Search

OBJECTIVES Neonatal cardiac surgery for congenital cardiac defects is associated with significant morbidity and mortality, and there is a need for early identification of patients at highest risk of adverse outcomes. Because vascular endothelial injury mediates damage across organ systems, we measured serum biomarkers of endothelial injury in neonates following cardiopulmonary bypass and examined their associations with short-term outcomes. DESIGN Prospective cohort study. SETTING Pediatric cardiac ICU. PATIENTS Thirty neonates less than 2 weeks old undergoing repair of congenital cardiac defects with cardiopulmonary bypass. INTERVENTIONS None. MEASUREMENTS AND MAIN RESULTS Biomarkers of endothelial integrity, angiopoietin-1 and angiopoietin-2, were measured preoperatively and at 24 hours postoperatively. A composite adverse outcome was defined as any of the following: stroke, need for renal replacement therapy, extracorporeal membrane oxygenation support, cardiac arrest, or death. Associations of biomarkers with adverse outcomes were examined using Wilcoxon rank-sum test. There was an increase in angiopoietin-2 from preoperatively to 24 hours postoperatively (p < 0.0001) and a decrease in angiopoietin-1 from preoperatively to 24 hours postoperatively (p < 0.0001). Patients with greater rise in angiopoietin-2 from preoperatively to 24 hours postoperatively had greater risk of composite adverse outcome (p = 0.04). They had a trend toward higher Vasoactive-Inotropic Score (p = 0.06) and a higher prevalence of low cardiac output syndrome (p = 0.06). Twenty-four hour postoperative angiopoietin-2 level was associated with the composite adverse outcome (p = 0.03). The rise in angiopoietin-2 level from preoperatively to 24 hours postoperatively directly correlated with cardiopulmonary bypass duration (r = 0.47; p = 0.01). CONCLUSIONS In neonatal cardiac surgery, longer duration of cardiopulmonary bypass is directly associated with greater endothelial injury as measured by increased serum levels of angiopoietin-2. Angiopoietin-2 levels 24 hours postoperatively were significantly associated with a composite adverse outcome. Postoperative angiopoietin-2 level may serve as an early indicator of patients in need of closer monitoring and protective intervention. Further research into endothelial protective strategies is warranted.

Neonates undergoing cardiac surgery are at risk for oral aversion (OA). OA is not well described outside of the index hospitalization and impacts patients and families. We evaluated the prevalence of OA at 1 year old after neonatal cardiopulmonary bypass (CPB) surgery.Retrospective cohort study.Single quaternary care hospital.Our cohort included 157 neonates who underwent CPB surgery from 2014 to 2017 and had follow-up data available at 1 year old.None.Three feeding experts reviewed the medical record to define children with OA; 30% of charts were evaluated in triplicate for validation. Neonates with and without OA were compared in univariate analysis, and risk factors for OA were explored in a limited multivariable analysis. OA was present at 1 year in 37 patients (23.6%) and other feeding difficulties were present in an additional 29 patients (18.5%). Thirty-eight patients (24.2%) had a feeding tube, including 12 (7.6%) with a gastrostomy tube. Factors associated with OA at 1 year included total ICU days, duration of mechanical ventilation, total number of nil per os days, and number of postoperative days (PODs) until oral feeding initiation (all p0.0001). Number of POD until oral feeding initiation remained independently associated with OA at 1 year in multivariable analysis (adjusted odds ratio, 1.08; 95% CI, 1.04-1.12; p0.0001). Infants with any oral intake at discharge had lower odds of OA at 1 year (0.21; 95% CI, 0.08-0.5; p = 0.0003). At hospital discharge, 132 patients (84.1%) were taking some oral feeds, and 128 patients (81.5%) received tube feeding.OA and other feeding difficulties are common at 1 year old in neonates undergoing CPB surgery. Delayed exposure to oral intake may be a modifiable risk factor for OA and efforts to improve early oral feeding could lead to better functional outcomes.

