Down syndrome is one of the most common genetic disorders in children with an incidence of 1 in 800.1 It is most commonly associated with trisomy 21, but can occur with mosaicism and translocation of chromosome 21. Up to 50% of children with this syndrome have congenital cardiac defects.2 Anesthesia in children including those with Down syndrome is often induced by having the child inhale potent anesthetics via a facemask, because it avoids the stress associated with placement of an IV catheter while awake.
Inhalation anesthetics have well-described cardiovascular depressant effects.3–6 The incidence of severe bradycardia associated with inhaled anesthetic induction in children with Down syndrome was reported as being 3.7% by Borland et al. in 2004.7 This study focused on patients receiving halothane and isoflurane. Murat et al. reported an incidence of bradycardia of 0.3% in a pediatric population of 24,165.8 There is greater hemodynamic stability during induction with sevoflurane than with the older drugs halothane and isoflurane, and most inhaled inductions of anesthesia in children in the United States are now done with sevoflurane.9 Although bradycardia during sevoflurane induction has been reported in children with Down syndrome,10 its incidence and severity are presently unknown.
The primary aim of this study was to compare the incidence and characteristics of bradycardia after induction of anesthesia with sevoflurane in children with Down syndrome with a cohort of healthy control subjects. Our secondary aim was to determine the factors associated with bradycardia, including the presence of congenital heart disease (CHD).
After approval from the IRB of the Stokes Research Institute of The Children's Hospital of Philadelphia, we performed a computerized search of our automated electronic anesthesia record-keeping system (CompuRecord, Philips Healthcare, Bothell, Washington) for anesthetic procedures in patients with Down syndrome that occurred between July 1, 1998, and November 15, 2006. The terms Down syndrome, trisomy 21, and their variants were searched for both in the diagnosis category and in the free text in the record. Demographic data— including age, weight, diagnosis, and surgical procedures performed—were recorded. The medical history and physical examination data for each subject with Down syndrome were reviewed for the presence of a history of significant CHD. For the purposes of this study, subjects with a history of the following isolated abnormalities were not considered to have significant CHD: patent ductus arteriosus without hemodynamic effects, mild tricuspid regurgitation, mitral valve prolapse, situs inversus totalis without cardiac structural abnormalities, ventricular septal defects that have spontaneously closed, bicuspid aortic valve without hemodynamic effects such as aortic stenosis, and patent foramen ovale. These defects were considered mild because these patients are often asymptomatic and may undergo spontaneous resolution of their lesions.11
A control group of patients was randomly selected from the remaining charts of patients without Down syndrome in the CompuRecord database for this time period. There were no attempts to match Down syndrome and control patients for age, gender, ASA physical status, or surgical procedure. In keeping with the aim of the study to compare the incidence of bradycardia during sevoflurane induction in Down syndrome and normal children, we excluded patients undergoing emergency surgery, IV induction of anesthesia, or induction with a drug other than sevoflurane from both groups. The number of patients in the control group was chosen to have an approximate 1:1 ratio with the Down syndrome group.
The use of anticholinergic premedication, its dose, and its route of administration were recorded. The electronic anesthesia records reviewed contain vital signs data stored at 15-second intervals. The time period starting 60 seconds before the start of anesthetic induction and ending 6 minutes after the start of anesthetic induction was studied, and data on the vital signs, end-tidal gas concentrations, drugs administered, and other interventions recorded were extracted from the record and entered into a database. The start of anesthetic induction was identified as the first time point at which either a tidal volume or an exhaled gas (carbon dioxide or sevoflurane) was recorded in the electronic record. Expired sevoflurane data were available at 15-second intervals for 104 of the Down syndrome subjects anesthetized after May 2004, when an updated CompuRecord system was installed. Before that date, stored exhaled anesthetic data were captured in intervals that were >90 seconds.
