Outcomes Following Cardiac Surgery in Patients with Trisomy 21. How Do They Compare with a Matched Non-Syndromic Cohort?

A. Dilawar, J. Lee, K. Gauvreau, H. Liu, D. Castellanos, J. Sasaki, N. Roy, D. Hoganson, M. Nathan
Boston Children's Hospital, Boston, Massachusetts

Purpose: While children with trisomy 21 (Tri 21) are at risk for congenital heart disease (CHD), there is limited data on non-cardiac factors that influence long-term outcome after cardiac surgery. The purpose of this study is to understand long-term outcomes and identify modifiable risk factors to improve outcomes in these patients

Methods: This is a single institutional, retrospective cohort study of Tri 21 patients under 18 years old at index surgery who underwent cardiac surgery between September 2009 and December 2021. A matched cohort based on age, Society of Thoracic Surgeons (STS) congenital heart surgery database primary diagnosis/primary procedure but without non-cardiac anomalies/syndromes/chromosomal abnormalities during the same period was identified. Outcomes including postoperative hospital length of stay (POHLOS) (primary), postoperative and post-discharge cardiac surgical/catheter intervention, non-cardiac surgery, hospital readmission, and mortality were compared between the matched groups using Fisher’s exact test, Wilcoxon rank sum test, or log-rank test as appropriate. Survival curves for post-discharge outcomes were estimated using the Kaplan-Meier method. A generalized linear model (GLM) with log link was used to further compare POHLOS between groups, adjusting for potential confounders which were not matched. Regression coefficients (b) and 95% confidence intervals (CI) estimate the magnitude of association.

Results: A total of 710 Tri 21 patients underwent cardiac surgery. Univariate analysis showed significantly longer median POHLOS for Tri 21 versus matched cohort for ages ≤ 30 days and 31 days to ≤ 1 year (26 vs. 13 days; 7 vs. 6 days), weights ≤ 2.5kg, 2.5 to < 5.0kg, and 5.0 to 9.9kg (37 vs 15 days; 8 vs. 7days; 7 vs. 5 days; respectively), delayed sternal closure (32 vs. 30 days), urgent/ emergency status (21 vs. 10 days). The unadjusted GLM showed Tri 21 patients had longer POHLOS (b=0.29, 95% CI: 0.12, 0.47, p=0.001). After adjusting for potential confounders (Table 1), Tri 21 patients continued to have longer POHLOS, however the association’s magnitude was lower (b=0.17, p=0.001, 95% CI: 0.06, 0.27). When Tri 21 patients were compared to their matched cohort, neither post-discharge surgical reintervention (10% vs. 10%, p=0.46), nor mortality (2.7% vs. 2.7%, p=0.81), nor pre-discharge catheter reintervention (1.4% vs.1.7, p=0.83) were significantly different. However, pre-discharge surgical reintervention (6.5% vs. 3.8%, p=0.030), post-discharge catheter intervention (5.6% vs. 7.8%, p=0.024), readmissions (42% vs. 25%, p< 0.001, Figure 1) were significantly different.

Conclusion: These findings highlight the challenges of longer POHLOS, increased readmission rates and earlier time to readmission that patients with Tri 21 face. Further research is necessary to identify factors that could reduce postoperative complications and improve in-hospital and long-term outcomes for Tri 21 patients undergoing cardiac surgery.

Identify the source of the funding for this research project: Boston Children's Hospital