key: cord-0314825-u1ww1229
authors: Brandner, J. M.; Boor, P.; Borcherding, L. S.; Edler, C.; Gerber, S.; Heinemann, A.; Hilsenbeck, J.; Kasajima, A.; Lohner, L.; Märkl, B.; Pablik, J.; Schröder, A. S.; Sommer, L.; Slotta-Huspenina, J.; Sperhake, J.-P.; von Stillfried, S.; Dintner, S.
title: Contamination of personal protective equipment during COVID-19 autopsies
date: 2021-07-15
journal: nan
DOI: 10.1101/2021.07.12.21260357
sha: 5c86fdfe1b2c22a658da9ed2ea781022bcf8c4f2
doc_id: 314825
cord_uid: u1ww1229

Confronted with an emerging infectious disease, the medical community faced relevant concerns regarding the performance of autopsies of COVID-19 deceased at the beginning of the pandemic. This attitude has changed, and autopsies are now recognized as indispensable tools for elucidating COVID-19; despite this, the true risk of infection for autopsy staff is still debated. To elucidate the rate of SARS-CoV-2 contamination in personal protective equipment (PPE), swabs were taken at nine locations of the PPE of one physician and an assistant each from 11 full autopsies performed at four different centers. Further samples were obtained for three minimally invasive autopsies (MIA) conducted at a fifth center. Lung/bronchus swabs of the deceased served as positive controls. SARS-CoV-2 RNA was detected by RT-qPCR. In 9/11 full autopsies PPE samples were tested RNA positive with PCR, in total 21% of all PPE samples taken. The main contaminated parts of the PPE were the gloves (64% positive), the aprons (50% positive), and the upper sides of shoes (36% positive) while for example the fronts of safety goggles were only positive in 4.5% of the samples and all face masks were negative. In MIA, viral RNA was observed in one sample from a glove, but not in other swabs. Infectious virus isolation in cell culture was performed in RNA positive swabs from full autopsies. Of all RNA positive PPE samples, 21% of the glove samples were positive for infectious virus taken in 3/11 full autopsies. In conclusion, in >80% of autopsies, PPE was contaminated with viral RNA. In >25% of autopsies, PPE was found to be even contaminated with infectious virus, signifying a potential risk of infection among autopsy staff. Adequate PPE and hygiene measures, including appropriate waste deposition, are therefore mandatory to enable safe work environment.

The results obtained through autopsies of the Severe acute respiratory syndrome Coronavirus-2 (SARS-2 CoV-2)-infected deceased are of crucial importance for understanding Coronavirus disease 2019 3 . Viral pneumonia with diffuse alveolar damage (DAD) is the most frequent cause of death 4 in fatal cases of COVID-19. In addition to the dramatic changes in the lungs, the affection of multiple 5 other organs is currently interpreted mainly as a systemic inflammatory reaction. Several authors have 6 described endothelial impairment with consecutive activation of the coagulation system (1, 2). 7

However, concerns about the safety of autopsy staff hampered the autopsy activities of surgical, 8 forensic and neuro-pathologists. Since the beginning of the pandemic, several reports and guidelines 9 from different authors and organizations have been published (3, 4) concerning this topic. 10

The environmental viability of the virus has been investigated in experimental conditions and real-11 world settings of the domestic or clinical surroundings of SARS-CoV-2-positive persons in a few studies 12 which were recently summarized by Meyerowitz et al. (5) . The presence of viable virus has been 13 identified for up to three hours in aerosols and 72 hours on surfaces. Half-lives were calculated to be 14 up to six hours (6) . 15

In Germany, the Robert Koch Institute has recommended compliance with protection level 3, which 16 requires wearing appropriate protective equipment (surgical hood cap, eye/face-protection with fully 17 protective safety goggles or visors, filtering face piece (FFP) 2/3 masks, long-sleeved and impermeable 18 protective clothing, waterproof apron, additional forearm protection, a second layer of latex/nitrile 19 gloves with long cuffs, and appropriate shoes), when handling COVID-19 deceased (7). 20

Only a few reports addressing topics related to the infectiousness of dead bodies of the SARS-CoV-2-21 infected deceased and the risk for autopsy staff have been published. All authors report detection of 22 the virus with reverse transcription quantitative PCR (RT-qPCR) in swabs taken from the airways at 23 different time intervals after death (8-11). Schroeder et al. detected viral RNA on various body surfaces 24 of the deceased, as well as body bags; however, no viable viruses were detected (12). Viral RNA 25 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 15, 2021. ; https://doi.org/10.1101/2021.07.12.21260357 doi: medRxiv preprint detection on the surfaces of the autopsy tables and autopsy room walls as well as face shields was 1 reported by Pomara et al. (11) . 2

