541 - Feeding Recovery in Post-PICU Patients: Case Series in an Intensive Feeding Program

Publication Number: 1520.541

Tariq Almanaseer, Helen DeVos Children's Hospital, Grand Rapids, MI, United States; Ellen Hayhurst, Michigan State University College of Human Medicine, Grand Rapids, MI, United States; Jessica B. Doorn, Helen DeVos Children's Hospital, Ada, MI, United States; Ashley Bonebrake, Helen DeVos Children's Hospital, Grand Rapids, MI, United States; Brooke Dudick, Helen DeVos Children's Hospital, Grand Rapids, MI, United States; Elizabeth A. Rosner, Helen DeVos Children's Hospital, Grand Rapids, MI, United States; Nancy F. Bandstra, Helen DeVos Children's Hospital, Grand Rapids, MI, United States; Mara L. Leimanis-Laurens, Helen DeVos Children's Hospital, Grand Rapids_D Level, MI, United States

Background: Survival after pediatric intensive care unit (PICU) admission has improved, yet many children experience post-intensive care syndrome in pediatrics (PICS-p), including persistent feeding difficulties that impair growth and quality of life, and caregiver burden for the family. Adverse nutritional and oral experiences during critical illness (e.g., intubation, tube feeding) are linked to oral aversion and feeding challenges. Intensive feeding program (IFP) also known as Intensive Interdisciplinary Behavioral Treatment (IIBT) reduces tube dependence and improves oral intake; however, outcomes in PICU survivors remain understudied.

Objective: to evaluate feeding outcomes in children with prior PICU admission who completed IIBT and explore associations between PICU-related factors and treatment response.

Design/Methods: We conducted a retrospective case series of children (0-18 years) admitted to the HDVCH, Grand Rapids, Michigan- a high-volume, free-standing ,urban, tertiary-care children's hospital-who subsequently completed IIBT. Data were collected from 2007-2024. Variables included demographics; PICU course (admission indication, complications, length of stay, ventilation, nutrition status); and IIBT outcomes (feeding modality, oral skills, malnutrition status). Feeding outcomes were compared pre- and post-IIBT.

Results: 16 patients were included (62.5% female; mean age 7.4±1.9 years). Primary PICU admission causes were post-operative recovery (68.8%) and acute respiratory failure (25%). PICU complications included acute respiratory failure (43.8%) and need for respiratory support beyond baseline (62.5%). At PICU discharge, 75% remained tube-fed and 18.8% were malnourished. The mean time from PICU discharge to IIBT initiation was 641 ± 385 days. At IIBT baseline, 75% were tube-fed and all were non-self-feeders. Following IIBT completion (mean LOS 4.8±0.9 weeks), 58% of tube-fed patients achieved tube removal eligibility; 44% transitioned to partial or full self-feeding; problematic mealtime behaviors decreased (45.7%→9.9%); oral acceptance improved (62%→95%); and mouth clearance improved (59%→96%). Malnutrition prevalence decreased (20%→12%).

Conclusion(s): Children with prior PICU admission demonstrated substantial feeding and behavioral improvement during IIBT participation, with over half achieving tube-weaning eligibility. Time from referral to program start reflects barriers that delay intervention. These findings highlight the potential of IIBT to address feeding-related morbidity in PICS-p, and provide a unique lens into the long-term patient outcomes, after merging inpatient and outpatient data sets.

Table 1: Demographics. Table 2: causes of PICU admission
Categorical variables are summarized as count (%). Numeric variables are summarized as mean ± standard deviation when normally distributed. Numeric variables are summarized as median [25th percentile, 75th percentile] when non-normally distributed as median is a better measure of central tendency when a data point is not normally distributed.

Table 3. PICU admission complications. Table 4. PICU feeding characteristics - Baseline and discharge
Categorical variables are summarized as count (%). Numeric variables are summarized as mean ± standard deviation when normally distributed. Numeric variables are summarized as median [25th percentile, 75th percentile] when non-normally distributed as median is a better measure of central tendency when a data point is not normally distributed.

Table 5. PICU admission characteristics. Table 6. IFP characteristics. Table 7. IFP feeding characteristics - Baseline and discharge
Categorical variables are summarized as count (%). Numeric variables are summarized as mean ± standard deviation when normally distributed. Numeric variables are summarized as median [25th percentile, 75th percentile] when non-normally distributed as median is a better measure of central tendency when a data point is not normally distributed.