Categories
Nevin Manimala Statistics

Reference Values for Ventricular Volumes and Pulmonary Artery Dimensions in Pediatric Patients with Transposition of the Great Arteries After Arterial Switch Operation

J Magn Reson Imaging. 2021 Mar 21. doi: 10.1002/jmri.27602. Online ahead of print.

ABSTRACT

BACKGROUND: Pulmonary artery (PA) anatomy in patients with transposition of the great arteries (TGA) after arterial switch operation (ASO) with Lecompte manoeuvre is different compared to healthy subjects, and stenoses of the PA are common. Magnetic resonance imaging (MRI) is an excellent imaging modality to assess PA anatomy in TGA patients. However, disease-specific reference values for PA size are scarce.

PURPOSE: To establish disease-specific reference ranges for PA dimensions and for biventricular volumes and mass.

STUDY TYPE: Retrospective.

SUBJECTS: A total of 69 pediatric patients with TGA after ASO (median age 12.6 years; range 5-17.8 years; 13 females and 56 males).

FIELD STRENGTH/SEQUENCE: 3.0 T, steady-state free precession (SSFP) and gradient echo cine sequences and four-dimensional time-resolved magnetic resonance angiography with keyhole.

ASSESSMENT: Right and left PA (RPA, LPA) were each measured at three locations during its course around the aorta. Ventricular volumes, mass, and ejection fraction were measured from a stack of short axis cine images.

STATISTICAL TESTS: The lambda-mu-sigma (LMS) method of Cole and Green, univariate and multivariate linear models, and t-test.

RESULTS: Centile graphs and tables for PA dimensions, biventricular volumes, mass, and ejection fraction were created. Univariate linear analysis showed significant associations (P < 0.05) between body surface area (BSA), height, and weight with systolic MPA and RPA diameter. In multivariate linear analysis, only BSA remained a strong predictor for main PA and RPA diameters. For biventricular volumes, the univariate linear model revealed a strong influence of BSA, height, weight, and age (all P < 0.05). On multivariate linear analysis, only body height remained associated.

DATA CONCLUSION: Uni- and multivariate linear analyses showed a strong association between BSA and PA diameters, as well as between height and biventricular volumes, and therefore, centile tables and graphs are presented accordingly. Our data may improve MR image interpretation and may serve as a reference in future studies.

LEVEL OF EVIDENCE: 4 TECHNICAL EFFICACY STAGE: 2.

PMID:33749058 | DOI:10.1002/jmri.27602

By Nevin Manimala

Portfolio Website for Nevin Manimala