Congenital Heart Surgeons' Society

Back to 2019 Program


Adult Congenital Heart Disease: A Framework For Classifying And Capturing The Spectrum Of Surgical Complexity
Tara Karamlou, MD, MSc, Pradeep Krishnamohan, MD, Xiaoje Huang, MD, Aisha Zia, MD, Eugene H. Blackstone, MD, David Majdalany, MD, Joanna Ghobrial, MD, Hani Najm, MD, MSc, Elizabeth Saarel, MD, Gosta Pettersson, MD, PhD.
Cleveland Clinic, Cleveland, OH, USA.

Background: Adults with congenital heart disease (ACHD) present complex problems encompassing more than the fundamental CHD. Current registries do not capture this complexity. To capture surgical complexity, we developed a framework for classifying ACHD fundamental diagnoses, procedural components, and indication for each. Methods: From 1/2000-1/2013, 7,428 adults with ACHD, mean age 55±16 years, underwent cardiac surgery. Indication for each surgical component was categorized as 1) initial CHD repair; 2) residual CHD repair; 3) re-intervention for CHD repair failure/complication; 4) intervention for associated CHD lesion; 5) intervention for unrelated acquired disease. Complexity was described by number of indications per fundamental diagnosis. Atrial septal defect (ASD) was utilized to depict associations between patient characteristics and complexity spectrum. Results: Most common fundamental CHD diagnoses were bicuspid aortic valve (N=3974), ASD (N=354), coronary anomalies (N=323), and aortic coarctation (N=136). Initial CHD repair was the most prevalent surgical indication for all fundamental diagnoses except coronary anomalies, for which unrelated cardiac pathology was most prevalent. For ASD, 47% (N=167) presented for primary isolated, 43% (N=153) with two indications, and 9% (N=32) with 3; the most common was initial repair and repair of unrelated acquired disease. Increased complexity was associated with older age (Figure). Conclusions: A conceptual framework incorporating complexity of surgery for ACHD and relating this to patient characteristics is critical to capture ACHD accurately and develop risk-adjustment models.