Postoperative Renal Failure, Shunt Type, and Mortality After Norwood Palliation

Abstract

Background: The Single Ventricle Reconstruction Trial demonstrated increased risk of death or heart transplant 1 year after Norwood among subjects randomly assigned to modified Blalock-Taussig shunts compared with right ventricle to pulmonary artery (RV-PA) shunts. We used the Single Ventricle Reconstruction Trial public use database to evaluate incidence and risk factors for postoperative renal failure and relationships between renal failure, shunt type, and outcomes after Norwood.

Methods: Postoperative renal failure was defined a priori as a threefold rise in creatinine from baseline, or dialysis use, within 7 days of Norwood. We used multivariate logistic regression to evaluate risk factors for postoperative renal failure and Cox hazard regression to determine the association between postoperative renal failure and 1 year after Norwood mortality.

Results: Overall, postoperative renal failure occurred in 8.4% (46 of 544) patients with risk factors including receipt of a modified Blalock-Taussig shunt (adjusted odds ratio [aOR] 3.3, P = .02), low center volume (aOR 2.7, P = .005), two or more preoperative complications (aOR 4.0, P < .001), low birth weight (aOR 3.2, P = .002), postoperative heart block (aOR 8.5, P = .001), and delayed sternal closure (aOR 5.3, P = .026). Renal failure was an independent risk factor for 1-year mortality (aHR 1.9, P = .019). Assessing interaction by shunt type, mortality risk associated with renal failure was greatest in the RV-PA shunt group (aHR 3.3 vs RV-PA shunt without renal failure, P = .001), but was also increased in the modified Blalock-Taussig shunt group (aHR 1.9, P = .03).

Conclusions: Postoperative renal failure is common after Norwood and is independently associated with mortality. Although renal failure is more common after modified Blalock-Taussig shunt, the highest mortality risk with renal failure occurs after RV-PA shunt.

Figure 1:

Histogram demonstrating the distribution and timing of post-operative renal failure after Norwood operation during SVR trial, 2005-2009.

Figure 2:

Kaplan-Meier curve for 1-year survival following Norwood operation stratified by presence of post-operative renal failure during SVR trial, 2005-2009. Shaded regions represent 95% confidence interval.

Figure 3:

Adjusted Kaplan-Meier curve for 1-year survival following Norwood operation stratified by shunt type and presence of post-operative renal failure during SVR trial, 2005-2009.