Cardiac Point-of-Care Ultrasound and Multi-Disciplinary Improvement Opportunities in Acute Systolic Heart Failure Management in a Pediatric Emergency Center

Abstract

Cardiac point-of-care ultrasound (POCUS) has the ability to rapidly assess function and identify systolic heart failure (HF), an often-missed diagnosis. POCUS has the potential to expedite medical intervention, improving overall outcomes. There have been limited studies describing pediatric emergency center (EC) utilization of cardiac POCUS and its impact on outcomes in pediatric patients. Authors performed a retrospective chart review at a tertiary children’s hospital to identify all patients admitted from the EC to the Cardiac Intensive Care Unit (CICU) with acute systolic HF between January 2017 and August 2019. Outcome measures included EC length of stay (LOS), CICU LOS, and time until first IV HF medicine was administered. A total of 21 patients and 24 encounters meeting criteria were identified. Cardiac POCUS agreed with standard echocardiography in 8 of 9 cases. Patients who had a cardiac POCUS in the EC seemed more likely to receive their first dose of intravenous heart failure medication while in the Emergency Center (70% vs 43%). There was a trend toward significance, but it did not reach statistical significance (p = 0.1). EC and CICU LOS were not significantly different between POCUS and non-POCUS groups. Cardiac POCUS has the potential to have a valuable role in the early diagnosis of acute systolic HF in children. However, early diagnosis by POCUS did not translate into shorter EC or CICU LOS. This pilot data serves as a baseline for efforts to promote earlier clinical recognition of acute HF and more efficient collaboration between clinical services.

Introduction

Hospitalizations for acute decompensated heart failure (ADHF) in the United States affect about 14,000 children annually and are associated with significant morbidity, mortality, and healthcare costs [1]. Clinical presentation of pediatric heart failure varies based on underlying heterogeneous etiologies seen in children [2]. Nonspecific symptoms such as fatigue, pallor, feeding difficulty, abdominal pain, and vomiting are common [3], thus heart failure is often a missed diagnosis in initial presentations to providers in the primary care clinic or emergency center (EC) [4].

Cardiac point-of-care ultrasound (POCUS) has emerged as a potential solution to delayed diagnosis and treatment of heart failure in the emergency center. Cardiac POCUS is a focused ultrasound-based imaging strategy performed by a healthcare provider at the bedside to identify a limited list of cardiac findings which could be the etiology of the patient’s often critical presentation. Typically limited to qualitative, binary assessments of systolic function and the presence of pericardial fluid, cardiac POCUS is not comprehensive nor consultative cardiac imaging, e.g., echocardiography [5–7]. Multiple single center pediatric research studies show non-cardiologists utilizing cardiac POCUS, predominantly as a screening tool, to diagnose depressed systolic function with reasonable accuracy compared to the gold standard of echocardiography [7–12]. These studies are either evaluations of specific cardiac POCUS training programs for non-cardiologists [9, 10, 12] or they are evaluations of cohorts of credentialed cardiac POCUS providers working within POCUS imaging programs [7, 8], typically compromised of longitudinal training prerequisites, credentialing systems, image storage and reporting, and quality assurance programs. As a result, training and clinical use of cardiac POCUS are nearly ubiquitous in pediatric EC academic programs and supported by organizations such as the American Academy of Pediatrics, the American College of Emergency Physicians, and the American Society of Echocardiography [6, 7, 13].

Early diagnosis in the emergency center can lead to goal-directed HF medication initiation shortly after presentation, which has been shown to be associated with improved clinical outcomes in adult HF management [14]. To our knowledge, there are no studies measuring the clinical utility of cardiac POCUS in a cohort of pediatric heart failure patients. In this retrospective study, we aimed to evaluate the accuracy of cardiac POCUS use by a tertiary pediatric EC program with credentialed providers during the management of acute HF and measure the impact of cardiac POCUS on important clinical outcomes.

