Original communication Evaluation of coronary stenosis with the aid of quantitative image analysis in histological cross sections Kate Dulohery MSc a, * , Asteria Papavdi MD b, c , Manolis Michalodimitrakis MD, JD a , Elena F. Kranioti MD, PhD a a Forensic Anthropology, School of History Classics and Archaeology, University of Edinburgh, Doorway 4, Teviot Place, Edinburgh EH8 9AG, Scotland, UK b Department of Forensic Sciences, Medical School, University of Crete, 71110 Heraklion, Greece article info Article history: Received 5 September 2011 Received in revised form 29 March 2012 Accepted 21 April 2012 Available online xxx Keywords: Coronary stenosis Quantification Image analysis abstract Coronary artery atherosclerosis is a hugely prevalent condition in the Western World and is often encountered during autopsy. Atherosclerotic plaques can cause luminal stenosis: which, if over a significant level (75%), is said to contribute to cause of death. Estimation of stenosis can be macro- scopically performed by the forensic pathologists at the time of autopsy or by microscopic examination. This study compares macroscopic estimation with quantitative microscopic image analysis with a particular focus on the assessment of significant stenosis (>75%). A total of 131 individuals were analysed. The sample consists of an atherosclerotic group (n ¼ 122) and a control group (n ¼ 9). The results of the two methods were significantly different from each other (p ¼ 0.001) and the macroscopic method gave a greater percentage stenosis by an average of 3.5%. Also, histological examination of coronary artery stenosis yielded a difference in significant stenosis in 11.5% of cases. The differences were attributed to either histological quantitative image analysis underestimation; gross examination over- estimation; or, a combination of both. The underestimation may have come from tissue shrinkage during tissue processing for histological specimen. The overestimation from the macroscopic assessment can be attributed to the lumen shape, to the examiner observer error or to a possible bias to diagnose coronary disease when no other cause of death is apparent. The results indicate that the macroscopic estimation is open to more biases and that histological quantitative image analysis only gives a precise assessment of stenosis ex vivo. Once tissue shrinkage, if any, is accounted for then histological quantitative image analysis will yield a more accurate assessment of in vivo stenosis. It may then be considered a comple- mentary tool for the examination of coronary stenosis. Ó 2012 Elsevier Ltd and Faculty of Forensic and Legal Medicine. All rights reserved. 1. Introduction In order for significant atherosclerosis (coronary artery disease) to be considered a contributing factor to the death of an individual, at least one of the major coronary arteries must have more than 75% luminal stenosis. 1 Reliable and accurate assessment of stenosis is essential for establishing cause of death and in some cases manner of death. 2e5 Exact percentages are especially relevant in cases where there is moderate stenosis with questions arising on life expectancy. There are two methods used to determine the percentage of stenosis. One is a gross visual assessment performed by the pathologist during the autopsy and the other requires the use of histological methods. Gross visual examination of coronary arteries is simple and inexpensive but is subjective. Pathologists can only assign stenosis with accuracy when it is under 30% or if it is over 70%. 6,7 Womack 7 demonstrated that pathologists classify stenosis as severe when the actual percentages ranged between 40% and 90% (average 67%). This may be attributed to various examiner biases. In order to examine the potential observer biases inherent in macroscopic visual assessment a reliable objective method of assessing coronary artery stenosis is required. A study focussing solely on the value of the histological assessment of coronary disease autopsy has not been found in the literature. With the macroscopic method, the pathologist has to visually correct for the different types of coronary narrowing (concentric or eccentric) as well as the lack of pressure in post-mortem arteries. 8,9 Diagrammatic schemes have been developed to indicate the * Corresponding author. Forensic Anthropology School of History, Classics and Archaeology, The University of Edinburgh, Doorway 4, Teviot Place, Edinburgh EH8 9AG, UK. Tel.: þ44 (0) 131 650 2368; fax: þ44 (0) 131 650 2378. E-mail address: katedulohery@yahoo.co.uk (K. Dulohery). c Equal contribution to the paper. Contents lists available at SciVerse ScienceDirect Journal of Forensic and Legal Medicine journal homepage: www.elsevier.com/locate/jflm 1752-928X/$ e see front matter Ó 2012 Elsevier Ltd and Faculty of Forensic and Legal Medicine. All rights reserved. doi:10.1016/j.jflm.2012.04.024 Journal of Forensic and Legal Medicine xxx (2012) 1e5 Please cite this article in press as: Dulohery K, et al., Evaluation of coronary stenosis with the aid of quantitative image analysis in histological cross sections, Journal of Forensic and Legal Medicine (2012), doi:10.1016/j.jflm.2012.04.024