Context.—

The first case of COVID-19 in the United States was confirmed in January 2020. Initially, little was known about the epidemiology and clinical course of the disease, and diagnostic testing was limited in the United States until March/April 2020. Since then, many studies have speculated that SARS-CoV-2 may have preexisted undiagnosed outside China before the known outbreak.

Objective.—

To evaluate the prevalence of SARS-CoV-2 in adult autopsy cases performed just before and during the beginning of the pandemic at our institution, where autopsy was not performed on known COVID-19 cases.

Design.—

We included adult autopsies performed in our institution from June 1, 2019, to June 30, 2020. Cases were divided into groups based on the likelihood of cause of death being related to COVID-19, presence of a clinical respiratory illness, and histologic findings of pneumonia. Archived formalin-fixed, paraffin-embedded lung tissue of all COVID-possible cases and COVID-unlikely cases with pneumonia was tested for SARS-CoV-2 RNA, using Centers for Disease Control and Prevention 2019-nCoV quantitative real-time reverse transcription–polymerase chain reaction (qRT-PCR).

Results.—

Eighty-eight cases were identified, and among those, 42 (48%) were considered COVID-possible cause of death, with 24 of those 42 cases (57%) showing respiratory illness and/or pneumonia. COVID-19 as cause of death was considered unlikely in 46 of 88 cases (52%), with 34 of those 46 cases (74%) showing no respiratory illness or pneumonia. SARS-CoV-2 real-time reverse transcription–polymerase chain reaction was performed on a total of 49 cases, 42 COVID-possible and 7 COVID-unlikely with pneumonia, and all cases were negative (0 of 49).

Conclusions.—

Our data suggest that autopsied patients in our community who died between June 1, 2019, and June 30, 2020, without known COVID-19 were unlikely to have had subclinical and/or undiagnosed COVID-19 infection.

In December 2019, news began to spread of a deadly respiratory illness in Wuhan, China, which was reportedly caused by a novel coronavirus of the betacoronavirus group, the same group as the 2003 severe acute respiratory syndrome coronavirus. The virus was named SARS-CoV-2, and the disease, COVID-19, spread rapidly. In the United States, the first case was confirmed1  on January 20, 2020. SARS-CoV-2 continued to spread, and a pandemic was declared by the World Health Organization (WHO)1,2  on March 11, 2020.

Hospitals around the United States began to see patients dying of COVID-19, and pathologists encountered these patients at autopsy. The majority of the patients who died with COVID-19 had severe respiratory illness, but the extent of disease and other organ involvement by viral infection was not known early in the pandemic. As diagnostic molecular tests were developed to identify SARS-CoV-2 in nasopharyngeal swabs, the diagnosis could be confirmed, and pathologists could be informed about the presence or absence of COVID-19 in the patients undergoing autopsy.

Beginning in March 2020, our autopsy service performed autopsies only on patients who tested negative for SARS-CoV-2 via quantitative real-time reverse transcription–polymerase chain reaction (qRT-PCR) from premortem and/or postmortem preautopsy nasopharyngeal swab. Although we relied on this diagnostic test to inform our autopsy procedure, many autopsied patients died with severe respiratory illnesses, and we hypothesized that some patients may have had undiagnosed SARS-CoV-2 infection prior to our testing protocol or had false-negative results. The goal of the study was to evaluate the presence (or absence) of SARS-CoV-2 infection in an adult autopsy population dying without known COVID-19 in the time period leading up to and during the beginning of the known SARS-CoV-2 outbreak.

Patient Selection and Autopsy Review

This was a retrospective cohort study of autopsied patients at a single institution where autopsies are performed for in-house patients and patients from the outlying regional hospitals. We searched our pathology database for autopsies of adults (≥18 years old) performed at our institution from June 1, 2019, to June 30, 2020, including cases both before and after the WHO pandemic declaration. Autopsied patients were included if they had undergone full unrestricted autopsies or in cases where the autopsy was restricted but included at least the heart and lungs. Using the autopsy reports and clinical histories available, we extracted data regarding the presence of a clinical respiratory illness, pulmonary and cardiac (cardiac tamponade, congestive heart failure, myocardial infarction and associated complications) pathology, other organ pathology, and the cause of death (COD). After review of each case, we assigned the COD as possibly related to COVID-19 (COVID-possible) if acute bronchopneumonia, diffuse alveolar damage, and/or multisystem organ failure without definitive explanation were present. Cases with COD of pulmonary embolism (PE) with concomitant pneumonia or PE without known risk factors for thromboembolism were also categorized as COVID-possible. Cases with myocarditis as COD, which has been reported in viral infections including COVID-19,3  were also included as COVID-possible.

