Necropsy Findings in Sars-CoV-2 Infections—A Retrospective Study from Iasi, Romania

Abstract

Introduction: The global spread of the SARS-CoV-2 pandemic led to a serious health, social and economic global crisis. This pandemic was and remains the most important health emergency worldwide, for which all professionals have been called to provide diagnosis and treatment support. Despite early concerns about safety, forensic medicine has contributed to a better understanding of the pathological mechanisms involved. Objective: This study aims to describe and analyze the postmortem pathological findings in confirmed SARS-CoV-2 cases, emphasizing the contribution of forensic autopsies to elucidating the mechanisms of death and associated comorbidities. Methods: A retrospective study was conducted on 279 autopsies between 2020 and 2022. Demographic, clinical, and pathological data were collected and statistically analyzed. Results: Following the descriptive analysis of the cases included in the study, as well as the analysis of the relevant scientific literature, the major impact of the COVID-19 pandemic was highlighted in terms of the death mechanisms involved, occurring consequences and induced changes. Conclusions: Autopsies remain the essential tools for investigating COVID-19-related deaths. The findings confirm that SARS-CoV-2 primarily affects the pulmonary and cardiovascular systems. However, the overlap between “death from” and “death with” COVID-19 highlights the need for standardized postmortem diagnostic criteria and comprehensive clinical–pathological correlation.

1. Introduction

At the beginning of 2020, the entire world was affected by the SARS-CoV-2 (Severe Acute Respiratory Syndrome Coronavirus 2) pathogen, the WHO (World Health Organization) declaring a pandemic in March 2020. The spectrum of COVID-19 symptoms was extremely varied, ranging from asymptomatic to mild respiratory symptoms to multi-organ failure and death [1]. Even from the first cases of pneumonia identified in Wuhan, China at the end of 2019, the whole world has had an up and down trend. Despite the vaccination program, at the end of November 2021, there were more than 260 million cases and about 5.2 million deaths worldwide. This up and down trend lasted from March 2020 until May–June 2023, with a steady decrease thereafter. Despite this downward trend, the health and economic impact of the SARS-CoV-2 infection can still be felt in many areas around the world, with almost half a million new cases per week and a total number of infections reaching 7.7 × 10⁶ as of November 2024, according to WHO statistics [2].

The SARS-CoV-2 infection causes the release of a significant number of proinflammatory cytokines that aggravate the interstitial pneumonia and ARDS. Other clinical conditions which were linked to this infection included pharyngitis, conjunctivitis, Sjogren’s syndrome, and others [1]. If left untreated, symptoms can progress to sepsis with hypercoagulability and multiple organ dysfunction [3].

The medico-legal system worldwide was forced to quickly adapt in order to respond adequately to the changes imposed by the pandemic and continue its activity. Undoubtedly, the forensic autopsy had an important role in unraveling how this virus affects the human body and, of course, in accurately establishing the person’s diagnosis of death (due to COVID-19 or whether he died with COVID-19, but the main cause of death being another one), the causal link between death and the presence of the virus, between death and comorbidities and the action of the virus on an organism with pre-existing pathologies [4].

In this context, the role of forensic medicine has been crucial. Although initially hindered by biosafety concerns, forensic autopsies provided essential data on how SARS-CoV-2 affects the human body and on the interplay between viral infection and pre-existing comorbidities [5,6,7]. Autopsies have helped differentiate between deaths directly caused by COVID-19 and those in which SARS-CoV-2 was an incidental finding, thereby improving the accuracy of mortality statistics.

International research has highlighted recurring histopathological features, such as diffuse alveolar damage, micro-thrombosis, and vascular inflammation [8,9,10]. Nevertheless, regional differences in autopsy findings suggest variability in clinical management, viral variants, and population health profiles.

In addition, the introduction of vaccination campaigns has significantly modified the clinical and pathological landscape of COVID-19. Recent studies comparing vaccinated and unvaccinated individuals have demonstrated reduced severity of pulmonary lesions, lower incidence of diffuse alveolar damage, and a decreased direct contribution of SARS-CoV-2 infection to the cause of death in vaccinated patients. These findings suggest that vaccination not only reduces mortality but may also alter postmortem morphological patterns, emphasizing the importance of considering immunization status in forensic investigations.

During the early phases of the pandemic, autopsies were substantially limited in many countries due to biosafety concerns, lack of personal protective equipment, and the absence of standardized protocols for handling infectious bodies. This led to delays in the acquisition of postmortem data and hindered the early understanding of COVID-19 systemic pathogenesis. Subsequent international efforts highlighted the need for harmonized autopsy guidelines and interdisciplinary collaboration to ensure both safety and scientific rigor in forensic practice.

Furthermore, recent systematic reviews and meta-analyses have synthesized autopsy findings from multiple regions, including Europe, the United States, and Asia, providing a comprehensive overview of the histopathological features associated with SARS-CoV-2 infection. These studies have confirmed the predominance of diffuse alveolar damage, endothelial injury, and thrombotic phenomena, while also revealing regional variability influenced by clinical management, viral variants, and population characteristics. However, in our study, vaccination status was unavailable in most cases, representing an important limitation [3,4,5].

The present study was undertaken to fill a regional knowledge gap by analyzing the postmortem findings in SARS-CoV-2-positive cases investigated in a north-east territory of Romania, during the pandemic period (2020–2022). The specific objectives were to (i) describe the demographic and clinical characteristics of deceased individuals with confirmed SARS-CoV-2 infection; (ii) identify and analyze the main comorbidities, clinical diagnoses, and causes of death; (iii) correlate postmortem pathological findings with demographic and clinical variables; (iv) asses the contribution of forensic autopsies to clarifying the mechanisms of death associated with SARS-CoV-2 infection.

By addressing these objectives, the study aims to contribute to a better understanding of COVID-19 mortality patterns and to emphasize the enduring importance of forensic pathology in pandemic research. This study contributes to a regional forensic perspective from Eastern Europe, based on a Romanian cohort spanning three pandemic years. Beyond describing common pulmonary and cardiovascular lesions, it identifies notable deviations from international findings—such as the absence of fibrous thrombi in alveolar vessels and lymphomononuclear myocarditis—and examines these through statistical associations between comorbidities, clinical diagnoses, and causes of death. By combining traditional pathology with exploratory data analysis, the study adds a quantitative dimension rarely applied in forensic autopsy literature.

