Abstract
Objective:
To assess the safety of topical epinephrine during pediatric functional endoscopic sinus surgery (FESS).
Methods:
After obtaining IRB approval, electronic medical records of patients aged 0 to 14 undergoing FESS in 2021 were retrospectively reviewed and divided into 2 cohorts based on the topical vasoconstrictive agent utilized during the case: oxymetazoline or epinephrine. Patients without documentation of vasoconstrictor utilization or those who received both or neither vasoconstrictive agent were excluded. Outcome variables consisted of preoperative and maximum intraoperative heart rate, systolic blood pressure, diastolic blood pressure, mean arterial pressure, as well as administration of intraoperative propofol, fluids, and antihypertensive medications. Patients without documented preoperative HR, SBP, and DBP were additionally excluded from the study. Mean differences and adjusted mean differences were estimated using linear models adjusted for baseline.
Results:
Among a total cohort of 123 FESS patients, 57 met inclusion criteria and were primarily male (63%, n = 36) with a median age at procedure of 10.94 years (IQR 7.66, 13.31). Oxymetazoline was administered to 71.93% (n = 41) of the cohort, while 28.07% (n = 16) received epinephrine. The unadjusted and adjusted mean differences between all preoperative and intraoperative hemodynamic parameters were not significantly different between cohorts (P > .05 in all cases). No intraoperative antihypertensives were administered to patients in the study.
Conclusion:
Topical epinephrine and oxymetazoline have similar risk profiles in pediatric functional endoscopic sinus surgery. A prospective study should be developed to evaluate the efficacy of topical epinephrine and to further assess its safety profile.
Introduction
Functional endoscopic sinus surgery (FESS) is frequently performed in pediatric otolaryngology for a variety of acute and chronic sinus conditions, with the most common indication being chronic rhinosinusitis (CRS). 1 Due to the nasal cavity’s highly vascularized nature, the use of topical vasoconstrictors for localized hemostasis is critical to improve intraoperative visualization and to reduce the risk of complications.2-5 Currently, oxymetazoline is the most commonly used vasoconstrictive agent in pediatric FESS, due to its well-established safety profile and effectiveness. However, the use of topical epinephrine is of growing interest for its powerful arteriolar and venous-sinusoidal constriction, which may improve visualization and reduce the frequency of application intraoperatively.6,7 Despite local application, systemic absorption of topical epinephrine may increase the risk of cardiovascular side effects such as tachycardia and hypertensive crisis, precluding its use in some patient populations.4,6,8
Topical epinephrine has demonstrated safety in the adult population with cardiovascular complications isolated to those with preexisting cardiovascular disease or those receiving injected epinephrine.2,5,9,10 Although topical epinephrine has demonstrated clinical effectiveness and a favorable safety profile in most adult patients, its safety profile in pediatric populations remains unclear, as few studies have specifically evaluated its use in children.6,7 Existing pediatric studies are characterized by limited reporting of dosing parameters and by evaluation of topical agents in isolation without accounting for concomitant use of injected vasoconstrictor, anesthetic agents, fluid administration, or operative blood loss. While the clinical efficacy of topical epinephrine may extend to the pediatric population, it is critical to first evaluate its safety. Thus, the aim of this study is to evaluate the safety profile of topical epinephrine in combination with injected epinephrine compared to topical oxymetazoline in combination with injected epinephrine during pediatric FESS.
