Impact of Early Fortini Initiation on Growth and Interstage Outcomes in Infants With Single Ventricle Physiology

Infants with single ventricle physiology are among the highest-risk pediatric cardiac patients, particularly during the interstage period between Stage 1 and Stage 2 surgical palliation. Suboptimal growth during this time is a well-established predictor of increased morbidity, prolonged hospitalization, and adverse surgical outcomes. Despite advances in interstage monitoring and nutritional support, many infants fail to achieve adequate weight gain, and caregiver burden related to complex feeding regimens remains high.

Current feeding strategies often require fortification to 24-30 kcal/oz using breast milk or standard infant formulas (including hydrolyzed formulas), which may necessitate increased volume or complex mixing that is difficult to sustain at home. Fortini, a commercially available formula providing 30 kcal/oz, is commonly used in older children with failure to thrive or feeding intolerance. It offers the theoretical benefit of providing higher caloric density in a lower volume, which may be advantageous in infants with single ventricle physiology, who often have limited volume tolerance and high metabolic demands.

Preliminary experience and retrospective studies in infants with congenital heart disease suggest that increasing caloric density without increasing volume may support improved weight gain. However, Fortini's use in infants under 6 months-especially those with single ventricle physiology-remains unstudied in a prospective, controlled manner. No data currently exist on its safety, efficacy, or impact on caregiver experience in this vulnerable population. This study seeks to address this gap by examining whether introducing Fortini formula at discharge improves growth and caregiver-reported feeding outcomes in single ventricle infants, compared to current standard nutritional care.

DATA ANALYSIS PLAN:

1. Descriptive Statistics: Means +/- SD or medians with IQR will summarize demographic and clinical characteristics
2. Independent t-tests or Mann-Whitney U tests will compare continuous variables (e.g. WAZ change) between groups depending on normality
3. Chi-square or Fischer's exact tests will compare categorical outcomes (e.g. readmissions, WAZ categories)
4. Multivariable linear regression will adjust for key confounders such as feeding method (oral vs tube) and WAZ at discharge
5. Repeated measures ANOVA or linear mixed models may be used if interim weight data points are analyzed longitudinally
6. Caregiver stress scores will be analyzed with t-tests or Wilcoxon rank-sum tests and categorized for odds ratio as appropriate All statistical tests will use two-sided alpha = -.05 for significance. Analyses will be conducted using SASv9.4 or R.

POWER ANALYSIS:

Based on preliminary data and existing literature:

1. Expected difference in WAZ change between groups: 0.4 SD
2. Assumed SD of WAZ change: 0.6
3. Power: 80%
4. Alpha: 0.05
5. Attrition: 15%

To detect a clinically meaningful difference of 0.4 in WAZ:

1. Required sample size: 34 infants per group (68 total)
2. Accounting for 15% attrition: 80 infants enrolled Sample size calculations performed using G*Power 3.1

CONFOUNDING VARIABLES:

Key anticipated confounders:

1. Feeding method (oral vs NG/GT)
2. Baseline WAZ at discharge
3. Presence of genetic syndromes
4. Interstage complications (e.g. infection, hospitalizations)

Plan to address:

1. Stratified randomization by feeding type and WAZ (<-2 vs > or equal to -2)
2. Multivariable regression models to adjust for baseline imbalances

HANDLING MISSING DATA AND OUTLIERS:

1. Missing outcome data will be assessed for randomness

• If missing at random (MAR): handled using multiple imputation
• If not MAR will perform sensitivity analyses to assess impact

1. Outliers (e.g. extreme WAZ changes or caloric intake values) will be:

• Verified for data accuracy
• Retained unless biologically implausible
• Analyzed with and without to test influence on outcomes (robustness checks)