIS.41 Furthermore, it has been postulated that analgesic concentrations of opioids produce minimal electrophysiological alterations on the cerebral cortex,41 probably because they exert their effects primarily on noncortical structures, such as the locus coeruleus-noradrenergic system.47

Ketamine has been shown to similarly depress level of consciousness without lowering BIS values.24 In fact, some studies have found a paradoxical increase in BIS despite deepening levels of hypnosis after ketamine administration in patients anesthetized with sevoflurane or propofol.48,49 Hans et al48 hypothesized that this effect may reflect a desynchronization of the EEG signal from the dissociative action of ketamine. Data from the present study demonstrate a poor correlation between BIS and UMSS scores in children sedated with ketamine, providing further evidence that BIS does not effectively reflect sedation depth during the use of this agent. Taken together, the above results demonstrate that BIS does not accurately reflect sedation depth during ketamine or opioid use, and if used to guide sedative administration in the presence of these agents, excessive dosing and deeper-than-intended levels of hypnosis may result.

Not surprisingly, BIS values were significantly lower in children who were moderately to deeply sedated with pentobarbital or propofol, suggesting that these agents have the propensity to cause very deep levels of sedation. Furthermore, the fact that no observations were captured at a UMSS score of 3 in this sample of children who received propofol indicates the tendency of this agent to produce a rapid transition from mild sedation to unresponsiveness. Previous investigators have similarly reported BIS values that were consistent with levels of general anesthesia in children who received propofol for procedural sedation.50 In addition, others reported levels of sedation that could not be distinguished from general anesthesia in children who were given propofol by nonanesthesiologists.51 Such findings emphasize the importance of judicious use of propofol only by personnel trained to rescue patients from general anesthesia.

Reliable and valid assessment of sedation depth is important in the clinical setting to detect inadequate sedation, as well as oversedation, either of which may result in adverse patient outcomes.52 Data from Berkenbosch et al13 suggested that maintenance of BIS <70 would ensure that 80% of patients in the ICU are adequately sedated based on Ramsey scores of 4 and 5 (ie, defined as deep sedation). These findings, however, were based on data from patients aged 1 month to 20 years. The present study, in contrast, found a higher BIS cutoff for deep sedation (ie, 77) in children ≥6 months to 18 years of age and a lower cutoff in infants (ie, 62), suggesting considerable differences in the interpretation of BIS values in children. The specific goals for sedation and a need to err on the conservative side for safe practice should be considered when interpreting BIS cutoff values in the clinical setting. Our data further suggest that, whereas BIS correlates fairly well with observed sedation depth during the use of common sedatives, it best differentiates the extremes of responsiveness but poorly demarcates the middle range of the sedation spectrum. These findings are similar to those of Mason et al,53 who recently demonstrated the limited ability of BIS in distinguishing between moderate and deep sedation levels in children.

Findings from this study may be limited in that data were collected in multiple settings, which may have introduced variability in the observations. In addition, the possibility of observer bias cannot be ruled out, because data were collected by multiple observers at multiple sites. However, good interrater reliability was established for observer scoring of sedation in 3 of the 4 studies used in this analysis, as well as in previous studies. Furthermore, all of the observers were blinded to BIS values during assignment of UMSS scores, and each site used monitors with the same BIS algorithms, which should have minimized the variability. It remains unclear whether use of the recently introduced pediatric quatro sensor with improved artifact reduction would yield less variability in similar data. Findings from this study may be further limited in that data were unavailable for some of the levels of sedation in certain sedative agent groups. Although this may reflect the rapid transition from mild to deep sedation levels with intravenous administration of these agents in the clinical setting, it did not allow for complete evaluation of BIS values across the sedation continuum in these groups. Additional study where these agents are slowly titrated may permit such analysis.