In single-ventricle patients with native-to-neoaortic anastomoses, the native aortic root serves as a conduit to the coronary arteries. Thrombosis of the native aortic root has been described only in small, limited reports. We described our center's experience with this rare adverse event. All single-ventricle patients who underwent native-to-neoaortic anastomosis from 2002 to 2017 were compiled from institutional databases. Chart review identified cases of native aortic root thrombosis. Of 467 patients, there were 9 (2%) cases of native aortic root thrombosis; all had hypoplastic left heart syndrome. Timing of thrombosis varied and occurred following each stage of single-ventricle palliation. For treatment, 8 patients received systemic anticoagulation and one patient did not receive any treatment. One patient also underwent percutaneous thrombectomy. Three patients (33.3%) died during the follow-up period. Among survivors, 4/6 had right ventricular systolic function that was mildly depressed or worse at a median follow-up of 8.3 years (interquartile range 1.3-10.8). Native aortic root thrombosis is a rare complication occurring at various stages of single-ventricle palliation. In this series, 6 of 9 patients (66.6%) survived; however, ventricular function was often compromised. No risk factors were identified to focus preventative therapies.

Introduction: After anatomic repair (AR) of congenitally corrected transposition of the great arteries (ccTGA), the left ventricle (LV) is re-established as the systemic ventricle. As such, LV dysfunction is an important factor in long-term patient morbidity. While predictors exist for successful AR, little is known regarding measures of LV function before and after AR, and no known predictors exist of worsening global longitudinal strain (GLS). Methods: All patients with ccTGA who survived to discharge after AR at a single institution from 2000-2018 were included. Pre-operative and discharge echocardiograms were reviewed and GLS analysis was performed. LV sphericity index was measured to determine the degree of ventricular septal shift; a lower index denotes more septal bowing into the LV. Logistic regression was used to identify factors associated with worsened GLS after AR. Results: Of 91 survivors (65% male), 51 underwent Senning/arterial switch (ASO) and 40 had Senning/Rastelli. Mean LV ejection fraction (EF) decreased from 66% to 61% after AR (p = 0.0003) with 26% having LVEF less than or equal to 55% (increased from 4% pre-op, p = 0.0003). Mean LV GLS was normal pre-op (-20.0% ± SD 4.1%) but worsened following AR (-16.5% ± 4.6%; p -18%) increased to 62% (vs 32% pre-op, p Conclusions: Following AR of ccTGA, LVEF decreases but remains in the normal range, while LV GLS detects abnormalities. Risk factors for worsening GLS include abnormal pre-op LV GLS, lower LV sphericity index, and a Senning/ASO repair. Lower LV sphericity index, or more septal shift into the LV, may represent a “less trained” LV, highlighting the need for careful patient selection for AR.

Introduction: Children with complex congenital heart disease (CHD) are at increased risk for neurodevelopmental impairments and reduced quality of life (QOL). Patients (pts) with congenitally corrected transposition of the great arteries (ccTGA) often undergo anatomic repair to make the left ventricle the systemic ventricle, with the goal of long term cardiac benefits. Understanding long-term outcomes and QOL may inform preoperative counseling and treatment strategies for ccTGA. Methods: We measured QOL using PedsQL generic and cardiac modules for pts following anatomic repair of ccTGA with Senning/Arterial Switch or Senning/Rastelli operations. Impaired QOL by the PedsQL is defined by total scores Results: Of 88 eligible pts, 56 (64%) returned surveys (29% female, median age 11.8 years). Respondents were more likely to have pre-operative arrhythmias or pacemaker implantation compared to non-respondents. Mean total PedsQL score was 72 + 19 for pt report and 75 + 18 for parent report; 22 of 56 (39%) pts had impaired QOL scores, compared to 20% impaired scores in pts with any type of CHD, previously described. Mean physical functioning score was 72 + 22 for pt report and 76 + 21 for parent report. Mean psychosocial summary score was 73 + 18 and 75 + 18 for pt and parent report, respectively. Communication was the lowest scored subscale of the Cardiac module by pt (63 + 26) and parent (66 + 31) report. The treatment II subscale addressing medication compliance was the highest scored subscale with scores of 90 + 8 for pts and 91 + 15 for parent, respectively. Longer cardiopulmonary bypass time (p = 0.01), surgical or catheter-based reintervention following hospital discharge (p = 0.02) and heart failure symptoms (p = 0.04) were associated with impaired QOL. Conclusions: After anatomic repair of ccTGA, pts are at risk of impaired QOL compared to healthy controls and to pts with a broad range of CHD. Indicators of more complex surgical and post-operative courses were associated with impaired QOL.

