Background
Propofol is commonly used for total intravenous anesthesia (TIVA) due to its characteristic ease of titration, rapid onset and offset of action, reduced incidence of postoperative nausea/vomiting [
1] and emergence agitation [
2]. In the morbidly obese (MO) paediatric population, despite propofol’s desirable characteristics, appropriate drug administration is complicated by numerous anatomic and physiological factors that accompany obesity, including increases in total body mass, blood volume, cardiac output and regional blood flow [
3]. Inavailability of evidence-based clinical guidelines and an adequate dosing scalar for individualized propofol dosing in MO children and adolescents could adversely impact the quality of TIVA administered to these patients [
4].
Recent evidence has highlighted drug dosing issues in obese adults raising concerns at both extremes of drug administration: inadequate anesthesia resulting in intra-operative awareness due to under-dosing propofol [
5] and excessive anesthetic administration, resulting in organ hypoperfusion and low processed electroencephalographic index values which could be associated with poor outcomes [
6‐
9]. Although the Bispectral Index/BIS monitor provides quantifiable and continuous assessment of propofol cortical effects in children and adolescents [
10‐
12], it is a common to practice TIVA with propofol in children without BIS monitoring. In this descriptive study in a cohort of MO paediatric patients, we evaluated the effects of propofol TIVA on perioperative outcomes.
Methods
A prospective study was conducted in MO children and adolescents between July 2009 and July 2010. The study protocol was approved by Cincinnati Children’s Hospital institutional review board and written informed assent / consent was obtained from all participants and/or their guardians as appropriate.
Study subjects
Inclusion criteria: 1) Males and females between the ages of 5 and 18 years, 2) Body Mass Index (BMI) for age > 95
th percentile {> 95
th percentile (obese), >99
th percentile (MO) [
13]}; 3) Patients undergoing elective surgery scheduled for a duration of at least 60 minutes.
Exclusion criteria: 1) Severe developmental delay, 2) Known cardiac anomaly, neurological, renal or hepatic disorders, 3) Known allergy to propofol, 4) Skin condition which would preclude placement of BIS sensor on the forehead.
Study protocol
The patient was brought to the operating room, electrocardiograph, non-invasive blood pressure and pulse oximeter were applied, and an intravenous catheter was established. Before or immediately after induction, an age and head-size appropriate disposable BIS sensor® XP, (Aspect Medical Systems, Norwood, MA) was placed on each patient’s forehead and connected to the BIS monitor. The BIS monitor screen was covered throughout the procedure to blind the anesthesia personnel to the BIS score and trend screen. Anesthesia was induced with propofol at a standardized infusion rate of 1000 μg.kg
-1.min
-1 after intravenous injection of lidocaine 30 mg. Infusion rates were based on adjusted body weight (ABW) which was calculated using total body weight (TBW) and ideal bodyweight (IBW), as described by Servin
et. al. [
14], substituting 22 kg/m
2 as Ideal BMI (in Servin’s formula) with 50
th percentile BMI for age and gender, obtained from Centers for Disease Control and Prevention, National Center for Health Statistics growth charts, United States. (
http://www.cdc.gov/growthcharts/. May 30, 2000).
(1)
(2)
Patients were asked to count, or called repeatedly in a normal voice until the induction end-point of loss of verbal contact; this was recorded as ‘time to induction’. Succinylcholine was administered and the trachea was intubated with an appropriate cuffed endotracheal tube. Anesthesia was maintained with propofol infusion. Vecuronium was titrated to Train-of-Four response (goal: one of four twitches). The induction dose of propofol was followed by propofol infusion at a rate of 250-350 μg/kg/min for 10 minutes and titrated in 25-50 μg/kg/min steps (reduced to prevent drop in systolic arterial blood pressure and heart rate below 30% of baseline values and titrated up when greater than 30% increase in heart rate or blood pressure occurred in the absence of new painful stimuli). Propofol was infused using calibrated pumps with internal memory and downloading capability. This allowed all real-time rates and rate changes to be recorded, including start and stop time of propofol dosing, propofol infusion rates, and propofol dose adjustments. Fentanyl 50-100 μg was administered after induction and 50 μg doses were administered in case of inadequate analgesia (defined as increase in heart rate and/or blood pressure above 30% of baseline with surgical incision or manipulation). When inadequate anesthesia or analgesia was not considered to be the reason for increase in blood pressure or heart rate, medications to correct hemodynamics were administered. The propofol infusion was decreased by 50% about 15 minutes before conclusion of surgery and discontinued when skin sutures were being placed. Muscle relaxants were reversed and once the patient was breathing, morphine/hydromorphone was dosed incrementally towards the end of the surgery, titrated to respiratory rate of 14-16 breaths per minute. After clinical confirmation of reversal, the trachea was extubated awake. Patients were then transferred to the recovery area (PACU) and followed until they achieved PACU discharge criteria.