   CONCLUSIONS
TOP
 ABSTRACT
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
An understanding of the potential limitations of BIS is important to correctly interpret data obtained from such monitors in sedation settings. Findings from this study suggest that BIS provides a fair reflection of mild and deep sedation during the use of commonly used agents, including chloral hydrate, midazolam, pentobarbital, or propofol, in children >6 months of age. However, BIS poorly reflects sedation depth during the use of opioids and ketamine and across the middle range of the sedation continuum. Finally, BIS values were significantly lower at all levels of sedation in infants ≤6 months of age. These data suggest that BIS values must be interpreted cautiously in sedated children, with particular consideration given to patient age and use of sedative agents.


   ACKNOWLEDGMENTS
 
This study was supported by the Department of Anesthesiology at UM. Aspect Medical Systems Inc (Newton, MA) provided BIS sensors and equipment for the original studies conducted at UM and partial financial support for the study at CHOP. No outside funding was provided for this study.

We thank Brady West, senior statistician at UM, for expert advice regarding the analyses.


   FOOTNOTES
 
Accepted Feb 16, 2007.

Address correspondence to Shobha Malviya, MD, Section of Pediatrics, Department of Anesthesiology, University of Michigan Health Systems, F3900 Box 0211, 1500 E Medical Center Dr, Ann Arbor, MI 48109-0211. E-mail: smalviya{at}umich.edu

Financial Disclosure: Dr Malviya and Ms Voepel-Lewis are currently receiving funding from Aspect Medical Systems for clinical trials and received BIS sensors for past studies; Drs Watcha and Sadhasivam received past funding from Aspect Medical Systems; and Drs Tait and Friesen have indicated they have no financial relationships relevant to this article to disclose.


