Factors associated with systemic to pulmonary arterial collateral flow in single ventricle patients with superior cavopulmonary connections

• Published in: Heart (British Cardiac Society) Vol. 101; no. 22; pp. 1813 - 1818
• Main Authors: Glatz, Andrew C, Harrison, Neil, Small, Adam J, Dori, Yoav, Gillespie, Matthew J, Harris, Matthew A, Fogel, Mark A, Rome, Jonathan J, Whitehead, Kevin K
• Format: Journal Article
• Language: English
• Published: England BMJ Publishing Group LTD 01.11.2015
• Subjects: Blood Flow Velocity - physiology; Chest tubes; Child, Preschool; Collateral Circulation - physiology; Cross-Sectional Studies; Female; Heart Defects, Congenital - physiopathology; Heart Ventricles - abnormalities; Hemoglobin; Hospitalization; Humans; Infant; Intensive care; Magnetic Resonance Angiography; Male; Pulmonary arteries; Pulmonary Artery - pathology; Pulmonary Veins - abnormalities; Retrospective Studies; Risk factors; Veins & arteries; Vena Cava, Superior - abnormalities; Ventilators
• ISSN: 1355-6037; 1468-201X
• DOI: 10.1136/heartjnl-2015-307703

ObjectiveSystemic to pulmonary arterial collateral flow (CollF) is common in single ventricle patients with superior cavopulmonary connections (SCPC), although associations with CollF are not well understood. We previously described a method to quantify CollF by cardiac MRI (CMR). We sought to identify factors associated with CollF in a large cross section of patients with SCPC.MethodsA retrospective observational cohort study of events from birth to study CMR was performed for all patients with SCPC who had CollF quantified by CMR.ResultsCollF was quantified in 96 patients at a median age of 2.6 (IQR 1.9–3.1) years and 2.1 (1.4–2.7) years after SCPC and measured 1.6±0.7 L/min/m2 (33±11% of aortic flow and 48±16% of pulmonary venous flow). Significantly higher amounts of indices of CollF were associated with: duration of chest tubes (p≤0.05 for all), intensive care unit and hospital length of stay (p≤0.04 for all), higher O2 saturation at Stage 2 discharge (p=0.04 for CollF/aortic), female sex (p≤0.007 for CollF/aortic and CollF/pulmonary venous), and history of a Blalock-Taussig shunt (p<0.04 for CollF and CollF/aortic). Multivariable models were constructed to identify factors independently associated with CollF measures and included: female sex (p≤0.006 for all), O2 saturation at Stage 2 discharge (p=0.013 for CollF/aortic) and total chest tube days (p=0.001 for all). These models explained 20–22% of the variance in the outcomes.ConclusionsThese data support hypotheses that perioperative morbidity and pleural inflammation play a role in CollF development and that CollF affects pulmonary blood flow.