Tetralogy of Fallot trial protocol pits neonatal repair against delayed infant surgery
A multicentre randomized trial in China is set up to compare primary repair during the neonatal period with repair later in infancy for clinically stable term neonates with isolated tetralogy of Fallot. The study protocol, known as RIFAN, takes on one of the most difficult timing questions in congenital heart surgery: whether earlier definitive repair can offer a cleaner physiologic course than waiting, or whether the neonatal operation simply shifts risk into a smaller, more fragile patient.
The trial is focused on a carefully selected group. Clinically stable term neonates are eligible, while infants who need urgent neonatal intervention, were born preterm, or have low birth weight fall outside the planned population. That restriction matters because it narrows the question to a subset in whom equipoise is most plausible and in whom either operation timing could be defended on clinical grounds. It also means the findings, once available, will speak most directly to stable newborns rather than to the sickest infants who often dominate day-to-day decision-making in cardiac units.
Child health sits at the center of the protocol’s framing, and congenital heart disease remains a major contributor to infant morbidity and mortality. Tetralogy of Fallot is especially relevant because the lesion combines ventricular septal defect, right ventricular outflow obstruction, overriding aorta, and right ventricular hypertrophy, producing a physiology that can deteriorate as cyanosis, hypoxemia, and hypercyanotic spells emerge. Early complete repair removes the anatomy driving that physiology, but the operation in the neonatal period brings small-circulation challenges, narrower operative margins, and the burden of surgery in the first weeks of life.
Why the timing question remains unresolved in stable newborns
The protocol is built around a real surgical dilemma rather than a theoretical one. Neonatal repair may avoid the interval risks that come with waiting for infant surgery, including cyanotic episodes, feeding difficulty, and other interim decompensations that can occur before definitive repair. Delayed infant repair, by contrast, gives the child more time to grow, which can make operative management and postoperative recovery easier in practical terms.
No single timing strategy solves every problem. Neonatal repair compresses the entire burden of diagnosis, stabilization, and definitive intervention into a short window, while infant repair leaves room for interval events that must be tracked separately from the index operation itself. The protocol recognizes that distinction by requiring separate reporting of waiting-period events rather than folding them into the surgical endpoint alone.
That design choice is clinically important because a child who remains stable until planned infant surgery is not the same as a child who deteriorates while waiting. The former reflects a successful bridge to repair; the latter reveals the hidden cost of delay. By separating those events from the surgery-focused outcomes, the trial is structured to capture both the operative and preoperative consequences of each timing strategy.
Co-primary outcomes anchored to surgery and perioperative follow-up
The trial’s co-primary outcomes are tied to surgery, not merely to hospital discharge. Standardized perioperative and longitudinal assessments will follow participants beyond the index admission, allowing investigators to look past the immediate operative course and into the months that follow. For congenital cardiac surgery, that approach is especially useful because the story rarely ends when the chest is closed. Recovery, rehospitalization, growth, and evolving hemodynamics can all shape the real-world value of one timing strategy over another.
A short hospital stay can mask later problems. Longitudinal follow-up is the only way to separate an uncomplicated early postoperative recovery from a durable clinical advantage.
The protocol does not promise precision where event rates are low. Low numbers of outcomes can limit the certainty of effect estimates, especially in a population that is intentionally restricted to stable term neonates with isolated disease. Even a rigorously designed randomized study can struggle when the total number of major events is small, because wide confidence intervals can leave the surgeon with more caution than closure.
Masking limits, but outcome assessment remains protected where feasible
Blinding is not possible for the surgical team or caregivers because operative timing is obvious. Neonatal repair and infant repair are not interventions that can be hidden behind a placebo or masked schedule. The protocol instead uses masking where it can, with outcome assessment and analysis blinded when feasible. In surgical research, that is often the most realistic compromise: treatment allocation cannot be concealed from the operating room, but downstream adjudication can still be protected from bias.
This matters because timing studies are especially vulnerable to expectation effects. Teams may unconsciously interpret early recovery differently if they believe neonatal repair should outperform delayed repair, or vice versa. Masked analysis cannot erase all bias, but it can reduce the chance that subjective judgments about recovery, complications, or longitudinal status are tilted by prior belief.
The study also reflects a multicentre structure, which strengthens generalizability within the target population and reduces the risk that one center’s practice style will dominate the findings. Multicentre enrollment is particularly valuable in congenital heart surgery, where institutional experience, postoperative pathways, and technical preferences can influence outcomes as much as the timing decision itself.
What this protocol can answer—and what it cannot
If the trial is completed as designed, it should clarify whether planned neonatal repair offers a measurable advantage over waiting for infant repair in stable term newborns with isolated tetralogy of Fallot. It should also define the burden of adverse events during the waiting period and place those events beside the operative outcomes rather than treating them as background noise.
The protocol cannot answer every question that families and surgeons face. It will not directly inform care for preterm infants, low-birth-weight newborns, or those who require urgent neonatal intervention, because those children are outside the study population. It also cannot remove the clinical uncertainty created by low event rates, which may leave some comparisons underpowered for fine distinctions.
Even so, the study addresses a practical decision that is encountered repeatedly in congenital cardiac care: whether definitive repair belongs in the nursery or later in early infancy. In a lesion where physiology can worsen before the planned operation, the time interval itself becomes part of the treatment strategy. This trial is designed to measure that interval rather than ignore it, and that is where its value will ultimately lie.