D28-10 Potts Shunt in Pulmonary Arterial Hypertension Timing and Risk Assessment
Z Xu, X Yu, Y Chen, Y Sun, L Fu, H ZhangAbstract
Rationale
Potts shunt represents a strategy for children with refractory pulmonary arterial hypertension (PAH). Procedural timing remains undefined, and interventions may be associated with perioperative mortality and limited clinical benefit. Evidence integrating advanced imaging, multidisciplinary decision-making, and risk stratification is lacking.Objective: To evaluate the impact of a multidisciplinary approach and cardiac magnetic resonance imaging (CMR)-guided assessment on outcomes for Potts shunt in pediatric PAH, and investigate the association between timing of intervention, preoperative right ventricular (RV) function, and clinical improvement.
Methods
Prospective, single-center study of 36 consecutive Group 1 PAH children (median age 11.2 years) undergoing surgical (n = 20) or transcatheter (n = 16) Potts shunt. The primary endpoint was all-cause mortality or persistent high-risk status 6 months post-procedure. Logistic regression identified predictors of clinical improvement, and a pre-procedural risk score was developed to inform procedural timing.
Measurements and Main Results
Overall, 75% of patients improved in risk category (55.6% to low-risk, and 82% weaned off prostacyclin therapy). Preoperative RV dysfunction and increased RV volumes were associated with lower likelihood of clinical improvement. Delayed intervention correlated with diminished benefit and increased perioperative mortality (11.1% - limited to surgical patients early in the series). A six-parameter, CMR-based risk score demonstrated high discriminatory capacity (AUC 0.89) for predicting clinical improvement.
Conclusions
A multidisciplinary, CMR-guided approach enables optimized Potts timing, improved patient selection, and favorable clinical outcomes in children with PAH. Potts shunt should not be delayed until advanced right ventricular failure, and the proposed risk score may assist in guiding optimal intervention timing.
This abstract is funded by: None