We assigned the COD as unlikely related to COVID-19 (COVID-unlikely) when a clear COD other than a respiratory illness, such as acute myocardial infarction, was present or when a well-characterized non–COVID-19 respiratory illness was present. Cases with respiratory COD such as PE that occurred secondary to other nonrespiratory diseases and known risk factors were also categorized as COVID-unlikely.

Both COVID-possible and COVID-unlikely categories were divided into 2 each, based on the presence or absence of clinical respiratory illness, and then the 4 groups were further subgrouped into 2 each, based on the presence or absence of histologic pneumonia. There were 8 groups in total (Figure 1): groups 1 through 4 were subgroups of the COVID-possible category, whereas groups 5 through 8 were subgroups of the COVID-unlikely category. Cases in groups 1 and 5 had both clinical respiratory illness and histologic pneumonia; groups 2 and 6 consisted of cases with clinical respiratory illness but no histologic pneumonia; cases in groups 3 and 7 lacked clinical respiratory illness but had histologic pneumonia; and cases in groups 4 and 8 had neither clinical respiratory illness nor histologic pneumonia.

Figure 1

Categorization of cases based of selection criteria. Asterisk (*) represents groups selected for polymerase chain reaction testing. Abbreviation: COD, cause of death.

Figure 1

Categorization of cases based of selection criteria. Asterisk (*) represents groups selected for polymerase chain reaction testing. Abbreviation: COD, cause of death.

Close modal

Nucleic Acid Extraction and Viral Testing

Lung tissue in autopsy cases is routinely sampled at our institution. Before the onset of the pandemic, lungs were typically examined fresh and samples were also taken fresh. However, if a case's circumstances warranted it, the intact lungs with attached bronchi and trachea were infused with formalin and then submerged in formalin for fixation prior to examination and sampling. At the onset of the pandemic, all lungs were fixed using this method for 24 to 48 hours prior to examination and sampling. A single random section was taken from all lobes in all cases and additional sections were taken from gross lesions, where identified.

After the cases were categorized as COVID-possible and COVID-unlikely, cases from COVID-possible groups (1–4) as well as COVID-unlikely groups with histologic pneumonia (groups 5 and 7) were selected for testing by PCR. Glass slides were reviewed and a representative block from lung lesional tissue of groups 1, 3, 5, and 7, which had histologic pneumonia, was selected for PCR. For cases of groups 2 and 4, which had no histologic pneumonia, a random block was selected. Selected blocks for testing were retrieved from archived formalin-fixed, paraffin-embedded (FFPE) tissue.

From each tissue block, eight 10-μM sections were cut and placed into an RNase/DNase-free tube. Total RNA was then extracted using an FFPE RNA Purification Kit (Norgen Biotek, Thorold, Ontario, Canada) as instructed by the manufacturer and eluted in a 50-μL elution buffer. After RNA was obtained, it was measured using the NanoDrop ND1000 spectrophotometer. Samples were stored at −80°C until ready for analysis. The FFPE samples were tested with the 2019-nCoV_N1 Combined Prime and Human RNase P (RP) Combined Primer for an extraction control (Integrated DNA Technologies, Coralville, Iowa). SARS-CoV-2 qRT-PCR was performed using TaqPath 1-Step RT-qPCR Master Mix (Thermo Fisher, Waltham, Massachusetts). Reagents and samples were placed on wet ice to slowly thaw samples on the day of qRT-PCR. Following the instructions as stated in the Centers for Disease Control and Prevention 2019-Novel Coronavirus (2019-nCoV) Real-Time RT-PCR Diagnostic Panel for SARS-CoV-2 N1 and RP internal control, the samples were run on the LightCycler 480 System (Roche Diagnostics Corporation, Indianapolis, Indiana) as a laboratory-developed procedure. Molecular-grade water was used as a no-template control, and an extracted sample from the Illinois Department of Public Health was used as the positive control. Results of the qRT-PCR were analyzed using the LightCycler 480 System Analysis program, with sample resulted positive if the Ct was less than or equal to 38 for both N1 and RP signal and not detected if the Ct was less than or equal to 38 for the RP signal but not detected for the N1 reaction (as long as both the positive control and negative control for the run were acceptable). If the RP was Ct less than or equal to 38 and the N1 was Ct greater than 38, the sample was resulted as indeterminate, and if the RP signal was not detected, the result for the sample was invalid.