Methods
Study Participants
This single-institution, retrospective study was approved by the Institutional Review Board (IRB) at Nationwide Children’s Hospital (NCH). Inclusion criteria consisted of all patients aged 0 to 14 years who underwent FESS (as determined by CPT codes 31254, 31255, 31253, 31257, 31259, 31256, 31267, 31276, 31287, and 31288) in 2021 at NCH, a tertiary care hospital (Figure 1). Patient charts were retrospectively reviewed to determine the topical vasoconstrictor applied prior to surgery. Cottonoid pledgets were soaked with oxymetazoline HCl 0.05% or topical epinephrine 1 mg/10 ml (1:10 000) and placed in the middle meatus bilaterally post-induction and prior to surgery and reapplied intermittently as needed to maintain operative hemostasis throughout the procedure. Per prior quantitative assessments, a standard 0.5 × 3-inch cottonoid pledget absorbs approximately 1.5 ml of 0.05% oxymetazoline and 0.0931 mg of 1:10,000 epinephrine.11,12 When the dosage was recorded in cottonoid units, the amount was estimated from the number of pledgets used (this occurred in 1 oxymetazoline-treated patient). Following pledget placement and prior to the procedure, epinephrine 0.01 mg/ml (1:100 000) was injected submucosally along the lateral nasal wall and middle turbinate bilaterally. Standard functional endoscopic sinus surgery tools were utilized, although instruments varied based on the needs of each case.
Study flow diagram description: patients outside the age range of 0 to 18.5 years, those who received both or neither vasoconstrictive agent, and patients lacking preoperative and intraoperative data were excluded from the study.
Patients were divided into 2 cohorts based on the type of vasoconstrictive agent utilized [topical epinephrine 1 mg/10 ml + injected epinephrine 0.01 mg/ml] or [topical oxymetazoline HCl 0.05% + injected epinephrine 0.01 mg/ml]. Patients were excluded from the study if they were outside the age range of 0 to 14 years, received both or neither vasoconstrictive agent, or lacked preoperative and/or intraoperative data.
Data Elements
Preoperative (PO) and maximum intraoperative (IO) heart rate (HR), systolic blood pressure (SBP), and diastolic blood pressure (DBP) were captured for each patient. Intraoperative hemodynamic factors, including doses of injected epinephrine, propofol, and net fluid volume were recorded, as well as any antihypertensive medications administered (including nitroprusside, nitroglycerin, labetalol, esmolol, fenoldopam, nicardipine, and clevidipine). In addition, demographic features, and clinical characteristics, such as surgeon name, indication for surgery, history of cystic fibrosis (CF), primary ciliary dyskinesia (PCD), and any known cardiovascular conditions (CD) were recorded. Cardiovascular conditions were documented if listed on a patient’s chart problem list and included arrhythmias, structural defects, murmurs, dyslipidemia, and Kawasaki disease.
Statistical Analysis
Demographic and perioperative clinical characteristics were summarized using count and percentage for categorical variables and median/interquartile range (IQR, 25th and 75th percentile) for continuous variables. All descriptive tables for patient characteristics are stratified by topical vasoconstrictor.
Primary outcomes were HR, SBP, DBP, and MAP from baseline to maximum intra-operative level. Unadjusted mean differences in increase between topical vasoconstrictors were estimated for each outcome using linear models adjusted for the corresponding baseline measure. Additionally, staged adjusted mean differences were first estimated by adding hemodynamic factors, including injected epinephrine dose, propofol dose, and net operative fluid volume, and then by further adding comorbidities, including the presence of cystic fibrosis, primary ciliary dyskinesia, and cardiovascular conditions.
Following best statistical practice and STROBE reporting guidelines, P-values are not presented in descriptive tables. For the main analyses, estimates and 95% CI intervals are presented, and P-values are interpreted on a continuum, with smaller P-values indicating stronger evidence that the data are incompatible with the null hypothesis.13-15 All analyses were performed using R version 4.3.0.16-19
Results
The study flow diagram depicts 123 observations included in the initial dataset, of which 6 MRNs were duplicated, and only the first chronological instance was included in the cohort, leaving 117 observations. Among the remaining 117 patients, 5 received neither agent, 11 received both topical agents, and 3 had no documented use of a topical vasoconstrictor to leave 97 observations. After applying the exclusion criteria, 63 patients remained eligible and received either [topical epinephrine 1 mg/ml + injected epinephrine 0.1 mg/ml] or [topical oxymetazoline HCl 0.05% + injected epinephrine 0.1 mg/ml]. Of these, 6 patients lacked operative hemodynamic data, for a total of 57 patients: 16 in the topical epinephrine cohort and 41 in the topical oxymetazoline cohort.
