Reproducibility of CT measurements of aortic valve calcification, mitral annulus calcification, and aortic wall calcification in the multi-ethnic study of atherosclerosis

University of Washington Seattle, Seattle, Washington, United States
Academic Radiology (Impact Factor: 2.08). 03/2006; 13(2):166-72. DOI: 10.1016/j.acra.2005.09.090
Source: PubMed

ABSTRACT Extracoronary calcifications may have clinical significance. The error in extracoronary calcification measurements is still unknown. Accurate quantification of calcifications of the aortic valve (AVC), mitral annulus (MAC), and aortic wall (AWC) may be possible by using cardiac computed tomography (CT). We sought to establish the interscan, interobserver, and intraobserver reproducibility of these measures in all cardiac CT scans in the Multi-Ethnic Study of Atherosclerosis.
We measured extracoronary calcifications in 100 randomly selected participants to assess interobserver, interscan, and intraobserver variability. Two scans were available for analysis in 99 of these participants, and we quantified thoracic aorta and valvular calcifications.
Mean interscan variability of AVC was 9.7% +/- 11.4% and 8% +/- 10.3% for Agatston and volume scores, with variability of the median at 6.4% and 5.5%, respectively (P > .05). MAC inter-reader variability was 8.2% and 8.9%, with interscan variability of 28% and 33% and intrareader variability of 4% and 4.1%, respectively. For AWC, inter-reader variability was 3%-7.1%, interscan variability was 17%-18%, and intrareader variability was 0.4%-1.4%.
AVC, MAC, and AWC measurements are sufficiently reproducible to allow serial investigations over a time suitable for clinical studies.

1 Follower
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Aortic valve calcification (AVC), even without haemodynamic significance, may be prognostically import as an expression of generalized atherosclerosis, but techniques for echocardiographic assessment are essentially unexplored. Two-dimensional (2D) echocardiographic recordings (Philips IE33) of the aortic valve in short-axis and long-axis views were performed in 185 consecutive patients within 1 week before surgery for aortic stenosis (n = 109, AS), aortic regurgitation (n = 61, AR), their combination (n = 8) or dilation of the ascending aorta (n = 7). The grey scale mean (GSMn) of the aortic valve in an end-diastolic short-axis still frame was measured. The same frame was scored visually 1-5 as indicating that the aortic valve was normal, thick, or had mild, moderate or severe calcification. The visual echodensity of each leaflet was determined real time applying the same 5-grade scoring system for each leaflet, and the average for the whole valve was calculated. Finally, a similar calcification score for the whole valve based on inspection and palpation by the surgeon was noted. Visual assessment of real-time images using the proposed scoring system showed better correlation with the surgical evaluation of the degree of valve calcification (r = 0·83, P<0·001) compared to evaluation of stop frames by visual assessment (r = 0·66, P<0·001) or the GSMn score (r = 0·64, P<0·001). High inter- and intra-observer correlations were observed for real-time visual score (both intraclass correlation coefficient = 0·93). Real-time evaluation of the level of AVC is superior to using stop frames assessed either visually or by dedicated computer grey scale measurement software.
    Clinical Physiology and Functional Imaging 11/2012; 32(6):470-5. DOI:10.1111/j.1475-097X.2012.01153.x
  • [Show abstract] [Hide abstract]
    ABSTRACT: The following note considers the complex case of Perry v. Louisiana, in which the Louisiana Supreme Court ordered an insane defendant on death row to be medicated against his will in order to render him sane, and therefore capable of being executed. In so doing the court pit judicial interests in effecting punishment of certain murderers against the physician's Hippocratic Oath, "first do no harm." In considering this conflict, the note identifies "first do no harm" as a guiding principle, explores the societal values underlying this basic principle, and concludes the judiciary must provide legal support for this medical ethical imperative. Similarly the legal profession must identify its organizing principles, its "first do no harm" proscriptions, and consider the application of those principles in the context of representing the insane. Some of the conflicts confronting the physician in the Perry situation have parallels for the attorney representing an insane client. Should the client be medicated in order to proceed to trial? Is it in the best interests of the client to remain a prisoner of her mental illness rather than to risk the possibility of conviction? How should the attorney address the paradoxical reality that a heavily medicated client may indeed become more lucid without becoming more competent? By publishing this note, the Journal hopes to engender discussion and clarification of the vague and sometimes incoherent guidance offered by the Medical Rules of Professional Responsibility, the Model Code of Professional Conduct, and the Criminal Justice Mental Health Standards for legal work with mentally disturbed clients.
    The Georgetown journal of legal ethics 02/1991; 4(3):707-29.