Contemporary outcomes of childhood aortic coarctation interventions: a national registry analysis of mortality, reinterventions and hospital resource use.

Abstract

Background: Coarctation of the aorta (CoA) has good modern results, but large multicentre longitudinal data on outcomes, especially hospital resource utilisation through childhood and adolescence, are not available.

Methods: All patients with CoA treated between 2000 and 2017 in England and Wales were linked to hospital and outpatient records through the Linking AUdit and National datasets in Congenital HEart Services (LAUNCHES) project. Mortality, reintervention and hospital stay were described, and associated risk factors were explored using multivariable regression models for each of these three outcomes (Cox regression, Fine-Gray subdistribution hazard model and quantile regression at median, respectively).

Results: A total of 3321 patients were included: n=669 (20.1%) had CoA with ventricular septal defect (VSD), n=331 (10.0%) had CoA with small VSD and n=2321 (69.9%) had isolated CoA. Mortality and cardiac reintervention at 10 years (from birth and CoA repair, respectively) were 3.7% (95% CI 3.0%; 4.4%) and 13.3% (12.1%; 14.5%), respectively. Compared with isolated surgical repair, isolated catheter repair (HR 3.7, (95% CI 2.2; 6)) and concomitant VSD closure (HR 1.34, (1; 1.9)) or pulmonary artery banding (HR 3.5, (2.4; 5.1)) had higher risk of reintervention. During the first year of life, the median time in hospital was 26 days (IQR 17; 44), decreasing to 1 (0; 2) day beyond 8 years. CoA with large VSD (-12, (-16; -8)), premature birth (-50, (-60; -40)), congenital comorbidity (-31, (-37; -25)), low weight (-23/kg, (-37; -11)) and younger age at first procedure (-6/year (-7; -5)) were associated with fewer days spent at home.

Conclusions: Subgroups of patients with CoA are still at risk of unfavourable outcomes during childhood and adolescence follow-up, especially cardiac reintervention at a distance from initial repair. Hospital resource utilisation remains low beyond the first year of life in the majority of patients. Identified factors, while non-modifiable, remain useful in risk stratification and counselling.