Threshold Analysis for the Effect of Elevated Pulmonary Vascular Resistance on One-Year Mortality after Heart Transplantation

J. Kwon¹, P. Broughton², L. Witer², N. Pope², B. Houston², R. Tedford², A. Kilic^1
1MUSC, Charleston, South Carolina ²Medical University of South Carolina, Charleston, South Carolina

Purpose: Elevated pulmonary vascular resistance (PVR) is associated with increased risk for early mortality after heart transplantation (HT).[1,2] However, international guidelines regarding PVR cutoffs have been liberalized as medical and mechanical support for elevated PVR have become more common.[3] This study identified a threshold PVR associated with one-year mortality after HT.

Methods: Adult HT recipients in the current allocation era (after October 2018, 2018) were identified in the United Network for Organ Sharing registry. Patients < 18 years of age, listed for multiorgan transplant, and with PVR outside the 5th and 95th percentile were excluded from analysis. Univariable Cox regression confirmed that increasing PVR at waitlist registration was associated with 1-year mortality after HT. Threshold regression analysis was performed to identify the PVR at which one-year mortality was significantly elevated. Patients were grouped according to their PVR at waitlist registration, either below or greater than or equal to threshold PVR. One-year survival for each group was estimated using Kaplan-Meier analysis and compared using log-rank tests. Multivariable Cox regression was used to calculate risk-adjusted hazards for 1-year mortality. Threshold regression and survival analysis were repeated for PVR at transplantation and PVR change while waitlisted.

Results: Among 16,447 HT recipients, median PVR at waitlist registration was 2.22 Woods Units (WU), IQR [1.43-3.33]. PVR was associated with 1-year mortality on univariable analysis (HR 1.06 per WU, 95% CI [1.02-1.10], p=0.005). Threshold regression identified that PVR≥2.75 WU predicted 1-year mortality, and patients were grouped according to this threshold. Patients with PVR≥2.75 were older and more likely to be female, on inotropes, and less likely to have durable LVAD (Table). Unadjusted 1-year survival was 89.0% vs 87.5% among patients below and above the PVR threshold (p=0.009) (Figure). After risk adjustment, PVR≥2.75 was associated with an increased risk for 1-year mortality (HR 1.16 (95% CI [1.03-1.30], p=0.016).

Among 13,367 patients with PVR reported at time of transplant, a threshold PVR of 1.69 predicted 1-year mortality. After risk adjustment, PVR≥1.69 was not associated with 1-year mortality (HR 1.12, 95% CI [0.98-1.28], p=0.084). While waitlisted, 13.4% of patients had an increase of ≥0.5, 67.6% had minimal change within 0.5, and 13.4% had a decrease of ≥0.5 WU of PVR. On linear regression, higher PVR at waitlist registration was associated with greater decreases in PVR while waitlisted (Figure). PVR change while waitlisted did not have a threshold which predicted 1-year mortality.

Conclusion: The majority of transplant recipients did not have a significant change in PVR while waitlisted. PVR assessment at the time of waitlist registration has a higher predictive value of post-HT outcomes compared to changes on the waitlist or assessment at the time of HT. The threshold PVR of 2.75 WU at waitlist registration may be useful for patient risk stratification and selection.

Identify the source of the funding for this research project: None