Cost-Effective Design of Pulmonary Arterial Hypertension Trials To Detect Changes in Right Ventricular Size and Function: 3D Echocardiography Versus Cardiovascular Magnetic Resonance

Abstract

Purpose Both cardiovascular magnetic resonance (CMR) and 3D transthoracic echocardiography (3DE) can quantify right ventricular (RV) end diastolic volume (EDV) and ejection fraction (EF) accurately. This study was designed to estimate sample sizes and imaging costs to detect changes in RV EDV and EF using 3DE and CMR for future pulmonary arterial hypertension (PAH) drug trials. Methods and Materials Same day CMR and 3DE were performed on 20 PAH patients (age: 53±13 yrs, 19 females) at 0 and 6 months. Consecutive short axis cines spanning the entire RV (Philips 1.5-T scanner) and full-volume acquisitions of the RV (Philips iE33+TomTec 4D RV-Function software) were used to measure EDV and EF. Sample size calculations using two-sample repeated measures to account for changes over time were performed modeling a placebo-controlled trial (Stata v12.1). Cost estimates were made using the 2012 Medicare physician fee schedule. Results No significant differences were noted between 0- and 6-month RV EF and EDV measurements with either CMR or 3DE, but the relatively large SDs reflected variable progression of disease in individual patients on standard therapy. Sample sizes required for 3DE were roughly 3 times greater than CMR. Although the per-study cost of CMR (~$400) is approximately double that of 3DE (~$200), the total cost of a trial using CMR is considerably lower. Conclusions Both CMR and 3DE can be used to detect the effects of new PAH therapies on RV EDV and EF, but CMR is more cost-effective. Population characteristics (average ± SD) Baseline 6 months CMR 3DE CMR 3DE RV EF (%) 34 ± 15 32 ± 12 36 ± 11 33 ± 9 RV EDV (mL) 245 ± 58 195 ± 59 244 ± 71 199 ± 48 Estimated per-group sample sizes (placebo-controlled trial - 90% power) Estimated cost (US$) CMR 3DE CMR 3DE EF: 3% change 54 169 89,000 143,000 EF: 5% change 20 61 33,000 52,000 RVEDV: 20 mL change 44 138 73,000 117,000 RVEDV: 30 mL change 20 61 33,000 52,000