Although numerous studies have analyzed the pathological features of COVID-19, regional forensic data remains limited, particularly in Eastern Europe. This study provides one of the few comprehensive datasets from Romania, encompassing three consecutive pandemic years (2020–2022) and integrating statistical analyses of comorbidities and clinical–pathological correlations. By examining both typical and atypical autopsy patterns, it adds novel epidemiological and methodological insights that complement existing international data.

The present study is designed as a retrospective descriptive-analytical and exploratory investigation, aiming to characterize pathological patterns and identify associations relevant to forensic interpretation, while also contributing to hypothesis generation.

2. Materials and Methods

The study design and methodology aimed to provide robust and meaningful insights into the necroptic findings observed in Sars-CoV-2 infections, contributing valuable data to the understanding of this complex disease.

Methodology

This retrospective, unicentric study was conducted between 2020 and 2022 in Iasi, a north-east city of Romania, to investigate necroptic aspects in individuals with SARS-CoV-2 infections. The study involved a thorough data collection process followed by a comprehensive analysis to explore various factors associated with the disease. All autopsies were performed according to Romanian legal and ethical standards, under the supervision of the Institute of Legal Medicine Ethics Committee (approval no. 35992/10.12.2024), and in compliance with WHO biosafety guidelines for handling potentially infectious material. The dataset was initially prepared for exploratory data analysis (EDA), ensuring the appropriate treatment of variables and handling of missing data.

Data Collection and Preparation

Case selection criteria

A total of 279 forensic autopsies were included in the study. The following inclusion criteria were applied: confirmed SARS-CoV-2 infection determined either antemortem, by a positive RT-PCR or antigen test within 14 days before death, or postmortem by positive RT-PCR testing on lung tissue or nasopharyngeal swabs obtained during autopsy; availability of complete necropsy data, including external and internal examination and histopathological evaluation of major organs; and sufficient clinical information, such as hospital records, test results, or medical summaries. Exclusion criteria were cases with incomplete autopsy reports or missing essential clinical or laboratory data, cases with inconclusive SARS-CoV-2 testing results, and decomposed bodies where the histopathological interpretation is impaired. The extent of documentation and sampling may vary between cases, reflecting routine forensic practice and the involvement of multiple practitioners. This variability is inherent to retrospective autopsy-based studies and reflects real-world medico-legal conditions.

Regarding the timing of autopsy, we make the following statement, according to standard medico-legal practice, autopsies were performed after confirmation of irreversible signs of death, typically after a minimum interval of several hours. However, due to the forensic nature of many cases, often occurring under unclear or unwitnessed circumstances, the exact postmortem interval could not be consistently established and was therefore not included as a standardized variable.

Although detailed biosafety procedures are not part of medico-legal documentation, autopsies were conducted under controlled conditions using appropriate personal protective equipment and designated facilities, in accordance with institutional and national safety practices applicable at the time.

Data was extracted from the Medico-Legal Institute’s autopsy registry and digital medical records. The dataset included: demographic variables (age, sex, and residence); clinical variables (hospitalization status, interval between diagnosis and death, comorbidities, and clinical diagnoses); autopsy findings (macroscopic and microscopic changes in pulmonary, cardiac, hepatic, renal and cerebral tissues); and causes of death (as recorded in the forensic reports). Vaccination status was not consistently available; in approximately 85% of cases, no data regarding prior immunization or disease history were recorded. This limitation is explicitly addressed in the discussion section. Histopathological examination was performed using standard laboratory procedures, including formalin fixation, paraffin embedding, and hematoxylin–eosin staining.

The dataset was cleaned and processed before being subjected to statistical analysis [5]. This included recoding of some variables from integer to string for easing the chart readability, unifying the values of some variables (in the initial dataset, sometimes the same comorbidity, clinic diagnosis, etc., was gathered with slightly different names), and also dealing with multi-valued variables such as comorbidities, clinic diagnosis, lesion diagnosis, histological diagnosis, and death causes. Since the graphical distribution of all numeric variables seemed far from normal, the Shapiro–Wilk test of normality was applied to confirm this assumption. Consequently, non-parametric statistical methods were chosen for further analysis.

In the initial dataset, the same comorbidities, clinic diagnoses, etc., were sometimes recorded under slightly different names. Also, some variables were multi-valued, such as comorbidities, clinic diagnoses, lesion diagnoses, histological diagnoses, and causes of death. Consequently, the dataset was cleaned and processed before being subjected to statistical analysis [5]. This included recording some variables from integer to string to improve chart readability and unify their values.

Autopsy procedure

All autopsies were performed according to standard medico-legal practice, including complete external and internal examination, macroscopic organ assessment, and histopathological sampling. A consistent institutional forensic workflow was followed, including external examination, systematic opening of the cranial, thoracic, and abdominal cavities, and evaluation of major organ systems.

All autopsies included a complete external and internal examination, as well as macroscopic assessment of major organs. Representative tissue samples were systematically collected for histopathological analysis. The organs routinely examined included the lungs, heart, liver, pancreas, spleen, kidneys, and brain. At minimum, lung and heart tissue samples were analyzed from a histopathological point of view in all cases, given their central role in COVID-19 pathology as these are both primary targets of SARS-CoV-2 pathology and commonly affected by pre-existing conditions.

The variables uniformly collected from medico-legal records included demographic data (age, sex), cause of death, associated comorbidities, and histopathological findings. These variables were consistently available across all included cases and formed the basis of the comparative analysis.

However, in forensic pathology, the extent and specific details of the autopsy procedure are inherently case-dependent. The examination may be adapted according to the circumstances of death, suspected cause, and findings observed during the procedure. Therefore, while a general workflow is maintained, specific aspects such as sampling sites and examination details may vary between cases.

All histopathological examinations were performed by qualified personnel according to routine forensic pathology practice.

This approach reflects standard medico-legal practice and the practical realities of retrospective autopsy-based studies.

Medico-legal interpretative framework

Forensic interpretation of SARS-CoV-2-positive deaths was performed by integrating clinical information, circumstances of death, autopsy findings, and histopathological examination. In order to provide a structured medico-legal approach, cases were interpreted within three conceptual categories:

(i)

deaths due to COVID-19, in which the primary mechanism of death is directly related to SARS-CoV-2-induced pathology, such as diffuse alveolar damage, acute respiratory distress syndrome, or thrombo-inflammatory complications;

(ii)

deaths in which SARS-CoV-2 infection represents a contributory factor, acting in conjunction with pre-existing comorbidities and influencing the thanatogenesis process;

(iii)

deaths with incidental SARS-CoV-2 infection, in which the primary cause of death is unrelated (e.g., trauma or independent pathological processes), and the viral infection does not play a decisive causal role.

