MRI of the arthritic finger joints: synovial membrane volume determination, a manual vs a stereologic method

Abstract

The aim of this study was to compare the stereology method for estimation of synovial volume on MR images with manually outlining method in inflammatory joint diseases. As manual outlining of the synovial volume is too time-consuming a method for clinical use, a less time-consuming methods are needed. Post-contrast 3D T1-weighted turbo gradient MR images of the finger (metacarpophalangeal and interphalangeal) joints (14 joints) were acquired from 11 patients with rheumatoid arthritis (n=8) and reactive arthritis (n=3). Manually outlined volume was taken as a gold standard and compared with stereologic volume estimation on transverse 1-mm-thick slices. The mathematical basis of the stereologic volume estimator is based on a two-step procedure: a 2D nucleator is used for estimation of the area of the synovial membrane on an MR slice and unbiased estimates of volumes are obtained by Cavalieri's principle. The 2D nucleator estimates the area of any object irrespective of its size, shape, and orientation by measuring the distance between a "central point" in the object and the intersections between the object boundary and radiating test lines. The total volume of 154 finger joints was estimated. A significant correlation between manual and stereologic estimations of synovial volumes was found (Spearman rho=0.71, P=0.002). Ninety-five percent limits of agreement were ±5–6 ml (14 finger joints per patient). The time used for volume determination in a patient varied from 1 to 2.5 h by manual outlining and from 0.5 to 1 h by stereologic determination. Stereologic volume estimation can provide measures of synovial volumes comparable to the manual outlining method and is less time-consuming. Stereologic volume estimation seems to be a clinically useful method, especially if it is integrated in the MR unit's workstation.