Introduction: Anatomic repair of congenitally corrected transposition of the great arteries (ccTGA) generally poses short-term risk for presumed long-term functional benefit by restoring the left ventricle to the systemic circulation. Understanding early and late surgical outcomes is crucial for counseling patients and families. Methods: All patients with anatomic repair at a single institution from 1993-2018 were included. Follow-up data was obtained by contact with patients’ primary cardiologists. The primary outcome was long-term survival and the secondary outcome was freedom from reintervention. Univariate Cox proportional hazard model was used to identify risk factors for mortality. In-hospital mortality and morbidity were compared between the most recent (2010-2018) and past surgical eras. Results: Anatomic repair was performed in 118 patients: 64 Senning/arterial switch operations (ASO) and 54 Senning/Rastelli operations. Survival was 85%, 82%, and 70% at 1, 10, and 20 years after repair, respectively, and was significantly lower in the Senning/Rastelli group compared to Senning/ASO (Hazard ratio 3.0, p=0.01; Figure 1). Preoperative factors associated with mortality were heterotaxy syndrome (p=0.03) and right ventricular systolic dysfunction (p=0.01). During a median follow up of 9.4 years, 47 of 98 patients (48%) required surgical or catheter-based reintervention; median freedom from reintervention was 11.7 years. For the most recent era, 30 of 32 patients (94%) with Senning/ASO and 19 of 22 patients (86%) with Senning/Rastelli operations survived to hospital discharge. In-hospital morbidity was similar between eras (30% in 1993-2009 and 28% in 2010-2018, p=0.82). Conclusions: Short- and long-term outcomes are favorable following Senning/ASO. Mortality following Senning/Rastelli operations remains high, highlighting the need for careful patient selection and family counseling.

Introduction: Limited data exist on the electrophysiologic outcomes of patients undergoing anatomic repair (AR) for congenitally corrected transposition of the great arteries (ccTGA). AR was defined as an atrial switch procedure plus either arterial switch (ASO) or Rastelli operation. Aims: To report mid and late electrophysiologic outcomes after AR and identify risk factors for those outcomes. Methods: Single center retrospective cohort study of patients undergoing AR between 1993-2017. Data were collected from available records. Transplant-free survival to 1 year post repair was required for inclusion. Standard descriptive statistical analysis and Cox proportional hazards were used. Results: Of 85 patients included, 95% had lesions in addition to ccTGA: most commonly VSD (84%) and pulmonary stenosis or atresia (58%). Median age at AR was 1.5y (IQR 0.9-2.8) with Senning/ASO in 56%, Senning/Rastelli in 38%, and hemi-Senning/Glenn/Rastelli in 6%. During a median follow-up of 10.6y, 45 (53%) patients developed an arrhythmia requiring intervention. Atrial tachycardia (AT) in 27 (32%) or ventricular tachycardia (VT) in 11 (13%) patients required intervention at a median of 7.4y (IQR 1.6-15.3y) and 15.9y (IQR 4.5-17.9) post-AR, respectively. Treatments included chronic medications in 29 (64%), cardioversion in 15 (33%) and catheter ablation in 10 (22%). Median freedom from AT and VT was 17.3y and 25y post-AR, respectively. D-looped ventricles (p=0.03) and multiple operations prior to AR (p=0.02) were associated with increased AT risk; and native pulmonary stenosis with increased VT risk (p=0.01). Those needing heart failure/transplant referral had increased risk of both AT and VT (both p=0.04). Pacemaker was implanted for heart block and/or SND prior to or during AR in 14 (16%), immediately post-op in 9 (11%), and late (median 6y post-AR) in 24 (28%). ICDs were implanted in 5 (6% of cohort), 4 for primary prevention. No patient had an appropriate shock. Conclusions: Anatomic ccTGA repair is associated with significant electrophysiologic morbidity. AT, VT, and SND develop at a similar incidence to that reported for d-TGA patients after atrial switch. The incidence of AV block follows a similar trajectory to that of physiologically palliated ccTGA.