Patient Characteristics Associated with Hospital Admission or Antiarrhythmic Medication Changes After Emergency Department Evaluation of Supraventricular Tachycardia

Supraventricular tachycardia (SVT) is a relatively frequent diagnosis in the pediatric emergency department (ED). However, there are no consensus guidelines for ED disposition, and there are limited data on ED outcomes. Better understanding of those who are admitted or have antiarrhythmic medication changes may avoid potentially unnecessary transfers or admissions. Our objective was to identify patient factors associated with discharge from the emergency department without medication initiation or modification after management of SVT in the pediatric ED. A retrospective review of children aged 0–18 years seen in the emergency department for SVT was conducted using electronic medical record data over a ten-year period at a single academic tertiary children’s hospital. Patients with congenital cardiac disease or prior cardiac surgeries were excluded. Multivariable logistic regression analysis was used to determine association between patient factors of interest and the primary outcome of admission and secondary outcome of change to antiarrhythmic medications. We analyzed 197 patients encounters. The mean age was 7 years. Of these 104 (52.8%) were admitted to the hospital or discharged with antiarrhythmic medication changes. This primary outcome was associated with younger age (aOR 0.77, 95% CI 0.67–0.86), history of pre-excitation (aOR 5.82, 95% CI 2.01–18.8), intercurrent illness (aOR 3.75, 95% CI 1.27–12.1), number of adenosine doses prior to arrival (aOR 5.45, 95% CI 1.55–22.3), and in-person cardiology consultation (aOR 6.42, 95% CI 2.43–19.4). Nearly half of children treated in a pediatric ED for SVT are discharged without changes in medications. We identified patient factors associated with hospital admission or antiarrhythmic medication changes. These factors represent high value care and can be assessed when considering transfer from a referring facility. Risk stratification using these patient characteristics may reduce potentially avoidable transfers and admissions.


Background
Supraventricular tachycardia (SVT) is a frequent diagnosis in the pediatric emergency department (ED), affecting between 1 and 250 and 1 in 1,000 children [1].The management of pediatric SVT varies widely in EDs, [2] especially with regards to patient disposition after resolution of SVT.There are no consensus guidelines on which patients require a higher level of care, admission for further monitoring and management, or are safe for discharge after medication initiation or adjustment.Prior research demonstrates a discharge rate of 60% for pediatric patients treated for SVT with adenosine in pediatric EDs [3].Similar data are not available for patients treated in general EDs.
Avoidance of unnecessary transfers has become a point of interest in emergency medicine and prehospital research [4].Potentially avoidable transfers represent a significant burden to limited hospital-based care, emergency medical services (EMS), patients, and families [5].Previous studies have investigated ED transfers from urgent care centers, to define essential versus non-essential transfers based on a variety of patient care characteristics and diagnostic studies performed at the receiving center [6].To our knowledge, there are no studies investigating the demographics or outcomes of pediatric patients presenting with SVT who do not require hospitalization or direct management by cardiology.After the acute management of pediatric SVT, centers with limited access to pediatric resources are often challenged with determining safe disposition plans.
The objective of this study is to identify patient factors associated with discharge from the emergency department without medication initiation or modification after management of SVT in the pediatric ED.

