The incidence of hypotension during sevoflurane induction in children with Down Syndrome; a prospective observational study

Inhalation induction using sevoflurane is most preferable in children. However, children with Down syndrome more frequently experience bradycardia with induction of anesthesia using sevoflurane when compared to healthy controls.^1,2

Previous studies by Kraemer and Bai identified increased incidence of bradycardia in children with Down syndrome while only the former concomittent hypotension.¹ However, the findings of either study cannot be considered definitive due to their retrospective study designs.^1,2

Isolated bradycardia without hypotension does not require treatment which may be detrimental to patients with underlying cardiac anomalies. We hypothesized that hypotension during induction would be independent of bradycardia. The primary aim of this study was to identify the incidence of bradycardia induced hypotension in children with Down syndrome during sevoflurane induction. Secondary aims include quantifying the association of sevoflurane concentration and bundle branch block on the incidence of bradycardia.

Following institutional review board approval and registration at clinicaltrials.gov (#NCT04603469;10/21/2020); patients were consented and enrolled at Texas Children’s Hospital between 05/26/21-10/28/2021. Inclusion criteria were surgical patients with Down syndrome, American Society of Anesthesiologists (ASA) status 1-3 and inhalation induction. Exclusion criteria were incomplete/mosaic Down syndrome, preoperative bradycardia, implantable pacemaker, unrepaired CHD, cardiac surgery, ventricular dysfunction, use of antihypertensive/anticholinergics, pulmonary hypertension and intravenous induction.

Sevoflurane concentration, titration, and use of nitrous oxide were at the discretion of the anesthesiologist. Standard ASA monitors were applied with blood pressure measured every 1 minute for five minutes. Heart rate was based on the ECG with comparison made to pulse oximetry reading. The NIBP, heart rate, and end-tidal sevoflurane concentration was recorded every minute for the first five minutes of induction from the initial use of sevoflurane. Providers were instructed to maintain their standard practice which included administration of anticholinergic and vasoactive drugs and use of manual stimulation. Following induction, intravenous access was established prior to airway manipulation. If the patient was ready for airway placement within 5 minutes with the use of intravenous medication they were excluded unless hypotension occurred.

Age, sex, weight, and the use of midazolam premedication previous ECG, echocardiogram (for residual defects) and cardiac surgery were recorded. The ECG was recorded as normal, left or right bundle branch block, or none available. The underlying cardiac lesion was recorded and classified as corrected or uncorrected. Simple unrepaired atrial and ventricular septal defects were not considered as CHD. Bradycardia and hypotension were defined a priori using Kraemer et al’s definitions.¹ Patients were classified as bradycardic or hypotensive with any heart rate or blood pressure below these thresholds.

Continuous variables were summarized as means and standard deviations while categorical variables were summarized as counts and frequencies. Student t-test and Fisher’s exact test was used to compare means and categorical variables by bradycardia status respectively. Unadjusted odds ratios with the outcome of bradycardia were derived on select variables. A random effects logistic regression model accounting repeated nature of the data was used to identify the association of sevoflurane on bradycardia. Based on time to heart rate nadir in patients with Down syndrome of (190±94 seconds) as previously reported, our data collection period of 5 minutes provided a study duration >1SD above the previously reported mean.¹ Assuming 60% as the rate of hypotension in bradycardic patients and 20% in non-bradycardia patients, we would have 80% power to detect a difference this large or larger with 23 patients in each group at alpha=0.05. We considered p-values <0.05 to be significant with analyses done using Stata/MP 15.1(StataCorp LLC, College Station, TX).

