Outcomes in adult congenital heart surgery: Analysis of the Society of Thoracic Surgeons Database
Mascio CE, Pasquali SK, Jacobs JP, Jacobs ML, Austin EH
The Journal of Thoracic and Cardiovascular Surgery 2011; 142: 1090-1097.
Reviewer: Nicole S. Wilder, MD, FAAP
University of Michigan C.S. Mott Children’s Hospital
This study represents the first clinical multi-institutional report to describe the patient characteristics, most common operations, and operative morbidity and mortality of adults undergoing surgery for congenital heart defects (CHD). Improvements in surgical techniques and advancements in supportive disciplines over the past two decades have led to dramatically higher survival rates following surgery for CHD.
However, outcomes data for adults undergoing CHD surgery are limited with previous analyses focused on single-center experiences or administrative data. The Society of Thoracic Surgeons (STS) Congenital Heart Surgery Database collects perioperative data on all patients at participating institutions undergoing surgery for CHD.1 The investigators used this comprehensive dataset as well as the congenital procedures listed in the STS Adult Cardiac Surgery Database to provide new insights into the care and outcomes of this high-risk cohort.
Design:
This retrospective descriptive study included adults (aged ≥ 18 years) listed in the STS Congenital Heart Surgery Database (2000-2009). Data analysis was restricted to 68 participating centers with more than 85% complete data for all study variables. Data collected included demographic information, cardiac diagnoses, presence of noncardiac/ genetic abnormality, and preoperative clinical factors. Operative data included the surgical procedure performed, number of prior cardiothoracic operations, and cardiopulmonary bypass (CPB) times. Outcomes data included in-hospital mortality, total postoperative length of stay, and postoperative complications, including cardiovascular, pulmonary, neurologic, renal, infectious, and hematologic complications, as well as the need for unplanned surgical reintervention during the hospitalization.
Additionally, congenital procedures listed in the STS Adult Cardiac Surgery Database during the same time period were examined to permit consideration of the primary dataset within a larger context. Formal statistical comparisons were not made.
Results:
A total of 5256 patients from the STS Congenital Heart Surgery Database (68 centers) were included in the primary analysis. Median patient age was 25 years. Common perioperative risk factors were the presence of any noncardiac/genetic abnormality (17%) and preoperative arrhythmia (14%). The most common categories of operations were right ventricular outflow tract/pulmonary artery procedure (21%), pacemaker/arrhythmia procedure (20%), left ventricular outflow tract/aortic valve procedure (18%), ASD closure (7%), mitral valve repair or replacement (5%), and Fontan operation (4%, any type). In-hospital mortality for the overall cohort was 2.1%, 27% had 1 or more postoperative complications, and median postoperative length of stay was 5 days.
Pulmonary valve replacement was the most common operation with low in-hospital mortality (0.7%) and 5 day median length of stay. Patients with ASD were relatively older at time of operation (median, 33 years), had a mortality of 0% and a short length of stay (median, 4 days). The conduit operation cohort was notable for having a 4% rate of preoperative neurologic deficit, a high number of previous cardiothoracic operations (median, 2) and a relatively high rate of postoperative pulmonary complications. The aortic aneurysm repair subgroup had a high rate of noncardiac/genetic abnormality (24% vs. 17%) and the highest risk of postoperative heart block requiring a permanent pacemaker (2.2% vs. 1.3%).
The patients undergoing mitral valvuloplasty were predominately female (72%), with a high rate of noncardiac/genetic abnormality (27% vs. 17%). These patients also had a high risk of postoperative arrhythmia (16% vs. 9.4%) and of postoperative pleural effusion (8.2% vs. 3.1%). The Ross operation subgroup was relatively young (median, 22 years) and had a low rate of noncardiac/genetic abnormality (6.5% vs. 17%) and other preoperative risk factors (14% vs. 31%). The Ross subgroup also had a long CPB time and a high rate of bleeding requiring reoperation (8.3% vs. 1.9%). The Fontan revision/conversion cohort was notable for a high rate of any preoperative risk factor, long CPB times, high in-hospital mortality (11% vs. 2.1%), relatively longer postoperative length of stay (median, 10 days), and frequent postoperative complications (51%).
Review of the STS Adult Cardiac Surgery Database revealed 39,872 patients who had undergone ASD repair, pulmonary valve replacement, or other ‘‘congenital defect repair’’ either in isolation or along with coronary artery bypass grafting, aortic valve procedure, or mitral valve procedure. ASD repair patients were older (median, 48 years) with an in hospital mortality of 2.1%-5%. Patients undergoing pulmonary valve repair had an in hospital mortality 2.8%-3.5% while patients undergoing “congenital defect repair” had an in hospital mortality rate of 3.7%-4.5%.
Comment:
The strength of this study is the large patient cohort (> 5000) of adults undergoing CHD surgery. The overall mortality rate of 2.1% in the STS Congenital Heart Surgery Database and 2.1% to 5.0% in the STS Adult Cardiac Surgery Database is consistent with previously reported figures derived from an administrative database.2 The highest mortality (11%) in the single ventricle group is not unexpected and highlights the challenge of caring for this tenuous patient population.
The major limitations of Mascio et al’s study are related to the retrospective nature of a database study. Although the nature of the study is descriptive, statistical analysis of the data where appropriate would have strengthened the manuscript. More importantly, clarification of the reason for inclusion in the two separate databases would be helpful. For instance, can one presume that the surgeries entered into the STS Adult Cardiac Surgery Database were performed by adult cardiac surgeons versus congenital heart surgeons?
It was surprising to learn that the majority (88%) of the adult CHD cardiac surgery cases are presumably being done by adult cardiac surgeons. This is concerning when one compares the mortality rates for ASD and pulmonary valve operations in the two databases. The finding of 0% mortality for ASD repair in the STS Congenital Heart Surgery Database is expected and the 2.1% mortality rate for isolated ASD repair in the STS Adult Cardiac Surgery Database appears high. Likewise, the in hospital mortality rate for pulmonary valve replacement is 0.7% in the STS Congenital Heart Surgery Database versus 2.8% in the STS Adult Cardiac Surgery Database.
It remains uncertain whether limited comfort with the care of congenital heart patients by adult cardiac surgeons or non-pediatric tertiary care settings is related to the observed differences in mortality. Existing literature in this growing subpopulation is limited and this study provides important data regarding morbidity and mortality for a small subset of adults with CHD undergoing cardiac surgery. Generalization of these findings to the entire adult CHD population would grossly underestimate the mortality for the majority adults with CHD undergoing cardiac surgery.
Conclusions:
The population of adults with CHD requiring cardiac surgery will continue to grow and the authors should be commended on their effort to provide illumination on the perioperative outcomes in this important surgical population. Recent modifications to both the adult and congenital heart STS databases will permit more detailed analyses of the entire cohort of adults with CHD undergoing surgery in the years to come. Inspection of the differences between patient selection, surgical technique, and perioperative care of patients entered into these separate databases may yield important areas for future improvements in patient care.
References:
- STS Database Full Specifications. http://www.sts.org/documents/pdf/Congenital_DataSpecs_250.pdf.
- Karamlou T, Diggs BS, Person T, Ungerleider RM, Welke KF. National practice patterns for management of adult congenital heart disease: operation by pediatric heart surgeons decreases in-hospital death. Circulation. 2008;118:2345-52.
