Study Design
A retrospective observational study was conducted in a high-complexity University hospital, La Fundación Valle Del Lili in Cali, Colombia. All pediatric patients (under 18 years old) undergoing any CHS between November 2019 and May 2022 were included. Patients with any oncoproliferative disease were excluded. We only analyzed the patient´s first cardiovascular procedure. There was no change in personnel during the study period.
Variables
Patient demographics, preoperative clinical status, surgery conditions, and outcomes were searched in the patient´s medical records. Each patient was assigned a study ID to ensure anonymity. The variables were annotated in the institution's clinical database management system by the researchers, a platform for protected data collection. For demographic variables, we identified the patients' age, sex, weight, and height. Preoperative clinical variables included risk factors and established pathologies including low weight for age, low weight for height, prematurity, history of cardiac surgeries, number of cardiac surgeries in the actual hospitalization, and history of pulmonary, hepatic, neurologic, or endocrine disease. For the surgery condition variables, we established the diagnosis, the type of surgery performed, the RACHS-1, and the level of urgency (elective or urgent/emergency) for each procedure.
The main outcomes evaluated in the study were complication, mortality, and FTR rates. The definition of each complication was based on the classification established in the STS-CHS database (version 2.50) [6]. We included cardiovascular (cardiac dysfunction, cardiac failure, cardiac arrest, arrhythmias, pericardial effusion, systemic vein obstruction, pulmonary vein obstruction), respiratory (respiratory insufficiency, pneumonia, pneumothorax, pleural effusion, pulmonary hypertension, chylothorax), renal (renal failure), infectious (sepsis, endocarditis), reintervention (secondary to bleeding, cardiac or non-cardiac causes, and unplanned catheterization), neurological (deficits, seizures, stroke, subdural hematoma, intraventricular hemorrhage), wound-related (dehiscence and surgical site infection), anesthetic-related, rehospitalization within 30 days, multiorgan failure and other postoperative complications. Delayed sternal closure was not included as a complication because the sternum was often left open due to institutional protocol and the patient's hemodynamic status.
Statistical method
All statistical analyses were performed using R statistical software version 4.2.1 (R Foundation for Statistical Computing) through RStudio 2022.12.0. For descriptive analysis, the normality of continuous variables was evaluated using the Shapiro-Wilk test. If the assumption of normality was rejected, these variables were presented as a median and interquartile range, and if not rejected, they were presented as a mean and a standard deviation. Qualitative variables were described using absolute frequencies and percentages. Operative mortality, rate of at least one complication, and FTR rate were calculated and presented with 95% confidence intervals. Univariate and multivariate logistic regression analysis was performed with mortality as the outcome variable. For the multivariate model, variables that were statistically significant based on their p-value and confidence intervals were selected. A statistical significance level of 5% was used in all tests.
Ethical considerations
This study was approved by the Institutional Ethics Committee (approval number 628–2022 Act No. I22-112 of November 02, 2022) following national and international recommendations for human research. In accordance with resolution 8430 of 1993, this study was considered risk-free, and the waiver of informed consent was requested and obtained.