Given the retrospective design, this framework is interpretative and based on clinicopathological correlation rather than strict classification criteria.

The study period includes the early phases of the COVID-19 pandemic, during which the number of autopsies performed was initially limited due to biosafety concerns and the absence of standardized protocols. As the pandemic evolved, autopsy practices were adapted in accordance with national regulations and emerging recommendations.

The analytical workflow involved grouping cases based on key variables, including demographic characteristics (age, sex), presence of comorbidities, major pathological findings, and medico-legal cause-of-death categories. Patterns were identified through comparative evaluation of these variables, supported by exploratory statistical analysis.

Bias and limitations

Given the retrospective nature of this research, several sources of bias were recognized: selection bias, since only cases subjected to forensic autopsy and postmortem testing were included; missing vaccination data, which may influence interpretations related to disease severity and mortality; and classification bias, as all cases were SARS-CoV-2-positive, limiting differentiation between deaths from and with COVID-19.

Statistical Analysis

After the exploratory data analysis [5], the bivariate associations between categorical variables were assessed with the Chi-square test of independence [6]. Since the Shapiro–Wilk test confirmed the non-normality of the numerical variables distribution, relations between continuous and nominal variables were examined with the Kruskal–Wallis test, which is a non-parametric alternative to ANOVA [6]. Exploratory data analysis (EDA) [5] identified the main patterns in data distribution for nominal and numerical variables. The bivariate associations between categorical variables were assessed with the Chi-square test of independence [6].

Since EDA revealed that the graphical distribution of all numeric variables seemed far from normal, the Shapiro–Wilk test of normality was applied and confirmed the non-normality assumption. Therefore, relationships between continuous and nominal variables were examined using the Kruskal–Wallis test, a non-parametric alternative to ANOVA [6].

For all Chi-square and Kruskal–Wallis tests, three metrics were collected: (1) the p-value, which measures the statistical significance of the association between variables; (2) the effect size, which signals the intensity (magnitude) of the association; and (3) the confidence interval for the effect size, which signals the stability of the association between variables.

Software and Tools

All data processing and statistical analyses were carried out using the R programming language version 4.4.0 [7] and its tidyverse ecosystem for data manipulation and visualization [11]. From tidyverse we used mainly the packages stringr (to manage multi-valued variables), lubridate (to process variables of type date and/or timestamp), and dplyr (to recode variables, perform aggregations, other checks). The charts in Figure 1, Figure 2, Figure 3 and Figure 4, Figure 6, Figure 7 and Figure 10 were created using the ggplot2 package (also part of the tidyerse), whereas packages ggstatsplot [12] and effectsize [13] were employed to perform, visualize and interpret the results of the Chi-square and Kruskal–Wallis tests. Package ggstatsplot is particularly useful since it provides (Figure 5, Figure 6, Figure 8, Figure 9, Figure 11 and Figure 12) all three important metrics for test results interpretation (p-value, effect size, confidence interval).

Taken together, the methodological approach supports the descriptive-analytical and exploratory nature of the study. The analysis was based on the identification of patterns within demographic, clinical, and pathological variables, without the intention of establishing causal relationships.

3. Results

The results are presented in a descriptive-analytical framework, combining quantitative description of the dataset with inferential statistical analyses aimed to identify associations between demographic, clinical and pathological variables.

For interpretative purposes, the findings were analyzed in relation to key variables such as comorbidities, pathological patterns, and medico-legal cause-of-death categories, allowing a structured comparison of observed trends.

3.1. General Overview

Between 2020 and 2022, a total of 279 forensic autopsies were performed in the north-east of Romania, on deceased individuals with confirmed or suspected SARS-CoV-2 infection. Figure 1 illustrates the monthly distribution of COVID-19-related deaths. The first cases appeared in April 2020, with major peaks in October–November 2020 and 2021. In 2022, most deaths occurred early in the year, corresponding to the final pandemic wave in Romania.

Figure 1. Number of recorded SARS-CoV-2, by month.

Figure 1. Number of recorded SARS-CoV-2, by month.

As shown in Figure 2, half of all deaths were recorded in 2021, 36% in 2020, and 14% in 2022. Approximately two-thirds of the deceased were male, and the urban–rural distribution was balanced (52% vs. 48%). Vaccination data were unavailable for most individuals (approximately 85%), and in 75% of cases, the postmortem RT-PCR test on lung fragments confirmed SARS-CoV-2 infection.

Distribution of low-cardinality variables (from top left to bottom right: immunization status, monthly distribution of cases, results of postmortem COVID testing, sex distribution, areas of residence (rural/urban), yearly cases distribution).

Figure 2. Distribution of low-cardinality variables (from top left to bottom right: immunization status, monthly distribution of cases, results of postmortem COVID testing, sex distribution, areas of residence (rural/urban), yearly cases distribution).

Figure 2. Distribution of low-cardinality variables (from top left to bottom right: immunization status, monthly distribution of cases, results of postmortem COVID testing, sex distribution, areas of residence (rural/urban), yearly cases distribution).

3.2. Demographic Characteristics and Survival Interval

Most deaths occurred in individuals aged 40–90 years, with few cases below this range (Figure 3, left). The survival interval, defined as the number of days between the positive test and death, varied widely; however, most deaths occurred within a few days of diagnosis, with a notable peak on the same day of testing (Figure 3, right). These findings reflect the rapid clinical progression observed in severe COVID-19 cases.

Figure 3. Distribution of variables age (left) and number of days of survival after the positive COVID test (right).

Figure 3. Distribution of variables age (left) and number of days of survival after the positive COVID test (right).

3.3. Comorbidities

To improve interpretability, the results are analyzed in relation to key variables, including comorbidities, clinical presentation, and cause of death, allowing the identification of relevant patterns within the cohort.

A total of 694 comorbidities were recorded across 279 individuals, indicating multiple pre-existing conditions per case. Figure 4 shows that cardiovascular diseases were most frequent (260 cases), followed by digestive, metabolic, cerebral, psychiatric, urogenital, respiratory, and tumoral disorders.