Bradycardia and hypotension were defined by age groups in keeping with the criteria that would trigger activating a rapid response team for preventing an impending cardiac arrest, as established by the Pediatric Affinity group of the American Academy of Pediatrics, the National Association of Children's Hospitals and Affiliated Institutions, and the National Initiative for Children's Healthcare Quality (Table 1).a12–14 These criteria are well accepted in pediatric critical care as indications for the need for quick intervention. Children were determined to have bradycardia if the above criteria were met for 2 or more consecutive electronic anesthesia record entries (captured in 15-second intervals). Heart rate was obtained primarily from pulse oximetry data. The potential for monitor artifact being identified as bradycardia was reduced by confirming that the heart rate on the electrocardiogram was also low whenever possible.
In those children who experienced bradycardia, the first arterial blood pressure that was measured after the start of the episode of bradycardia was recorded. Any treatment of bradycardia noted in the record was recorded. The onset time of bradycardia after the start of anesthetic induction was recorded. The heart rate and pulse oximetry data downloaded from the CompuRecord were available at 15-second intervals for all subject encounters in both the Down syndrome and control groups.
A power analysis was performed by assuming that the incidence of bradycardia in the Down syndrome patient was similar to that reported by Borland et al. (3.7%) and that the incidence in the control population would be similar to that reported by Murat et al. (0.13%).7 A sample size of 230 would have an 80% chance of detecting such a difference at the 0.05 level of significance.
Results were expressed as means (±SD) and median with (interquartile range) for continuous variables and number (percentage) for categorical variables. A univariate analysis was performed to assess the association of the following factors with bradycardia: age, gender, ASA physical status (PS) as a dichotomous variable (ASA PS 1 or 2 vs. ASA PS ≥3), diagnosis of Down syndrome, anticholinergic premedication, presence of significant congenital heart disease, oxyhemoglobin saturation <90% during induction, peak expired sevoflurane concentration, and mean expired sevoflurane concentration during induction. Additional analyses were performed separately to determine whether bradycardia was more common in patients who were undergoing cardiac surgery, had previous cardiac surgery, had a history of right ventricular outflow tract incision, or had unrepaired cardiac lesions. Univariate analyses were performed using χ2 and Fisher's exact tests for categorical variables and t test for independent samples for continuous data. Variables with a P value of <0.1 were entered into a multivariate analysis, for which a backward stepwise multivariable logistic regression was performed to seek independent factors associated with bradycardia. Statistical analysis was performed with STATA version 10.1 (StataCorp, College Station, Texas). P values <0.05 were considered statistically significant. All P values were 2 sided. Both the univariate and multivariate analyses were conducted on the entire combined study populations of Down syndrome and control patients. The Hosmer–Lemeshow test was performed to determine the goodness of fit of the model.
A total of 567 anesthetic records of children with Down syndrome were identified. However, 295 were excluded for the following reasons: IV induction of anesthesia (139 cases), inhaled induction with halothane (28 cases), age older than 18 years (35 cases), heart rate data not present for >105 seconds after anesthetic induction (12 cases), and absent stored vital signs data in 81 patients anesthetized at locations where an automated computerized anesthetic record that included vital signs was not available. There were 63 repeat anesthetic encounters in the remaining 272 patients, which were excluded from the univariate and multivariate analyses, which was limited to the first anesthetic encounter in 209 subjects with Down syndrome and 268 controls. The age, weight, ASA PS, and gender data from all the patients are summarized in Table 2.
The overall incidence of bradycardia and hypotension was significantly higher in the Down syndrome group than in the control group for all ages (57% vs. 12%, respectively; odds ratio [OR] 9.56, 95% confidence interval [CI] 6.06 to 15.09). In the univariate analysis, other variables identified as being associated with bradycardia during induction of anesthesia included ASA physical status 1 or 2, a history of CHD, children with uncorrected CHD, anesthetics for children undergoing cardiac surgery, mean sevoflurane concentration, and occurrence of SPO2 <90% during induction (Table 3). When multiple logistic regression analysis was performed, the only variables that remained in the model as independent predictors of bradycardia were diagnosis of Down syndrome and ASA PS 1 or 2 (Table 4). The Hosmer–Lemeshow goodness-of-fit test statistic failed to reject the null hypothesis that there was no difference between observed and model-predicted values, implying that the model's estimates fit the data at an acceptable level (P = 0.208).15
The time to the nadir of the heart rate nadir was similar in the Down syndrome and control groups (190 ± 94 seconds vs. 196 ± 116 seconds, respectively; P = 0.52). More patients with Down syndrome received anticholinergic drugs after induction in comparison with the control group (24% vs. 0%; P < 0.001), with 14%, 5%, and 5% receiving IV atropine, IM atropine, and IV glycopyrrolate, respectively. No patient was treated with a sympathomimetic drug. Three Down syndrome subjects received oral premedication with atropine, but all 3 subsequently experienced bradycardia on anesthetic induction. None of the patients in the control group received an anticholinergic premedication.