Eighty-eight adult autopsies were performed during the study period, and all 88 cases, including 44 men and 44 women with an average age of 66 years, met the inclusion criteria. Forty-five cases were in-house cases and 43 were consult cases from regional hospitals; 65 of 88 patients (74%) died before the WHO pandemic declaration in March 2020 (Figure 2). The postmortem interval ranged from 0 to 9 days, with a median and mode of 2 days and a mean of 2.36 days.

Figure 2

Number of cases in autopsy categories per month. Abbreviations: COD, cause of death; PCR, polymerase chain reaction.

Figure 2

Number of cases in autopsy categories per month. Abbreviations: COD, cause of death; PCR, polymerase chain reaction.

Close modal

The distribution of COD for all the study cases is shown in the Table. Cardiac and respiratory illnesses were the most common CODs, together accounting for 45 of 88 of the CODs (51.1%). We considered the COD COVID-possible in 42 of 88 cases (48%), with more than half of those cases (24 of 42; 57%) showing a clinical respiratory illness and/or acute pneumonia. Of these 42 COVID-possible cases, 8 (19%) were in group 1, 7 (16.7%) in group 2, 9 (21.4%) in group 3, and 18 (42.9%) in group 4. COVID-19 as COD was considered unlikely in the remaining 46 of 88 (52%), with 34 of 46 (74%) showing neither respiratory illness nor acute pneumonia. However, histologic pneumonia was seen in 7 of the COVID-unlikely cases. Also, of the 46 COVID-unlikely cases, there were 4 group 5 cases (8.7%), 5 group 6 cases (10.9%), 3 group 7 cases (6.5%), and 34 group 8 cases (73.9%).

Frequency of Categories of Causes of Death

Frequency of Categories of Causes of Death
Frequency of Categories of Causes of Death

SARS-CoV-2 qRT-PCR was performed in a total of 49 cases: 42 COVID-possible and 7 COVID-unlikely with histologic pneumonia (Figure 1). Thirteen of the 49 cases tested involved patients who died after the WHO pandemic declaration, whereas 36 died before the pandemic declaration. Seven of these 36 patients died between the first confirmed US case on January 20, 2020, and the pandemic declaration on March 11, 2020. SARS-CoV-2 qRT-PCR was negative in all 49 cases, with appropriate positive control and RNA integrity in all cases.

Our data from an adult autopsy cohort between June 1, 2019, and June 30, 2020, revealed that SARS-CoV-2 was not detected in FFPE lung tissue from any autopsy cases, including those with undefined respiratory illness and/or histologic pneumonia. Furthermore, none of the autopsy cases had a clinical diagnosis of COVID-19, and our data suggested that patients in our community who died between June 1, 2019, and June 30, 2020, without known COVID-19 disease were unlikely to have had a subclinical and/or undiagnosed SARS-CoV-2 infection. Therefore, clinical examination and accurate laboratory testing likely identified most cases of COVID-19.

In the early stages of the pandemic, with the increase in reported cases and limited availability of widespread diagnostic testing and possible treatment strategies, countries went into lockdown to control the spread of the disease. Also, there was a growing concern of sick patients dying without an established clinical diagnosis of SARS-CoV-2 infection. Various studies have demonstrated that the death count was underreported for certain reasons, such as the rapidly growing infection rate and limited hospital capacity and/or resources to definitively diagnose and treat the patients.47  The Department of Public Health in Brazil demonstrated underreporting of COVID-19–related deaths by 22.62%.3  In another study5  published in March 2021, similar interpretations were made regarding deaths worldwide caused directly by SARS-CoV-2.