   REFERENCES
TOP
 ABSTRACT
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 

  1. Doi M, Gajraj RJ, Mantzaridis H, Kenny GN. Relationship between calculated blood concentration of propofol and electrophysiological variables during emergence from anaesthesia: comparison of bispectral index, spectral edge frequency, median frequency and auditory evoked potential index. Br J Anaesth. 1997;78 :180 –184[Abstract/Free Full Text]
  2. Liu J, Singh H, White PF. Electroencephalogram bispectral analysis predicts the depth of midazolam-induced sedation. Anesthesiology. 1996;84 :64 –69[CrossRef][Web of Science][Medline]
  3. Kearse LA Jr, Rosow C, Zaslavsky A, Connors P, Dershwitz M, Denman W. Bispectral analysis of the electroencephalogram predicts conscious processing of information during propofol sedation and hypnosis. Anesthesiology. 1998;88 :25 –34[CrossRef][Web of Science][Medline]
  4. Katoh T, Suzuki A, Ikeda K. Electroencephalographic derivatives as a tool for predicting the depth of sedation and anesthesia induced by sevoflurane. Anesthesiology. 1998;88 :642 –650[Web of Science][Medline]
  5. Olofsen E, Dahan A. The dynamic relationship between end-tidal sevoflurane and isoflurane concentrations and bispectral index and spectral edge frequency of the electroencephalogram. Anesthesiology. 1999;90 :1345 –1353[CrossRef][Web of Science][Medline]
  6. Struys MM, Vereecke H, Moerman A, et al. Ability of the bispectral index, autoregressive modelling with exogenous input-derived auditory evoked potentials, and predicted propofol concentrations to measure patient responsiveness during anesthesia with propofol and remifentanil. Anesthesiology. 2003;99 :802 –812[CrossRef][Web of Science][Medline]
  7. Struys M, Versichelen L, Byttebier G, Mortier E, Moerman A, Rolly G. Clinical usefulness of the bispectral index for titrating propofol target effect-site concentration. Anaesthesia. 1998;53 :4 –12[CrossRef][Web of Science][Medline]
  8. Sandler NA, Sparks BS. The use of bispectral analysis in patients undergoing intravenous sedation for third molar extractions. J Oral Maxillofac Surg. 2000;58 :364 –368[CrossRef][Web of Science][Medline]
  9. Bell JK, Laasch HU, Wilbraham L, England RE, Morris JA, Martin DF. Bispectral index monitoring for conscious sedation in intervention: better, safer, faster. Clin Radiol. 2004;59 :1106 –1113[CrossRef][Web of Science][Medline]
  10. Degoute CS, Macabeo C, Dubreuil C, Duclaux R, Banssillon V. EEG bispectral index and hypnotic component of anaesthesia induced by sevoflurane: comparison between children and adults. Br J Anaesth. 2001;86 :209 –212[Abstract/Free Full Text]
  11. Bannister CF, Brosius KK, Sigl JC, Meyer BJ, Sebel PS. The effect of bispectral index monitoring on anesthetic use and recovery in children anesthetized with sevoflurane in nitrous oxide. Anesth Analg. 2001;92 :877 –881[Abstract/Free Full Text]
  12. McCann ME, Bacsik J, Davidson A, Auble S, Sullivan L, Laussen P. The correlation of bispectral index with endtidal sevoflurane concentration and haemodynamic parameters in preschoolers. Paediatr Anaesth. 2002;12 :519 –525[CrossRef][Web of Science][Medline]
  13. Berkenbosch JW, Fichter CR, Tobias JD. The correlation of the bispectral index monitor with clinical sedation scores during mechanical ventilation in the pediatric intensive care unit. Anesth Analg. 2002;94 :506 –511[Abstract/Free Full Text]
  14. Davidson A. The correlation between bispectral index and airway reflexes with sevoflurane and halothane anaesthesia. Paediatr Anaesth. 2004;14 :241 –246[CrossRef][Web of Science][Medline]
  15. McDermott NB, VanSickle T, Motas D, Friesen RH. Validation of the bispectral index monitor during conscious and deep sedation in children. Anesth Analg. 2003;97 :39 –43[Abstract/Free Full Text]
  16. Malviya S, Voepel-Lewis T, Ludomirsky A, Marshall J, Tait AR. Can we improve the assessment of discharge readiness: a comparative study of observational and objective measures of depth of sedation in children. Anesthesiology. 2004;100 :218 –224[CrossRef][Web of Science][Medline]