Down syndrome patients were more likely to have an oxyhemoglobin saturation of <90% during the induction period studied (23/209 vs. 2/268, OR 16.5, 95% CI 4.3 to 63.8). However, hypoxemia was not an independent factor for bradycardia in the multivariate analysis.
In this study we found the incidence of bradycardia and hypotension during inhaled induction of anesthesia with sevoflurane in children with Down syndrome to be 57% in comparison with 12% in the control group. This is considerably higher than what was previously described in both the Down syndrome patient population and the healthy pediatric population.7,8 There are a number of reasons for this difference in incidence. The previously published studies relied on self-reporting through quality assurance systems and provider identification of bradycardia, which likely resulted in the lower reported incidence rate. Because of the fidelity and high frequency of data capture with our electronic anesthesia record-keeping system, the present study provides an objective evaluation of this phenomenon without reporting bias. Others have commented on the value of automated anesthetic records in such research.16
In addition these published observational studies did not provide a clear definition of bradycardia and hypotension, leaving it to the individual anesthesiologist to decide whether a given patient had developed that complication. In a recent article, Nafiu et al. have pointed out that the literature is without a robust definition of intraoperative hypotension in children.17 Age-related norms of blood pressure data in children focus on defining hypertension, not hypotension18,19; furthermore, these data are not from anesthetized children. However, during the last few years, many children's hospitals have publicized age-related heart rates and blood pressure values that should trigger the activation of a rapid response team to administer therapeutic interventions. Cardiac arrests outside the intensive care unit have decreased with the advent of these teams.b12–14 We chose these values as the criteria for bradycardia and hypotension in our study, realizing that intraoperative interventions by anesthesiologists differ from those by the rapid response team outside the operating room. For example, bradycardia often did not require treatment beyond decreasing the inspired sevoflurane concentration.
Data on the recorded blood pressure after episodes of bradycardia in Down syndrome subjects merit discussion. Our study depended on data in the automated anesthesia record in which heart rate, oxygen saturation, and gas concentrations were collected at 15-second intervals. In the absence of continuous arterial blood pressure monitoring via an arterial line, blood pressure data are collected only when the noninvasive blood pressure monitor is cycled. This is usually programmed to occur at 2- to 3-minute intervals, and so there may be only 2 to 3 blood pressure recordings made during the first 6 to 7 minutes of induction. Whereas a relatively high incidence of bradycardia was found in the Down syndrome subjects, the clinical importance of this finding ultimately relates to whether pharmacological interventions are required to manage bradycardia. The threshold for intervention with anticholinergic drugs is variable among individual anesthesiologists, because only 24% of patients with bradycardia received these drugs, and the others were managed with a reduction in inspired sevoflurane concentration. However, children with Down syndrome were more likely to be treated with anticholinergic drugs than were children in the control group (29/209 vs. 0/268; P < 0.001). We speculate that although there is often little hemodynamic consequence of bradycardia during sevoflurane induction in Down syndrome, it can be associated with profound hypotension in some patients with a rate-dependent cardiac output, particularly if other drugs such as dexmedetomidine are used. Investigation of strategies such as anticholinergic premedication to prevent this phenomenon of bradycardia on induction may be warranted. However, it is worth noting that 3 patients with Down syndrome received anticholinergic premedication in our study, but all 3 developed bradycardia on induction.
