Cardiothoracic ratio in postmortem computed tomography: reliability and threshold for the diagnosis of cardiomegaly

Winklhofer, Sebastian; Berger, Nicole; Ruder, Thomas; Elliott, Marina; Stolzmann, Paul; Thali, Michael; Alkadhi, Hatem; Ampanozi, Garyfalia (2014). Cardiothoracic ratio in postmortem computed tomography: reliability and threshold for the diagnosis of cardiomegaly. Forensic science, medicine and pathology, 10(1), pp. 44-49. Springer US; 10.1007/s12024-013-9504-9

[img] Text
art%3A10.1007%2Fs12024-013-9504-9.pdf - Published Version
Restricted to registered users only
Available under License Publisher holds Copyright.

Download (278kB) | Request a copy

The aim of this study was to evaluate the reliability of the cardiothoracic ratio (CTR) in postmortem computed tomography (PMCT) and to assess a CTR threshold for the diagnosis of cardiomegaly based on the weight of the heart at autopsy. PMCT data of 170 deceased human adults were retrospectively evaluated by two blinded radiologists. The CTR was measured on axial computed tomography images and the actual cardiac weight was weighed at autopsy. Inter-rater reliability, sensitivity, and specificity were calculated. Receiver operating characteristic curves were calculated to assess enlarged heart weight by CTR. The autopsy definition of cardiomegaly was based on normal values of the Zeek method (within a range of both, one or two SD) and the Smith method (within the given range). Intra-class correlation coefficients demonstrated excellent agreements (0.983) regarding CTR measurements. In 105/170 (62 %) cases the CTR in PMCT was >0.5, indicating enlarged heart weight, according to clinical references. The mean heart weight measured in autopsy was 405 ± 105 g. As a result, 114/170 (67 %) cases were interpreted as having enlarged heart weights according to the normal values of Zeek within one SD, while 97/170 (57 %) were within two SD. 100/170 (59 %) were assessed as enlarged according to Smith's normal values. The sensitivity/specificity of the 0.5 cut-off of the CTR for the diagnosis of enlarged heart weight was 78/71 % (Zeek one SD), 74/55 % (Zeek two SD), and 76/59 % (Smith), respectively. The discriminative power between normal heart weight and cardiomegaly was 79, 73, and 74 % for the Zeek (1SD/2SD) and Smith methods respectively. Changing the CTR threshold to 0.57 resulted in a minimum specificity of 95 % for all three definitions of cardiomegaly. With a CTR threshold of 0.57, cardiomegaly can be identified with a very high specificity. This may be useful if PMCT is used by forensic pathologists as a screening tool for medico-legal autopsies.

Item Type:

Journal Article (Original Article)


04 Faculty of Medicine > Department of Radiology, Neuroradiology and Nuclear Medicine (DRNN) > Institute of Diagnostic, Interventional and Paediatric Radiology

UniBE Contributor:

Ruder, Thomas


600 Technology > 610 Medicine & health




Springer US;




Aisha Stefania Mzinga

Date Deposited:

10 Oct 2014 10:06

Last Modified:

23 Aug 2018 14:43

Publisher DOI:


PubMed ID:





Actions (login required)

Edit item Edit item
Provide Feedback