Figure 4. Frequency of comorbidities.

Figure 4. Frequency of comorbidities.

Next, we were interested in whether the comorbidities are associated with the sex of the deceased. The Chi-square test of independence assesses the statistical significance of the association between two nominal variables. Its null hypothesis (H0) is that there is no association. Figure 5 shows that the p-value returned by the test is smaller than the accepted 0.05 threshold (0.03), so the null hypotheses can be rejected. Nevertheless, the effect size (Cramer’s V) of 0.12 is small (as qualified by the effectsize package) with a 95% confidence interval of [0, 0.17] which incorporates 0. In our dataset, sex appears to be weakly associated with comorbidities.

These findings indicate statistically significant associations between the analyzed variables, supporting the identification of relevant patterns within the cohort.

Age distribution across comorbidity categories is summarized in

Table 1. Mean and median of age, by comorbidity.

Comorbidity	No. of Occurrences	Age Mean	Age Median
urogenital	39	75	77
psychiatric	37	74	77
cerebral	40	74	78
cardiovascular	260	69	70
metabolic	60	68	70
digestive	187	65	66
respiratory	28	63	66
tumoral	15	61	56
other	28	53	59

. The largest age medians were recorded for the cerebral comorbidity (78 years), followed by urogenital and psychiatric comorbidity (77 years). Tumoral is the comorbidity with the lowest age median (56 years).

Figure 5. Results of the test of independence between sex and comorbidity.

Figure 5. Results of the test of independence between sex and comorbidity.

The Kruskal–Wallis test (Figure 6) confirmed a significant association between age and comorbidities (p < 0.001; η² = 0.07), showing that certain comorbidities, particularly urogenital, psychiatric, and cerebral conditions, were more prevalent in older individuals.

These findings indicate statistically significant associations between the analyzed variables, supporting the identification of relevant patterns within the cohort.

The effect size package qualifies the intensity of the association (as measured by eta squared) as medium.

Figure 6. Results of the test of association between age and comorbidity.

Figure 6. Results of the test of association between age and comorbidity.

3.4. Clinical Diagnoses

Clinical diagnoses recorded before death (n = 401) revealed a consistent pattern of respiratory and infectious complications. As shown in Figure 7, the most frequent diagnoses were SARS-CoV-2 infection (115 cases), acute respiratory failure (69 cases), and COVID-19 pneumonia (68 cases).

Figure 7. Frequency of clinical diagnoses.

Figure 7. Frequency of clinical diagnoses.

When analyzed by sex, the Chi-square test (p = 1.0) indicated no association between clinical diagnosis and sex (Figure 8).

Table 2. Mean and median of age, by clinical diagnosis.

Clinical Diagnosis	No. of Occurrences	Age Mean	Age Median
other diagnostic	34	75	78
COVID-19 pneumonia	68	69	71
Sars-CoV-2 infection	115	67	71
acute respiratory failure	69	67	71
no hospitalization	101	65	65
bronchopneumonia	11	64	69
pulmonary embolism	3	52	56

summarizes the relationship between clinical diagnosis and age. Although age varied slightly among diagnostic categories, the Kruskal–Wallis test (Figure 9) revealed only a small effect size (η² = 0.05), suggesting that age did not significantly influence diagnostic category distribution. Notably, acute respiratory failure and bronchopneumonia were also recorded in a few young adults under 30 years of age.

Figure 8. Results of the test of independence between sex and clinical diagnosis.

Figure 8. Results of the test of independence between sex and clinical diagnosis.

Kruskall–Wallis test results (Figure 9) show a small effect size (0.05) in the differences among all diagnosis groups, and the only significant difference appears between two generic (and, in this case, irrelevant) groups, no hospitalization vs. other diagnostic.

Figure 9. Results of the test of association between age and clinical diagnosis.

Figure 9. Results of the test of association between age and clinical diagnosis.

3.5. Causes of Death

A total of 734 pathologies involved in thanatogenesis were identified across the 279 autopsy cases. Since all cases involved individuals who tested positive for SARS-CoV-2 either antemortem or postmortem, the majority of deaths were directly or indirectly linked to COVID-19-related complications (Figure 10).

Figure 10. Frequency of death causes.

Figure 10. Frequency of death causes.

Since this study was based on identifying deceased individuals who had a prior or postmortem SARS-CoV-2-positive diagnosis, almost all causes of death also referred to SARS-CoV2 infection, whether it was the primary or associated cause of death, as seen in Figure 10. We note that bronchopneumonia/pneumonia was found in many cases in association with associated organic diseases that also played a role in causing death. Almost 1/3 of deaths were also due to ARDS. In some cases, with a positive COVID test, but without lung changes, the other pathologies or trauma that the person presented, by their severity, caused death. We also note that thrombotic organ damage with consecutive infarctions is also found among the causes of death.

As we previously mentioned, we can state that the leading causes were COVID-19 pneumonia, acute respiratory distress syndrome (ARDS), and bronchopneumonia, often associated with comorbid conditions. Thrombotic events, such as pulmonary embolism and organ infarctions, were also noted in several cases. In a minority of individuals, trauma or unrelated diseases were determined to be the primary cause of death, despite positive COVID-19 status.

For the subsequent tests of association between death cause and sex, and the death case and age, we recoded the variable death cause, placing all its values with smaller frequency (less than 10 occurrences) into a general (other) category.

In Figure 11 the distribution of death cause appears to be similar for males and females. Except for traumatic group, for which the percentage of females is 25%, for all other death causes, the female percentage varied between 33% and 46%. With a p-value of 0.69 (much larger than the 0.05 significance threshold), the Chi-square test of independence supports the idea that death cause is not associated with the sex.

Figure 11. Results of the test of independence between sex and death cause.

Figure 11. Results of the test of independence between sex and death cause.

Next, we were interested in whether the causes of death are associated with the deceased’s age.

Table 3. Mean and median of age, by causes of death.

