Intra-Abdominal Adipose Tissue Quantification by Alternative Versus Reference Methods: A Systematic Review and Meta-Analysis.

Abstract

This meta-analysis aimed to assess the agreement between intra-abdominal adipose tissue (IAAT) quantified by alternative methods and the reference standards, computed tomography (CT) and magnetic resonance imaging (MRI). MEDLINE and EMBASE electronic databases were systematically searched to identify studies that quantified IAAT thickness, area, or volume by a comparator method and CT or MRI. Using an inverse variance weighted approach (random-effects model), the mean differences and 95% limits of agreement (LoA) were pooled between methods. The meta-analysis included 24 studies using four comparator methods. The pooled mean differences were -0.3 cm (95% LoA: -3.4 to 3.2 cm; P = 0.400) for ultrasound and -11.6 cm2 (95% LoA: -43.1 to 19.9 cm2 ; P = 0.004) for bioelectrical impedance analysis. Dual-energy x-ray absorptiometry (DXA) quantified both IAAT area and volume with mean differences of 8.1 cm2 (95% LoA: -98.9 to 115.1 cm2 ; P = 0.061) and 10 cm3 (95% LoA: -280 to 300 cm3 ; P = 0.808), respectively. Ultrasound and DXA measure IAAT with minimal bias from CT or MRI, while bioelectrical impedance analysis systematically underestimates IAAT. However, with the exception of DXA for IAAT volume, the wide LoA caution against clinical or research use of the comparator methods and emphasize the need to optimize alternatives to the reference standards.