Methods

We conducted a retrospective chart review at a tertiary children's hospital to identify all patients admitted from the emergency center (EC) to the Cardiac Intensive Care Unit (CICU) with a diagnosis of acute HF between January 2017 and August 2019. Patients were included if HF was identified at the time of admission by CICU providers either as a primary diagnosis or comorbidity in the electronic medical record. Chart review was conducted to confirm newly decreased systolic ventricular function or persistence of severe systolic dysfunction with subsequent receipt of new intravenous HF medication within the first 24 hours of admission to the CICU. HF medications included intravenous diuretics, steroids (for acute transplant rejection), and vasoactive medications, which are recognized as first-line treatments for children and adults with symptomatic HF [15]. Patients with non-cardiogenic shock, primary arrhythmias, or medical disease were excluded. Patients with congenital heart disease (CHD) were excluded from the analysis as well as those presenting with ventricular assist devices. Basic demographic and clinical data were obtained including age, gender, and previous cardiac history.

Encounters that met inclusion criteria were then searched in the EC POCUS image storage and reporting database, Q-path (Telexy Healthcare, British Columbia, Canada), to identify patients who underwent cardiac POCUS in the EC by a pediatric emergency medicine physician. Trainees at our institution are considered eligible for POCUS credentialing after completion of pre-requisite education (e.g., > 30 h of didactics and 150 completed supervised scans). Credentialing for cardiac POCUS, at the time of this study, was achieved after the completion of at least 25 supervised cardiac exams with adequate quality images, accurate interpretation, and completed documentation [16]. Cardiac POCUS diagnostic targets and evaluations were limited to binary assessments (presence or absence) of ventricular systolic dysfunction, ventricular dilation, and/or pericardial effusion [5]. Decision to perform cardiac POCUS was determined by EC physicians and was based on availability of POCUS-credentialed staff.

Outcome measures included EC length of stay (LOS), CICU LOS, time until cardiac imaging, and time until the administration of the first intravenous HF or anti-rejection medication. Data were analyzed using SPSS Statistics 28 software. Mann–Whitney U-test was used for continuous and ordinal nonparametric variables to compare the cardiac POCUS group vs. non-POCUS.

Results

Patient Characteristics

A total of 228 unplanned admissions to CICU from the EC were evaluated. Enrollment criteria were met for 21 patients and 24 separate admissions; 3 patients presented twice during the time period (Fig. 1). The median age of included patients at presentation was 11.8 years (IQR: 4.9–16.8 years). Demographics are listed in Table 1. Cardiac encounter diagnoses include primary cardiomyopathy (n = 12), status post orthotopic heart transplantation (n = 8), and acute myocarditis (n = 4). Acute HF symptoms were new onset in 10 (42%) encounters.

Fig. 1.

Inclusion and exclusion criteria

Table 1.

Demographic information

Patient characteristics	Total N = 24	POCUS N = 10	Non-POCUS N = 14
Sex
Female	14 (58.3%)	5 (50%)	9 (64%)
Age at admission
Median (years)	11.8	11.8	10.5
IQR	4.9–16.8	6.5–15.8	4.8–17.5
Race
Caucasian	10 (41.6%)	3 (30%)	7 (50%)
Afr American/black	9 (37.5%)	5 (50%)	4 (29%)
Asian/Pacific islander	5 (20.8%)	2 (20%)	3 (21%)
Diagnosis
1o Cardiomyopathy	12 (50%)	4 (40%)	8 (57%)
Myocarditis	4 (16.7%)	3 (30%)	1 (7%)
S/p orthotopic heart transplant	8 (33.3%)	3 (30%)	5 (36%)
EC triage level
Triage median	2	2	2
1- Resuscitation	3	0	3
2- Emergent	19	8	11
3- Urgent	2	2	0

Ultrasound Evaluation in Emergency Center

EC staff trained in POCUS performed cardiac POCUS in 10 encounters (42%) with 9 formal reports documented. All patients had formal echocardiograms within 24 hours of arrival to the EC. Formal echocardiography was performed in the emergency center during 13 encounters (54%), while the remainder were completed after transfer to the CICU. POCUS examinations and reports were completed by or supervised by one of the credentialed EC attending physicians (n = 5). Regarding biventricular systolic function and the presence of pericardial effusion, cardiac POCUS agreed with contemporaneous standard echocardiography in 8 of 9 cases. POCUS missed moderately depressed function in one child with myocarditis, but correctly identified the concomitant pericardial effusion in the same patient. Median time to POCUS was 33 min from arrival to the EC (IQR: 27–71 min) with 8 of 10 being performed within 76 minutes of arrival to EC. POCUS was obtained significantly earlier than standard echocardiography (median: 126 min, IQR: 93–167 min); p < 0.01. Abnormal POCUS findings were re-confirmed by standard echocardiography in the EC in 4 encounters with a median wait time following POCUS of 81 minutes (range: 34–139 min). There was no difference in age (p = 0.78) or initial EC triage level (p = 0.14) between POCUS and non-POCUS groups.

