One size does not fit all: the influence of age at surgery on outcomes following Norwood operation

Background: Given our large catchment area that often results in later presentation age, we sought to understand our institutional outcomes for the Norwood operation in the context of published data. Specifically, we studied whether operative and late death post-Norwood are dependent on age at operation. Methods: Retrospective review of 105 consecutive infants undergoing Norwood (2004-2011) at our institution. Patients were divided into those undergoing Norwood <= 7 days of age (N = 43; 41%) and those undergoing Norwood > 7 days of age (N = 63; 59%). Operative mortality (= 30 days), interstage mortality (between Norwood and superior bidirectional Glenn), STS-mortality (operative death + in-hospital death), and late mortality, occurring any time following hospital discharge were compared among groups. Multivariable factors for mortality at each time-point were compared using logistic regression models. Results: Underlying diagnosis was HLHS in 67 (64%) with the remainder (N = 38; 36%) being other single ventricle variants. Median age at surgery was 8 days (range 1-63 days) and mean weight at surgery was 3.2 +/- 0.6 kg. Pulmonary blood flow was provided by a right ventricle-pulmonary artery conduit in 94% (N = 99). Overall operative survival was 92%, with 73% (N = 66) undergoing bidirectional Glenn. Median age was higher for operative survivors compared to non-survivors (12 days vs. 5 days; P = 0.036), with operative mortality higher for infants = 7 days at Norwood compared to infants > 7 days at Norwood (14% vs. 3%; P = 0.04). After censoring for in-hospital death, age = 7 days was also associated with increased risk for late death (26% vs. 5%; P = 0.005). Conclusions: In contrast to other institutional series, infants at our center undergoing Norwood operation at an earlier age have worse outcomes. Adoption of published practice patterns could lead to different local outcomes because of intangible center-specific effects, underscoring the principle that results from one institution may not be generalizable to others. Targeted center-specific internal review, if possible, should precede externally recommended changes in practice.