The aim of this study is to evaluate the extent of viral RNA contamination of the personal protective 3 equipment (PPE) of autopsy staff during autopsies of the COVID-19 deceased. Special focus was placed 4 on the infectivity of the samples with positive SARS-CoV-2 RNA detection. 5 with ultrasound-conducted biopsies were performed at MU. Written consent was obtained from the 15 next of kin to perform the autopsies. The inclusion criterion for decedents was confirmed diagnosis of 16 SARS-CoV-2 infection, as evidenced by PCR test of the nasopharyngeal swab during the hospital stay 17 and by either rapid PCR or antigen testing during the full autopsies. Three autopsies each from AU, DR, 18 HH, and MU were included, and two cases were contributed by AA. All autopsy rooms were maintained 19 at negative pressure, with a minimum of 10 air changes per hour. 20

Commercially available swab sets were used (COPAN eSwab B 80482CE, Mast Group, Reinfeld, 22

Germany). Before swabbing, the tips were moistened with the transport medium, and then the PPE 23 surfaces were thoroughly swabbed in a meandering manner for at least 15 seconds. Finally, the tips 24 were placed in the transport container. The swab locations are shown in Fig 1. Two swabs per location 25 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 15, 2021. ; https://doi.org/10.1101/2021.07.12.21260357 doi: medRxiv preprint were taken next to each other, one for RT-qPCR testing, one for testing of virus infectivity (virus 1 isolation). Samples for RT-qPCR testing were stored in refrigerators at 4°C, while swabs intended for 2 eventual isolation of infectious virus were frozen at -80°C. PCR testing of all samples was performed at 3 AU. For samples from full autopsies that showed positive results in the PCR testing, the corresponding 4 samples collected for viral isolation were sent to HH, where another PCR testing and virus isolation 5 were performed. 6

In all full autopsies, swabs from the PPE of the autopsy-conducting physician and one autopsy assistant 7 were performed after finishing the autopsy. In MIAs, only the team member with the closest contact 8 to the body of the deceased -the ultrasound-guiding physician -was evaluated. 9

To generate reference samples (positive controls), swabs from the plane-cut surface of the lungs and, 10 in one case, the bronchus (case AU2) or from the lung biopsies (MIA, MU) were collected during each 11 autopsy. 12 13 RT-qPCR 14 All samples were processed and analyzed primarily at the Institute of Pathology and Molecular 15

Diagnostics at the Medical Center in Augsburg. The method has been described in the literature (13) . 16

In brief, RNA was extracted using the Promega Maxwell® 16 MDx system and the Promega Maxwell 16 17 LEV RNA FFPE Purification Kit (AS1260, Promega Corporation, Madison, WI, USA). Quantitative real-18 time PCR for SARS-CoV-2 was performed on the extracts with one-step multiplex RT-qPCR targeting 19 the SARS-CoV-2 ORF1ab, N Protein, and S Protein using the TaqPath COVID-19 CE-IVD RT-PCR Kit 20 (A48067, Thermo Fisher, Pleasanton, TX, USA). The RT-qPCR was conducted using the QuantStudio 5 21 Dx real-time PCR Instrument, and the data were analyzed and interpreted using QuantStudio™ design 22 and analysis software (v.1.2x, Thermo Fisher, Carlsbad, CA, USA). Results with two or more positive 23 targets were considered as valid. A singular failure of the curve for the S protein was used as indirect 24 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. For control of RNA integrity, confirmatory RT-qPCR of the samples was performed as described (14) . 7

Vero E6 cells were maintained and cultivated under standard conditions (15). For virus isolation, 500 8 μl of swab medium was used, and infection was performed as described (12, 15). Supernatants were 9 harvested at 72 hours post-infection, and virus growth was analyzed, as described before (16). Virus 10 isolation experiments were restricted to samples from full autopsies. 11

To compare data measured by order or rank, the Mann-Whitney-U-Test and the One Way Repeated 13 Measures ANOVA test were used. The Spearman´s rank order correlation was applied to calculate 14 correlations between ranked data. Depending on the proportion numbers, tabulated nominal data 15 were compared using either the chi-square test or Fisher's exact test. A p-value < 0.05 was considered 16 significant. All calculations were performed using the statistics package Sigmaplot 13.0 (Systat, San 17

