Causes of death of forensic autopsy cases tested positive for COVID-19 in Tokyo Metropolis, Japan

Abstract

Background

The profile of deaths related to coronavirus disease of 2019 (COVID-19) that occurred outside the hospital in Japan remains unclear because of cautious stance on performing autopsies of COVID-19 positive cases.

Methods

Autopsy cases that tested positive for COVID-19 in the Tokyo Metropolis from April 2020 to July 2022 were handled by medical examiners (n = 41). Age, sex, medical history, autopsy findings, cause of death, postmortem computed tomography (PMCT) findings, and the causal relationship between death and COVID-19 were examined.

Results

The mean age of the deceased was 58.0 years (range: 28–96 years), and the study sample consisted of 33 males (80.5%) and 8 females (19.5%). The most frequent medical histories were hypertension (n = 7) and diabetes (n = 7), followed by mental disorders (n = 5). Nineteen cases showed a body mass index ≧25.0 (46.3%). The leading cause of death was pneumonia (n = 17), in which diffuse ground-glass opacification and/or consolidation was noted on PMCT. There were 26 deaths directly related to COVID-19 (63.4%), including pneumonia, myocarditis, laryngotracheobronchitis, and emaciation. The proportion of deaths directly related to COVID-19 was lower after 2022 (42.1%) than prior to 2022 (81.8%).

Conclusion

Pneumonia was the leading cause of death in this study sample; however, the causes of death in COVID-19 positive cases varied, especially after 2022, when the omicron variant was dominant. Mortality statistics may be affected by viral mutations, and the results of this study further emphasize the need for autopsy because more differential diagnoses should be considered in the phase of the omicron variant.

1. Introduction

Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has spread rapidly across the globe since the end of 2019, and is responsible for 629 million infections and 6.5 million deaths as of October 28, 2022 [1]. Despite the decrease in mortality rate and severity of COVID-19 due to the development of vaccines and therapeutics, some individuals deteriorate rapidly and develop acute respiratory distress syndrome (ARDS) [2].

In Japan, the number of patients increased dramatically in January and August 2021, and July 2022, making medical resources scarce. As a result, even older patients or those with underlying diseases are forced to remain at home if they do not have severe symptoms, leading to the appearance of out-of-hospital sudden death [3]. Postmortem computed tomography (PMCT) revealed severe pneumonia as a cause of death in some cases; however, PMCT has not always been performed for deaths occurring outside of the hospital. In addition, the cautious stance on performing autopsies on COVID-19 positive cases has hampered an assessment of the causative relationship between COVID-19 and the cause of out-of-hospital deaths in Japan. Several case reports in which pathological autopsies were performed have been published; however, these reports are mostly single case reports and do not include out-of-hospital or non-natural death [4], [5], [6], [7], [8], [9], [10]. There are few reports of forensic autopsy cases in Japan at the time of writing this manuscript [11]. Therefore, the profile of COVID-19 related deaths that occurred outside the hospitals in Japan remains unclear.

Tokyo Metropolis is a metropolitan prefecture, and the medical examiner system has been implemented in its special wards. All medicolegal deaths, including natural, non-natural, and undetermined manner of death, that occurred in the special wards of Tokyo Metropolis are reported to the Tokyo Medical Examiner’s Office. Medical examiners performed postmortem examinations to determine the manner and cause of death in these cases. Medical examiners also observed that the deceased tested positive for SARS-CoV-2, therefore making them cases suspected of COVID-19. They evaluated the results of situational investigation, external examination, viral RNA test, and PMCT and performed an autopsy if the cause of death was undetermined. In this study, we investigated forensic autopsied cases that tested positive for COVID-19 handled by medical examiners in the Tokyo Metropolis to clarify the profile of COVID-19 related deaths that occurred outside the hospital in Japan. In addition, we investigated whether there is a difference in the involvement of COVID-19 in the cause of death between cases before and after 2022, when the omicron variant became dominant.