Demographics
Patient demographics, age, gender, weight (TBW) and height were collected. After computing the BMI, IBW and ABW were calculated according to equations
1 and
2. Ideal BMI in Equation
2 is defined as the 50
th percentile values from age and sex – specific BMI for age charts at
http://www.cdc.gov. A calculator available at
http://www.bcm.edu/cnrc/bodycomp/bmiz2.html was used to calculate BMI for age percentiles. Lean body mass (LBM) was calculated using the formula described by Peters
et. al. by first estimating Extracellular Volume (ECV) from weight and height [
15] according to the following equation.
Propofol and opioid doses
Posthoc calculation of induction dose required to achieve loss of verbal contact was performed by multipying the rate of infusion and time taken to reach the end-point. Means and SD of propofol infusion rates during maintenance were analyzed from pooled data. Propofol maintenance infusion rates were plotted against BIS values and time since start of propofol infusion. Infusion rates of eight patients corresponding to BIS values of 40-60 were then analyzed to derive mean and SD. Hourly opioid use as fentanyl equivalent doses were calculated, based on an equivalence of morphine 10 mg = 2 mg hydromorphone = 100 μg fentanyl.
Hemodynamics
Clinical data including mean, systolic, diastolic blood pressure (MAP, SBP and DBP respectively) and heart rate/HR were recorded electronically every 5 minutes intraoperatively. For each of the measured hemodynamic parameters, percentage difference from baseline (value recorded 5 minutes before propofol induction) was calculated according to the following equation.
BIS
BIS data were transferred electronically to a computerized record in one-second increments. This included the date and time of BIS data collection, minimum and maximum BIS values, average Signal Quality Index (SQI) and average electromyography (EMG). The smoothing rate of the BIS monitor was set at 15 seconds. Evaluable BIS values were defined as those with Signal Quality Index > 70.
Blood sampling and propofol analysis
Blood samples (1.0 ml) were obtained from a dedicated intravenous catheter placed in the upper extremity contralateral to the propofol infusion site. Samples were obtained at baseline prior to the start of propofol, approximately 15, 30, 45, 60, 120, 180, 240 minutes after the start of the propofol infusion, at 5 and 20 minutes after dose adjustment, just before discontinuation of the propofol infusion and at 5, 10, 15, 30, 45 and 120 minutes after termination of the infusion. Whole-blood samples for propofol analysis were stored at 4°C until analysis (within 1 month) by high-performance liquid chromatography with fluorescence detection. The coefficients of variation for the intra-assay and interassay precision over the concentration range from 0.05 to 5.0 mg.l
-1 were less than 4.5% and 7.1% respectively. The lower limit of quantification was 0.05 mg.l
-1[
16].
Ramsay sedation scores
Ramsay Sedation Scores (RSS) were recorded post-operatively about every 10 minutes for the first 30 minutes and thereafter every 30 minutes while in the PACU [
17].
Other clinical data
‘
Time to eye opening’, defined as the time from cessation of propofol infusion to eye opening on verbal command, was noted.
Respiratory adverse events (RAE) defined as airway obstruction requiring airway manipulation, episodes of desaturation (< 90%) and/or need for oxygen for >120 minutes in the immediate postoperative period were also recorded. On postoperative day 1 and 3, patients were evaluated using the
Structured Awareness Screening Interview created by Davidson
et. al. [
18].
Statistical analysis
GraphPad Prism 5 software (GraphPad Software Inc., La Jolla, CA) was used to generate descriptive statistics (mean, standard deviation, median and range for continuous variables and frequencies for categorical variables). Linear, quadratic and cubic trends were tested to detect correlation of weight scalars (TBW, ABW and LBM) with induction dose, in addition to calculation of root mean square errors (MSE) and the regression lines fitted. SAS software © (SAS version 9.2, Cary, North Carolina) was used to perform logistic regression between occurrence of respiratory adverse events and explanatory variables (TBW, IBW, ABW, BMI, propofol amount and duration of propofol infusion) to detect two-tailed p values with a 95% Confidence Intervals (CI).
Discussion
TIVA with propofol in MO pediatric patients can be challenging in the absence of weight and dosing guidelines.We evaluated the clinical response to propofol anesthesia in this population.
While hemodynamic parameters during propofol TIVA were largely unchanged, BIS values for MO adolescents were below 40 for 93% of the maintenance phase. We believe that the increased anesthetic depth was a result of clinical overestimation of propofol requirements. Although our study did not have a BIS control group, our findings that MO adolescents undergoing clinically titrated propofol TIVA received high propofol doses, is in accordance with what has been reported in obese adults. Gaszynski
et. al. demonstrated that obese adults undergoing clinically titrated propofol TIVA without BIS monitoring received higher propofol infusions (10 vs. 5.8 mg.kg-1/h), consumed more total propofol (2012 ± 310 mg vs. 1210 ± 370 mg) and had longer awakening times [
19], compared to those who were BIS monitored.
