Decreased Pulmonary Perfusion in Pulmonary Vein Stenosis After Radiofrequency Ablation: Assessment With Dynamic Magnetic Resonance Perfusion Imaging

Abstract

The functional impact of pulmonary vein (PV) stenosis on pulmonary perfusion after radiofrequency ablation (RFA) for atrial fibrillation (AF) has not been systematically evaluated previously. Therefore, we correlated magnetic resonance (MR) pulmonary perfusion imaging with single-photon emission CT (SPECT) perfusion and with the degree of PV stenosis (PVS) apparent on MR angiography (MRA) after RF ablation. Joint radiology-cardiology collaborative magnetic resonance unit at the Kerckhoff Heart Center. This was a cohort study of 110 patients who were routinely examined by MRA after RFA for AF, whereby 51 patients with a PV diameter reduction of > 25% or with clinical symptoms (ie, dyspnea and cough) were enrolled into the study. Patients were examined at follow-up by MR perfusion imaging and MRA, and the results were compared to those from patients who underwent SPECT scanning and from a control group of 26 untreated patients. Twelve patients underwent PVS dilatation as well as 22 sequential follow-up examinations. Pulmonary perfusion was evaluated using a dynamic contrast-enhanced three-dimensional MR perfusion sequence (1.5 T, 2.5-s temporal resolution, and 0.05 cm spatial resolution), and high-resolution, contrast-enhanced MRA was performed to measure PV diameter. PV dilatation was performed using an angioplasty catheter that was 8 to 10 mm in diameter. The localization and extent of perfusion defects measured by MRI or SPECT scanning were precisely matched. MR perfusion imaging detected 20 of 21 perfusion defects (sensitivity, 95.2%; specificity, 100%). PVSs and perfusion deficits correlated closely and showed the following threshold: perfusion decreased substantially in PVs <or= 6 mm in diameter (21 of 25 areas; 84.0%) compared to 2 of 180 areas (1.1%) with PVs > 6 mm in diameter. After PVS dilatation, perfusion was restored partially after weeks, and complete normalization was seen in 4 of 12 patients (33%). PVSs caused severe perfusion deficits, which were reliably demonstrated by MR perfusion imaging. Clinical symptoms correlated better with MR perfusion than they did with MRA. The combination with MRA to assess underlying PVS allowed a "one-stop-shopping" MRI procedure to be carried out. The results led to alterations of RFA techniques, and therefore MRA and MR perfusion imaging may be beneficial in patient follow-up and in evaluating new ablation techniques.