2. Materials and methods

From April 1, 2020, to July 31, 2022, 41 autopsies of persons who tested positive for SARS-CoV-2 ante- or postmortem were performed at the Tokyo Medical Examiner’s Office. Among the 41 patients, 8 (19.5%) died in 2020, 14 (34.1%) died in 2021, and 19 (46.3%) died in 2022. All documents available concerning the deaths of these cases (medical history, police investigation reports, death certificates, PMCT, and autopsy reports) were evaluated descriptively. We examined age, sex, medical history, PMCT findings, autopsy findings, cause of death, and the causal relationship between death and COVID-19.

Autopsy findings included macroscopic findings, histopathological findings, and toxicological analyses. Histopathological examination was performed in all cases, and blood ethanol levels were measured in all cases except for case (No. 19). Toxicological analysis was performed for 21 cases (No. 1,2,5,7,8,13,15–17,20,21,23–25,27,30,32–34,37,41). For the viral RNA test, nasopharyngeal swab samples were subjected to a transcription-reverse transcription concerted reaction using TRC Ready-80 (Tosoh Techno-System, Tokyo, Japan), except for two cases in which COVID-19 had already been diagnosed before death (No. 15, 20). Regarding causal relationships, we divided the cases into three groups (death directly related to COVID-19, death indirectly related to COVID-19, and death unrelated to COVID-19), according to the death certificates and autopsy findings. The proportion of deaths directly related to COVID-19 was compared between the cases that occurred before and after 2022.

Whole-body PMCT was performed before autopsy in all cases using a 64-row CT scanner (Somatom Definition AS; Siemens Healthcare, Forchheim, Germany) with the following parameters: 120 kV; quality reference, 400 mAs; thickness, 64 × 0.6 mm. Image data were analyzed using the syngo.via software (Siemens Healthcare). Lung patterns were evaluated and divided into two subcategories: ground glass opacities (GGO) and consolidation. The location of each lung pattern (diffuse, patchy, hypostatic, anterior, or posterior) was also determined. The Ethics Committee of the Tokyo Medical Examiner’s Office approved the study protocol and data use (approval number: 2020–3).

3. Results (Table 1)

Table 1.

Detailed information concerning the investigated 41 cases.