Given these concerns, our hospital system suspended performing autopsies on patients with suspected or confirmed COVID-19 to minimize exposure and the potential spread of the infection. Interestingly, this study on our autopsy cohort, which includes cases from our hospital located north of Chicago and consult cases from regional hospitals, performed shortly before and during the early pandemic revealed that no death was attributed to undiagnosed SARS-CoV-2 infection. Literature on this issue with similar autopsy-based PCR studies has mixed findings. A study performed in Switzerland8  retrospectively tested the autopsy cases for the presence of SARS-CoV-2 using qRT-PCR and also did not find the virus. However, that study tested only 11 prepandemic autopsy cases using qRT-PCR, whereas the data set in the present study is 4 times larger.

Another autopsy-based study from Switzerland7  revealed a 16% higher SARS-CoV-2 infection rate and an 8% higher SARS-CoV-2–related mortality rate than premortem rates reported by clinicians during the first and second waves of COVID-19. However, this study did not include prepandemic autopsy cases. A similar study from north Italy showed the presence of sporadic cases of COVID-19 in the region before the region's first confirmed COVID-19 case when researchers simultaneously tested blood and lung tissue (subject to availability) to identify SARS-CoV-2.9  These autopsy-based studies also used FFPE lung tissue for qRT-PCR testing. The present study used the largest data set to perform SARS-CoV-2 qRT-PCR testing on patients who died of clinical causes other than COVID-19. Currently, to our knowledge, there have been no studies reporting variable sensitivity of SARS-CoV-2 qRT-PCR in FFPE tissue or its potential false-negative rate compared with detection in fresh samples. However, the utility of FFPE tissue has been validated for other viruses.10  Sekulic et al11  in a recent analysis were able to detect viral load in FFPE tissue, and they reported the highest SARS-CoV-2 RNA levels in lung tissue. We also used the FFPE tissue from the lungs to perform qRT-PCR testing to detect the viral RNA of SARS-CoV-2.

To our knowledge, this is the first and only study to date in the United States that examined the results of autopsies performed before the earliest reported COVID-19 case in January 2020. We believe that our study has unique findings, as US cases were discovered later than European cases, and very quickly the daily report of newly infected cases in the United States surpassed the number in other countries. Hence, there was a high suspicion among peers that the disease may already have been present but remained undiscovered because of the unavailability of widespread testing in the United States.

This single-center autopsy study further illustrates that there was likely limited fatal spread of undiagnosed COVID-19 in the regional Chicago population served by our hospital system before the initial report of the first US case in January 2020. The clinical symptomatology and organ-specific histologic findings attributable to SARS-CoV-2 infection are still debatable in the literature.12  Although infection was known to spread via the respiratory route, organ involvement was not limited to the lungs, and some patients had renal failure and thromboembolic events.1315  Given the lack of established clinical and/or histologic criteria that could be diagnostic of COVID-19 infection at the time of the study, we recruited all the adult autopsy cases in our system and categorized them as COVID-possible or COVID-unlikely according to their CODs, as explained in the Methods section. Cases with COD known to be closely related to clinicopathologic findings reported in COVID-19 disease, such as bronchopneumonia, multisystem organ failure, and PE with no anatomical or clinical predisposition to hypercoagulability, were categorized as COD-possible. There were 6 cases identified to have thromboembolic events, but they were considered COVID-unlikely given the presence of identifiable anatomic/clinical causes of thromboembolism other than COVID-19, such as widespread metastatic tumors, and PCR testing was not done on these cases.

