ABSTRACT
Catheter ablation in children requires placement of multiple large femoral venous sheaths and catheters. Magnetic resonance angiography (MRA) was used to evaluate the effect of indwelling lines on femoral venous blood flow. Between October 1993 and February 1994 a total of 17 patients scheduled for catheter ablation underwent venous MRA. Two-dimensional time-of-flight MRA was performed 12-70 hours after catheterization on all patients. All patients received intravenous heparin during the procedure and had aspirin therapy instituted after ablation. Eighteen catheter ablations and MRA studies were performed on the 17 patients (one patient underwent repeat ablation). There were 7 females and 10 males, with a mean age of 14.8 +/- 4.2 years (range 8-21 years). Patients had three venous sheaths inserted in the left femoral vein (5F, 6F, and 7F with external diameters measuring 1.7, 2.0, and 2.3 mm, respectively) and one sheath in the right femoral vein (7F). Four patients (22%) had altered venous flow (two complete obstructions and two partial obstructions) following catheterization. None of these patients experienced symptoms or complications. It was concluded that there is an increased incidence (22%) of venous obstruction following catheter ablation, but there are no related complications. Venous MRA provides a rapid, noninvasive method for evaluating venous flow abnormalities and possibly detects patients at risk for complications.
Subject(s)
Arrhythmias, Cardiac/surgery , Catheter Ablation/instrumentation , Femoral Vein/pathology , Magnetic Resonance Angiography , Postoperative Complications/diagnosis , Thrombosis/diagnosis , Adolescent , Adult , Child , Constriction, Pathologic/diagnosis , Female , Humans , Image Processing, Computer-Assisted , MaleABSTRACT
RATIONALE AND OBJECTIVES: To investigate whether accurate right ventricular volumes could be obtained using an ellipsoidal shell model with magnetic resonance (MR) image measurements from a single imaging plane. METHODS: An initial retrospective patient study included 10 patients with congenital cardiac defects who had undergone same-day or next-day radiographic contrast ventriculography. An expanded study included MR scans of a total of 29 patients with congenital cardiac defects. Magnetic resonance scans of 10 healthy volunteers were also included in part of the study. Right ventricular volumes were calculated from (1) model-based calculations using single-plane cine MR imaging, and (2) multislice calculations based on contrast angiography. RESULTS: Regression of angiography-based volumes against MR-based volumes showed high correlation (r = 0.97, see = 12.5 mL) and slope near unity. Regression of right against left stroke volumes, both calculated from MR data, showed excellent correlation (r = 0.90, see = 11.6 ml) and slope near unity. CONCLUSIONS: The ellipsoidal shell model can be used to reliably estimate right ventricular volume using single-plane MR images.
Subject(s)
Cardiac Volume , Image Processing, Computer-Assisted , Magnetic Resonance Imaging , Ventricular Function, Right , Adolescent , Adult , Angiography , Animals , Cardiac Catheterization , Child , Child, Preschool , Cineradiography , Dogs , Heart Defects, Congenital/diagnosis , Heart Defects, Congenital/diagnostic imaging , Humans , Infant , Magnetic Resonance Imaging, Cine , Middle Aged , Regression Analysis , Reproducibility of Results , Retrospective Studies , Stroke Volume , Swine , Ventricular Function, LeftABSTRACT
In patients with congenital and other heart disease, measurement of right ventricular (RV) volumes would be as useful as left ventricular (LV) volume measurement has been for diseases of the LV. Model-based techniques have had limited success. Simpson's rule (multislice) techniques require lengthy data collection and reduction. We investigated a technique for volume estimation with a new but simple geometric model. A retrospective patient study compared RV volumes from model-based calculations with dual-plane cine magnetic resonance (MR) imaging and multislice calculations with biplane cineangiography. Linear regression showed high correlation (r = 0.98, standard error of the estimate = 11.8 ml) between the two techniques, with a slope near unity. Comparison of calculated right and left stroke volumes also showed an excellent correlation (r = 0.93, standard error of the estimate = 10.4 ml) and a slope near unity. It is concluded that the ellipsoidal shell model can be used to estimate RV volume reliably and practically with dual-plane MR imaging.