No.	Age	Gender	BMI	Cause of death	Autopsy findings relevant to cause of death	Other autopsy findings	Causal relationship
1	55	M	34.3	Pneumonia	Diffuse alveolar damage in the lungs	Cardiomegaly (738 g)	1
2	44	F	43.7	Pneumonia	Diffuse alveolar damage in the lungs	Coronary sclerosis, fatty liver	1
3	48	M	35.8	Pneumonia	Diffuse alveolar damage in the lungs	Cardiomegaly (750 g)	1
4	74	M	25	Pneumonia	Diffuse alveolar damage in the lungs	Cardiomegaly (450 g), chronic thyroiditis	1
5	52	M	35.6	Pneumonia	Diffuse alveolar damage in the lungs	Cardiomegaly (640 g)	1
6	74	M	22.8	Pneumonia	Diffuse alveolar damage in the lungs	Cardiomegaly (647 g), coronary sclerosis	1
7	77	M	20.6	Drowning	Aspiration of water in the airway and the stomach	Partial appearance of hyaline membrane and hyperplasia of type Ⅱ alveolar epithelial cells, chronic interstitial pneumonia	3
8	47	M	24.4	Poisoning (psychiatric drugs)	Detection of Olanzapine in blood (1.4 μg/mL)	Aspiration of vomitus in the bronchi, Pneumonia	3
9	38	M	26.5	Alcohol intoxication	Blood ethanol level (4.18 mg/mL)	Fatty liver	3
10	52	M	18.9	Pneumonia	Diffuse alveolar damage in the lungs	–	1
11	68	F	18	Pneumonia	Diffuse alveolar damage in the lungs	Cardiomegaly (405 g)	1
12	71	M	26.7	Pneumonia	Diffuse alveolar damage in the lungs	Cardiomegaly (533 g), fatty liver	1
13	53	M	20.3	Pneumonia	Diffuse alveolar damage in the lungs	Emphysema, coronary sclerosis	1
14	53	M	24	Pneumonia	Diffuse alveolar damage in the lungs	Emphysema, cardiomegaly (479 g), fatty liver	1
15	44	F	23.8	Myocarditis	Edema and infiltration of monocyte and neutrophil in the interstitial space of cardiomyocytes, appearance of contraction band in cardiomyocytes	Partial infiltration of macrophage and lymphocyte in the alveolar septum	1
16	50	M	28.4	Myocarditis	Edema and infiltration of monocyte and neutrophil in the interstitial space of cardiomyocytes, appearance of contraction band in cardiomyocytes	Cardiomegaly (574 g), fatty liver, enlargement of thyroid gland (93 g)	1
17	60	F	18	Myocarditis	Edema and infiltration of monocyte and neutrophil in the interstitial space of cardiomyocytes, appearance of contraction band in cardiomyocytes	Diffuse alveolar damage in the right segment 6, chronic thyroiditis	1
18	28	M	38.9	Pneumonia	Diffuse alveolar damage in the lungs	Engorgement of pharyngeal and laryngeal mucosa, cardiomegaly (520 g), fatty liver	1
19	47	M	21.6	Pneumonia suspected	Suspicious of diffuse alveolar damage in the lungs, no findings other than COVID-19	Severe putrefactive changes	1
20	60	M	20.7	Myocarditis, epicarditis	Infiltration of monocyte and neutrophill in the interstitial space of cardiomyocytes and epicardium, appearance of contraction band in cardiomyocytes	Cardiomegaly (471 g), liver cirrhosis	1
21	55	F	21.6	Cor pulmonale due to COVID-19 pneumonia	Enlargement of the right ventricle, lung fibrosis,	Chronic thyroiditis, uterine myoma	1
22	60	M	24.7	Alcoholic liver cirrhosis	Yellowish, swollen liver (2368 g), significant lipid deposition and proliferation of collagen fibers	Cardiomegaly (470 g), engorgement of bronchial mucosa, chronic pancreatitis	2
23	72	M	13.3	Heat stroke	High rectal temperature, thermal denaturation of whole organs	Emphysematous change, moderate putrefactive change	3
24	62	M	25.2	Ischemic heart disease	Cardiomegaly (1053 g), old myocardial infarction, coronary sclerosis	Lung edema, scattered hyaline membrane along the alveolar wall	2
25	89	M	17.6	Dehydration with mild pneumonia	Dehydration, malnutrition (BMI 17.6), neutrophil infiltration in the alveoli in the right lower lobe	Cardiomegaly (433 g), coronary sclerosis, benign nephrosclerosis	1
26	61	M	27.1	Pneumonia	Diffuse alveolar damage in the lungs	Cardiomegaly (552 g), coronary sclerosis, fatty liver, diabetic nephrosclerosis	1
27	54	M	21.3	Choking	Aspiration of vomitus in the larynx, trachea and bronchi	Blood ethanol level (1.02 mg/mL), bronchitis(mild)	3
28	70	M	14.1	Cor pulmonale due to emphysema	Significant enlargement of the air cavity due to collapse of the alveolar septum, enlargement of the right ventricle	Malnutrition, coronary sclerosis	2
29	45	M	26.4	Laryngotracheobronchitis, pharyngitis	Dehydration, lymphocyte and macrophage infiltration in engorged mucosa of the pharynx, larynx, trachea and bronchi	Cardiomegaly (688 g), coronary sclerosis, fatty liver	1
30	56	M	27.1	Laryngotracheobronchitis, pneumonia	Dehydration, neutrophil infiltration in engorged mucosa of the larynx, trachea and bronchi, focal infiltration of neutrophil in the alveoli	Cardiomegaly (448 g), atrophy of adrenal cortex	1
31	56	M	25.6	Pneumonia	Diffuse alveolar damage in the lungs	Cardiomegaly (542 g), fatty liver	1
32	96	F	12.8	Choking	Aspiration of semiliquid food in the bronchi	Aspiration pneumonia, malnutrition, benign nephrosclerosis	2
33	35	F	42.5	Laryngotracheobronchitis, epiglottitis, hypoxic encephalopathy	Neutrophil infiltration in engorged mucosa of the epiglottis, larynx, trachea and bronchi	Aspiration pneumonia suspected, Fatty liver, enlargement of the thyroid gland	1
34	44	M	24.6	Ischemic heart disease	Significant stenosis of the left anterior descending coronary artery	Bronchitis (mild), lung edema	3
35	85	F	16.8	Peritonitis	Perforation of the ileum, redness of the surface of the small intestine, muddy ascites	–	3
36	71	M	27.9	Pneumonia	Diffuse alveolar damage in the lungs	Cardiomegaly (430 g), fatty liver, aortic stenosis	1
37	44	M	38.9	Myocarditis	Infiltration of monocyte and neutrophil in the interstitial space of cardiomyocytes, appearance of contraction band in cardiomyocytes	Cardiomegaly (693 g), bronchopneumonia, fatty liver	1
38	68	M	17	Acute subdural hematoma	Bilateral (left > right) subdural hematoma, traumatic subarachnoid hemorrhage in the left parietal lobe	Diabetic nephrosclerosis	3
39	27	M	26.2	Multiple injury	Multiple rib fractures, pulmonary contusion, injury of the spleen and kidney, traumatic subarachnoid hemorrhage	Blood ethanol level (4.46 mg/mL)	3
40	72	M	16.6	Rectal cancer	Posterior peritoneal abscess around the rectal cancer (6 × 6 cm sized)	Cardiomegaly (546 g)	3
41	61	M	21	Drowning	Water inhalation and microbubbles in the trachea	Detection of levomepromazine in blood (0.39 μg/mL)	3