Jose, CA, USA). 18

Case collection 20 The case characteristics are given in Table 1 . Fourteen autopsies were included, of which three were 21 conducted as MIA. The median age of the deceased was 71 years (range: 52 -91 years), with a male 22

to female ratio of 1.8: 1. The postmortem interval (PMI) had a broad range (15 to 144 hours, median 23 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 15, 2021. ; https://doi.org/10.1101/2021.07.12.21260357 doi: medRxiv preprint 7 of 18 55 hours). The median period from the first positive SARS-CoV-2 RT-PCR test to death was 10.5 days 1 (range: 0 -51 days). 2

In total, 209 swabs were performed for 11 full autopsies for RT-qPCR testing. Eleven of these samples 4 were taken from the lungs/bronchus of the dead body to serve as a reference (positive control) for 5 each case. All lung/bronchus swabs (11/11) were positive. 6

The remaining 198 swabs were collected from nine locations on the PPE of one physician and one 7 assistant per autopsy. Of these, 41 (21%) were SARS-CoV-2 RNA-positive, 24 from physicians and 17 8 from assistants ( Fig. 2A) . In only two autopsies (2/11, 18%) all PPE swabs were SARS-CoV-2 RNA 9 negative, ( Fig. 2A) , while in 9/11 autopsies (82%) RNA contamination was detected with the number 10 of SARS-CoV-2 RNA-positive PPE swabs per autopsy ranging from 3 to 7 (median 4, Fig. 2A the goggle-samples, while lateral sides of safety goggles, FFP masks, and the back of the protective 20 clothing were negative. In the 14 events when gloves were positive, also aprons were positive in 9 21 events (64%) and the upper sides of the shoes in 7 events (50%) (Fig. 2C ). There was a correlation trend 22 between contamination of gloves and aprons (p =0.08). Furthermore, a correlation trend between 23 parallel RNA detection on the apron and the upper sides of the shoes (R = 0.38; p = 0.08) could be 24 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 15, 2021. ; https://doi.org/10.1101/2021.07.12.21260357 doi: medRxiv preprint 8 of 18 observed. The latter was also trend-wise associated with the positivity of the shoe soles (R = 0.39; p = 1 0.08; Figures 2B and 2C ). 2 A highly significant difference (p < 0.001) was observed between the Ct values of the samples obtained 3 from the lungs or bronchus (median Ct: 17; range: 14 -28) and from the PPE (median Ct: 28; range: 18 4 -32). The viral density decreased highly significantly in a sequel from the lung/bronchus samples 5 (positive controls) via the gloves and the aprons to the shoes ( Figure 2D ). 6

In total, 30 swabs were performed for three autopsies, including one lung control from each autopsy. 8

In one autopsy (MIA-1), the lung control was tested positive for SARS-CoV-2, the lung controls of the 9 other two autopsies (MIA-2 and MIA-3) were negative (Table 1 ). In MIA-1 viral RNA could be detected 10 from the gloves of the ultrasound-physician, all other swabs were negative. For MIA-2 and MIA-3 all 11 swabs were tested negative for SARS-COV-2 RNA. 12 Concerning lung/bronchus samples, virus isolation was successful in 4/11 (36%) samples taken at three 20 centers (AU, DR, and HH) (Fig. 3) . 21

Concerning PPE samples, 11/41 samples used for virus isolation were negative in confirmatory qPCR performed in HH and thus probable RNA degradation was suspected. Hence, they were excluded 23 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. were obtained in four of six autopsies with a lung/bronchus sample Ct value below 18, suggesting a 10 high viral load, while only one of five cases with lower virus concentration (Ct>21) was infectious (for 11 Ct values, see Table 1 ). However, this distribution trend did not reach significance (p = 0.242). is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 15, 2021. ; https://doi.org/10.1101/2021.07.12.21260357 doi: medRxiv preprint biopsy) was tested positive. Thus, the number of evaluable MIA autopsies was only one. The fact that 1 in this one positive case the glove sample was positive showed that even in MIA, PPE contamination 2 could not be completely excluded. However, further investigations with a higher number of autopsies 3 are necessary to elucidate its real extent. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 15, 2021. ; https://doi.org/10.1101/2021.07.12.21260357 doi: medRxiv preprint

In concordance with other reports investigating the persistence of viral material on and in people 1 deceased due to COVID-19 ' (10-12, 17, 18) , the PIM of up to 144 hours did not reduce the risk of PPE 2 contamination with SARS-CoV-2 RNA. In our study, we did not aim to evaluate the stability of the virus 3 on PPE over time. Data addressing the topic of SARS-CoV-2 stability have been summarized by 4