Demographic characteristics are summarized in Table 1. The cohort totaled 57 patients and was primarily male (63%, n = 36), White (73%, n = 41), and non-Hispanic (94.7%, n = 54) with a median age of 10.94 years (IQR 7.66, 13.31) at the time of surgery. Oxymetazoline was administered to 71.93% (n = 41) of the cohort, while 28.07% (n = 16) received epinephrine. A single surgeon performed all of the cases that utilized topical epinephrine. In contrast, the 41 procedures using oxymetazoline were completed by 14 different surgeons, including the surgeon who performed the topical epinephrine cases. Half of the entire cohort’s encounters (n = 29) were ambulatory surgeries, while 18% (n = 10) of the encounters were inpatient. Of the ambulatory surgeries, 56% (n = 9/16) received epinephrine, and all inpatient surgeries (n = 10/10) received oxymetazoline (Table 1).
Patient Demographics, Visit Information, and Relevant Conditions Stratified by Vasoconstrictor.
n (%); Median (IQR).
Fourteen percent of the cohort had cystic fibrosis, while 8.8% had primary ciliary dyskinesia, and 12% had a preexisting cardiovascular condition. Cardiovascular conditions included a dilated aortic root, left axis deviation, hypertriglyceridemia, mitral valve prolapse, dextrocardia, tricuspid valve insufficiency, a history of patent ductus arteriosus, and an unspecified murmur. Patients with cystic fibrosis and cardiovascular conditions had a higher incidence of receiving oxymetazoline than epinephrine, while patients with primary ciliary dyskinesia had a higher incidence of receiving epinephrine than oxymetazoline (Table 1).
Of the overall cohort, 68.42% (n = 39/57) underwent FESS for varying chronic rhinosinusitis management, of which 15 patients received epinephrine, and 24 received oxymetazoline. Other less common indications for FESS were acute sinusitis, tissue biopsy, orbital abscess, and frontal bone osteomyelitis (Table 2).
Indication for FESS as a Function of Topical Vasoconstrictive Agent Utilized.
n (%).
Median operative time for the overall cohort was 75 minutes, with shorter operative times observed in the epinephrine group compared with the oxymetazoline group, at 61 and 76 minutes, respectively. Net fluid was similar between groups, with 500 ml for the overall cohort (IQR: 345, 970), 585 ml for the epinephrine cohort (IQR: 335, 948), and 495 ml for the oxymetazoline cohort (IQR: 345, 970). Median injected epinephrine dose was comparable across groups (3.0 ml overall), and the mean propofol dose was similar among the groups at 100 mg (IQR: 45, 200) than in the oxymetazoline group at 100 mg (IQR: 70, 160). Topical epinephrine was administered only in the epinephrine group, with a median of 0.7 mg, and topical oxymetazoline in the oxymetazoline group, with a median of 10.0 ml (Figure 2). No intraoperative antihypertensives were administered to patients in either cohort (Table 3).
Distribution of topical and injected vasoconstrictor dosages. Description: Boxplots with individual data points depicting the distribution of topical oxymetazoline, topical epinephrine, and injected epinephrine dosages. Dosages are reported in ml for oxymetazoline and injected epinephrine and in mg for topical epinephrine.
Perioperative Hemodynamic Outcomes and Dosage Information Prior to and During Surgery Stratified by Vasoconstrictor.
Median (Q1, Q3).
Intra-operative measure is maximum during procedure.
