Objective To evaluate outcomes of systemic to pulmonary artery shunts (SPS) in patients weighing less than 3 kg with regard to shunt type, shunt size, and surgical approach. Methods Patients weighing less than 3 kg who underwent modified Blalock-Taussig or central shunts with polytetrafluoroethylene grafts at our institution from January 1, 2000, to May 31, 2011, were reviewed. Patients who had undergone other major concomitant procedures were excluded from the analysis. Primary outcomes included mortality (discharge mortality and mortality before next planned palliative procedure or definitive repair), cardiac arrest and/or extracorporeal membrane oxygenation (ECMO), and shunt reintervention. Results In this cohort of 80 patients, discharge survival was 96% (77/80). Postoperative cardiac arrest or ECMO occurred in 6/80 (7.5%), and shunt reintervention was required in 14/80 (17%). On univariate analysis, shunt reintervention was more common in patients with 3-mm shunts (11/30, 37%) compared with 3.5-mm (2/36, 6%) or 4-mm shunts (1/14, 7%) (P < .003). There were no statistically significant associations between shunt type, shunt size, or surgical approach and cardiac arrest/ECMO or mortality. Multiple logistic regression demonstrated that a shunt size of 3 mm (P = .019) and extracardiac anomaly (P = .047) were associated with shunt reintervention, whereas no variable was associated with cardiac arrest/ECMO or mortality. Conclusions In this high-risk group of neonates weighing less than 3 kg at the time of SPS, survival to discharge and the next planned surgical procedure was high. Outcomes were good with the 3.5- and 4-mm shunts; however, shunt reintervention was common with 3-mm shunts.