O1001: Mortality Among Young Adult Congenital Heart Disease Patients Hospitalized at Pediatric versus Adult Institutions in the United States, 2001-2019

Background: Data on current era trends in hospitalization and inpatient surgical mortality among young adults with congenital heart disease (ACHD) in both pediatric and adult institutions is limited. The objective of this study was to determine sociodemographic associations with surgical mortality among young ACHD patients admitted to a children’s (CH) and non-children’s hospital (NCH) in the United States.

Methods: We conducted a retrospective cohort analysis using the National Inpatient Sample (NIS) to capture a weighted estimate of total and cardiac-specific hospitalizations. Hospitals were categorized as CH if ≦25% of discharged patients were ≧18 years, and otherwise as NCH. Inclusion criteria were non-pregnant young ACHD patients aged 18-40 years hospitalized with both non-severe and severe CHD (excluding ASD and/or PFO). Cardiovascular-related (CV) admission diagnoses were based on ICD9/10 diagnostic codes, and included heart failure, endocarditis, arrhythmias, or valve disorders. Patients with procedure codes associated with open cardiac surgery were considered a surgical admission. Chi-square analyses compared cardiac and sociodemographic characteristics between NCH and CH. Multivariable logistic regression was performed using survey techniques to evaluate predictors of mortality (Stata v18; College Station, TX). The primary predictor variable was CH versus NCH hospital, and the primary outcome variable was hospital mortality. Secondary predictor variables included cardiac and sociodemographic characteristics.

Results: There were 291,182 young ACHD hospitalizations in the United States between 2001-2019, of which 51,109 (17%) were due to pregnancy. Among non-pregnant hospitalizations, 18,753 (7.8%) patients were admitted to CH and 221,320 (92%) to NCH. The majority of hospitalizations were among patients with non-severe ACHD (88%), and for non-CV related diagnoses (76%). A greater proportion of the CH ACHD hospitalizations had severe CHD (34%) than did the NCH hospitalizations (10%). ACHD hospital mortality rate was 1.9% overall and decreased significantly over time (OR per year 0.94, CI 0.93-0.96). Overall mortality rate was lower for CH than NCH (1.5% vs. 2%, p=0.0362) (Figure 1). Multivariable regression for all ACHD as well as severe CHD hospitalization revealed no significant differences between CH and NCH mortality after controlling for CV diagnoses, sociodemographic variables and CV surgery. Among non-severe CHD patients, hospital mortality approached significance (p=0.053) when accounting for cardiovascular surgery (OR 1.53, CI 0.99-2.32). All ACHD regression models demonstrated a higher odds of mortality among patients who were older, had the lowest income quartiles, had public insurance and lived in the southern and western geographic regions.

Conclusion: Hospital mortality has significantly decreased over time among young ACHD patients admitted to both CH and NCH, with a greater numeric decrease in inpatient mortality among CH than NCH. Among non-severe ACHD patients, multivariate regression approached significance for increased mortality in NCH compared to CH, and thus ACHD surgery at CH should be considered all CHD, regardless of severity.