Median preoperative HR, SBP, DBP, and MAP were 88 bpm (IQR: 74, 102), 113 mmHg (105, 105), 68 mmHg (63, 76), and 84 mmHg (80, 91) in the epinephrine group; and 85 bpm (79, 102), 109 mmHg (104, 119), 67 mmHg (60, 73), and 83 mmHg (76, 87) in the oxymetazoline group. Median intraoperative HR, SBP, DBP, and MAP were 122 bpm (IQR: 113, 142), 121 mmHg (105, 139), 77 mmHg (57, 91), and 94 mmHg (73, 103) in the epinephrine group; and 120 bpm (103, 131), 117 mmHg (104, 123), 70 mmHg (59, 81), and 87 mmHg (60, 120) in the oxymetazoline group (Table 3).
Prior to adjusting for confounding medications and comorbidities, the mean difference between cohorts was estimated after adjustment for baseline values of each outcome. Increase in heart rate was on average slightly higher for patients receiving topical epinephrine [Mean difference: 4.9 bpm, 95% CI: −7.5, 17, p: .4; Hemodynamic-adjusted difference: 5.5 bpm, 95% CI: −6.8, 18, p: .4; Hemodynamic- and comorbidity-adjusted difference: 8.9 bpm, 95% CI: −3.6, 21, p: .2]. Increase in systolic blood pressure was on average slightly higher for patients receiving topical epinephrine [Mean difference: 1.5 mmHg, 95% CI: −8.6, 12, p: .8; Hemodynamic-adjusted difference: 2.0 mmHg, 95% CI: −8.3, 12, p: .7; Hemodynamic- and comorbidity-adjusted difference: 5.7 mmHg, 95% CI: −4.5, 16, p: .3]. Increase in diastolic blood pressure was on average slightly higher for patients receiving topical epinephrine [Mean difference: 2.2 mmHg, 95% CI: −8.1, 12, p: .7; Hemodynamic-adjusted difference: 2.4 mmHg, 95% CI: −7.6, 12, p: .6; Hemodynamic- and comorbidity-adjusted difference: 5.4 mmHg, 95% CI: −4.9, 16, p: .30]. Increase in mean arterial pressure was on average slightly higher for patients receiving topical epinephrine [Mean difference: 1.5 mmHg, 95% CI: −8.1, 11, p: .8; Hemodynamic-adjusted difference: 1.9 mmHg, 95% CI: −7.6, 11, p: .7; Hemodynamic- and comorbidity-adjusted difference: 5.2 mmHg, 95% CI: −4.5, 15, p: .3]. No statistically significant differences were observed across all models (Figure 3).
Regression analysis of hemodynamic changes associated with topical epinephrine use description: forest plots depicting unadjusted and adjusted linear regression estimates comparing changes in hemodynamic parameters among patients receiving topical epinephrine. Models are presented as unadjusted, adjusted for baseline hemodynamic factors (injected epinephrine dose, propofol dose, and net volume), and adjusted for both baseline hemodynamic factors and comorbidities (cystic fibrosis, primary ciliary dyskinesia, and cardiovascular conditions). Effect estimates are shown as mean differences with 95% confidence intervals (CIs). No statistically significant differences were observed for heart rate, systolic blood pressure, diastolic blood pressure, or mean arterial pressure.
Discussion
Although data on topical epinephrine in pediatric FESS remain limited, our outcomes demonstrate a favorable safety profile compared with standard-of-care oxymetazoline. We found no significant differences in hemodynamic parameter changes between oxymetazoline and epinephrine. In unadjusted and adjusted linear regression analyses, topical epinephrine use was associated with small, non–statistically significant increases in heart rate and blood pressure parameters. After adjustment for hemodynamic factors (injected epinephrine dose, propofol dose, and net volume), estimated mean differences increased for HR, SBP, DBP, and MAP. After subsequent adjustment for hemodynamic factors and comorbidities (cystic fibrosis, primary ciliary dyskinesia, and cardiovascular conditions), estimated mean differences increased but remained non-significant for HR, SBP, DBP, and MAP, with all 95% confidence intervals crossing zero (all P > .20).