In our autopsy cohort, lung tissue from 29 patients who died between June 1, 2019 and January 20, 2020, when the first US case was confirmed, was tested for the SARS-CoV-2 virus. The negative qRT-PCR results in these cases support a lack of early undetected COVID-19 death/spread in our autopsy department patient population. Our autopsy data also highlight the accuracy of premortem PCR testing, which is the cornerstone of diagnosing COVID-19 infection via qRT-PCR detection of SARS-CoV-2 RNA. During the early days of COVID-19 spread, another dilemma was the limited availability of widespread COVID-19 qRT-PCR testing. Also, the performance of many SARS-CoV-2 qRT-PCR assays was not entirely known because of the lack of a gold standard, and institutions where the early testing was available had concerns about a low sensitivity. The false-negative rate from respiratory samples for SARS-CoV-2 reported by various studies ranges from 1% to 30%.16,17  A few identifiable causes of false-negative rates include testing too early in the disease process, low analytic sensitivity, low viral load, and inappropriate specimen type and collection.1820  Our laboratory was one of the first laboratories in the state of Illinois that went live with COVID-19 PCR testing, and we experienced similar concerns. However, in retrospect, our study supports consistency between premortem and postmortem PCR results, because no patients in our cohort with a premortem or preautopsy negative nasopharyngeal swab for SARS-CoV-2 qRT-PCR demonstrated a postmortem positive SARS-CoV-2 qRT-PCR in their lung tissue.

A few limitations do apply to our study. We studied a selective cohort that included only patients with negative premortem and/or postmortem preautopsy COVID-19 qRT-PCR after the start of the pandemic. Therefore, our results cannot be compared with other autopsy studies where patients with known premortem COVID-19 disease were studied. Also, we tested only the lung tissue for the qRT-PCR identification of SARS-CoV-2 viral RNA. However, given that COVID-19 spreads through the respiratory route and the majority of patients had respiratory illness, testing only lung tissue in our cohort is unlikely to have had a negative impact on the findings of this study.

Also, the lack of cases with positive SARS-CoV-2 qRT-PCR results in this study may raise concern about the performance of the test, but we have been able to identify SARS-CoV-2 in other FFPE tissues, such as placenta, in our laboratory, using the same protocol/methods used in this study. Furthermore, the assay included a positive internal control from the Illinois State Department of Health. Additionally, the use of postmortem FFPE tissue raises the possibility of RNA integrity issues, particularly with long postmortem intervals (such as at least one patient at 9 days in our study). However, the RNA integrity was consistently adequate in each of our autopsy cases, as indicated by the RP internal control results.

In this US-based retrospective study, we did not detect SARS-CoV-2 viral RNA, using FFPE lung tissue, in autopsy patients surrounding the time of pandemic onset. Our results indicate that subclinical fatal COVID-19 infection was unlikely in our hospital population subjected to autopsy approximately 6 months before and 6 months after the first confirmed US case.