There are two other findings of significance. Firstly, prolonged emergence from anesthesia was observed in our study patients, with an average ‘time to eye opening’ of 25.9 ± 22.6 min, compared to 10.3 ± 5.4 minutes reported in non-obese children after clinically titrated propofol TIVA [
19,
20]. This was also reflected by deeper levels of sedation (RSS > 4) during the first 30 minutes in the PACU. Although there is some evidence for propofol accumulation and slow washout after continuous propofol infusions in MO adults [
21], this has not been supported by clinical data in adults [
14]. We believe the prolonged emergence is due to the high propofol doses our study subjects received (mean = 3244 mg or 11.5 mg kg
-1 h
-1), which positively correlated with the ‘time to eye-opening’ (p = 0.03). Secondly, we note a 30% incidence of RAE in the immediate postoperative period with a 14% increased risk of RAE for every unit increase in BMI. Increased risk of RAE after propofol TIVA in obese patients, is supported by Zoremba
et. al.’s finding of excessive impairment of pulmonary function in obese adults, two hours after propofol anesthesia [
22].
Despite the fact that in clinical settings, propofol is generally administered as a bolus for induction, we chose to use a standardized infusion method for induction. This allowed us to calculate an induction dose based on a clinical endpoint rather than an arbitrary weight-based dose. We noted a high correlation for induction dose to LBM (similar to findings of Ingrande
et. al.) [
23] and ABW which suggests that the dosing for induction be based on these scalars and not TBW. These findings need to be confirmed with large prospective studies and a formal pharmacokinetic-pharmacodynamic analysis. Pharmacokinetic analysis following this study has been completed and results have been published in an earlier report wherin TBW proved to be the most significant determinant for clearance, while no predictive covariates for volume of distribution were identified [
24]. Our infusion regimen was based on ABW as Servin
et. al. had used this weight in morbidly obese adults without evidence of propofol accumulation [
14]. Our finding that an average infusion rate of 7 mg kg
-1 h
-1 TBW during 20-90 minutes of propofol maintenance phase correlates with a BIS of 40-60 (Figure
2A), is higher than the recommended rate of 4.6 to 6 mg kg
-1 h
-1 TBW to maintain BIS of 50 in obese adults during the same time period [
14,
25]. Considering that concentrations of 4.3 ± 1.1 mg.l
-1 in non-obese children [
11] and 3-4 mg.l
-1 in obese adults, have been reported to correlate with a BIS of 50 [
26], our findings of higher propofol concentrations during maintenance of anesthesia and corresponding lower BIS values suggests that clinical titration of propofol anesthesia in MO adolescents is not optimal.
Ramsay sedation scores were used to assess sedation in the postoperative period. We used a single non-anesthesia observer to rate RSS in all study subjects to limit inter-observer variance. However, caution is required in interpreting correlation of propofol concentrations with RSS as these sedation scores reflect the combination of propofol and opioid effects. We also note that the observational study design allowing clinical titration of propofol doses prevented standardization of dosages. Although dosing of propofol could be affected by differences in opioid doses, it has been reported to not affect the relation between propofol concentrations and BIS [
27]. Hence, the lack of standardization of opioid doses would likely not affect our conclusions. Finally, our premise that BIS values below 40 represent very ‘deep’ anesthesia is debatable, but there is no evidence to the contrary, as none of our patients suffered any awareness. The other dilemma concerning the risks associated with excessive anesthesia dosing is still unresolved [
28].
Conclusion
In conclusion, this study presents a detailed descriptive analysis of propofol anesthesia in MO adolescents. Although BIS has been found to improve clinically important outcomes in children undergoing inhalation anesthesia [
29], it is not a standard monitor in paediatric anesthetic practice. In MO adults, La Colla
et. al. concluded that it is advisable to administer propofol to MO patients by titration to targeted processed-EEG values [
30]. Our findings suggest that in the absence of evidence based dosing guidelines for propofol administration in this MO paediatric population, use of only clinical parameters to dose TIVA with propofol can result in excessive depth of anesthesia. In this setting, BIS monitoring provides anesthesiologists information about real time trend of anesthetic depth and helps prevent excessive propofol administration and associated negative consequences. Our findings also emphasize the need for improved propofol dosing guidelines and monitoring during TIVA in MO adolescents to minimize relative overdosing and its negative consequences.
Funding disclosure
This study was funded by a Translational Research Initiative grant from Cincinnati Children’s Research Foundation, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH. The findings of this study were presented in part as abstract/poster/oral presentation at the Society of Paediatric Anesthesia Annual meeting at San Antonio, 2011 and American Society of Anesthesiology Annual Meeting, Chicago, 2011.
Acknowledgements
We would like to acknowledge the Teen Longitudinal Assessment of Bariatric Surgery, Cincinnati Children’s Hospital Medical Center (reference
http://www.Teen-LABS.org) for their support; and Aspect Medical Systems, Norwood, MA for loan of BIS monitor for this study. We also acknowledge the help of Elke HJ Krekels, M.Sc. and Lily Ding, PhD. In biostatistics in reviewing the manuscript.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
VC was involved in design, conduct of the study, analysis of the data, and manuscript preparation, SS helped design, conduct the study, and write the manuscript, JD helped design the study and write the manuscript, HE was involved in conduct of the study, SC helped conduct the study and analyze propofol, BS analyzed the data, PS participated in the conduct of the study and manuscript writing, TI helped conduct of the study, AAV and CAK were involved in designing the study, analysis of the data, and preparation of the manuscript. All authors read and approved the final manuscript.