Adjustment for relevant covariates resulted in differing point estimates; while most associations were clinically plausible, none were clinically or statistically significant. One potential explanation for the increase in HR is the higher affinity of epinephrine for β1-adrenergic receptors at low doses, compared to oxymetazoline, which primarily exerts α1-adrenergic agonism and may induce reflex bradycardia. The small increases in blood pressure, none of which were greater than a 0.5 mmHg increase, after adjustment for hemodynamic factors were expected, as fluid administration combined with minimal blood loss during FESS increases intravascular blood volume, thus attributing the initial increase in blood pressure to fluid states, rather than to the topical epinephrine. In this cohort, patients received a net positive fluid balance of 500 ml over an average operative time of 75 minutes. However, the slight increase in the mean difference in blood pressure outcomes, all of which were less than a 4 mmHg increase, after adjustment for both hemodynamic factors and comorbidities, is not physiologically intuitive. After adjusting for comorbidities, the mean difference increased. However, of the cardiac comorbidities adjusted for, only the dilated aortic root observed in 1 patient within the oxymetazoline cohort could theoretically lower blood pressure. The other cardiovascular comorbidities, such as conduction abnormalities, valvular disease, congenital heart disease, and hypertriglyceridemia, would not be expected to substantially influence acute intraoperative heart rate or blood pressure, as seen in our results. Moreover, the majority of patients with these comorbidities more often received topical oxymetazoline rather than epinephrine, further arguing against a true underlying effect. Consistent with this interpretation, no patients in either cohort required intraoperative antihypertensive administration for hemodynamic instability.
The higher use of topical epinephrine among patients with primary ciliary dyskinesia likely reflects surgeon-specific practice patterns rather than patient-related factors, as nearly all PCD cases were performed by a single surgeon who frequently uses topical epinephrine. This provider did not operate on patients with cystic fibrosis, contributing to the decreased incidence of topical epinephrine use in this subgroup. As the selection of topical vasoconstrictor largely relies on the surgeon’s preference, we were unable to account for inter-surgeon variability in epinephrine use and the influence of surgeon-specific factors, including technique and experience, on outcomes such as the shorter operative times observed in the epinephrine group.
Despite concerns regarding adverse cardiovascular events, particularly in adults with cardiac comorbidities, topical epinephrine has repeatedly demonstrated safety in endoscopic sinus surgery, supporting its use to achieve hemostasis, enhance visualization, and optimize operative efficiency. Gunaratne et al 6 conducted a prospective study of 19 patients, with a mean age of 43, assessing cardiovascular effects of topical and injected epinephrine, and found no significant hemodynamic changes associated with topical administration, and only injected epinephrine statistically affected hemodynamic parameters. Upon subsequent retrospective review, Gunaratne et al. compared topical epinephrine (1:2000) to injected epinephrine (1:100 000), in which 2 patients of the 1260-cohort experienced cardiovascular complications, both being arrhythmias without postoperative residual ECG changes or cardiac enzyme abnormalities. 6 Importantly, both adverse events were attributed to injected epinephrine, supporting the safety of topical epinephrine (1:1000) for adults undergoing endoscopic sinonasal surgery. 6 To further assess the safety of topical epinephrine in adults, Dow et al 9 conducted a non-inferiority trial in 19 patients, with a mean age of 36, excluding all patients with pre-existing cardiovascular comorbidities, to compare 1:1000 to 1:10 000 concentrations of topical epinephrine. Dow et al 9 found that 1:1000 topical epinephrine provided comparable intraoperative hemodynamic stability while offering superior visualization. Despite these findings among adult populations, our study investigated the lower concentration (1:10 000) of topical epinephrine, as this dose is utilized at our institution, given that systemic absorption is always a possibility with mucosal administration of a drug, and administering the lowest dose required to achieve the desired effect is important, especially among pediatric populations. More recently, Nesbitt et al 2 supported the safety of topical epinephrine for hemostasis and visualization in FESS, cautioning against use only in patients with preexisting cardiac disease. This adult safety profile, together with emerging pediatric data, supports the use of topical epinephrine in children to achieve hemostasis.