Medical Management in Emergency Center

Prior to transfer to the CICU, 13 patients (54%) received their first dose of intravenous heart failure medication while in the emergency center. Patients who had a cardiac POCUS in the EC seemed more likely to receive their first dose of intravenous heart failure medication while still in the emergency center (70% vs 43%). There was a trend toward significance, but it did not reach statistical significance (p = 0.1).

Patient Outcomes

Two patients died after CICU admission (one from each cohort), and no patients died in the EC. Ten patients in the cohort required some form of mechanical support during admission to the CICU (table 3). Four patients required emergent extracorporeal membrane oxygenation (ECMO) support on the same day of admission to the CICU (two POCUS, two non-POCUS). Overall, EC and CICU LOS were not significantly different between POCUS and non-POCUS groups (p = 0.66, 0.54) (Table 2).

Table 2.

Patient outcomes

Patient outcomes	Total	POCUS N = 10	Non-POCUS N = 14	p value
Medical management	N = 24
Median time to first IV medication (minutes)	165	166.4	245.5	NS
Length of stay	N = 24
EC LOS (hours)	2.9	2.5	3.6	NS
CICU LOS (days)	11.8	12.4	10.6	NS
Clinical outcomes
Death	N = 2	1	1	NS
VAD	N = 6	3	3	NS
Emergent ECMO	N = 4	2	2	NS

NS non-significant

Discussion

This single center cohort is the first study evaluating not just the accuracy of POCUS but its impact on patient management and clinical outcomes in a large pediatric emergency center. Cardiac POCUS has the potential to have a highly valuable role in the early diagnosis of acute systolic HF in children as a screening tool in conjunction with gold standard tools such as echocardiography and other cardiac imaging. In the study, patients were able to get POCUS significantly sooner than conventional echocardiogram while still maintaining accurate findings. This is due in part to the relative ease of access to cardiac POCUS in the EC while a patient is undergoing intake and triage. Although this did not correlate to shorter EC or CICU lengths of stay, there was a trend toward significance in earlier administration of specified HF therapy in those who received POCUS versus patients who did not. This is an underpowered single center study, and we would expect this trend may achieve significance in larger studies given how early POCUS was able to facilitate the diagnosis of acute systolic HF.

Our study adds to the literature supporting the accuracy of pediatric cardiac POCUS in the hands of well-trained pediatric emergency medicine providers and is similar to their outcomes [11, 17, 18]. The only diagnostic discrepancy with contemporary echocardiography in our study was in a patient where the primary cardiac emergency (pericardial effusion) was correctly identified, and the systolic function was mischaracterized. The presence of the pericardial effusion also likely impeded the visualization of ventricular wall contractility. This rate of missed diagnosis is consistent with published literature regarding the accuracy of POCUS [11, 12]. It is important to remember that the high level of accuracy in this study comes from an experienced core of POCUS-credentialed providers in a high-volume pediatric EC with a robust system of image storage, reporting, and quality assurance. When evaluating POCUS accuracy, it is also important to remember its adjunctive role to formal cardiology or radiology imaging. The most well-established non-cardiac POCUS protocols, e.g., Focused Assessment in Trauma (FAST), have high false negative rates as a stand-alone test [19]. However randomized controlled studies have shown that the FAST exam has its impact when it is partnered with supportive clinical and radiology services. When placed into this type of collaboration, the FAST exam can directly improve patient transit times, lengths of stay, and other important patient outcomes during the triage of unstable trauma patients towards the operating room or to formal radiology imaging [20]. Therefore, we feel that describing the case as a “missed diagnosis” in our cohort does not correctly portray the clinical impact of cardiac POCUS. This previously healthy 10-year-old patient presented with cardiogenic shock due to myopericarditis at 2:30AM. Cardiac POCUS was performed within 15 min of arrival to EC triage desk and identified the pericardial effusion. The cardiology team was immediately called to the EC, and the formal echocardiogram was performed 35 min later, diagnosing the effusion, as well as depressed systolic function. He was transferred to the CICU and had an eventual complete recovery without needing pericardiocentesis or mechanical support. We feel that this case demonstrates that the screening cardiac POCUS at triage correctly identified the cardiac emergency to expedite clinical care. Later, formal cardiac imaging was used in the CICU to guide the cardiology teams to guide remainder of medical management.

