Reoperations in Adolescents and Adults After Prior Arterial Switch Operation: The Society of Thoracic Surgeons Congenital Heart Surgery Database Analysis.

Reoperations in Adolescents and Adults After Prior Arterial Switch Operation: The Society of Thoracic Surgeons Congenital Heart Surgery Database Analysis.

Cedars A, Jacobs ML, Gottlieb-Sen D, Jacobs JP, Alejo D, Habib RH, Parsons N, Tompkins BA, Mettler B. Ann Thorac Surg. 2024 Nov;118(5):1080-1087. doi: 10.1016/j.athoracsur.2024.05.038. Epub 2024 Jun 13.PMID: 38878952

Take-home Points:

  1. Late reoperation (> 10 years after ASO) is more likely to involve the LVOT, including the neoaortic valve and root, followed by RVOT and coronary artery procedures
  2. The frequency of left heart interventions is increasing relative to non-left heart procedures after ASO in late follow up
  3. Reoperation outcomes are good with low rates of post-operative mortality and complications (excluding those requiring mechanical circulatory support or transplantation)

Commentary By:

Dr Timothy Roberts

Dr. Timothy Roberts (Melbourne, Australia), section editor of ACHD Journal Watch

Introduction

Outcomes after arterial switch operation (ASO) are superior to those after atrial switch operation, but nonetheless present a unique set of complications including progressive neoaortic root dilatation, aortic insufficiency, pulmonary artery stenosis, and complications related to the coronary arteries. The incidence of subsequent cardiac surgery is increasing, with some single-center studies observing 15-20% of patients requiring repeated intervention by 15 to 20 years after ASO. Early interventions (within the first decade after ASO) are predominantly directed at the right ventricular outflow tract (RVOT), whereas the relative frequency and types of late reoperations during adolescence and adulthood remain less well defined.

Study Design:

  • Retrospective cohort study utilizing data from The Society of Thoracic Surgeons Congenital Heart Surgery Database (STS CHSD)
  • Identification of cardiac surgical interventions in patients aged 10 years and older who had previously undergone ASO for TGA or DORV/TGA type
  • Hospital encounters with admission date between January 1, 2010, and December 31, 2021
  • Inclusion criteria: any patient undergoing any cardiovascular surgical operation during the study period with age at surgery 10 years or older, with a fundamental diagnosis of:
    • TGA with IVS
    • TGA, IVS with LVOT obstruction
    • TGA with VSD
    • TGA, VSD, LVOT obstruction
    • DORV, TGA type. with a diagnosis code indicating one of the following: status post ASO; status post ASO and VSD repair; status post ASO and aortic arch repair; or status post ASO and VSD repair and aortic arch repair.
  • Exclusion criteria: any diagnosis of status post Mustard, status post Senning, congenitally corrected TGA, or single ventricle of any type.
  • Hierarchical stratification of procedure categories established a priori by investigators to identify the procedure of “principal intent” in those undergoing multiple procedures.
  • Primary outcome: discharge mortality
  • Secondary outcomes:
    • any 1 of more than 6 STS CHSD major complications: renal failure, neurologic deficit, arrhythmia necessitating permanent pacemaker placement, post operative mechanical circulatory support, paralyzed diaphragm, or unplanned reoperation. 
    • Composite of mortality/morbidity 
    • Postoperative length of stay for those who survived to discharge.

Key Finding

  • Cohort demographics:
    • 698 hospital encounters, comprising 651 patients
    • Mean age 16 (IQR 13-21); 33% female
    • Prior history: status post ASD (55%), ASO with VSD repair (33%)
    • Fundamental diagnoses: TGA with IVS (42%), TGA with VSD (36%), DORV with TGA (18%); 36 patients with history of prior palliative pulmonary artery banding
  • Most procedures in patients aged 10-18 years, with frequency declining thereafter; 78 (11.1%) patients aged 25 years or older at time of surgical intervention
  • Most common procedural categories: RVOT and pulmonary artery procedures (n=235, 34%), aortic valve procedures (n=146, 21%), and aortic root procedures (n=117, 17%)
  • Mortality rare, except for hospital encounters limited to isolated mechanical circulatory support procedures (4/8, 50%)
  • Overall discharge mortality 2.3%; no difference across procedural categories
  • Significant difference between groups for the combined end point of morbidity/mortality (P < 0.001) due almost exclusively to occurrence of major complications (P < 0.001)
  • Differences in complications between groups driven by renal failure (P = 0.002) and unplanned reoperation (P = 0.003)
    • Renal failure occurred exclusively in patients undergoing transplant procedures (n=3, 18%), aortic root procedures (n=3, 2.6%), aortic valve procedures (n=1, 0.7%), and other left-sided heart procedures including mitral valve procedures (n=1, 7.7%)
    • Unplanned reoperation occurred in most groups, most common for those who underwent transplantation procedures (n=4, 24%), aortic root procedures (n=19, 16%), and aortic valve procedures (n=15, 10%)
  • Length of stay significantly different between groups (P < 0.001) driven by long length of stay in the transplantation procedures group (27 days; IQR 17-58 days)

Strengths:

  • Data contributed by over 90% of pediatric cardiothoracic surgical programs in the United States
  • Relatively large sample size for ASO population
  • Sensible hierarchical procedural categorization of operations

Limitations:

  • Total number of patients having undergone ASO without a requirement for reoperation cannot be determined from this database
  • Strong potential for missed adult patient procedures at centers participating only in the STS Adult Cardiac Surgery Database
  • Unable to evaluate either anatomic or surgical risk factors for reoperations after ASO
  • Data on catheter-based interventions not able to be considered due to the available data source

Discussion:

This retrospective analysis provides further data to consider and discuss with families of children being considered for ASO.  The frequency of left heart interventions 10 years or more after ASO increases, with relatively low perioperative mortality, and the likelihood of requiring reoperations to address LVOT problems is expected to become more common with an aging post-ASO population.

Conclusion:

Patients post ASO are more likely to require reoperation for abnormalities in the LVOT (aortic root/valve procedures) rather than the RVOT, 10 or more years after ASO. The frequency is increasing over time, with generally very good outcomes regardless of the procedure performed.