Identification of SARS‐CoV‐2 RNA in healthcare heating, ventilation, and air conditioning units

Abstract Evidence continues to grow supporting the aerosol transmission of severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2). To assess the potential role of heating, ventilation, and air conditioning (HVAC) systems in airborne viral transmission, this study sought to determine the viral presence, if any, on air handling units in a healthcare setting where coronavirus disease 2019 (COVID‐19) patients were being treated. The presence of SARS‐CoV‐2 RNA was detected in approximately 25% of samples taken from ten different locations in multiple air handlers. While samples were not evaluated for viral infectivity, the presence of viral RNA in air handlers raises the possibility that viral particles can enter and travel within the air handling system of a hospital, from room return air through high‐efficiency MERV‐15 filters and into supply air ducts. Although no known transmission events were determined to be associated with these specimens, the findings suggest the potential for HVAC systems to facilitate transfer of virions to locations remote from areas where infected persons reside. These results are important within and outside of healthcare settings and may present necessary guidance for building operators of facilities that are not equipped with high‐efficiency filtration. Furthermore, the identification of SARS‐CoV‐2 in HVAC components indicates the potential utility as an indoor environmental surveillance location.

movement patterns indoors induced through heating, ventilation, and air conditioning (HVAC) systems may contribute to transmission events. 16,17 Aerosolized SARS-CoV-2 RNA has been previously detected in the air of hospital rooms with symptomatic COVID-19 patients, [18][19][20][21] suggesting the possibility that SARS-CoV-2 viral RNA (and potentially virus) have the capacity to enter into building HVAC systems via evacuated room air after a shedding or aerosolization event from infected individuals. Although hospitals contain higher levels of mechanical filtration and room air exchange than almost all other buildings, which are important strategies to help prevent the transmission of disease, a growing body of evidence suggests that these precautions may not be adequate to completely eliminate SARS-CoV-2 10,22,23 in filtered air. Furthermore, the identification of viable SARS-CoV-2 in the air of COVID positive patient rooms further underlies the need to better understand aerosol dynamics of SARS-Cov-2 spread indoors. 12 Studies have shown the persistence of SARS-CoV-2 to be hours in the air and on surfaces. 24 Efforts to limit the transmission and continued spread of SARS-CoV-2 have mainly focused on social (spatial) distancing, increased cleaning regimens, mandated face coverings, and increased surveillance. 24,25 However, as more indoor spaces begin to reopen and increase in occupant density, more individuals will occupy shared air spaces serviced by In the past, ventilation has played a key role in the transmission of infectious disease. 17,[26][27][28] Similarly, there are several reported hospital-associated SARS-CoV-2 infections and outbreaks during the COVID-19 pandemic. 29,30 With the demonstration that ventilation systems may contribute to occupant transmission events, HVAC and ventilation guidelines and recommendations have been modified. [31][32][33] Here, we present data demonstrating the presence of SARS-CoV-2 RNA at several locations along mechanical ventilation air return and supply pathways, including multiple locations in air handling units (AHUs). During May and June 2020, samples were collected from three separate AHUs (Figure 1) serving wards with COVID-19 patients.

| Sample collection
COVID-19 positive patients were determined through either an initial rapid SARS-CoV-2 antigen test followed by a quantitative reverse-transcription polymerase chain reaction (RT-PCR) diagnostic test or simply a RT-PCR diagnostic test. The AHUs have a minimum outside air percentage of 40% by design. The percentage of outside air was determined using return air temperature, supply air temperature, and mixing air temperature (Table 1). COVID-19 positive patients were housed in ward 6A (emergency department), 7C (MICU), 12C (Labor and Delivery), and 14C (Internal Medicine Inpatient) that were serviced by the three sampled AHUs. These AHUs were selected because they handled air from areas with confirmed COVID-19 positive patients. When space allowed, COVID-19 positive patients were first placed in airborne infection isolation (AII) rooms that vent directly to the outside. Additional patients were cohorted on 6A, 7C, 12C, and 14C that employ mixing ventilation with air supply diffusers and return grills located on the ceiling in patient rooms. Within each AHU, three areas along the path of airflow were sampled, including the pre-filters, final filters, and supply air dampers ( Figure 1). A summary of the most recent maintenance data can be found in Table 2. The pre-filters are rated at MERV10 and final filters are rated at MERV15, both in excess of minimum code requirements. 34 MERV10 and MERV15 filters carry a capture efficiency of 50% and 90% respectively for particles ranging from 1-3 microns in size. Based upon engineering calculations and equipment documentation available, the HVAC system is capable of cycling air from the ward, to the AHU, and back to the ward in a time between 90 sec and 5 min, depending on travel distance to room location. All filters used in the AHU filter blocks are manufactured by Camfil (Models 3V and 5V). Ultraviolet light disinfection was not utilized in the AHUs sampled during this study. Samples were collected using Puritan PurFlock Ultra swabs (catalog #25-3606-U) and swabs were taken

