Point-of-Care Ultrasound in Children With Medical Complexity

This was a descriptive study evaluating the use of POCUS performed for clinical purposes in CMC admitted to a 50-bed pediatric PACH located in the northeastern United States. The PACH has plain radiography and x-ray availability 4-hours per day. Advanced imaging modalities, including radiology performed ultrasound, CT, and MRI, are not available and require transport to another local hospital when clinically indicated. All children for whom a POCUS was requested by a medical team provider (pediatric pulmonologist, pediatrician, and/or pediatric nurse practitioner) were eligible for inclusion. The study was performed over a 9-month period from May 2021 to January 2022. The study protocol was approved by the Institutional Review Board.

Demographic and clinical data were collected from the electronic medical record and included sex, age at time of POCUS, and admitting diagnosis. Because of the complexity of the clinical profiles, PACH primary admitting diagnosis retrieved from the patient’s history and physical in the electronic health record was used to assign a previously reported diagnostic category (prematurity, multiple congenital anomalies, neurologic or neuromuscular disease, other) for CMC in a pediatric postacute care hospital.^12–14  Technology dependence included ongoing dependence to sustain vital functions or multiple organ system involvement.

The number and specific type of POCUS application(s) performed (eg, cardiac, diaphragm, soft tissue), reason for POCUS referral (baseline, new clinical question, follow-up), and POCUS findings (within normal limits [WNL], abnormal) were recorded. Technically limited studies were deemed inadequate for analysis. Baseline referrals were sought for points of future comparison with POCUS in those patients assessed to be at risk for future changes (eg, lung and cardiac evaluation for those requiring long-term mechanical ventilation [MV]). Examples of clinical questions leading to POCUS referral included suspicion of pneumonia, soft tissue infection, diaphragmatic dysfunction, or cholecystitis. Follow-up studies were performed for previously identified abnormalities during the study period (eg, newly diagnosed pleural or pericardial effusion, atelectasis, or diaphragmatic dysfunction) or for abnormalities identified before transfer to the PACH (eg, hydronephrosis, hydrocephalus). All data were entered into a REDCap database.

An emergency medicine ultrasound fellowship-trained faculty member with 4-years of POCUS experience in a large academic center performed all ultrasound exams with an M8 ultrasound system (Mindray North America, Mahwah, NJ). The POCUS physician participated in daily ward rounds 1 to 2 times per week with the inpatient team. When a specific POCUS question was posed, the POCUS physician provided relevant background regarding the specific utility of the POCUS application requested, performed the scan if appropriate and within the scope of his training, provided immediate feedback to the inpatient team regarding the results, and documented a note in the patient chart regarding the POCUS findings. All video clips and still images were stored digitally for review. POCUS studies were deidentified and reviewed independently by 4 emergency medicine ultrasound faculty at the primary investigator’s academic site. Disagreements in POCUS interpretation were resolved by consensus opinion.

All data were analyzed descriptively. Age at the time of POCUS is presented as the mean ± SD for the total sample. Clinical variables (eg, sex, diagnostic categories, hospitalization history, type of technology dependence) and POCUS-related data variables (eg, POCUS type, reason for referral, findings) are presented as number and percentage. The number of technically limited POCUS studies was calculated as a percent of the total.

One hundred and four POCUS evaluations were performed for 33 patients. Twenty-three percent (24 of 104) of POCUS studies were requested to answer a specific clinical question, 56% (58 of 104) for follow-up information, and 21% (22 of 104) for baseline evaluation. For follow-up information specifically, 5% (5 of 104) of POCUS evaluations were performed for abnormalities identified before transfer to the PACH and 95% (99 of 104) for abnormalities identified at the PACH. The mean age at POCUS was 5.23 years (SD = 6.56, range 0.18–20.41) and 55% were male. Diagnostic groups for the 33 children included multiple congenital anomalies (41%), neurologic or neuromuscular (31%), prematurity (25%) and cardiac (3%). All children met the criteria for medical complexity. Characteristics of the study population are presented in Table 1.