Meyerowitz et al. (5) . 5

To assess for infectivity, frozen swabs from all locations previously tested SARS-CoV-2 positive with RT-6 qPCR were selected for further cultivation, including the reference lung/bronchus samples. Notably, 7 only 41 of these 52 samples tested positive by confirmatory RT-qPCR in HH; therefore, they had to be 8 excluded from calculation of the overall infectivity rate. Retrospectively, it is difficult to identify the 9 reason for this RT-qPCR negativity. RNA degradation during storage and transport could have been an 10 issue. Of note, even though all samples were shipped on dry ice by an experienced courier service, it 11 is remarkable that none of the negative samples came from HH, where the samples could be shipped 12 in-house. Remarkably, the non-evaluable samples were mainly from the shoes (7/11) with high Ct-13

values. In addition, it is conceivable that sampling the swabs at locations directly next to each other 14 might contribute to these differences, suggesting that the contamination might be more locally 15 concentrated than distributed. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 15, 2021. ; https://doi.org/10.1101/2021.07.12.21260357 doi: medRxiv preprint

In the lung/bronchus samples that were chosen as positive control, virus isolation was successful in 1 4/11 samples (36%). There was a clear trend towards a higher viability rate in cases with low Ct -values 2 under 18 in the lung/bronchus samples, reflecting that higher viral load likely results in higher 3 probability of virus infectivity, as was shown before (19). There was no correlation between successful 4 virus isolation and PMI. Also Plenzig et al. reported the isolation of viable viruses from lungs in two out 5 of four cases independent from PMI (10). 6

The fact that only the lung/bronchus samples and samples from the gloves were infectious -with a 7 higher infectivity rate in lungs/bronchus -while no infectious virus could be isolated from RNA In conclusion, as recommended by several national and international instances, the use of adequate 23 PPE is mandatory because the risk of infection during autopsy is a matter of reality, not a theoretical 24 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 15, 2021. ; https://doi.org/10.1101/2021.07.12.21260357 doi: medRxiv preprint consideration. Together with hygiene measures, including appropriate waste disposal, they enable 1 the safe performance of COVID-19 autopsies, which are essential for a better understanding of this 2 disease. Also, for future infectious diseases advised selection of appropriate PPE and hygienic 3 measures will provide the basics to carry out autopsies as an important source for new knowledge. Table  7 Demographic and autoptic data of all cases. Rf = reference organ (positive control); Gl = Gloves; n.a. 8 = no data available; n.i. = no information concerning the viral lineage available; * evaluated during 9

RT-qPCR -loss of the S-curve as hint for variant of concern; ** no testing prior to death #:positive 10 tested in confirmatory RT-qPCR in HH. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 15, 2021. ; https://doi.org/10.1101/2021.07.12.21260357 doi: medRxiv preprint

Associated With SARS-CoV-2: Systematic Review and Meta-analysis

Findings and Venous Thromboembolism in Patients With COVID-19: A Prospective Cohort Study. Ann 6

Autopsy of patients with COVID-19: A 10 balance of fear and curiosity. Pathology -Research and Practice

Transmission of SARS-CoV-2: A Review of 12

Viral, Host, and Environmental Factors

Aerosol and Surface Stability of SARS-CoV-2 as Compared with SARS-CoV-1

Recommendations for the management of SARS-CoV-2-infected 17 decedents

Risk of COVID-19 Transmission During Autopsy

Strategies in Clinical Forensic Autopsies of Confirmed COVID-19 Cases. Diagnostics (Basel)

The handling 4 of SARS-CoV-2 associated deaths -infectivity of the body

Viral mapping in

Evaluation of a quantitative RT-PCR assay 9 for the detection of the emerging coronavirus SARS-CoV-2 using a high throughput system

Genome Sequence of a SARS-CoV-2 Strain Isolated in Northern Germany

Dying with 18

SARS-CoV-2 infection-an autopsy study of the first consecutive

Analysis of the persistence 21 time of the SARS-CoV-2 virus in the cadaver and the risk of passing infection to autopsy staff

Estimating infectiousness throughout SARS-CoV-2 26 infection course

Modeling the Stability of 28 SARS-CoV-2 on Personal Protective Equipment (PPE)

the author/funder, who has granted medRxiv a license to display the preprint in perpetuity

CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)The copyright holder for this preprint this version posted July 15, 2021. ; https://doi.org/10.1101/2021.07.12.21260357 doi: medRxiv preprint Demographic and autoptic data of all cases. Rf = reference organ; Gl = Gloves; n.a. = no data available; n.i. = no information concerning the viral lineage available; * evaluated during RT-qPCR -loss of the S-curve as hint for variant of concern; ** no testing prior to death # : positive tested in confirmatory RT-qPCR in HH.