Gender: M; male, F; female.

Causal relationship: 1; death directly related to COVID-19, 2; death indirectly related to COVID-19, 3; death unrelated to COVID-19.

The mean age of the deceased individuals was 58.0 years (range: 28–96 years), and the most frequent age group was 50–59 years (n = 10), followed by 70–79 years (n = 8), 60–69 years (n = 8) and 40–49 years (n = 8). The study sample consisted of 33 males (80.5%) and 8 females (19.5%). Twenty-seven cases (65.9%) had a medical history, and the most frequent diseases were hypertension (n = 7), diabetes (n = 7), mental health disorders (n = 5), and cardiac diseases (n = 3). The mean body mass index (BMI) was 24.8 kg/m². Approximately half of the cases (19 cases, 46.3%) had a BMI ≧25.0, and 7 cases (17.1%) had a BMI ≧30.

The manner of death was categorized as natural in 32 cases and non-natural in 9 cases. Causes of death due to natural causes included pneumonia (n = 17), myocarditis (n = 5), laryngotracheobronchitis (n = 3), ischemic heart disease (n = 2), dehydration (n = 1), emphysema (n = 1), liver cirrhosis (n = 1), rectal cancer (n = 1), and peritonitis (n = 1). Natural deaths included 26 cases in which COVID-19 was directly related to death, including pneumonia (Fig. 1 ), myocarditis (Fig. 2 ), and laryngotracheobronchitis (Fig. 3 ) (No. 1–6, 10–21, 25, 26, 29–31, 33, 36, 37). There were three cases in which COVID-19 was indirectly related to death (No. 22, 24, 28). The main pathology in these cases was alcoholic liver cirrhosis, severe cardiomegaly with old myocardial infarction, and emphysema. There were 3 cases in which COVID-19 was not related to death (No. 34, 35, 40), and causes of death in those cases were ischemic heart disease, peritonitis due to perforation of the ileum, and rectal cancer. Causes of death due to non-natural causes (n = 9) included drowning (n = 2), choking (n = 2), psychiatric drug poisoning (n = 1), acute alcohol intoxication (n = 1), heat stroke (n = 1), acute subdural hematoma (n = 1), and multiple injuries (n = 1) (case 7–9, 23, 27, 32, 38, 39, 41). All non-natural deaths were unrelated to COVID-19, except for one case (No. 32). Among the participants, there were 26 deaths directly related to COVID-19 (63.4%), 4 deaths were indirectly related to COVID-19 (9.8%), and 11 deaths were unrelated to COVID-19 (26.8%). The proportion of deaths directly related to COVID-19 was higher in cases before 2022 (81.8%) than in those post 2022 (42.1%).

