Impact Of Ventricular Assist Device Use On Pediatric Heart Transplant Outcomes In A Linked Cohort Of The Advanced Cardiac Therapies Improving Outcomes Network (ACTION) And Pediatric Heart Transplant Society (PHTS) Databases.

PURPOSE: Pediatric ventricular assist devices (VAD) improve waitlist mortality. We aimed to compare outcomes after heart transplant (HT) based on pre-HT support in patients <18 years at listing.

METHODS

VAD and post-HT data were linked between ACTION and PHTS databases, respectively (4/1/2018-6/30/2022). Support groups were defined as medical (MG) and VAD (VG). Within VG, device groups were defined as paracorporeal pulsatile (PP), paracorporeal continuous (PC), and implantable continuous (IC). Kaplan-Meier analysis, the log rank test, and Cox proportional hazards modeling analyzed graft loss post-HT, with and without propensity matching.

RESULTS

The 1360 patients included VG=405 and MG=955. At HT, VG was younger (4.2 [0.8-13.3 vs 6.1 [0.9-16.7] years; p=0.03) and more likely to receive mechanical ventilation (15.3% vs 11.0%; p=0.03). The proportion of congenital heart disease (CHD) was higher in MG (64.8%) vs VG (35.1%; p<0.0001). Overall graft survival did not differ between VG and MG (p=0.17). In unmatched multivariable analysis, PC and PP devices, CHD, Black or Other race, mechanical ventilation at transplant, and absence of induction therapy were independently associated with graft loss (all p<0.05), while VG was associated with lower risk of post-HT infection (HR 0.75 [95% CI 0.56-0.99]). In propensity-matched cohorts, VAD was not associated with graft loss. Within VG, VAD major adverse events of infection, bleeding, and respiratory failure were associated with inferior post-HT outcomes.

CONCLUSION

VAD support was not associated with graft loss after propensity matching. Differences by device type in unmatched analyses likely reflect differences in patient complexity rather than device-specific effects.