Portable UV-C device to treat high flow of infectious aerosols generated during clinical respiratory care

Our UV-C₂₅₄ prototype was attached to a bioaerosol generating circuit to simulate an infected patient (see Fig. 1). At the beginning of the bioaerosol transmission circuit, a misting nebulizer (Aeroneb Lab Micropump Nebulizer, Kent Scientific, Torrington, CT) was inserted to aerosolize the A/PR/8/34 H1N1 influenza viral suspension of 10⁷ PFU mixed with 10 mL Phosphate Buffer Solution (PBS) into the airstream that entered the tubing connected to the prototype. The misting nebulizer allowed less turbulent nebulization conditions than a previous respiratory nebulizer, which enhanced virus viability at the end of the circuit. One or two parallel standard CPAP generators (Resmed S9, San Diego) were used to create suction as an airflow generator. These provided various flow rates (suction was used rather than positive airway pressure to minimize leaks from the circuit) through the entire aerosol sampling circuit. A setting on the flow generators intended to generate a CPAP pressure of 10 cm H₂0 generated a high flow when connected to the circuit. The gas was humidified to approximately 100% relative humidity (RH) with distilled water as is typically done in clinical situations of CPAP and nasal NHF (HC 150 Fisher and Paykel Healthcare, New Zealand). This high RH was intended to mimic clinical usage and to keep the viral suspension from desiccation. Various humidification levels were tested for their possible effect on virus viability. The flow rate through the circuit during the experiments was measured (TSI 5300 Series Digital flow meter, TSI, Shoreview, Minnesota) both upstream and downstream of the prototype to identify any leaks in the system. Ambient relative humidity, as well as at the RH at the beginning and end of the circuit, were measured with a calibrated psychrometer (General Tools No. EP8709, China). As the virus contaminated gas exited the prototype (either in single pass or recirculation mode (Fig. 1a, b)) with increasing numbers of UV lamps on or off, it was collected with an SKC Biosampler, 3 nozzles with sonic orifices, (SKC, Inc, Eighty Four, PA). The SKC Biosampler was connected to a SKC BioLite pump to achieve sonic flow within the Biosampler. A separate trap was placed in line after the Biosampler circuit to collect any liquid overflows. Airflow speed through the entire circuit was regulated by changing the setting for CPAP pressure, and ultimately by adding a second parallel CPAP device to increase flow (ResMed, San Diego) depending on the velocity (low, medium, high) to be tested. A HEPA filter was placed in line before the CPAP flow generator to prevent virus from entering the device. As this was downstream from all collection of samples, it had no impact on the data collected.

Virus stock

150 plaque forming units (PFU) of the A/PR/8/34 (H1N1) influenza virus were injected into the allantoic fluid of 8-days old embryonated chicken eggs and was incubated for 48 h at 37 °C followed by an overnight incubation at 4 °C. Allantoic fluid, containing the virus, was carefully collected, briefly centrifuged, aliquoted and stored at – 80 °C. The virus stock was titrated by plaque assay on pre-seeded confluent monolayers of Madin-Darbin Canine Kidney (MDCK) cells (ATCC CCL-34).

Virus titration by plaque assay

For plaque assays, 250 µL of tenfold dilutions in PBS of collected samples were incubated on confluent monolayers of MDCK cells at 37 °C. After 1 h of incubation, the inoculum was removed by aspiration and cells were overlaid with 2% oxoid agar (Oxoid, Basingstoke, UK) mixed with an equal volume of NaHCO3-buffered 2xMEM supplemented with DEAE/Dextran and TPCK-treated trypsin (1 µg/mL). Cells were incubated for 48 h at 37 °C and 5% CO2. Plaque formation was visualized by staining of cell surfaces after fixation with 4% formaldehyde (5 min at room temperature). For staining, cells were incubated with a 1/1000 dilution of post challenge mouse serum followed by incubation with 1/1000 diluted sheep anti-mouse serum conjugated to horse radish peroxidase (GE Healthcare) and addition of TrueBlue substrate (KPL—Seracare, Milford, MA, USA).

Nebulizer suspensions

H1N1 was suspended in Phosphate Buffer Solution (PBS). The titer of freshly prepared nebulizer suspensions was calculated to be 10⁷ PFU/mL and confirmed by plaque assay titration.

RNA extraction and quantitative reverse-transcription polymerase chain reaction (qRT-PCR)

RNA was extracted using the E.Z.N.A Viral RNA kit, according to manufacturer’s protocol. RNA analysis was performed by one step qRT-PCR with the Luna Universal One-Step RT-qPCR Kit (NEB) using PR8-NP primers (forward—GCACGGTCTGCACTCATATTGAG and reverse—GTGTGCTGGATTCTCATTTGGTC). A PdZ-NP plasmid was used as an internal control to assess the copy number of NP RNA.

Test procedure

Three rounds of experiments were conducted to determine if, before adding UV irradiation, viable virus could be recovered in sufficient concentration after physical losses within the circuit, after either a single pass or recirculation (Fig. 1a, b) through the prototype device. We conducted separate experiments with nebulized virus at room humidity and at initial relative humidity of 100%. The experiments were conducted at ISMMS in a Biosafety Level 2 (BSL-2) safety hood (Baker Co, Sanford, ME) (Fig. 2). After demonstrating we could culture viable virus from the apparatus, our goal was to provide viral inactivation with gas passing through the portable UV-C₂₅₄ device at up to100 L/min. Our experiments tested various combinations of low (26 l/min), medium (41 l/min) and high (126 l/min) rates of total flow with various combinations of 1 to 6 high-output UV-C lamps switched on. Each experiment ran between 20 and 30 min.

UV-C prototype device in test rig within BSL-2 Hood and shown in recirculation condition.