  17. Shields CH, Styadi-Park G, McCown MY, Creamer KM. Clinical utility of the bispectral index score when compared to the University of Michigan Sedation Scale in assessing the depth of outpatient pediatric sedation. Clin Pediatr (Phila). 2005;44 :229 –236[Abstract/Free Full Text]
  18. Powers KS, Nazarian EB, Tapyrik SA, et al. Bispectral index as a guide for titration of propofol during procedural sedation among children. Pediatrics. 2005;115 :1666 –1674[Abstract/Free Full Text]
  19. Grindstaff RJ, Tobias JD. Applications of bispectral index monitoring in the pediatric intensive care unit. J Intensive Care Med. 2004;19 :111 –116[Abstract/Free Full Text]
  20. Kim HS, Oh AY, Kim CS, Kim SD, Seo KS, Kim JH. Correlation of bispectral index with end-tidal sevoflurane concentration and age in infants and children. Br J Anaesth. 2005;95 :362 –366[Abstract/Free Full Text]
  21. Davidson AJ, McCann ME, Devavaram P, et al. The differences in the bispectral index between infants and children during emergence from anesthesia after circumcision surgery. Anesth Analg. 2001;93 :326 –330[Abstract/Free Full Text]
  22. Malviya S, Voepel-Lewis T, Tait AR. A comparison of observational and objective measures to differentiate depth of sedation in children from birth to 18 years of age. Anesth Analg. 2006;102 :389 –394[Abstract/Free Full Text]
  23. Davidson AJ, Huang GH, Rebmann CS, Ellery C. Performance of entropy and Bispectral Index as measures of anaesthesia effect in children of different ages. Br J Anaesth. 2005;95 :674 –679[Abstract/Free Full Text]
  24. Sakai T, Singh H, Mi WD, Kudo T, Matsuki A. The effect of ketamine on clinical endpoints of hypnosis and EEG variables during propofol infusion. Acta Anaesthesiol Scand. 1999;43 :212 –216[CrossRef][Web of Science][Medline]
  25. Tirel O, Wodey E, Harris R, Bansard JY, Ecoffey C, Senhadji L. The impact of age on bispectral index values and EEG bispectrum during anaesthesia with desflurane and halothane in children. Br J Anaesth. 2006;96 :480 –485[Abstract/Free Full Text]
  26. Wodey E, Tirel O, Bansard JY, et al. Impact of age on both BIS values and EEG bispectrum during anaesthesia with sevoflurane in children. Br J Anaesth. 2005;94 :810 –820[Abstract/Free Full Text]
  27. Eeg-Olofsson O. Longitudinal developmental course of electrical activity of brain. Brain Dev. 1980;2 :33 –44[Medline]
  28. Watcha MF. Investigations of the bispectral index monitor in pediatric anesthesia: first things first. Anesth Analg. 2001;92 :805 –807[Free Full Text]
  29. Sleigh JW, Barnard JP. Entropy is blind to nitrous oxide: can we see why? Br J Anaesth. 2004;92 :159 –161[Free Full Text]
  30. Sadhasivam S, Ganesh A, Robison A, Kaye R, Watcha MF. Validation of the bispectral index monitor for measuring the depth of sedation in children. Anesth Analg. 2006;102 :383 –388[Abstract/Free Full Text]
  31. Malviya S, Voepel-Lewis T, Tait AR, Merkel S, Tremper K, Naughton N. Depth of sedation in children undergoing computed tomography: validity and reliability of the University of Michigan Sedation Scale (UMSS). Br J Anaesth. 2002;88 :241 –245[Abstract/Free Full Text]
  32. Verbeke G, Molenberghs G. Linear Mixed Models for Longitudinal Data. New York, NY: Springer; 2000
  33. Altman DG, Bland JM. Diagnostic tests 3: receiver operating characteristic plots. BMJ. 1994;309 :188[Free Full Text]
  34. Munro B. Correlations. In: Munro B, ed. Statistical Methods for Health Care Research. 4th ed. Philadelphia, PA: Lippincott; 2001:223 –244
  35. Detsch O, Schneider G, Kochs E, Hapfelmeier G, Werner C. Increasing isoflurane concentration may cause paradoxical increases in the EEG bispectral index in surgical patients. Br J Anaesth. 2000;84 :33 –37[Abstract/Free Full Text]
  36. Mi WD, Sakai T, Kudo T, Kudo M, Matsuki A. Performance of bispectral index and auditory evoked potential monitors in detecting loss of consciousness during anaesthetic induction with propofol with and without fentanyl. Eur J Anaesthesiol. 2004;21 :807 –811[CrossRef][Web of Science][Medline]
  37. Barr G, Jakobsson JG, Owall A, Anderson RE. Nitrous oxide does not alter bispectral index: study with nitrous oxide as sole agent and as an adjunct to i.v. anaesthesia. Br J Anaesth. 1999;82 :827 –830