CHD was present in 120 of 209 (57%) Down syndrome subjects, which is in keeping with previous reports.20 In the univariate analysis, the presence of CHD, the presence of uncorrected CHD, and anesthetics for children undergoing cardiac surgery were all identified as being associated with bradycardia during anesthetic induction with sevoflurane. However, none of these were independent factors in the multivariate analysis. This suggests that the bradycardia on induction phenomenon occurs independently of overtly manifest structural heart defects and that other factors are responsible. Structural and ultrastructural myocardial abnormalities have been previously described in subjects with Down syndrome. Recalde et al., studying the size of cardiac muscle fibers in 15 Down syndrome patients without CHD, showed both an increased cell size and a reduced cell number per unit area. These data seem to support the previously reported suggestion that chromosome 21 controls the size and number of at least certain cell types.21,22 Others have shown that Down syndrome patients without CHD have left ventricular hyperkinesia, which does not seem to reflect intrinsic abnormalities of myocardial properties but a reduced afterload.23 Others have noted a reduced heart rate and blood pressure response to sympatho-excitatory tasks in patients with Down syndrome, indicating reduced sympathetic nervous system activity in these patients.24 We are not aware of in vitro data on the effects of inhaled drugs on contractility in myocardial fibers from normal and Down syndrome patients, but this may be worth studying.
It is interesting to note that the mean expired sevoflurane concentrations but not the peak concentrations were lower in Down syndrome subjects and in those with higher ASA PS than they were in healthy controls. This may reflect attempts by the more experienced anesthesiologist to prevent or attenuate the severity of the bradycardia phenomenon by reducing inspired concentrations quickly after reaching a peak concentration, particularly in patients with higher ASA PS. This may explain why a low ASA PS remained an independent factor for bradycardia.
This study has a number of limitations. It could be argued that the limits chosen for defining bradycardia and hypotension were not appropriate for all of the subjects. However, the literature describing bradycardia as an anesthetic-related complication includes vague or subjective definitions.6,7 We used heart rates and blood pressure values that would trigger a rapid response team. We believe that a different series of heart rate limits would not change our findings, namely that bradycardia and hypotension occurred more frequently in children with Down syndrome who underwent anesthetic induction with sevoflurane.
Another potential defect was that the study was retrospective and depended on the availability and accuracy of electronic data that are associated with artifacts such as those from patient movement during induction. The potential for monitor artifact being identified as bradycardia was reduced by using 2 consecutive anesthetic record entries to define bradycardia and by confirming whenever possible that the heart rate on the electrocardiogram and pulse oximeter were similar.
The study can also be criticized for not using controls matched for demographic data, surgical history, and coexisting medical conditions. However, the use of univariate and multivariate analyses of these variables showed that only Down syndrome and low ASA PS were independent predictors of bradycardia. This retrospective study has shown the increased susceptibility of a specific subset of pediatric patients (those with Down syndrome) to bradycardia during induction of anesthesia with sevoflurane. The clinical implication of this finding is that anesthesiologists taking care of these patients should be aware of the potential for this complication and attenuate its occurrence by a judicious reduction in the concentration of inhaled drug and also be ready to treat this condition quickly in the event that bradycardia does occur.
In summary, we have demonstrated that the incidence of bradycardia and hypotension in children with Down syndrome after inhaled induction of anesthesia with sevoflurane is high, and this phenomenon is independent of the presence of CHD.
a See http://www.nichq.org/pdf/RRTsupplementversionAug8th.pdf. Accessed October 22, 2009.
b See http://www.nichq.org/pdf/RRTsupplementversionAug8th.pdf. Accessed October 22, 2009.