The frequency of normal and abnormal findings by type of POCUS application are seen in Fig 1. Lung (n = 30, 29%), cardiac (n = 18, 17%), and diaphragmatic (n = 11, 11%) ultrasound accounted for 57% of the POCUS applications requested. Seventy-three percent (22 of 30) of lung ultrasounds were abnormal and revealed bilateral B-lines (n = 19), unilateral B-lines (n = 1), bilateral pleural effusion (n = 1), and/or atelectasis (n = 1). Eleven percent (2 of 18) of cardiac ultrasounds were abnormal, revealing pulmonary hypertension (n = 1) and pericardial effusion (n = 1). Eighty-two percent (9 of 11) of diaphragmatic ultrasounds were abnormal and findings included unilateral diaphragmatic paralysis (n = 1), unilateral diaphragmatic dysfunction (n = 4), and bilateral diaphragmatic dysfunction (n = 4). Forty three percent (3 of 7) of cranial ultrasounds were abnormal, indicating Grade II intraventricular hemorrhages (n = 1) and hydrocephalus (n = 2). All soft tissue, biliary, and liver-spleen measurements were normal. Examples of POCUS findings in our study are seen in Fig 2. Three of 104 (<3%) POCUS evaluations were technically limited studies.

Representative examples of POCUS applications, indications, findings, and integration into clinical follow-up care are seen in Table 2. In the examples listed, POCUS impacted clinical follow-up decisions including medication changes, imaging requirements, and hospital transfers.

The applications of POCUS in pediatric emergency medicine, pediatric critical care, and neonatology are extensive.^15–17  There are several reports on its use in diagnosis and management of pediatric patients with pneumonia,^18–20  bronchiolitis,^21,22  pneumothorax,²³  pleural effusion,²⁴  diaphragmatic dysfunction,^10,11  pericarditis and pulmonary hypertension,¹⁶  hydrocephalus,²⁵  intussusception,²⁶  appendicitis,²⁷  bronchopulmonary dysplasia (BPD),²⁸  and soft tissue infection.²⁹  The objective of this study was to describe the use of POCUS in CMC at 1 PACH. In addition, we describe the indications, applications, and POCUS findings at our single center PACH. This is the first description of the general applicability of POCUS for CMC.

Lung ultrasound was the most frequent POCUS application requested to evaluate for clinical questions, such as presence of pneumonia, pulmonary edema, or pneumothorax. When clinically indicated, lung ultrasound is a feasible and safe substitute for chest radiography when evaluating children for pneumonia^18–20  and has high sensitivity and specificity for the detection of pneumothorax and pleural effusions in children.^23,24  In the dyspneic patient, there is a substantial body of literature supporting the use of lung ultrasound for the presence of B-lines. B-lines (Fig 2B) are reverberation artifacts originating from the pleural line and can represent interstitial edema in clinical situations concerning for pulmonary edema, pneumonia, or viral disease in pediatrics including bronchiolitis and coronavirus disease 2019.^20,21,30 

It was an unexpected finding that 61% of lung ultrasounds were found to have predominantly bilateral B-lines. B-lines have also been described in BPD as a consequence of interlobar septal scarring secondary to MV or oxygen supplementation.^28,31  As no studies of lung ultrasound have been performed in this specific group of children, it is unclear at this time whether the findings of B-lines represent a forme fruste of chronic lung disease similar to BPD or other clinical entity.

Diaphragmatic ultrasound was requested for MV weaning difficulty in all cases and 82% identified unilateral or bilateral diaphragmatic dysfunction or paralysis. Previous work by our group evaluated 24 children requiring long-term MV and identified 54% with ventilator-induced diaphragmatic dysfunction with POCUS, a finding that may have important implications for weaning readiness and prognosis.¹⁰ 

Renal, vascular (DVT), and cranial POCUS exams were performed in 16% of cases. For clinical concerns of obstructive uropathy, point-of-care renal ultrasound has a high specificity for detecting hydronephrosis because of obstructive ureteral stones.³²  In patients with risk factors for DVT, POCUS has high sensitivity and specificity compared with radiology performed ultrasound.^33,34  For clinical concerns of hydrocephalus in infants (poor feeding, vomiting, fussiness, full fontanelle), point-of-care cranial ultrasound use in the acute setting has been described for the detection of hydrocephalus.²⁵  At this time, the literature is limited to descriptions of performing this POCUS application, case reports, and pilot studies.^25,35–38 

In our patient cohort, abnormal renal, DVT, and cranial ultrasounds were performed for follow-up of conditions diagnosed before transfer to the PACH and included hydronephrosis, DVT, hydrocephalus, and intraventricular hemorrhage. In these situations specifically, POCUS findings were compared with the referral hospitals ultrasound scans or reports. For the 3 abnormal cranial ultrasounds and 1 renal ultrasound, no change was noted on follow-up POCUS and no further interventions performed. For one of the previously known positive DVT ultrasounds, POCUS demonstrated persistence of a femoral DVT, and the patient was transferred back to the referring hospital for a repeat formal ultrasound study and further management of this finding.