  38. Marshall PJ, Bar-Haim Y, Fox NA. Development of the EEG from 5 months to 4 years of age. Clin Neurophysiol. 2002;113 :1199 –1208[CrossRef][Web of Science][Medline]
  39. Rampil IJ. A primer for EEG signal processing in anesthesia. Anesthesiology. 1998;89 :980 –1002[CrossRef][Web of Science][Medline]
  40. Todd MM. EEGs, EEG processing, and the bispectral index. Anesthesiology. 1998;89 :815 –817[CrossRef][Web of Science][Medline]
  41. Dahaba AA. Different conditions that could result in the bispectral index indicating an incorrect hypnotic state. Anesth Analg. 2005;101 :765 –773[Abstract/Free Full Text]
  42. Vanluchene AL, Struys MM, Heyse BE, Mortier EP. Spectral entropy measurement of patient responsiveness during propofol and remifentanil: a comparison with the bispectral index. Br J Anaesth. 2004;93 :645 –654[Abstract/Free Full Text]
  43. Kodaka M, Okamoto Y, Handa F, Kawasaki J, Miyao H. Relation between fentanyl dose and predicted EC50 of propofol for laryngeal mask insertion. Br J Anaesth. 2004;92 :238 –241[Abstract/Free Full Text]
  44. Schmidt GN, Bischoff P, Standl T, Voigt M, Papavero L, Schulte am Esch J. Narcotrend, bispectral index, and classical electroencephalogram variables during emergence from propofol/remifentanil anesthesia. Anesth Analg. 2002;95 :1324 –1330[Abstract/Free Full Text]
  45. Guignard B, Menigaux C, Dupont X, Fletcher D, Chauvin M. The effect of remifentanil on the bispectral index change and hemodynamic responses after orotracheal intubation. Anesth Analg. 2000;90 :161 –167[Abstract/Free Full Text]
  46. Lysakowski C, Dumont L, Pellegrini M, Clergue F, Tassonyi E. Effects of fentanyl, alfentanil, remifentanil and sufentanil on loss of consciousness and bispectral index during propofol induction of anaesthesia. Br J Anaesth. 2001;86 :523 –527[Abstract/Free Full Text]
  47. Pan YZ, Li DP, Chen SR, Pan HL. Activation of mu-opioid receptors excites a population of locus coeruleus-spinal neurons through presynaptic disinhibition. Brain Res. 2004;997 :67 –78[CrossRef][Web of Science][Medline]
  48. Hans P, Dewandre PY, Brichant JF, Bonhomme V. Comparative effects of ketamine on Bispectral Index and spectral entropy of the electroencephalogram under sevoflurane anaesthesia. Br J Anaesth. 2005;94 :336 –340[Abstract/Free Full Text]
  49. Vereecke HE, Struys MM, Mortier EP. A comparison of bispectral index and ARX-derived auditory evoked potential index in measuring the clinical interaction between ketamine and propofol anaesthesia. Anaesthesia. 2003;58 :957 –961[CrossRef][Web of Science][Medline]
  50. Reeves ST, Havidich JE, Tobin DP. Conscious sedation of children with propofol is anything but conscious. Pediatrics. 2004;114 (1). Available at: www.pediatrics.org/cgi/content/full/114/1/e74
  51. Barbi E, Gerarduzzi T, Marchetti F, et al. Deep sedation with propofol by nonanesthesiologists: a prospective pediatric experience. Arch Pediatr Adolesc Med. 2003;157 :1097 –1103[Abstract/Free Full Text]
  52. Tonner PH, Wei C, Bein B, Weiler N, Paris A, Scholz J. Comparison of two bispectral index algorithms in monitoring sedation in postoperative intensive care patients. Crit Care Med. 2005;33 :580 –584[CrossRef][Web of Science][Medline]
  53. Mason KP, Michna E, Zurakowski D, et al. Value of bispectral index monitor in differentiating between moderate and deep Ramsay Sedation Scores in children. Paediatr Anaesth. 2006;16 :1226 –1231[CrossRef][Medline]
PEDIATRICS (ISSN 1098-4275). ©2007 by the American Academy of Pediatrics

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Facebook Facebook   Add to Reddit Reddit   Add to Technorati Technorati   Add to Twitter Twitter    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Malviya, S.
Right arrow Articles by Friesen, R. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Malviya, S.
Right arrow Articles by Friesen, R. H.
Related Collections
Right arrow Surgery
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Facebook   Add to Reddit   Add to Technorati   Add to Twitter  
What's this?