726 - Caregiver and Patient Insights on the Use of Low-Field Point-of-Care Brain MRI in Pediatric Emergency, Hospital, and Critical Care

Publication Number: 3702.726

Christina Lee, Keck School of Medicine of the University of Southern California, Los Angeles, CA, United States; Jason M. Toliao, University of Southern California, Los Angeles, CA, United States; Kayla Guzman, Children's Hospital Los Angeles, Los Angeles, CA, United States; Samantha Lozano, Children's Hospital Los Angeles, Los Angeles, CA, United States; Payal Shah, Children's Hospital Los Angeles, Los Angeles, CA, United States; Frank Gonzalez, Children's Hospital Los Angeles, Los Angeles, CA, United States; Marvin Nelson, CHLA/USC, los angeles, CA, United States; Chia-Shang J. Liu, Children's Hospital Los Angeles, Los Angeles, CA, United States; Bradley De Souza, Children's Hospital Los Angeles, Los Angeles, CA, United States; Ryan Spurrier, Children's Hospital Los Angeles, Hermosa Beach, CA, United States; Peter Chiarelli, Phoenix Children's Hospital, Phoenix, AZ, United States; Natasha Lepore, Children's Hospital Los Angeles, Los Angeles, CA, United States; Pradip P.. Chaudhari, Children's Hospital Los Angeles, Los Angeles, CA, United States

Background: Current neuroimaging modalities for children such as CT and conventional MRI have limitations related to radiation exposure, sedation requirements, and limited scanner availability. Point-of-care, low-field MRI (POC LF-MRI) systems such as the Hyperfine Swoop have recently been developed, combining MR images acquired at low magnetic field strengths with automated optimization algorithms to generate clinically useful images. Understanding patient and family perspectives on use of POC LF-MRI will allow for improved implementation of this novel technology in acute care settings.

Objective: To describe patient and family attitudes toward POC LF-MRI and compare their experiences with standard clinical neuroimaging (CT or MRI) with POC LF-MRI.

Design/Methods: We conducted a prospective, single-center cohort study enrolling patients < 18 years undergoing standard-of-care neuroimaging (brain CT or conventional MRI) in the emergency department (ED), inpatient units, or intensive care unit (ICU) at a tertiary children's hospital. We performed a research POC LF-MRI study within 72 hours of a patient's clinical care neuroimaging. Participants/families completed a de novo oral survey after undergoing a POC LF-MRI. The survey consisted of 5-point Likert and yes/no questions assessing satisfaction, comfort, and preferences between modalities. Descriptive and comparative analyses were performed using Wilcoxon signed-rank and chi-square tests.

Results: We studied 137 participants who completed surveys (median age 12 years [interquartile range (IQR) 8-14]), including 94 (68.6%) who underwent CT and 43 (31.4%) who underwent conventional MRI (Table 1). The median bedside scan duration was 21.5 minutes (IQR 12-26). Overall, participants rated their experience with POC LF-MRI more favorably than with conventional imaging (median 4 [IQR 4-5] vs. 4 [IQR 3-5], p=0.003) and reported greater comfort during the POC LF-MRI scan (median 4 [IQR 3.5-5] vs. 4 [IQR 3-4], p=0.004) (Table 2). Fewer participants found the bedside scan intimidating compared to conventional neuroimaging (15% vs 32%). Most patient/families (n=86, 63.2%) preferred POC LF-MRI, while 16.9% (n=23) preferred conventional imaging and 19.9% (n=27) reported equal preference. Avoidance of radiation was a key factor for families, with 69% rating it as very or extremely important in their imaging preferences.

Conclusion(s): POC LF-MRI was perceived by patients and families as more comfortable and preferable to conventional neuroimaging. An additional preference on avoiding radiation highlights patient/family-centered support for POC LF-MRI as an emerging diagnostic tool.

Sample characteristics by type of imaging
Footnotes: all values represent n (%) or median [IQR]

Survey responses
Footnotes: all values represent n (%) or median [IQR]
1Wilcoxon signed rank; 2aWilcoxon signed rank one-sample test H₀: median=3 vs. H₁: median <3, 2bOne-sample test H₀: median=1 vs. H₁: median >1, 3Chi squared