Forty-five patients were evaluated for inclusion with 38 included for analysis (suppl Figure 1). Demographic data is summarized in Table 1 (Table 1). Bradycardia was observed in 23(60.5%,95%CI(43.4%,76.0%)) patients within the first five minutes following anesthesia induction. Hypotension occurred in 5(13.2%,95%CI (4.4%, 28.1%)) patients within the first five minutes following anesthesia induction. There was no association between bradycardia and hypotension. In 15 patients without bradycardia, 2/15(13.3%,95% CI (1.7%, 40.55%)) developed hypotension compared to 3/23(13.0%,95%CI(2.8%,33.6%)) with bradycardia developing hypotension p=1.00 (Table 1). There was no difference in the incidence of bradycardia based on age, bundle branch block, or corrected CHD (suppl table 1). Exploratory analyses of variables associated with bradycardia is in supplemental table 1. While all patients experienced a decline in heart rate over the first 5 minutes, bradycardia patients had lower heart rates even at 1 minute compared with non-bradycardic patients (mean(SD): 91(29) vs. 119(30) beats per min; p=0.006) (suppl fig 2). There was no association between bradycardia and sevoflurane concentration (p=0.998) (suppl fig 3).

Table 1:

Patient Variables Stratified by Bradycardia Status. Abbreviations; ASA=American Society of Anesthesiologists.

Variable	Overall N = 38	No Bradycardia N = 15	Bradycardia N = 23	P-value
Age (years); Mean (SD)	8.6 (5.7) 95% CI (6.7, 9.5)	8.9 (6.8) 95% CI (5.1, 12.7)	8.4 (4.9) 95% CI (6.3, 10.5)	0.799
Weight (kg); Mean (SD)	30.6 (19.9) 95% CI (24.0, 37.1)	30.4 (22.0) 95% CI (18.2, 42.6)	30.7 (19.0) 95% CI (22.5, 38.9)	0.964
Female; (n, %)	16 (42.1%) 95% CI (26.3%, 59.2%)	4 (26.7%) 95% CI (7.8%, 55.1%)	12 (52.2%) 95% CI (30.6%, 73.2%)	0.182
ASA physical status; (n, %)
ASA = 1	0 (0.0%) 95% CI (0%, 9.3%)	0 (0.0%) 95% CI (0%, 21.8%)	0 (0.0%) 95% CI (0%, 14.8%)
ASA = 2	3 (7.9%) 95% CI (1.7%, 21.4%)	2 (13.3%) 95% CI (1.7%, 40.5%)	1 (4.3%) 95% CI (0.1%, 21.9%)	0.550
ASA = 3	35 (92.1%) 95% CI (78.6%, 98.3%)	13 (86.7%) 95% CI (59.5%, 98.3%)	22 (95.7%) 95% CI (78.1%, 99.9%)
Electrocardiogram Abnormalities; (n, %)
Normal	29 (76.3%) 95% CI (59.8%, 88.6%)	13 (86.7%) 95% CI (59.5%, 98.3%)	16 (69.6%) 95% CI (47.1%, 86.8%)
No Prior Electrocardiogram	2 (5.3%) 95% CI (0.6%, 17.7%)	1 (6.7%) 95% CI (0.2%, 31.9%)	1 (4.3%) 95% CI (0.1%, 21.9%)	0.568
Left bundle branch block	2 (5.3%) 95% CI (0.6%, 17.7%)	0 (0%) 95% CI (0%, 21.8%)	2 (8.7%) 95% CI (1.1%, 28.0%)
Right bundle branch block	5 (13.2%) 95% CI (4.4%, 28.1%)	1 (6.67%) 95% CI (0.2%, 31.9%)	4 (17.39%) 95% CI (5.0%, 38.8%)
Prior Congenital Heart Disease
No	21 (55.3%) 95% CI (38.3%, 71.4%)	10 (66.7%) 95% CI (38.4%, 88.2%)	11 (47.8%) 95% CI (26.8%, 69.4%)	0.736
Yes	17 (44.7%) 95% CI (28.6%, 61.7%)	5 (33.3%) 95% CI (11.8%, 61.6%)	12 (52.2%) 95% CI (30.6%, 73.2%)
Percent uncorrected among those with prior congenital heart disease *	7 (41.2%) 95% CI (18.4%, 67.1%)	3 (60.0%) 95% CI (14.7%, 94.7%)	4 (33.3%) 95% CI (9.9%, 65.1%)	0.593
Prior Correction of Structural Heart Disease; (n, %)
No	27 (71.1%) 95% CI (54.1%, 84.6%)	13 (86.7%) 95% CI (59.5%, 98.3%)	14 (60.9%) 95% CI (54.1%, 84.6%)
Yes	10 (26.3%) 95% CI (13.4%, 43.1%)	2 (13.3%) 95% CI (1.7%, 40.5%)	8 (34.8%) 95% CI (16.4%, 57.3%)	0.195
Unknown	1 (2.6%) 95% CI (0.1%, 13.8%)	0 (0%) 95% CI (0%, 21.8%)	1 (4.3%) 95% CI (0.1%, 21.9%)
Premedication (midazolam); (n, %)
No	30 (78.9%) 95% CI (62.7%, 90.4%)	8 (53.3%) 95% CI (26.6%, 78.7%)	22 (95.7%) 95% CI (78.1%, 99.9%)	0.003
Yes	8 (21.1%) 95% CI (9.6%, 37.3%)	7 (46.7%) 95% CI (21.3%, 73.4%)	1 (4.3%) 95% CI (0.1%, 21.9%)
Hypotension; n, %	5 (13.2%) 95% CI (4.4%, 28.1%)	2 (13.3%) 95% CI (1.7%, 40.5%)	3 (13.0%) 95% CI (2.8%, 33.6%)	1.000