Importantly, although POCUS was requested for specific clinical questions, it was predominantly requested to obtain baseline and follow-up information for previously performed POCUS studies at this PACH or at a referring facility. This finding is in marked contrast to current practice of POCUS in most settings (including emergency medicine and critical care) where POCUS use has been limited to addressing either acute clinical questions or changes in clinical status.³⁹  POCUS may offer a broader scope of practice in the postacute care or ICU setting where children with complex chronic conditions, technology dependence, and recurrent or prolonged hospitalizations are cared for. In this study, lung and cardiac ultrasound were commonly requested to obtain baseline or follow-up information in children with long-term MV, or to follow clinically relevant pleural effusions, atelectasis, or persistent pneumothorax.

POCUS can provide information acquired immediately at the bedside that is noninvasive, rapidly performed, cost-effective, and does not require patient transport or ionizing radiation.⁴⁰  However, several limitations and pitfalls in the use of POCUS specific applications have been identified and delineated extensively in the literature that can lead to misdiagnosis in the interpretation of POCUS findings with potentially negative consequences.^15,41,42  Most importantly, appropriate use of POCUS requires knowledge and training by the practitioner. Training and proficiency in POCUS have been addressed in policy statements from national and international organizations, including the American College of Emergency Physicians, the American Academy of Pediatrics, the Society of Academic Emergency Medicine, the World Interactive Network Focused on Critical Ultrasound, as well as expert guidelines.^39,43,44  Criteria for competency in the use of POCUS include: (1) recognizing the clinical indications and contraindications, (2) developing the technical skills for image acquisition, (3) having the ability to interpret images by distinguishing between normal anatomy, common variants, and a range of pathologic conditions, and (4) ability to integrate findings into clinical management.⁴⁵  Competency assessment benchmarks for trainees includes completion of 25 to 50 quality-reviewed POCUS exams for each application (eg, cardiac, lung, gallbladder) with 20% to 30% of examinations demonstrating pathology.⁴⁴ 

With the rapid adoption and widespread use of POCUS across several subspecialities, recent concerns of overuse of this technology has given rise to the concept of point-of-care ultrasound stewardship. The goal of POCUS stewardship is to ensure safe and effective integration of POCUS in routine clinical practice by considering appropriate clinical indications and integration of POCUS results to prevent overuse or misuse, with consideration of the pretest probability of disease.^46,47  Although this study is the first description of the use of POCUS in technology dependent CMC, POCUS stewardship is particularly applicable to this group of vulnerable children. In our study, we describe varied clinical indications for POCUS (eg, predominantly for baseline and follow-up information), as well as the use of less commonly performed POCUS applications (eg, cranial and diaphragmatic). Safe and impactful use of POCUS in CMC requires knowledge of potentially more diverse indications (as compared with the outpatient setting) as well as experience and skill in less commonly performed POCUS applications.

There were important limitations to our study. First, this was a single site study in a postacute care hospital, and the results may not be generalizable to other institutions or providers. Second, only 1 physician trained in POCUS performed all the ultrasound exams and although all scans were reviewed by 4 trained physicians, interrater reliability was not assessed. Third, our study does not address questions regarding the impact of POCUS on patient management decisions or further diagnostic testing.

POCUS use in a pediatric postacute care hospital can potentially be integrated into patient care decisions and may be useful in settings without access to radiology performed ultrasound. Lung, diaphragmatic, and cardiac ultrasound were the most prevalent POCUS studies requested in our postacute care hospital. POCUS may offer an expanded role for CMC in a PACH by answering clinical questions and, also, by providing baseline and follow-up information. POCUS training and proficiency, knowledge of indications and contraindications, and ability to integrate findings into clinical management are essential components of POCUS stewardship. Multicenter prospective studies should be performed to evaluate the impact of POCUS on patient management decisions and diagnostic testing in the postacute care setting for children with medical complexity.