1.
Centers for Disease Control and Prevention.
CDC Museum COVID-19 timeline.
https://www.cdc.gov/museum/timeline/covid19.html. Published August 16, 2022. Accessed February 17, 2023.
2.
Ciotti
M,
Ciccozzi
M,
Terrinoni
A,
Jiang
WC,
Wang
CB,
Bernardini
S.
The COVID-19 pandemic
.
Crit Rev Clin Lab Sci
.
2020
;
57
(6)
:
365
388
.
3.
Mele
D,
Flamigni
F,
Rapezzi
C,
Ferrari
R.
Myocarditis in COVID-19 patients: current problems
.
Intern Emerg Med
.
2021
;
16
(5)
:
1123
1129
.
4.
Kupek
E.
How many more? Under-reporting of the COVID-19 deaths in Brazil in 2020
.
Trop Med Int Health
.
2021
;
26
(9)
:
1019
1028
.
5.
Lau
H,
Khosrawipour
T,
Kocbach
P,
Ichii
H,
Bania
J,
Khosrawipour
V.
Evaluating the massive underreporting and undertesting of COVID-19 cases in multiple global epicenters
.
Pulmonology
.
2021
;
27
(2)
:
110
115
.
6.
Wang
H,
Paulson
KR,
Pease
SA,
et al
Estimating excess mortality due to the COVID-19 pandemic: a systematic analysis of COVID-19-related mortality, 2020–21
.
Lancet
.
2022
;
399
(10334)
:
1513
1536
.
7.
Schwab
N,
Nienhold
R,
Henkel
M,
et al
COVID-19 autopsies reveal underreporting of SARS-CoV-2 infection and scarcity of co-infections
.
Front Med
.
2022
;
9
:
868954
.
8.
Haslbauer
JD,
Perrina
V,
Matter
M,
Dellas
A,
Mihatsch
MJ,
Tzankov
A.
Retrospective post-mortem SARS-CoV-2 RT-PCR of autopsies with COVID-19-suggestive pathology supports the absence of lethal community spread in Basel, Switzerland, before February 2020
.
Pathobiology
.
2021
;
88
(1)
:
95
105
.
9.
Lai
A,
Tambuzzi
S,
Bergna
A,
et al
Evidence of SARS-CoV-2 antibodies and RNA on autopsy cases in the pre-pandemic period in Milan (Italy)
.
Front Microbiol
.
2022
;
13
:
886317
.
10.
Bodewes
R,
van Run
PRWA,
Schürch
AC,
et al
Virus characterization and discovery in formalin-fixed paraffin-embedded tissues
.
J Virol Methods
.
2015
;
214
:
54
59
.
11.
Sekulic
M,
Harper
H,
Nezami
BG,
et al
Molecular detection of SARS-CoV-2 infection in FFPE samples and histopathologic findings in fatal SARS-CoV-2 cases
.
Am J Clin Pathol
.
2020
;
154
(2)
:
190
200
.
12.
Sridhar
S,
Nicholls
J.
Pathophysiology of infection with SARS-CoV-2—what is known and what remains a mystery
.
Respirology
.
2021
;
26
(7)
:
652
665
.
13.
Menter
T,
Haslbauer
JD,
Nienhold
R,
et al
Postmortem examination of COVID-19 patients reveals diffuse alveolar damage with severe capillary congestion and variegated findings in lungs and other organs suggesting vascular dysfunction
.
Histopathology
.
2020
;
77
(2)
:
198
209
.
14.
Wichmann
D,
Sperhake
JP,
Lütgehetmann
M,
et al
Autopsy findings and venous thromboembolism in patients with COVID-19: a prospective cohort study
.
Ann Intern Med
.
2020
;
173
(4)
:
268
277
.
15.
Martines
RB,
Ritter
JM,
Matkovic
E,
et al
Pathology and pathogenesis of SARS-CoV-2 associated with fatal coronavirus disease, United States
.
Emerg Infect Dis
.
2020
;
26
(9)
:
2005
2015
.
16.
Long
DR,
Gombar
S,
Hogan
CA,
et al
Occurrence and timing of subsequent severe acute respiratory syndrome coronavirus 2 reverse-transcription polymerase chain reaction positivity among initially negative patients
.
Clin Infect Dis
.
2021
;
72
(2)
:
323
326
.
17.
Arevalo-Rodriguez
I,
Buitrago-Garcia
D,
Simancas-Racines
D,
et al
False-negative results of initial RT-PCR assays for COVID-19: a systematic review
.
PLoS One
.
2020
;
15
(12)
:
e0242958
.
18.
Kinloch
NN,
Ritchie
G,
Brumme
CJ,
et al
Suboptimal biological sampling as a probable cause of false-negative COVID-19 diagnostic test results
.
J Infect Dis
.
2020
;
222
(6)
:
899
902
.
19.
Kucirka
LM,
Lauer
SA,
Laeyendecker
O,
Boon
D,
Lessler
J.
Variation in false-negative rate of reverse transcriptase polymerase chain reaction–based SARS-CoV-2 tests by time since exposure
.
Ann Intern Med
.
2020
;
173
(4)
:
262
267
.
20.
Yang
Y,
Yang
M,
Yuan
J,
et al
Laboratory diagnosis and monitoring the viral shedding of SARS-CoV-2 Infection
.
Innovation (Camb)
.
2020
;
1
(3)
:
100061
.

Author notes

The authors have no relevant financial interest in the products or companies described in this article.

The abstract of this work was presented as a poster at the United States and Canadian Academy of Pathology Annual Meeting; March 21, 2022; Los Angeles, California.