The finding that consulting teams repeated cardiac imaging in 40% of the POCUS cohort in the EC suggests suboptimal information sharing between teams. In these cases, the diagnosis of newly depressed systolic function was made early in the EC course, but consulting teams proceeded down a parallel path diagnosing the same finding with standard echocardiography in the EC at a much later time. This reflects either a failure of EC staff to effectively communicate the findings, but more likely a refusal or reluctance of consulting teams to view POCUS images and reports. During the era of this study, POCUS stored images were not directly linked to the medical record system at the institution, but they could very easily be viewed at the bedside on the POCUS machine. We suspect that this lack of communication in several cases likely contributed in part to the inability to demonstrate a difference in EC LOS between the cohorts in our study. With POCUS democratizing ultrasound into more non-traditional pediatric subspecialties, our study demonstrates how collaboration and communication of even the most critical findings are not guaranteed. Adult hospital systems are building systems of collaboration and information sharing for POCUS findings, and some of the biggest success stories are with cardiac POCUS during the coronavirus pandemic [21].

Our trend towards significance for the initiation of appropriate HF medications in the EC after POCUS (70% vs 43%, p = 0.1) bolsters the potential utility of cardiac POCUS. HF medications are rarely administered in the pediatric EC; one pediatric tertiary center reported the use of inotropic agents only 15 times in a year out of more than 75,000 visits [22]. The earlier diagnosis of depressed systolic function may encourage EC providers, collaboratively with cardiology consultative service, to start medications in the EC prior to transfer to CICU. While it remains unproven in children, the initiation of appropriate IV HF medications has the potential to affect overall outcomes in pediatrics as has been shown in adult HF. A multicenter study of over 100,000 adults with acute HF found that early initiation of therapy in the EC was associated with improved hospital LOS even after adjusting for other risk factors. Patients who received IV therapy for HF in the EC were also less likely to transfer to the ICU and were more likely to be asymptomatic at the time of discharge [14].

The major limitation was the small sample size limiting the power of the study. It was conducted at a single center, which limits generalizability. The decision to perform POCUS was not protocol-based in the EC, therefore the time to POCUS was variable, which could have in turn affected the time to receive medications, disposition, and overall LOS. The retrospective, non-randomized nature of the study makes the comparison of cohorts of debatable value. However, the two groups seem to be relatively similar when comparing basic demographics, triage levels, and cardiac diagnoses. The overall cohort is likely a reasonable representation since the baseline demographics and eventual outcomes (including mortality and mechanical support use) are similar compared to the non-congenital heart disease patients of a large, multi-center registry-based pediatric series of acute HF patients admitted to the cardiac ICU [23].

This pilot data serves as a baseline for multi-disciplinary efforts to promote earlier clinical recognition of acute HF, more efficient communication with consultative services, and improved transit times to an appropriate unit. Further investigation, perhaps involving multiple centers or with randomization, regarding the clinical impact of cardiac POCUS in a pediatric EC is warranted.

Author Contributions

The authors confirm contribution to the manuscript as follows: CS and AR prepared the main manuscript text. Study conception and design: AR and AM. Data collection and database maintenance: KA, SL. Data analysis and interpretation: CS, AR. All authors reviewed the results and approved the final version of the manuscript.

Declarations

Conflict of interest

The authors declare that they have no relevant financial or non-financial conflicts of interest to disclose.

Ethical Approval

Approved by Baylor College of Medicine Institution Review Board.

Informed Consent

Not applicable.