^*Values are the percentage of patients uncorrected among those with prior congenital heart disease.

In this study, we did not see an association of bradycardia with hypotension in children with DS undergoing inhalation induction. Sympathetic nervous system undergoes rapid development throughout infancy with development of the parasympathetic system occurring throughout early childhood results in heart rates comparable to adults by adolescence. However, development of the sympathetic nervous system can be abnormal in children with Down syndrome resulting in relatively more pronounced parasympathetic (vagal) tone. Therefore, there may be both blunting of the sympathetic response occurring during inhalation induction as well as a more pronounced vagal influence resulting in bradycardia.

Previous studies examined the incidence of bradycardia and hypotension in children with Down syndrome.^1,2 Kraemer et al reported an incidence of bradycardia and hypotension to be 57% compared with 12% of controls; albeit the reporting is unclear as the majority of the data reported is only for bradycardia.¹ Our observed rate of bradycardia and hypotension of 13.0% is significantly different than Kraemer et al’s (p<0.001) and notably similar to their control rate.¹ A retrospective study by Bai et al. identified a similar incidence in hypotension in children with Down syndrome when compared with controls.² However, they identified bradycardia to have occurred in 25% of DS vs. 9% of controls and hypotension to have occurred in 73% of Down syndrome patients and 64% of controls.² These differences may be related to thresholds used to define bradycardia which were lower in the study by Bai et al.² Our cohort was examined using the blood pressure thresholds identified by Graaf et al. to determine how many patients fell below specific age, sex adjusted blood pressures by SD.³ 8 patients were at the mean, 19 were below −1SD, 1 patient was below −2SD while 6 and 2 patients were +1 and +2 SD above the mean respectively (1=−2SD;19=−1SD; 8=0SD;6=+1SD;2=+2SD).³ For those patients identified as hypotensive, 4 were within −1SD with 1 below −2SD.

In our cohort, prior conduction anomalies or cardiac surgery were not associated with bradycardia on induction with sevoflurane. This supports similar findings in prior retrospective studies in which CHD was not associated with bradycardia during induction.^1,2 Adults with Down syndrome have been shown to experience autonomic dysfunction irrespective of structural heart disease.^4–6 Therefore, it may be underlying autonomic dysfunction causing bradycardia during inhalational induction. Similar to previous studies, we observed that sevoflurane concentration was not associated with bradycardia.¹ Inhalation induced bradycardia is likely idiosyncratic and not reduced by incremental titration of sevoflurane. Limitations include use a study duration of 5 minutes and opting for routine practice vs. protocolized anesthetic concentrations. Future study would provide even more expansive conclusions.

In conclusion, bradycardia during inhalation induction was not associated with hypotension in children with Down syndrome. Bradycardia was not associated with the concentration of sevoflurane, age, weight, bundle branch block or previous CHD surgery.

Supplementary Material

fS1

Figure 1: Attrition table

fS2

Figure 2: Heart rate by study minute stratified by presence of bradycardia

fS3

Figure 3: Sevoflurane concentration (vol.%) by study minute stratified by presence of bradycardia