In the evolving landscape of cardiovascular medicine, adults living with congenital heart disease (CHD) represent a uniquely complex and expanding patient cohort. Recent research unveiled at the 2026 Society of Thoracic Surgeons (STS) Annual Meeting by the Mayo Clinic underscores the pressing necessity for refined surgical risk prediction tools tailored specifically to this vulnerable population. Despite substantial advances in pediatric cardiac surgery that have enabled most children with CHD to survive into adulthood, these individuals often require multiple cardiac interventions throughout their lives, including high-risk reoperations that present challenges beyond the scope of existing adult cardiac risk assessment models.
Adults with CHD carry the burden of anatomical alterations and altered physiology resultant from their initial congenital defects and prior surgeries performed early in life. These patients’ cardiovascular systems have often undergone extensive remodeling, both structurally and functionally, rendering conventional surgical risk models—designed predominantly for acquired adult heart disease—inadequate for accurately predicting postoperative outcomes in this group. The need for personalized and reliable prognostic tools has become paramount, particularly to guide clinical decision-making and improve patient counseling regarding risks and benefits of complex cardiac procedures.
The study, spearheaded by Dr. Elaine Griffeth, a surgical resident specializing in general and thoracic surgery at Mayo Clinic, represents a seminal effort to bridge this knowledge gap using data-driven methodologies. By analyzing a comprehensive dataset derived from the STS Adult Cardiac Surgery Database (ACSD), encompassing procedures performed between July 2017 and December 2023 across diverse healthcare institutions nationwide, the research team harnessed advanced machine-learning algorithms alongside traditional logistic regression techniques to elucidate predictive factors for operative mortality and major postoperative complications in adults with CHD undergoing redo cardiac surgery.
A striking finding from this analysis reveals that approximately 16.7% of adults with congenital heart disease who undergo reoperative cardiac procedures are classified as high-risk for adverse postoperative events. These risks include increased mortality as well as severe complications such as the requirement for mechanical circulatory support, renal failure necessitating dialysis, neurologic events including stroke, and cardiac arrest. Such data underscore the critical nature of enhanced risk stratification models that can anticipate these outcomes more reliably than existing generalized scoring systems.
Central to the study was the identification of fifteen key clinical variables that demonstrated the greatest influence on postoperative risk prediction. The integration of these variables into a predictive model yielded robust discrimination capability, suggesting that a specifically tailored tool for adult CHD surgical risk assessment is attainable. This milestone validates the potential of combining machine-learning analytics with classical statistical frameworks to generate clinically applicable models that transcend institutional limitations, providing a versatile resource adaptable across diverse surgical settings.
Importantly, the research deliberately excluded certain patient subsets—such as individuals undergoing initial cardiac surgery, those with isolated bicuspid aortic valve disease, heart transplants, or isolated coronary artery bypass grafting (CABG)—to focus on the more representative adult CHD population encountering complex reoperations. This methodological refinement enhances the model’s relevance by concentrating on those with the intricate and heterogeneous surgical profiles characteristic of this field.
One of the inherent challenges in developing accurate predictive models for this demographic is the incomplete capture of certain critical clinical features in the ACSD, notably the presence of single-ventricle physiology—a significant risk determinant in long-term CHD outcomes. Dr. Griffeth and colleagues addressed this limitation by employing innovative analytical adjustments calibrated to the nuances of the nationally sourced dataset, thereby ensuring that the model faithfully reflects the surgical risk profile of the broader adult CHD population.
Dr. Griffeth emphasizes that surgical outcomes are influenced by a multifaceted interplay of patient-specific factors and the expertise of the multidisciplinary team involved in care. The heterogeneity in surgical practice patterns and resource availability across institutions can complicate the extrapolation of risk factors identified at a single center. Leveraging the extensive breadth of the STS ACSD, which aggregates data from thousands of surgeons and hospitals, allows for the identification of universally applicable predictors, thus standardizing risk evaluation and facilitating equitable clinical decision-making.
The enduring success of pediatric cardiac interventions now positions congenital heart disease as the most prevalent birth defect, with an estimated 1.4 million affected adults in the United States alone. This demographic surge accentuates the urgent demand for specialized risk calculators that integrate comprehensive clinical datasets to guide surgeons and patients through the intricate decision-making process surrounding reoperative cardiac procedures.
This research paves the way for the eventual development of a dedicated surgical risk calculator for adults with congenital heart disease—a tool designed to complement the suite of procedure-specific calculators already developed by the Society of Thoracic Surgeons. These calculators exploit the unparalleled scope and depth of the STS National Database™, fostering evidence-based, personalized care pathways that improve surgical outcomes across a spectrum of cardiovascular disease presentations.
Already, the STS National Database™ stands as one of the world’s most extensive clinical repositories, documenting nearly 10 million procedures by over 4,300 surgeons and encompassing approximately 95% of adult cardiac surgeries nationwide. The integration of adult CHD–specific risk models into this infrastructure represents a significant leap forward in precision medicine, offering tangible benefits for clinicians and patients alike by facilitating transparent, data-driven discussions on operative risks.
In conclusion, this groundbreaking work marks a critical stride toward enhancing the safety and efficacy of surgical management in adult congenital heart disease. As the field continues to evolve, the fusion of machine learning with traditional analytical methodologies promises transformative advances in perioperative risk evaluation. These innovations will ultimately empower patients and their care teams with the knowledge needed to make informed decisions, optimizing surgical outcomes for a growing population with complex lifelong cardiovascular needs.
Subject of Research: Prediction of surgical risk in adults with congenital heart disease using machine learning and statistical models.
Article Title: Study Examines Prediction of Surgical Risk in Growing Population of Adults with Congenital Heart Disease.
News Publication Date: January 31, 2026.
Web References:
https://www.cdc.gov/heart-defects/data/index.html
Keywords:
Health and medicine, Clinical medicine, Congenital disorders, Vascular diseases, Cardiovascular disorders
Tags: adults with congenital heart diseaseanatomical alterations in CHD patientscardiovascular medicine advancementsclinical decision-making in cardiac surgeryfunctional remodeling in congenital heart diseasehigh-risk cardiac reoperationsimproving outcomes in congenital heart disease patientsMayo Clinic research on CHDpediatric to adult cardiac care transitionpersonalized prognostic tools for CHDSociety of Thoracic Surgeons Annual Meetingsurgical risk prediction in congenital heart disease
