Trends in Ductus Arteriosus Stent Versus Blalock-Taussig-Thomas Shunt Use and Comparison of Cost, Length of Stay, and Short-Term Outcomes in Neonates With Ductal-Dependent Pulmonary Blood Flow: An Observational Study Using the Pediatric Health Information Systems Database

Abstract

Background: The modified Blalock-Taussig-Thomas shunt is the gold standard palliation for securing pulmonary blood flow in infants with ductal-dependent pulmonary blood flow. Recently, the ductus arteriosus stent (DAS) has become a viable alternative.

Methods and results: This was a retrospective multicenter study of neonates ≤30 days undergoing DAS or Blalock-Taussig-Thomas shunt placement between January 1, 2017 and December 31, 2020 at hospitals reporting to the Pediatric Health Information Systems database. We performed generalized linear mixed-effects modeling to evaluate trends in intervention and intercenter variation, propensity score adjustment and inverse probability weighting with linear mixed-effects modeling to analyze length of stay and cost of hospitalization, and generalized linear mixed modeling to analyze differences in 30-day outcomes. There were 1874 subjects (58% male, 61% White) from 45 centers (29% DAS). Odds of DAS increased with time (odds ratio [OR] 1.23, annually, P<0.01 [95% CI, 1.10-1.38]) with significant intercenter variation (median OR, 3.81 [95% CI, 2.74-5.91]). DAS was associated with shorter hospital length of stay (ratio of geometric means, 0.76 [95% CI, 0.63-0.91]), shorter intensive care unit length of stay (ratio of geometric means, 0.77 [95% CI, 0.61-0.97]), and less expensive hospitalization (ratio of geometric means, 0.70 [95% CI, 0.56-0.87]). Intervention was not significantly associated with odds of 30-day transplant-free survival (OR,1.18 [95% CI, 0.70-1.99]) or freedom from catheter reintervention (OR, 1.02 [95% CI, 0.65-1.58]), but DAS was associated with 30-day freedom from composite adverse outcome (OR, 1.51 [95% CI, 1.11-2.05]).

Conclusions: Use of DAS is increasing, but there is variability across centers. Though odds of transplant-free survival and reintervention were not significantly different after DAS, and DAS was associated with shorter length of stay and lower in-hospital costs.

Keywords: Blalock‐Taussig‐Thomas shunt; catheterization; cost of hospitalization; ductal dependent pulmonary blood flow; ductal stent; ductus arteriosus stent; length of stay.

Figure 1. Adjusted and unadjusted proportion/probability DAS versus BTTS over time.

Line Bars represent the unadjusted proportion of DAS vs BTTS by year with gray whiskers representing SEs. Line represents fitted probability of DAS vs BTTS (reference) by year based on the adjusted linear mixed model with black whiskers representing 95% CIs. Covariates including presence of genetic anomaly, concomitant condition, prematurity, low birth weight, ventilation prior to intervention, and prematurity, were all set to zero (absence of covariate). BTTS indicates Blalock‐Taussig‐Thomas shun; and DAS, ductus arteriosus stent.

Figure 2. Results of generalized linear mixed model quantifying the effect of calendar year on odds of DAS vs BTTS use adjusted for measurable confounders; BTTS is reference.

BTTS indicates Blalock‐Taussig‐Thomas shunt; DAS, ductus arteriosus stent; and OR, odds ratio.

Figure 3. Unadjusted percent DAS (versus BTTS) by center across all years.

BTTS indicates Blalock‐Taussig‐Thomas Shunt; and DAS, ductal arteriosus stent.

Figure 4. Results of linear mixed models quantifying the effect of DAS vs BTTS on adjusted odds of 30‐day transplant‐free survival, freedom from catheter reintervention, and freedom from composite adverse outcome based on 3 different models.

Covariates for all models included genetic anomaly, concomitant condition, prematurity, low birth weight, ventilation before intervention, pulmonary atresia, and year of procedure with center as random intercept. BTTS indicates Blalock‐Taussig‐Thomas shunt; and DAS, ductus arteriosus shunt.