Fig. 1.

Different stages of COVID-19 pneumonia (a, b; No. 31, c, d: No. 36, e, f; No. 21) a. PMCT shows diffuse GGO and crazy-paving pattern. b. Histopathology shows a hyaline membrane along the alveolar wall and infiltration of lymphocytes/macrophages in the alveolar septum. c. PMCT shows diffuse GGO with patchy consolidation and crazy-paving pattern. d. Histopathology shows a thrombus with marginal organization in the small artery and a residue of hyaline membrane with neutrophil infiltration in the alveoli. e. Diffuse GGO and interstitial changes (especially in the bilateral lower lobes) are noted on PMCT. f. Proliferation of collagen fibers with infiltration of lymphocytes and fibroblasts is seen on histopathology.

Fig. 2.

PMCT and macroscopic/histopathological findings of myocarditis/epicarditis (No. 20) a, b. PMCT shows pericardial effusion, right pleural effusion, and GGO in the posterior lung. c. The epicardium is reddened (the circle). d. Histopathology showing infiltration of monocytes and neutrophils in the interstitial space of cardiomyocytes and the appearance of contraction bands in cardiomyocytes.

Fig. 3.

Macroscopic and histopathological finding of laryngotracheobronchitis (No. 33) a. The mucosa of the epiglottis, larynx, and trachea are seen engorged with pus. b. Neutrophil infiltration is noted in the mucosal layer of the larynx.

Lung pathology related to COVID-19 was observed in 25 cases, including diffuse alveolar damage (DAD) in a wide range of bilateral lungs (n = 17) (Fig. 1b), partial hyaline membrane along the alveolar wall and infiltration of lymphocytes/macrophages in the alveolar septum (n = 4), and focal infiltration of neutrophils in the alveolar space (n = 4). All deaths from pneumonia in this study sample showed DAD in a wide range of lungs. Thrombi were observed macroscopically in the peripheral branch of the pulmonary arteries in one case, and thrombi were observed microscopically in the small arteries of the lungs in 6 cases (Fig. 1d). Representative autopsy findings, other than lung pathology, included cardiomegaly (n = 20), fatty liver (n = 12), coronary sclerosis (n = 7), and nephrosclerosis (n = 4).

All deaths from pneumonia (n = 17) showed diffuse GGO and/or consolidation, except for one case (case 19; decomposition) (Fig. 1 a, c). A crazy-paving pattern was observed in 8 cases. Regarding PMCT findings in other cases (n = 24), patchy or localized GGO and/or consolidation was observed in 16 cases, only hypostatic changes were seen in 2 cases, and diffuse GGO and/or consolidation was observed in 6 cases.

4. Discussion

It has been reported that the presence of some chronic diseases besides COVID-19 pneumonia is a poor prognostic criterion and increases mortality [12], [13], [14]. Chen et al. reported that hypertension, cardiovascular disease, and diabetes were more prevalent among COVID-19 patients who died than among survivors [12]. Zhou et al. reported that hypertension, diabetes, coronary heart disease, chronic renal disease, and COPD were more frequent among non-survivors than among survivors [13]. Patients with obesity are reported to have a high risk of mortality from COVID-19 [14]. The major comorbidities detected in this study (e.g., hypertension, diabetes, cardiac diseases, obesity, cardiomegaly, fatty liver, coronary sclerosis, and nephrosclerosis) also tracked those in a previous study.

