Virtual autopsies based on computed tomography (CT) and magnetic resonance imaging (MRI) are now used in addition to the traditional ‘body-opening’ autopsies to determine the cause of death in humans (1–5).

Pulmonary thromboembolism (PTE) is a cause of sudden death that is often difficult to diagnose with conventional imaging. Contrast-enhanced CT (CECT) is often used to diagnose PTE in living patients (6,7), but no studies have established a role for post mortem CECT.

Here, we describe a case of PTE diagnosed by post mortem CECT in a patient who died as a result of post-operative cardiopulmonary arrest.

A 77-year-old woman suffering from dural arteriovenous fistulae was treated by transfemoral transvenous embolization. She had a cardiopulmonary arrest after getting out of bed on the first post-operative day. Despite receiving prompt cardiopulmonary resuscitation, the patient died.

Post mortem echocardiography showed mild enlargements of her right atrium and right ventricle, but no intracardiac thrombus. We obtained permission from the patient’s family for a post mortem CT to determine the cause of death. We did not identify any specific lesion on a plain, whole-body, 64-multidetector row CT scan (Aquilion 64; Toshiba Medical Systems, Tokyo, Japan) (Fig. 1A and B). We then performed cardiac compressions at a rate of 70/min for 4 min while administering 100 ml of nonionic contrast material (iopamidol 370 mg I/ml, Iopamiron 370; Bayer Yakuhin, Osaka, Japan) via a peripheral vessel at 0.5 ml/sec. A subsequent post mortem CECT of the thorax showed filling defects characteristic of a large pulmonary thrombus in the left lower pulmonary artery (Figs. 1C, D and 2). Her D-dimer was also elevated at 149.2 μg/ml. Therefore, we diagnosed PTE as the cause of death.

PTE is a relatively common cardiovascular emergency that is difficult to diagnose and is frequently missed (6,7). Acute occlusion of the pulmonary arterial bed may cause life-threatening complications, and high-risk PTE has a short-term mortality rate of more than 15% (7–9). Patient-related predisposing factors for PTE include increased age, a past history of venous thromboembolism (VTE), active cancer, neurological disease with extremity paresis, medical disorders causing prolonged bed rest, such as heart and respiratory failure, congenital or acquired thrombophilia, hormone replacement therapy and oral contraceptive therapy (7). Short-term immobilization also increases the risk of VTE (6). Although these risk factors are well established, many cases of PTE still go unrecognized and untreated (6). The prevalence of PTE at autopsy is approximately 12–15% in hospitalized patients and has not changed over the last three decades (10). The rate of undiagnosed PTE in patients at post mortem has not diminished either, even in individuals who die from massive or sub-massive PTE (6,11). In autopsy studies, the prevalence of unsuspected PTE, either fatal or contributing to death, ranges from 3 to 8% (6,10,12). The incidence of VTE increases exponentially with age, as do the rates of idiopathic and secondary PTE (13,14). The mean age of patients with PTE is 62 years and approximately 65% are 60 years of age or older. Eight-fold higher rates of PTE are observed in patients over 80 years, compared to those younger than 50 years (15). PTE has a wide range of clinical presentations including dyspnea, chest pain, syncope, hypotension and shock (6,7). First-line diagnostic tests, such as ECG, chest X-ray and blood-gas analysis, are indicated to assess the clinical probability of PTE and the general condition of the patient (6). A negative D-dimer result safely excludes the diagnosis in patients with a low or moderate clinical probability of PTE (6,7). However, D-dimer has very high sensitivity but low specificity, so a positive result requires imaging to confirm the diagnosis. Specific diagnostic imaging techniques for PTE include plain chest radiography, echocardiography, ventilation-perfusion scintigraphy, CT, MRI and pulmonary angiography (6,7). In recent years, technical advances in CT have prompted interest in this technique for the diagnosis of PTE (6,7). However, two systematic overviews of the performance of single detector spiral CT in suspected PTE reported wide variations in sensitivity (53–100%) and specificity (73–100%) (7).

Invasive ‘body-opening’ autopsy is the traditional post mortem investigation in humans (1). Modern cross-sectional imaging techniques, however, can supplement and even partially replace traditional autopsy (1–3,5). Conventional autopsies, which are often rejected by family members or certain religious groups, may eventually be replaced by noninvasive imaging (1). CT is the imaging modality of choice for two- and three-dimensional documentation and detects fractures, pathological gas collections (air embolism, subcutaneous emphysema after trauma, hyperbaric trauma and decomposition effects) and gross tissue injuries (1). The documentation and analysis of post mortem findings with CT and MRI and post-processing techniques (‘virtopsy’ or ‘autopsy imaging’) is investigator-independent, objective and noninvasive, and should lead to qualitative improvements in pathologic investigation (1,3). Potential applications for this technique include the assessment of morbidity and mortality in the general population and the routine screening of bodies prior to burial (1). A transdisciplinary research project, virtopsy, is dedicated to increase the use of modern imaging techniques in forensic medicine and pathology to augment current examination techniques and offer alternative methods (16).

In conclusion, our patient died suddenly in the hospital the day after her surgery. Although a previous study using post mortem CT reported a high success rate in detecting causes of sudden death (17), we did not find any lesion on a plain CT or CECT without cardiac compression in our patient. However, a post mortem CECT conducted after cardiac compression confirmed PTE as the cause of her sudden death.