1. Centers for Disease Control and Prevention. Improved national prevalence estimates for 18 selected major birth defects, United States, 1999–2001. Morbidity and Mortality Weekly Report 2006;54:1301–5
2. Van Praagh R, Papagiannis J, Bar-El Y, Schwint O. The Heart in Down Syndrome: Pathologic Anatomy. Baltimore, MD: Brookes Publishing Co., 1996
3. Eger EI. Anesthetic Uptake and Action. Baltimore, MD: Williams & Wilkins, 1974
4. Lerman J, Sikich N, Kleinman S, Yentis S. The pharmacology of sevoflurane in infants and children. Anesthesiology 1994;80:814–24
5. Sarner J, Levine M, Davis P, Lerman J, Cook D, Motoyama E. Clinical characteristics of sevoflurane in children. A comparison with halothane. Anesthesiology 1995;82:38–46
6. Kataria B, Epsein R, Bailey A, Schmitz M, Backus WW, Schoeck D, Hackl W, Govaerts MJM, Rouge JC, Kern C, Van Ackern K, Hatch DJ. A comparison of sevoflurane to halothane in paediatric surgical patients: results of a multicentre international study. Paediatr Anesth 1996;6:283–92
7. Borland LA, Colligan J, Brandom BW. Frequency of anesthesia-related complications in children with Down syndrome under general anesthesia for noncardiac procedures. Paediatr Anesth 2004;14:733–8
8. Murat I, Constant I, Maud'huy H. Perioperative anaesthetic morbidity in children: a database of 24,165 anaesthetics over a 30-month period. Paediatr Anaesth 2004;14:158–60
9. Saudan S, Beghetti M, Spahr-Schopfer I, Mamie C, Habre W. Cardiac rhythm and left ventricular function of infants at 1 MAC sevoflurane and halothane. Paediatr Anaesth 2007; 17:540–6
10. Roodman S, Bothwell M, Tobias JD. Bradycardia with sevoflurane induction in patients with trisomy 21. Paediatr Anesth 2003;13:538–40
11. Hoffman J, Kaplan S. The incidence of congenital heart disease. J Am Coll Cardiol 2002;39:1890–900
12. Sharek P, Parast L, Leong K, Coombs J, Earnest K, Sullivan J, Frankel L, Roth S. Effect of a rapid response team on hospital-wide mortality and code rates outside the ICU in a children's hospital. JAMA 2007;298:2267–74
13. Tibballs J, Kinney S, Duke T, Oakley E, Hennessy M. Reduction of paediatric in-patient cardiac arrest and death with a medical emergency team: preliminary results. Arch Dis Child 2005; 90:1148–52
14. Nowak J, Brilli R. Pediatric rapid response teams: is it time? JAMA 2007;298:2311–2
15. Hosmer D, Lemeshow S. Applied Logistic Regression. 2nd ed. New York: Wiley & Sons, 2000:150–1
16. Tremper K. Anesthesia information systems: developing the physiologic phenotype database. Anesth Analg 2005;101:620–1
17. Nafiu O, Voepel-Lewis T, Morris M, Chimbira W, Malviya S, Reynolds P, Tremper K. How do pediatric anesthesiologists define intraoperative hypotension? Paediatr Anesth 2009; 19:1048–53
18. Rosner B, Prineas R, Loggle J, Daniels S. Blood pressure nomograms for children and adolescents, by height, sex, and age, in the United States. J Pediatr 1993;123:871–86
19. Update on the 1987 Task Force Report on High Blood Pressure in Children and Adolescents: a working group report from the National High Blood Pressure Education Program. National High Blood Pressure Education Program Working group on hypertension control in children and adolescents. Pediatrics 1996;98:649–58
20. Freeman S, Taft L, Dooley K, Allran K, Sherman S, Hassold T, Khoury M, Saker D. Population-based study of congenital heart defects in Down syndrome. Am J Med Genet 1998; 80:213–7
21. Recalde A, Landing B, Lipsey A. Increased cardiac muscle fiber size and reduced cell number in Down syndrome: heart muscle cell number in Down syndrome. Pediatr Pathol 1986;6:47–53
22. Landing B, Shankle W. Reduced number of skeletal muscle fiber nuclei in Down syndrome: speculation on a “shut off” role of chromosome 21 in control of DNA and nuclear replication rates, possibly via determination of cell surface area per nucleus. Birth Defects Orig Artic Ser 1982;18:81–7
© 2010 International Anesthesia Research Society
23. Russo M, Pacileo G, Marino B, Pisacane C, Calabro P, Ammirati A, Calabro R. Echocardiographic evaluation of left ventricular systolic function in the Down syndrome. Am J Cardiol 1998;81:1215–7