Several studies have investigated the causes of death in consecutive forensic autopsy of COVID-19 positive cases [15], [16], [17], [18], [19], [20], [21]. Edler et al. investigated 80 autopsy cases of death with SARS-CoV-2 infection and reported that pneumonia was the cause of death in all cases of definite COVID-19 deaths (71%) and that pneumonia was present in all cases of probable COVID-19 deaths (12.5%). Four deaths were defined as non-COVID-19 deaths with virus-independent causes (5%). They also reported that DAD was observed histologically in 8 of 12 cases they evaluated and that pneumonia was combined with fulminant pulmonary thromboembolism (PE) in 8 cases [15]. Romanova et al. investigated forensic (n = 60) and clinical (n = 42) autopsies with positive postmortem SARS-CoV-2 PCR results. According to their results, COVID-19 caused or contributed to death in 71% of the clinical cases and 83% of the forensic autopsies. Regarding the cause of death, the vast majority of fatalities were related to DAD, and lymphocytic myocarditis was a rare finding (n = 2) [16]. Arslan et al. analyzed COVID-19 positive cases (n = 26), and the cause of death was determined to be viral pneumonia in 21 cases, blunt trauma in 4 cases, and hanging in one case. They also reported that DAD was prominent and the main pathology was pneumonia in autopsied cases (n = 7) [17]. Muchelenganga et al. investigated 21 COVID-19 autopsy cases and reported that PE (n = 16), DAD (n = 3), and pneumonia (n = 2) were the common causes of death [18]. Fanton et al. reported four COVID-19 related out of hospital cardiac arrest, and the cause of death in three individuals was acute respiratory failure due to DAD, while violent death due to suicidal acute alcohol intoxication was the cause of death in one case [19]. Keresztesi et al. investigated 15 autopsy cases and reported that massive bilateral pneumonia was the direct cause of death in 13 cases. The causes of death in the other two cases were pulmonary carcinoma and bronchopneumonia following femur fracture. DAD was observed histologically in five out of seven cases that they evaluated [20]. Danics et al. divided their 100 autopsy cases into three mortality categories by relevance of COVID-19 infection: strong association (n = 57), contributive association (n = 27) and weak association (n = 16), and lung pathology was the primary cause of death in the strong and contributive categories [21]. In this study, the most frequent cause of death was pneumonia, and the most common lung histopathology was DAD. These features are similar to the results of the previous studies mentioned above; however, the proportion of deaths from pneumonia in this study (41.5%) was not high compared with previous studies (e.g., 86.3% [15], 80.8% [17], and 86.7% [20]). In addition, this study showed that the causes of death from COVID-19 were more varied, including pneumonia, myocarditis, laryngotracheobronchitis, and dehydration (i.e., emaciation due to COVID-19).

In this study, there were three cases of laryngotracheobronchitis as the cause of death. Unlike previous forensic autopsy studies, our sample included deaths occurring in 2022 (n = 19, 46.3%), when the omicron variant spread quickly globally, replacing the delta variant. All deaths due to laryngotracheobronchitis in this study occurred in February 2022. Compared with previous variants, the omicron variant showed milder lower respiratory tract symptoms and olfactory or taste disturbances; however, upper airway symptoms, such as sore throat, rhinorrhea, and sneezing, were reported more frequently [22]. Piersiala reported a case series of COVID-19 positive patients with acute odynophagia, severe sore throat, and fever. All the patients developed COVID-19 associated acute laryngitis and/or pharyngitis [23]. In Japan, multiple cases of COVID-19 induced upper airway stenosis and related acute laryngitis were reported by the Japanese Society of Otorhinolaryngology Head and Neck Surgery in late February 2022 [24]. In our case, dehydration (No. 29, 30), pneumonia (No. 30), and airway stenosis (No. 33) may have been accompanied by laryngotracheobronchitis, resulting in death. On the other hand, several studies showed milder severity of the omicron variant compared to the previous variant [25], [26]. Regarding the causal relationship between death and COVID-19 in our sample, the proportion of deaths directly related to COVID-19 was higher in cases before 2022 (81.8%) than in those after 2022 (42.1%). This result suggests that the proportion of deaths directly related to COVID-19 might have decreased after the appearance of the omicron variant, although further studies are needed.

