The Importance of Postprocessing Methods for Assessing Right Ventricular Volumes and Function in Patients With Pulmonary Hypertension: Results From the Postoperative Right Heart Remodeling in Patients With Chronic Thromboembolic Pulmonary Hypertension After Endarterectomy, or Pulmonary Arterial Hypertension After Lung Transplantation Study

Abstract

Background: Right ventricular (RV) function and volumes assessment by cardiac MRI has emerged as a strong prognostic marker in patients with pulmonary hypertension (PH), but expert centers propose different thresholds, possibly explainable by methodologic differences.

Research question: What is the impact of RV trabeculation exclusion on RV metrics in patients with PH? Which level of pulmonary artery (PA) flow assessment best estimates RV stroke volume (RVSV) using 4-dimensional (4D) flow MRI.

Study design and methods: Forty-two patients with PH were recruited prospectively in the Postoperative Right Heart Remodeling in Patients With Chronic Thromboembolic Pulmonary Hypertension After Endarterectomy, or Pulmonary Arterial Hypertension After Lung Transplantation, study and underwent right heart catheterization and cardiac MRI including 4D flow MRI. Right ventricular stroke volume (RVSV), RV ejection fraction (RVEF), RV end systolic volume index (RVESVi), and RV mass were assessed either including or excluding trabeculations from the RV volume. RVSV including trabeculations from the right ventricular volume (RVSV_Tin(v)) and excluding trabeculations from the right ventricular volume (RVSV_Tex(v)) then were compared with the PA forward flow (PAFF) measured at the mid-PA trunk (Mid_PAFF) and at the PA annular level (Ann_PAFF), using 4D flow MRI. Twenty-three patients were re-evaluated after surgery: patients with chronic thromboembolic PH underwent pulmonary endarterectomy, whereas patients with refractory pulmonary arterial hypertension underwent lung transplantation.

Results: Excluding trabeculations from RV volumes led to higher RVEF, RVSV, and RV mass median values (47.3% [interquartile range (IQR) 34.6%-54.6%] vs 40.9% [IQR, 29.6%-50.6%]; 69.5 mL [IQR, 62.5-80.6 mL] vs 59.3 mL [IQR, 50.9-74.8 mL]; and 54.5 g [IQR, 41-69 g] vs 37 g [IQR, 29.2-50 g], respectively; P < .001 for all), and lower RVESVi median values (44.5 mL/m² [IQR, 29.5-63.7 mL/m²] vs 54.4 mL/m² [IQR, 33.5-77.2 mL/m²]; P < .001), reclassifying 15 patients as low risk. Compared with the standard approach of associating Mid_PAFF flow measurement with RVSV_Tin(v), the association between Ann_PAFF and RVSV_Tex(v) was stronger (r = 0.66 and r = 0.85 respectively; P = .04). Longitudinal analysis demonstrated strong changes for RV mass, RVESVi, and RVEF. After surgery, excluding trabeculations showed much less impact on RV metrics, reflecting reverse remodelling.

Interpretation: Our results show that postprocessing methodologic differences using cardiac and 4D flow MRI significantly impacted RV measurements when applying guideline-recommended thresholds.

Clinical trial registration: ClinicalTrials.gov; No.: NCT03205085; URL: www.

Clinicaltrials: gov.

Keywords: 4-dimensional flow MRI; cardiac MRI; chronic thromboembolic pulmonary hypertension; methodology; postprocessing; pulmonary arterial hypertension; pulmonary artery flow; pulmonary hypertension; reverse remodeling; right ventricle; right ventricle remodeling; right ventricular function; right ventricular volumes.