Methods
We conducted a retrospective cross-sectional analysis of pediatric subjects ages 0-18 years old evaluated at a tertiary children's hospital for SVT from 2013 to 2020.Our facility serves as the only pediatric hospital in the region.The study was approved by the institutional review board of the University of Pittsburgh.
Children presenting to the ED and patients transferred from another hospital for a tachycardia-related complaint were included.We identified subjects with SVT by query of an electronic medical records for patients treated from January 2003-January 2020 with International Classification of Diseases (ICD) 9 or ICD 10 codes 427.0 (SVT).To capture patients with SVT who may have received incorrect ICD codes, we also searched 427.89 (Other Arrhythmia), 427.1 (ventricular tachycardia), 427.41 (ventricular fibrillation), and 427.9 (cardiac arrhythmia) and included those confirmed to have a final diagnosis of SVT for this encounter upon hospitalization or follow up.Atrial rhythms such as atrial tachycardia, atrial flutter, atrial fibrillation, and/or preexcited atrial fibrillation were not included.We also identified emergency department patients who received adenosine in the emergency department to identify additional subjects who did not receive an ICD code for SVT.Subjects with structural heart disease or previous cardiac surgery were excluded.
Electronic medical records for each queried patient were manually reviewed by two of the study authors (KS, CM).Those not meeting inclusion criteria were excluded after chart review.Investigators reviewed available prehospital records, ED notes, consult notes, hospital and ICU notes.All cardiology and ED documentation available until the date of data extraction in 2021 were reviewed for follow up information to determine recurrence of SVT within 48 hours.Patient factors including age, sex, medical comorbidities, presence of intercurrent illness, medications and interventions performed, transfer status from an outside facility, and modality of transfer were extracted from the medical record.To avoid confounding with patients with SVT but admitted to the hospital solely for non-SVT indications, we excluded those with intercurrent illness receiving hospital level care (IV fluids, oxygen, breathing treatments) but without strong indications for SVT monitoring such as recurrent SVT, albuterol therapy, racemic epinephrine therapy, not tolerating oral medication, or blunt chest trauma.
Subjects were defined as receiving "additional care" if they were (1) admitted for inpatient management or (2) had a change in maintenance therapy or were started on a new antiarrhythmic medication.The retrospective review of all patients meeting inclusion criteria was performed at a visitlevel (i.e.multiple visits for the same patient were analyzed separately), allowing inclusion of patients who were seen in the ED on more than one occasion for SVT.Descriptive statistics are all presented as mean ± standard deviation unless otherwise noted in cases where non-parametric distributions were present.
Bivariate logistic regression analysis was performed between all variables in Table 1 (age, maintenance therapy, medical comorbidities, history of SVT, history of pre-excitation, intercurrent illness, transfer from outside facility, number of adenosine doses prior to arrival, total number of adenosine doses, electrical cardioversion, duration of SVT at the time of evaluation, seen in person by cardiology) and no collinearity was established.Multivariable logistic regression analysis was performed between the primary outcome of meaningful additional care (combined admission to hospital and/or change in cardiac medications) and all patient demographic variables.Significance was set apriori at p ≤ 0.05 for all statistical comparisons (Prism 9.3.0,GraphPad, San Diego, CA).
On initial evaluation at the study site, 55 subjects (27.9%) were reported to have an intercurrent illness at the time of presentation.No patients were admitted to the hospital solely for management of the intercurrent illness.The duration of SVT was less than one hour in 47 subjects (23.8%), between one and four hours in 31 (15.7%), between 4 and twelve hours in 11 (5.6%) and greater than 12 h in one subject (0.5%).The remaining 90 subjects (45.7%) had unknown duration of symptoms.

Interventions Performed
At the time of presentation to the study site, 59 of 197 subjects (29.9%) presented as transfers from another facility.The majority (n = 139, 69.0%) did not receive adenosine prior to arrival at the study site after transfer via EMS or outside referral.In total, 87 subjects (44.1%) did not receive any doses of adenosine from prehospital, outside facility, or the study site.Of those who were chemically cardioverted with adenosine, 57 (28.9%) received one dose, 33 (16.8%) received two doses, 18 (9.1%)received three doses, and two (1.0%) received four doses.Six subjects (3.0%) received electrical cardioversion; four of these were completed by emergency medical services or a referring hospital.Six subjects (4.3%) arrived via transfer in active SVT.
Cardiology was consulted for in-person evaluation in the pediatric ED in 105 cases (53.3%).Baseline characteristics of those evaluated by cardiology versus not evaluated were compared (Table 2).The median age of those evaluated by cardiology was 5 years compared to 10 years in those not evaluated in person (p = 0.0047).A smaller percentage (45%) of those evaluated by cardiology had a history of SVT compared to those not evaluated (54%) (p = 0.013).
There was no difference in other medical comorbidities between groups (46.0%) vs. (46.7%).One-hundred eighteen subjects (59.9%) were discharged to home from the ED and

Multivariable Logistic Regression
A multivariable logistic regression model was created with the primary outcome of combined admission to the hospital and/or change in cardiac medication (

Discussion
The results of the present study indicate that a majority (59%) of patients were discharged to home without a change to their maintenance therapy after pediatric ED management of SVT.Previous research has demonstrated wide variation in admissions ranging from 19% to 85% at comparable children's EDs nationwide [2].To our knowledge, this is the first study evaluating patients undergoing changes to cardiac medications as a primary outcome.Multiple factors were found to be significantly associated with admission or change in cardiac medication.Younger age was associated with receiving additional care in this study.Younger patients (< 1-year-old) have a higher risk of SVT, and previous literature has shown that younger patients are less responsive to adenosine [7].79 (40.1%) were admitted to the hospital.A total of 88 children (44.7%) had changes made to their cardiac medication regimen, either in the ED or after admission.The combined outcome of admission to the hospital or change in cardiac medications occurred in 104 visits (52.8%).Of the 93 subjects discharged without medication changes, two (2.0%) returned to the ED within 48 hours with recurrence of SVT.
Of subjects transferred from a referring facility (n = 59), 26 (44.1%) were admitted to the hospital.Five children (8.5%) underwent changes to their cardiac medications and were subsequently discharged home from the pediatric ED.The remaining patients (n = 28, 47.5%) who presented as transfers were discharged to home with no changes in cardiac medications.Of these, (n = 19, 67.9%) were evaluated by cardiology.