PMCT findings correlate with the severity of COVID-19 lung disease and have been proposed as a useful screening tool to identify COVID-19 related fatalities [27], [28], [29]. All cases of death from pneumonia in this study, except for one case (severe decomposition), showed findings characteristic of COVID-19 pneumonia, such as diffuse GGO, crazy paving, and areas of consolidation [27], [28], [29], [30], [31]. The use of PMCT in conjunction with postmortem RNA testing could be considered a reliable and safe modality for confirming COVID-19 pneumonia. However, reliable diagnosis of fatal COVID-19 can only be established by a combination of clinical, radiologic, microbiologic, and histopathologic correlations, with the latter two having the most discerning diagnostic value [28]. Indeed, GGO and/or consolidation were seen on PMCT in many cases of death unrelated to COVID-19, although they are not typical of COVID-19 pneumonia. Furthermore, COVID-19 related extra-pulmonary manifestation, such as myocarditis/epicarditis and laryngotracheobronchitis, in this study are difficult to prove by PMCT, although pericardial effusion on PMCT might hint at the diagnosis of myocarditis/epicarditis.

In Japan, physicians have to report cases of COVID-19 and cases suspected of dying from COVID-19 to the public health center according to the Infectious Diseases Control Law. However, reported cases may include not only definite COVID-19 deaths, but also cases dying with COVID-19, including deaths unrelated to COVID-19 because differentiating between deaths with COVID-19 and death from COVID-19 is difficult, especially out-of-hospital death in which ante-mortem information is limited. The clinical course of COVID-19 is highly heterogeneous, and the deceased may have pre-existing diseases that may contribute to or even cause death [32]. In addition to microbiological testing and PMCT, autopsy and histological analysis of COVID-19 positive cases plays a crucial role in assessing the causative relationship between death and COVID-19. The results of this study suggest that the proportion of definite COVID-19 deaths among SARS-CoV-2 positive cases might have decreased, and the causal relationship might have been more valid after the appearance of the omicron variant, which emphasizes the need for autopsying COVID-19 positive cases to obtain accurate mortality statistics.

This study had several limitations. First, the sample size was small, and the data were collected from an area inhabited by approximately 7% of the total Japanese population. Therefore, these results may not be generalizable to the entire Japanese population. Second, we selected only autopsied cases (41 of 365 SARS-CoV-2 positive cases; 11.2%) for this study because detailed investigation (e.g., causal relationship between death and COVID-19) was impossible in non-autopsied cases. Therefore, we cannot deny that there was selection bias in this study. Further large-scale studies are needed to address these limitations.

In conclusion, our study investigated forensic autopsy cases that tested positive for COVID-19 in Tokyo Metropolis and revealed that the leading cause of death was pneumonia, similar to previous studies; however, the causes of COVID-19 related death were various including myocarditis, laryngotracheobronchitis, and emaciation due to COVID-19. Furthermore, the proportion of deaths directly related to COVID-19 was lower in cases after 2022 when the omicron variant was dominant. Viral mutations may affect the pathology and mortality statistics. The results of this study further emphasize the need for autopsying COVID-19 positive cases to obtain accurate mortality statistics because physicians should consider more differential diagnoses in the phase of the omicron variant.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.