Cause of Death	No. of Occurrences	Age Mean	Age Median
pulmonary infarction	1	76.00	76.00
traumatic	36	73.11	74.00
myocardial infarction	7	69.29	67.00
bronchopneumonia/pneumonia	197	67.24	70.00
SARS-CoV-2 infection	205	66.75	69.00
acute respiratory distress syndrome	77	66.58	69.00
other associated pathologies	151	66.19	68.00
other pathological death causes	40	64.98	64.50
pulmonary embolism	3	64.67	66.00
stroke	13	63.77	65.00
acute hemoragic pancreatite	4	60.50	54.50

shows both age mean and median for each group of the death cause. The highest value for the average and median age was recorded for the traumatic group (since pulmonary infarction has only one observation, it cannot be interpreted), whereas the lowest mean and median were recorded for the stoke group. The mean age was distributed within the [60–73] range, whereas the median age varied between 54 and 74 years.

The lack of significance of the small differences among death cause groups is confirmed by the Kruskall–Wallis test results (Figure 12).

Figure 12. Results of the test of association between death cause and age.

Figure 12. Results of the test of association between death cause and age.

The difference in median age in Table 3 and Figure 12 appears since, for the association test, the death cause groups were recoded by merging the groups with cardinality lower than 10.

Taken together, these results highlight consistent patterns linking comorbidities, clinical presentation, and causes of death; these results are being supported by inferential statistical analyses, which allow the identification of significant associations between the analyzed variables.

3.6. Histopathological Morphology and Correlation with Literature Findings

To contextualize the autopsy data, the macroscopic and microscopic changes identified in this study were compared with published international findings (

Table 4. Comparative overview of pulmonary findings in the present study and in previously reported autopsy series on SARS-CoV-2 infection.

	Morphological Appearance	Institute of Legal Medicine Iasi Cases	Reported in International Autopsy Studies (Maiese et al., 2021 [3], Carsana et al., 2020 [14])
Macroscopic	Pleurisy Lobar pneumonia Pulmonary edema Bronchopneumonia	+ +++ +++ +++	++ + +++
Microscopic	Hyaline membranes Interstitial inflammatory lymphoplasmacytic infiltrate Alveolar eosinophilic edema Alveolar fibrin exudate Alveolar hemorrhages ARDS Fibrinous thrombus in alveolar vessels	++++ +++ ++ ++ ++ ++ x	++++ + + + +++ + ++

Note: Semi-quantitative grading: + (rare), ++ (moderate), +++ (frequent), ++++ (very frequent), x (absent). Data from literature are adapted from Maiese et al. (2021) [3] and Carsana et al. (2020) [14] and are presented for comparative purposes.

and

Table 5. Comparative overview of cardiovascular findings in the present study and in previously reported autopsy series on SARS-CoV-2 infection.

	Morphological Appearance	Institute of Legal Medicine Iasi Cases	Reported in International Autopsy Studies (Maiese et al., 2021 [3], Carsana et al., 2020 [14])
Macroscopic	Cardiomegaly Cardiac lipomatosis Cardiac hypertrophy Myocardial fibrosis	++ + ++ ++	++ ++
Microscopic	Coronary atherosclerosis Acute myocarditis Intramyocardial interstitial fibrosis Intimate hyperplasia Hypertrophic cardiomyocytes Coronary thrombosis Focal necrosis of cardiomyocytes Lymphomononuclear myocarditis	++++ ++ + + +++ ++ +++ x	+++ ++ ++ ++ + ++

Note: Semi-quantitative grading: + (rare), ++ (moderate), +++ (frequent), ++++ (very frequent), x (absent). Data from literature are adapted from Maiese et al. (2021) [3] and Carsana et al. (2020) [14] and are presented for comparative purposes.

). The pulmonary lesions most frequently observed included pleurisy, lobar pneumonia, pulmonary edema, and bronchopneumonia, all consistent with the literature [8,9,10,14]. Microscopically, diffuse alveolar damage (DAD), characterized by hyaline membranes, alveolar edema, and interstitial lymphoplasmacytic infiltration, was a constant finding (Figure 13 and Figure 14). Other microscopic findings occasionally included small vascular microthrombi within the alveolar capillaries, whereas organized fibrinous thrombi in alveolar vessels were notably absent in the Iasi series.

Figure 13. Lung (autopsy): diffuse intra-alveolar damage with hyaline membranes, intravascular fibrin thrombi, intra-alveolar proliferation of fibroblasts, type II pneumocyte hyperplasia (HE stain × 100).

Figure 13. Lung (autopsy): diffuse intra-alveolar damage with hyaline membranes, intravascular fibrin thrombi, intra-alveolar proliferation of fibroblasts, type II pneumocyte hyperplasia (HE stain × 100).

Figure 14. Lung—diffuse alveolar damage with hyaline membranes, organizing phase, intra-alveolar fibroblastic proliferation, enlarged pneumocytes with large nuclei, macrophage infiltration (HE stain × 100).

Figure 14. Lung—diffuse alveolar damage with hyaline membranes, organizing phase, intra-alveolar fibroblastic proliferation, enlarged pneumocytes with large nuclei, macrophage infiltration (HE stain × 100).

Cardiac findings included cardiomegaly, interstitial fibrosis, and myocyte hypertrophy, as well as coronary atherosclerosis and focal myocardial necrosis. However, lymphomononuclear myocarditis, frequently reported in German and Italian autopsy series, was not identified in this cohort. These differences may reflect regional variations in patient demographics, disease course, or postmortem timing.

In addition to the findings summarized above, the microscopic examination identified changes in other organs. Hepatic samples frequently showed centrilobular congestion and micro-vesicular steatosis, while renal tissue displayed acute tubular injury. The neuropathological findings were nonspecific but included vascular congestion and perivascular inflammatory infiltration, findings similar to those described in the literature [15,16]. These microscopic features taken together support the multisystemic involvement of SARS-CoV-2 observed in prior autopsy-based investigations.

3.7. Summary of Key Findings

Most deaths occurred in elderly males with pre-existing cardiovascular and metabolic comorbidities. Respiratory failure and pneumonia were the predominant causes of death. Age showed a moderate association with comorbidity type but not with clinical diagnosis or cause of death. Sex was not significantly associated with clinical or pathological variables. Histopathological findings were aligned with international data confirming that COVID-19 mortality is mainly linked to diffuse alveolar damage, without any significant thrombotic phenomena in our study.

4. Discussion

The autopsy has always played an important role not only in the forensic field, but also in clinical practice, being characterized by its utilization for a better understanding of diseases. All the information that the forensic doctor obtains from the autopsy and from the histopathological examination helps clinicians understand in more detail the changes caused in the body by an unknown disease. The initial discouragement of performing autopsies in the early stages of the pandemic, justified by the high contagiousness of the SARS-CoV-2 virus, has also slowed down the researchers’ findings of diagnostic and treatment elements useful in COVID-19 infection. Despite initial hesitation due to biosafety concerns, autopsy-based research has been instrumental in elucidating the mechanisms of COVID-19-related death, particularly in distinguishing between deaths “from” and “with” the infection.

