Effect of Gastrointestinal Malformations on the Outcomes of Patients With Congenital Heart Disease.

BACKGROUND: The goal of this study was to assess the effect of associated gastrointestinal malformations (GI) on the outcomes of patients undergoing congenital heart operations. METHODS: Neonates and infants with thoracic (esophageal atresia, tracheoesophageal fistula) and abdominal (duodenal stenosis/atresia, imperforate anus, Hirschsprung disease) GI malformations undergoing congenital heart operations between 1995 and 2015 were included. Two control groups were created, one for each group. Patients were matched by diagnosis, procedure, history of prematurity, presence of genetic syndrome, and a propensity score including weight and year of operation. RESULTS: The cohort included 383 patients: 52 (14%) with thoracic GI malformations and 98 (25%) thoracic GI controls, 80 (21%) with abdominal GI malformations and 153 (40%) abdominal GI controls. Median follow-up was 6 years (range, 16 days to 20 years). Patients with thoracic GI malformations had longer length of stay (p < 0.001), longer intubation times (p = 0.002), and higher perioperative death (p = 0.015) than controls. There was a tendency for worse overall survival than controls, mainly explained by the higher risk of early death (p = 0.06). No difference was found in outcomes between patients with abdominal GI malformations and controls. CONCLUSIONS: Patients with thoracic GI malformations have worse perioperative outcomes than controls, but their long-term survival does not seem to be significantly different. Abdominal GI malformations do not have a significant effect on outcomes. The presence of GI malformations should likely not preclude patients from undergoing congenital heart operations, but careful family counseling is necessary, especially for thoracic GI malformations.

Routine preoperative laboratory testing in elective pediatric cardiothoracic surgery is largely unnecessary.

OBJECTIVE: Routine preoperative laboratory testing (RLT) is common practice in pediatric cardiothoracic surgery and is associated with significant cost burden to patients and families. We sought to examine the value of RLT in patients undergoing elective pediatric cardiothoracic surgery. METHODS: We conducted a retrospective study of all scheduled elective pediatric cardiothoracic surgery at our institution from 2012 to 2014. Inpatients were excluded. Patient charts were reviewed to obtain preoperative laboratory values and determine relationship to case cancellation. RLT includes complete blood count, prothrombin time, partial thromboplastin time, urinalysis, 7 chemistry metabolic panel, electrocardiogram, and 2-view chest radiograph. RESULTS: RLT was completed for 1106 scheduled elective cases. Six (0.5%) cancellations were related to abnormal preoperative laboratory test results: 5 complete blood counts and 1 urinalysis. Hospital charge for RLT averaged $2064 per patient. Based on this incidence, we estimated that 184 routine preoperative laboratory tests, which generated a total hospital charge of $379,776, were required to capture 1 abnormal test significant enough to cancel surgery. An estimated charge of $2,169,552 was generated on prothrombin time, partial thromboplastin time, 7 chemistry metabolic panel, electrocardiogram, and 2-view chest radiograph, and none of these tests resulted in a cancellation. CONCLUSIONS: RLT does not significantly impact decision-making in elective pediatric cardiothoracic surgery. The decision to order a specific screening test should be clinically driven. Selective preoperative laboratory testing may have a positive impact on healthcare costs without affecting outcomes.

The Role of Echocardiography and Intracardiac Exploration in the Evaluation of Candidacy for Biventricular Repair in Patients With Borderline Left Heart Structures.

BACKGROUND: Predictors for single ventricle palliation (SVP) or successful biventricular repair (BVR) in patients with borderline left-side heart structures are not well defined. The goal was to evaluate the role of echocardiography and intracardiac exploration in determining feasibility of BVR. METHODS: All neonates surgically treated from 1995 to 2015 with mitral valve (MV), aortic valve, or left ventricle end-diastolic dimension z score of -2 or less for whom management was controversial were included. Data were analyzed using Fisher's exact test, Kruskal-Wallis test, and Kaplan-Meier analysis. RESULTS: The cohort consisted of 42 patients: 7 SVP (17%) and 35 BVR (83%). Median follow-up was 7 years (range, 6 months to 18 years). Intracardiac exploration was performed in 29 patients (69%). There was poor correlation between echocardiographic and intraoperative MV measurements (intraclass correlation coefficient 0.14). Preoperative echocardiography significantly underestimated MV size in 14 patients (54%). Two BVR patients were converted to SVP, and 4 (including 1 converted patient) had cardiac-related deaths. All patients with MV greater than 8 mm on preoperative echocardiography had successful BVR. An intraoperative MV less than 8 mm and an abnormal subvalvar apparatus was present in 5 of 6 SVP (83%) and 3 of 3 (100%) failed BVR patients who had intracardiac exploration, and in only 1 of 20 successful BVR patients (5%) who had an intracardiac exploration. CONCLUSIONS: The decision to proceed to BVR in patients with borderline left-side heart structures should not rely strictly on echocardiographic measurements. Intracardiac exploration of the MV and subvalvar apparatus is useful before committing a patient to SVP. Patients with low MV z scores, especially those with a normal subvalvar apparatus, may undergo BVR with good outcomes.