The most recent systematic review, Macmillan et al, 20 assessed the safety and dosage of intranasal decongestants exclusively in pediatric patients undergoing nasal surgery. This study included 10 studies consisting of 5 case reports, 3 observational studies, and 2 randomized control trials, totaling 209 patients with ages ranging from 1 to 17 years old. 20 Most of the studies utilized topical decongestants such as phenylephrine, oxymetazoline, and xylometazoline. 20 The only study within the review to assess topical epinephrine was by Korkmaz et al, 21 who investigated topical 1:1000 epinephrine in 1140 patients, including 14 patients under the age of 18 undergoing nasal surgery. The study used 1:1000 epinephrine-soaked neuropatties, and no adverse cardiovascular events were noted in any of the pediatric patients. 21 However, there was no specific dosing regimen included in the article, and it included patients ≥ 15, thus limiting its use in determining a safe dosing criterion for topical epinephrine for pediatric populations. 21 Therefore, to our knowledge, our study is the first study to date that has shown a dose-dependent safety of topical epinephrine (1:10 000) use in pediatric patients undergoing FESS when compared to the standard of care oxymetazoline.
Although the safety profile of oxymetazoline has been well established, its operative use is not without risk. Two case reports documented significant cardiovascular events, including severe hypertension and reflex bradycardia, in previously healthy children following routine perioperative administration of oxymetazoline for nasal vasoconstriction during a turbinate reduction and nasal intubation. 22 However, the dosage of oxymetazoline was not recorded. This study highlights the importance of careful dosing and documentation when administering topical vasoconstrictors during pediatric sinus surgery, even when using the standard-of-care agent. 22 These occurrences demonstrate that systematic absorption of oxymetazoline is possible, despite its widely recognized safety profile. However, our findings support that the safety profile of topical epinephrine is comparable to that of oxymetazoline, and the observed differences were not clinically significant in pediatric FESS.
This study is not without limitations, which are primarily a function of the single-institution, retrospective design. This limits the generalizability of the data beyond our institution and prevents us from controlling for anesthetic and fluid administration, factors that could theoretically confound hemodynamic changes. Congruently, age and indication for FESS varied, and preoperative medical management, anesthetic agents, and intraoperative analgesic use were not standardized. Thus, presurgical medical management, operative anesthetic agents, and operative analgesic administration were not prospectively controlled; these variables should be evaluated in future prospective studies. To mitigate the potential influence of these factors on variability in hemodynamic measures, we utilized the mean differences rather than absolute values to assess changes in hemodynamic status to account for individual baseline differences within subjects, and to reduce the impact of inter-patient variability, and completed a regression analysis to assess the possible impact of hemodynamic confounders.
Conclusion
We found no complications associated with the use of topical epinephrine in pediatric patients undergoing FESS. Specifically, we did not find evidence that the use of topical epinephrine significantly alters intraoperative hemodynamic parameters when compared to the standard of care, oxymetazoline. Based on these findings, it is reasonable to develop a randomized noninferiority study to evaluate the efficacy of topical epinephrine and to further assess its safety profile.
Footnotes
Acknowledgements
The authors would like to thank the Department of Otolaryngology at Nationwide Children’s Hospital for their support in facilitating this study.
Ethical Considerations
This study was reviewed by Nationwide Children’s Hospital Institutional Review Board and determined to be exempt from full review.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr. Charles Elmaraghy reports professional relationships with Zotarix LLC, Grace Medical, Otologic LLC, and Engineering Mother LLC, all of which are unrelated to the work presented in this manuscript.
Data Availability Statement
The datasets generated and analyzed are not publicly available due to institutional restrictions and patient confidentiality, but are available from the corresponding author on reasonable request.