Practical Implications
• More work is needed to further evaluate the risk of SARS-CoV-2 transmission via HVAC systems and to verify effectiveness of building operations mitigation strategies for the protection of building occupants.
• The results of this study suggest that building occupants, homeowners, and buildings and facility managers should review their air filtration and ventilation practices accordingly.
• Evaluating building HVAC operational practices and equipment and implementing enhancements could reduce future built environment transmission risks as more buildings reopen or increase in occupancy.
• This study suggests the potential for AHUs to serve as a location to conduct environmental viral surveillance to guide building operations, human behavior, and other mitigation actions according to the relevant risks identified.
from the left, middle, and right sides of each filter block along the path of airflow. Swabs were individually processed. Samples were only collected from the intake side of the filters. Swabs were premoistened using viral transport media (RMBIO, catalog #VTM-CHT).
Swabbing occurred for 20 sec on an area approximately 20 × 30 cm at each location and swabs were immediately placed into 15 ml conical tubes (Cole-Parmer, catalog #UX-06336-89) containing 1.5 ml viral transport media and stored on ice for transport to a BSL-2 laboratory with enhanced precautions (BSL2+) lab for processing, which typically occurred within 2 h after collection. Samples were collected by the same researcher each sampling time, and the researcher did not demonstrate any symptoms of COVID-19 and tested negative by qRT-PCR.

| Sample processing and molecular analysis
Samples were hand-carried to a research laboratory at OHSU for initial processing. In a class 2 biosafety cabinet (BSC), conical tubes were vortexed briefly, allowed to settle for 5 min, and 200 µl of the supernatant was removed and combined with 600 µl of the lysis/preservative buffer (DNA/RNA Shield, Zymo Research). spike glycoprotein gene. 3 An artificial gene standard from Integrated DNA Technologies with known copy number was utilized to create a dilution series and standard curve for the determination of viral gene copies in each sample, 35 with a limit of detection of 2.22 gene copies/ul. All qRT-PCR reactions were run in triplicate. Potential contamination during initial processing was assessed through the use of passive air settling plates and reagent controls within the BSL2 lab. To accomplish this, passive air settling plates were placed in the BSC and on the outside lab bench for the duration of processing.
Following the completion of specimen processing, the same swabs, F I G U R E 1 AHU sectional diagram illustrating the path of airflow, mixing of recirculated return air with outdoor air fraction and locations of swab sampling AIR HANDLING UNIT (AHU) All controls tested negative for the presence of SARS-CoV-2 RNA.
All samples that tested negative for the presence of SARS-CoV-2 RNA were included in the analysis by treating them as non-detection samples instead of true zero samples [36][37][38] in order to avoid creating a censored dataset. The full dataset can be found in Table S1. The samples were considered to contain one-half the assay's limit of detection in gene copies/ul (1.1 gene copies/ul).

| RE SULTS
In total, 56 samples from three different AHUs were collected; 25% (14/56) of samples contained detectable SARS-CoV-2 RNA (Table   S1). The highest abundance sample (~49 gene copies/ul) was found  Table 3). The least SARS-CoV-2 RNA was detected at the final filter and the most at the pre-filters (Table 3). No correlation between outside air percentage and detected genome copies was observed ( Figure S1).

| DISCUSS ION
This investigation demonstrates the presence of SARS-CoV-2 RNA at multiple locations within mechanical AHUs, and more specifically, Previous studies have demonstrated that SARS-CoV-2 can be found in aerosols and droplets ranging from 0.25 to 4 microns. 10,11 MERV15 filters have a capture efficiency of 85%-90% for particles ranging from 0.3 to 3 microns in size. It is possible that particles containing SARS-CoV-2 RNA passed through the filters or bypassed the filters and were impacted on the supply air dampers. Additionally, it has been previously observed that small gaps in between filters from suboptimal installation may allow for biological material to pass through filters in hospital HVAC systems. 1,2 However, we did not set out to experimentally determine how SARS-CoV-2 genetic material may have bypassed filtering steps in the HVAC system, and any assertions as to the mechanism of bypass would be nothing more than speculation. In experimentally generated aerosols, SARS-CoV-2 and other coronaviruses have been demonstrated to retain infectivity for between one and sixteen hours, [45][46][47][48] lending credence to the potential for aerosolized transmission to occur. Even more,

SARS-CoV-2 has been demonstrated to be infectious in aerosols in
rooms of SARS-CoV-2 positive patients, 12 suggesting the potential risk of transmission through HVAC systems, and warranting further investigation.
There are steps that can be taken to limit the potential impact of airborne dissemination of viruses in the built environment, including careful donning and doffing of personal protective equipment, 24 hand hygiene after hand contact with the environment, cleaning of contaminated surfaces, use of UV radiation, 49 ensuring indoor relative humidity is between 40% and 60%, 46 and providing indoor ventilation pathways that extract contaminated air with minimal in-room recirculation (such as displacement ventilation).
Use of the highest possible efficiency filters for the building type and HVAC system design, 5  census data were collected retrospectively for the sampling days.

ACK N OWLED G EM ENTS
The authors would like to thank Oregon Health and Science