According to Romanian legislation, any case of violent death or suspected of being violent or when the cause of death is not known exactly is subjected to a forensic autopsy by the criminal investigators or by the court.

The forensic autopsies were performed in accordance with the Romanian legislation. Autopsy procedures in forensic practice may involve case-specific adaptations depending on the circumstances of death and suspected pathology, which should be considered when interpreting retrospective findings. After the completion of the necroptic examination, the body was closed and stored in special conditions according to the WHO recommendations regarding the degree of infectivity of the new coronavirus.

Our study included both cases in which data on SARS-CoV-2 infection were known (people with a positive COVID test or suspected infection) and those in which organ changes suggestive of a possible SARS-CoV-2 infection were found during the autopsy and from which lung fragments were collected for further testing.

In the present study, the 279 autopsies performed in a forensic institute from a north-east territory of Romania between 2020 and 2022 provided insight into the demographic, clinical, and pathological characteristics of individuals infected with SARS-CoV-2. The results confirmed several well-established trends reported internationally [8,9,14,17,18,19]: (i) most decedents were elderly males with cardiovascular or metabolic comorbidities; (ii) the primary mechanisms of death involved acute respiratory distress syndrome (ARDS), diffuse alveolar damage, and embolic events; (iii) comorbidities significantly influenced disease outcomes, while sex did not appear to alter mortality patterns.

The study highlights the critical role of forensic autopsies in advancing the understanding of SARS-CoV-2’s impact on the human body. Despite the initial reluctance to perform autopsies due to the high contagion risk, postmortem examinations provided invaluable insights into the pathophysiology of COVID-19. The macroscopic and histological findings, such as pulmonary edema and acute respiratory distress syndrome (ARDS), align with global studies conducted in regions like Germany [8], France [9] and Japan [20].

The ability of forensic examinations to distinguish between “dying with COVID-19” and “dying due to COVID-19” was particularly vital in clarifying mortality statistics. The detailed analyses also underscored the contribution of pre-existing comorbidities, such as cardiovascular and metabolic disorders, to COVID-19 mortality, echoing findings from other international studies.

From a medico-legal perspective, the distinction between “death from” and “death with” COVID-19 reflects different thanatogenesis scenarios that must be carefully interpreted. In this context, SARS-CoV-2 infection may act as a primary cause of death, a contributory factor in the presence of significant comorbidities, or an incidental finding without direct causal impact.

This structured approach is essential in forensic practice, as the presence of viral infection alone does not establish causality and must be interpreted in correlation with autopsy findings, clinical data, and the overall pathological context.

To illustrate the practical applicability of this medico-legal framework, the analyzed cases can be interpreted through simplified case-based stratification. Several cases correspond to “death with SARS-CoV-2 infection,” particularly those involving violent deaths or clearly defined external causes, where the primary mechanism of death was unrelated to viral pathology and histopathological findings were nonspecific.

A second category includes cases in which SARS-CoV-2 infection may have played a contributory role. In these situations, the coexistence of significant comorbidities, respiratory pathology, and systemic involvement suggests a complex interaction between viral infection and pre-existing disease, without allowing a definitive attribution of causality.

Finally, certain cases illustrate incidental infection, where a clearly defined pathological mechanism (e.g., acute aortic dissection or advanced chronic disease) was sufficient to explain death independently of SARS-CoV-2 infection.

Notably, in the present cohort, no cases could be unequivocally classified as “death due to COVID-19” in the absence of significant comorbidities, which further highlights the complexity of cause-of-death attribution in forensic practice.

Although this classification is illustrative rather than quantitative, it demonstrates the applicability of the medico-legal framework to real-world forensic cases.

From a pathophysiological perspective, the findings observed in this cohort support the concept that severe COVID-19 is not limited to viral pneumonia, but represents a systemic disease involving endothelial dysfunction, dysregulated inflammation, and microvascular injury. The interaction between viral replication and the host immune response may lead to a thrombo-inflammatory state, in which vascular involvement plays a central role in disease progression [21].

Diffuse alveolar damage and the presence of hyaline membranes remain the most characteristic pulmonary correlates of severe SARS-CoV-2 infection. However, these lesions are frequently associated with vascular congestion, micro-thrombotic phenomena, and extrapulmonary involvement, supporting the concept of multisystem disease [21,22].

Previous observations and literature data have highlighted the role of endothelial injury and endothelitis, suggesting direct viral involvement of vascular structures and subsequent microvascular dysfunction. These mechanisms may contribute to the development of microthrombi and organ ischemia, even in the absence of large-vessel thrombosis. However, in the present cohort, such vascular lesions were inconsistently identified and, in some cases, were absent. This variability may reflect differences in disease stage, comorbidity profiles, or methodological limitations inherent to retrospective autopsy studies [23].

Interestingly, some of our findings differ from those reported in the literature. While several studies describe interstitial pulmonary edema as a predominant feature in COVID-19-related ARDS, our cases frequently showed intra-alveolar edema, often associated with hyaline membranes. This discrepancy may be related to differences in disease evolution, timing of autopsy, or associated comorbidities, and highlights the heterogeneity of pathological manifestations in SARS-CoV-2 infection.

At the same time, forensic interpretation requires caution in distinguishing lesions more directly attributable to COVID-19 from secondary or nonspecific processes. Diffuse alveolar damage and ARDS-related changes may support a direct role of SARS-CoV-2, whereas bronchopneumonia, hypoxic injury, superinfection, or decompensation of chronic diseases may represent secondary or contributory mechanisms.

It must also be emphasized that many of the histopathological findings observed in SARS-CoV-2-positive individuals are not entirely specific and may overlap with lesions caused by hypoxia, systemic inflammation, or pre-existing disease. Therefore, the attribution of certain lesions to direct viral injury remains partly interpretative, particularly in retrospective analyses lacking molecular confirmation.