Limitations
This is a single center study.As previously discussed, management of SVT is variable across sites and throughout the country, therefore the local practices reported in this study may not apply to every region.Diagnosis was by the ED physician's assessment.For some patients, this diagnosis may have been made based on monitor interpretation rather than EKG.Due to these factors, the authors do not have information on SVT subtypes.An additional limitation is the ease of pediatric cardiology consultation, which may increase the rate of which this consultation is obtained at the tertiary care center studied.Further, this study describes associations with receiving additional care, but we cannot determine how this care affects patient outcomes.A prospective quality improvement project looking at telemedicine and virtual consultation instead of transfer to a tertiary center for patients with SVT is currently underway.

Conclusions
Nearly half of children treated in a pediatric ED for SVT are discharged without changes in medications.Younger age, receiving multiple doses of adenosine, history of preexcitation, intercurrent illness, and in-person cardiology consultation are associated with hospital admission or antiarrhythmic medication changes.These factors represent high value care and can be assessed when considering transfer from a referring facility.Risk stratification using these patient characteristics may reduce potentially avoidable transfers and admissions.
Author Contributions KS wrote the first draft of the manuscript.KS, MM, GA, and JD provided study conception and design.JT provided statistical design, ran statistics, and created all tables.KS and CM provided chart review and extraction and manually screened charts for inclusion.All authors reviewed the manuscript and provided substantive feedback that were incorporated into the final product.Additionally, limited history in younger children make it challenging to determine the duration of SVT and gauge their risk of heart failure [8].For these reasons, many cardiologists recommend admission to educate families on heart rate monitoring.A history of pre-excitation was found to be associated with hospital admission or change in cardiac medication in this study.Underlying conditions, such as Wolff-Parkinson-White Syndrome, have been estimated to carry a lifetime mortality rate of 3-4% [9].Clinicians likely view instances of tachyarrhythmias in these patients as higher risk and therefore warrant additional monitoring.Patients with Wolff-Parkinson-White Syndrome also have a higher risk of recurrent SVT, perhaps influencing cardiologists to more closely adjust their medications if discharged [10].

Statements and Declarations
The presence of an intercurrent illness was also associated with admission to the hospital or change in cardiac medication.Acute illness is a known trigger of SVT in those prone to the condition.It is unclear from the data if these patients are admitted due to the acute illness or because SVT is viewed as a disease severity marker.
Patients receiving multiple doses of adenosine, representing refractory SVT, are likely to need prolonged critical care treatment time.We found that patients receiving additional doses of adenosine were more likely to be admitted to the hospital, likely due to concern for recurrence from the managing medical team.Further research is needed to determine if refractory SVT is associated with recurrence.
In person evaluation by a cardiologist was also associated with admission or discharge with medication changes.Patients evaluated by cardiology were younger, in part explaining this difference.Interestingly, cardiology evaluated in person a higher percentage of patients without a history of SVT.Perhaps this is again explained by younger age where evaluation of infants with first time SVT will not carry a prior diagnosis.Comorbidities were similar between groups, arguing against an assumption that cardiology preferentially evaluated more medically complex patients.Interestingly, factors such as time of day of arrival, distance traveled, transfer status, and air ambulance transfer were not associated with increased care.It may be interpreted that momentum at the time of transfer does not dictate care received after arrival.However, it does underscore the importance of critically evaluating these transfers as patients were often discharged after consultation with cardiology.This presents a unique opportunity for transfer centers to improve care by offering cardiology consultation to referring institutions.

Table 3
Results of multivariable regression analysis with combined endpoint of admission to hospital or change in maintenance therapy All authors contributed to the study conception and design.Material preparation, data collection and analysis were performed by Kyle Schmucker, Caroline Morris, Jennifer Dunnick, and Robert Tisherman.The first draft of the manuscript was written by Kyle Schmucker and all authors commented on previous versions of the manuscript.All authors read and approved the final manuscript.
Human Participants This retrospective chart review study involving human participants was in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.The Human Investigation Committee (IRB) of The University of Pittsburgh approved this study.