A more detailed analysis of the histopathological findings highlights several aspects that both align with and diverge from current models of COVID-19 pathophysiology. Acute respiratory distress syndrome (ARDS) is considered a terminal stage of SARS-CoV-2 infection, typically characterized by a combination of interstitial and alveolar edema. However, in contrast to reports in the literature describing the predominance of interstitial edema, the present study frequently identified intra-alveolar edema associated with abundant hyaline membrane formation. This discrepancy suggests potential differences related to disease evolution, comorbidities, or postmortem factors and warrants further investigation [22,23,24].

Current pathogenetic models emphasize endothelial dysfunction and thrombo-inflammatory processes leading to pulmonary vasculopathy and microvascular thrombosis. Although such mechanisms are widely described, endothelial alterations and extensive microvascular thrombi were inconsistently observed in the present cohort, which may reflect variability in disease stage, individual inflammatory response, or methodological limitations inherent to retrospective autopsy studies.

The inflammatory pattern was heterogeneous, with interstitial lymphoplasmacytic infiltrates frequently associated with polymorphonuclear components, suggesting superimposed bacterial infection in some cases. Intra-alveolar fibrinous exudates with varying degrees of organization were also observed, particularly in cases with prolonged clinical evolution or pre-existing pulmonary pathology.

Extrapulmonary findings further support the systemic nature of the disease, although most alterations were nonspecific and strongly influenced by pre-existing conditions. Cardiovascular changes were predominantly consistent with chronic pathology, while only limited cases showed features suggestive of possible viral myocarditis. Renal, hepatic, and cerebral findings were largely attributable to underlying diseases or postmortem changes, complicating the interpretation of direct viral involvement [14].

Overall, these findings underscore the heterogeneity of SARS-CoV-2-related pathology and highlight the difficulty of distinguishing between primary viral injury and secondary or comorbidity-related processes. In many cases, the histopathological changes are nonspecific, and their attribution to COVID-19 remains partly interpretative, particularly in the context of multiple associated pathologies.

Additional histopathological features further illustrate the complexity of disease progression. In several cases, intra-alveolar fibrinous exudates with varying degrees of organization were observed, in some instances progressing toward connective tissue proliferation and partial “carnification,” particularly in cases with prolonged clinical evolution or pre-existing pulmonary disease. Multinucleated cells with vacuolated cytoplasm were also identified in selected cases, suggesting advanced stages of cellular injury.

Table 4 highlights both similarities and discrepancies between our findings and those reported in international autopsy studies, particularly regarding the frequency of diffuse alveolar damage and the lower detection of microvascular thrombi in our cohort [3,14].

Table 5 illustrates that while chronic cardiovascular changes are comparable to those described in the literature, acute inflammatory lesions such as myocarditis were less frequent or absent in our cohort, suggesting possible regional or methodological variability [3,14].

Regarding pulmonary pathology, our findings corroborate data from another study conducted by Suran M [21], which emphasizes diffuse alveolar damage as hallmarks of COVID-19 deaths. A review by Maiese et al. from 2021 [3] further highlights ARDS and hyaline membranes as common autopsy findings, mirroring observations in this cohort. However, fibrinous thrombi, frequently reported in alveolar vessels by studies in Germany and Italy, were absent in the Iasi cases, indicating possible differences in detection protocols or postmortem timing. The absence of these thrombi and of lymphomononuclear myocarditis in our series, compared with other forementioned reports, may stem from several factors. First, autopsy timing (often within 24–48 h after death) might have influenced the detectability of fibrin organization. Second, regional differences in viral variant predominance and comorbidity structure (e.g., higher prevalence of chronic cardiovascular disease) could alter the inflammatory response. Lastly, limited access to advanced histochemical and molecular tests may have reduced sensitivity for subtle vascular or myocardial lesions.

The detected thrombotic events included pulmonary embolism and small non-fibrinous vascular microthrombi, and the two pathological entities were significant contributors to mortality [3].

Cardiovascular comorbidities were prevalent but did not consistently predict survival times, contrasting with studies showing direct myocardial injury and myocarditis in many COVID-19 autopsies. This study reveals that SARS-CoV-2 deaths were primarily associated with respiratory complications, with bronchopneumonia and ARDS as predominant findings. Cardiovascular, metabolic, and digestive comorbidities were prevalent, while no significant statistical relationship was found between comorbidities and survival time. These observations align with the systemic nature of COVID-19 pathogenesis and its interaction with pre-existing conditions.

Histopathological changes observed in the cases presented in this study align with those described in the literature, particularly highlighting the presence of hyaline membranes indicative of acute respiratory distress syndrome (ARDS) caused by COVID-19. However, in the analyzed cases, intra-alveolar edema was exclusively noted, with interstitial edema being absent, a finding that demands further investigation for clarification.

Microthrombi were observed in myocardial vessels, alongside cardiovascular changes such as necrosis and cardiomyocyte hypertrophy. Nevertheless, it remains uncertain whether these alterations predated SARS-CoV-2 infection. Given the overlap between SARS-CoV-2-related lesions and pre-existing pathological conditions, causal attribution must be approached with caution. In this context, the relationship between infection and death is often probabilistic rather than definitive, and should not be interpreted as direct causation in the absence of clear supporting evidence. Additionally, the histopathological changes observed are predominantly nonspecific, making it impossible to definitively conclude that COVID-19 was the primary cause of death in the studied cases. It is also uncertain whether the individuals, given their pre-existing comorbidities, would not have succumbed in the absence of SARS-CoV-2 infection [25,26].

The histopathological findings observed in this study highlight the complexity of COVID-19-related mortality and the difficulty of distinguishing between deaths “from” and “with” SARS-CoV-2 infection, particularly in the presence of significant comorbidities.

In this context, it is also important to distinguish between primary viral injury and secondary or nonspecific changes. Diffuse alveolar damage and ARDS-related lesions may support a direct role of SARS-CoV-2, whereas findings such as bronchopneumonia, hypoxic injury, or decompensation of chronic diseases may reflect secondary processes or comorbid conditions. Additionally, certain postmortem or agonal changes may further complicate interpretation, particularly in retrospective analyses.

Diffuse alveolar damage, hyaline membrane formation, and ARDS-related changes may support a primary role of viral infection in the mechanism of death. In contrast, findings such as bronchopneumonia, hypoxic visceral injury, thromboembolic events, or decompensation of chronic cardiovascular and metabolic diseases are more likely to reflect secondary or contributory processes. Where clinical information was available, these autopsy findings showed correspondence with clinical features, such as respiratory failure in cases with diffuse alveolar damage and vascular complications in cases with thrombotic findings.

In addition, certain histopathological changes may represent agonal phenomena or postmortem alterations, further complicating interpretation. Consequently, the attribution of specific lesions to direct viral injury must be approached with caution and requires integration of pathological findings with clinical and circumstantial data.

Recent studies emphasize that endothelial dysfunction and thrombo-inflammatory mechanisms play a central role in COVID-19 pathogenesis, contributing to multisystem involvement and organ injury [23]. Vascular alterations, including microthrombi and endothelial injury, have been consistently described in autopsy studies and are considered central features of COVID-19 pathology [27]. Emerging evidence also indicates that SARS-CoV-2 infection may lead to persistent dysfunction and multi-organ alterations, further complicating the interpretation of autopsy findings and causal relationship in forensic context [28].

While the major morphological patterns identified in this study are broadly consistent with those reported in the literature, several differences should be noted. In particular, the absence of fibrinous alveolar thrombi and lymphomononuclear myocarditis contrasts with findings reported in other European studies, suggesting potential regional, methodological, or disease-stage-related variability.

Recent studies have further supported the concept that COVID-19 is a systemic disease with multi-organ involvement, extending beyond the acute phase and potentially contributing to long-term pathological sequelae. This complexity reinforces the difficulty of establishing a direct causal relationship between SARS-CoV-2 infection and death in forensic situations [14,21,24,28]. Current evidence also emphasizes the central role of thrombo-inflammatory mechanisms and endothelial dysfunction in disease progression, although the extent and distribution of vascular lesions may vary significantly between cases and studies. At the same time, increasing evidence of methodological heterogeneity and biological variability across autopsy studies suggests that no single pattern can be considerably universally representative of SARS-CoV-2 infection [26,27,28,29].

Taken together, these findings are consistent with the broader forensic literature, which emphasizes the complexity of cause-of-death attribution in infectious diseases with multisystem involvement. Several autopsy-based studies have highlighted the difficulty of distinguishing between primary viral injury, secondary complications, and the contribution of pre-existing conditions, supporting a probabilistic rather than deterministic interpretation of SARS-CoV-2-related mortality.

Although the statistical analysis performed in this study is exploratory and does not establish relationships, it provides additional support for the observed associations between comorbidities, clinical features, and cause of death, reinforcing the descriptive-analytical nature of the study.

The present findings further emphasize that SARS-CoV-2-related mortality frequently occurs in the context of significant comorbidities, which complicates causal attribution. The predominance of nonspecific histopathological changes and the variability of organ involvement highlight the need for an integrative diagnostic approach.

Overall, this study reinforces the value of forensic autopsy in understanding SARS-CoV-2-related mortality, not only by establishing cause of death, but also by contributing to epidemiological accuracy and pathophysiological insight. The findings highlight the complexity and heterogeneity of COVID-19-related pathology, particularly in the presence of significant comorbidities.

At the same time, the study illustrates the inherent limitations of retrospective autopsy-based research, including variability in available data and the difficulty of establishing direct causal relationships. These results should therefore be interpreted within a descriptive-analytical and exploratory framework.

Future research integrating molecular, immunological, and clinical data, as well as vaccination status, would further refine the understanding of SARS-CoV-2-related mortality and improve the standardization of forensic diagnostic approaches.

The findings should be interpreted within the limitations of a retrospective study with an exploratory design and should not be used to infer definitive causal relationships.

Study limitations

Some aspects of the study reflect intrinsic limitations of retrospective forensic research, such as variability in autopsy practices and incomplete clinical data, which cannot be fully standardized. Other aspects, including methodological transparency and interpretative framework, have been addressed and improved in the present revision.

Several limitations must be acknowledged to correctly interpret these findings:

-

selection bias: only cases subjected to forensic autopsy and postmortem SARS-CoV-2 testing were included; this may not fully represent the broader population of COVID-19-related deaths, particularly hospital-based cases without forensic autopsy;

-

incomplete clinical and vaccination data: the absence of consistent vaccination records limited our ability to evaluate the impact of immunization on disease severity, mortality patterns, and histopathological findings; future studies incorporating vaccination status are essential to refine the distinction between deaths “from” and “with” COVID-19;

-

lack of molecular and immunohistochemical correlation: the absence of molecular assays limits the ability to understand whether observed lesions were directly caused by viral replication or by secondary inflammatory responses;

-

the lack of complete clinical data in some cases limited the extent of clinicopathological correlation.

Despite these limitations, the study provides valuable regional data and reinforces the critical role of autopsies in understanding COVID-19 pathophysiology. The weak statistical associations observed between sex, comorbidities, and cause of death likely reflect the homogeneity of vulnerability within the affected population. Most deceased individuals were elderly, with multiple chronic diseases that independently increase mortality risk, regardless of sex. The uniform severity of infection among advanced-age patients may have masked subtler demographic differences. Further stratified analyses, ideally incorporating vaccination and prior infection status, would help refine these associations.

5. Conclusions

This study supports the continued relevance of forensic autopsy as a key tool for understanding SARS-CoV-2-related mortality. The findings indicate that most deceased individuals had significant pre-existing comorbidities, and that the main pathological patterns, including diffuse alveolar damage, hyaline membranes, and intra-alveolar edema, are consistent with those reported in the literature.

At the same time, the predominantly nonspecific nature of many histopathological changes highlights the difficulty of establishing a direct causal relationship between SARS-CoV-2 infection and death, particularly in complex cases with multiple associated conditions. The absence of certain lesions frequently reported in other studies, such as fibrinous thrombi or myocarditis, further suggests potential variability related to regional, methodological, or disease-stage factors.

From a medico-legal perspective, the interpretation of SARS-CoV-2-positive deaths requires a structured and integrative approach, combining autopsy findings with clinical history, comorbidities, and circumstantial data. The distinction between deaths “due to” COVID-19 and those “with” SARS-CoV-2 infection remains a central challenge and cannot be based solely on viral detection.

These findings should be interpreted within the limitations of a retrospective, descriptive-analytical, and exploratory study. The results are primarily hypothesis-generating and do not establish definitive causal relationships. Future research integrating molecular, immunohistochemical, and clinical data, as well as vaccination status, is necessary to further clarify the mechanisms of SARS-CoV-2-related mortality and to improve the accuracy of forensic diagnosis.