Use of HEPA filters to reduce the risk of nosocomial spread of SARS-CoV-2 via operating theatre ventilation systems

Editor—Transmission of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) via respiratory droplets and surface contacts is clear. The increasing evidence of aerosol disease transmission raises concern of an additional potential route of nosocomial spread. It is within this context that our small obstetric theatre complex presented a novel infection control challenge that we believe other departments may also face.

Positive pressure theatre ventilation is commonplace in operating theatres worldwide. Positive pressure pushes air from our obstetric theatre through large ventilation grills into the adjacent recovery area where the extraction vents reside. This results in potentially contaminated airflow passing directly over the recovery beds (Fig. 1 ). Current UK guidelines state that positive pressure theatre ventilation should not be considered a risk for aerosol spread to adjacent areas because of the dilution of particles resulting from rapid air changes.¹ This is in conflict with other guidelines which advise that the operative management of coronavirus disease 2019 (COVID-19) patients should take place in a negative pressure or airborne infection isolation room to ensure viral transmission is contained.2, 3, 4

Fig 1.

Schematic of ventilation setup in our obstetric coronavirus disease 2019 (COVID-19) theatre showing positive pressure ventilation and bulk airflow direction from theatre to recovery.

The proximity of our obstetric recovery beds to aerosol-generating procedures occurring in the obstetric theatre is just over 3 m distant and therefore well within the environmental contamination range described.^{5 , 6} Furthermore, evidence of aerosol concentration downstream of infected subjects highlights the potential for disease spread to staff and patients in our recovery area.^{5 , 7} The potential to expose our staff and recovering non-COVID-19 women and their babies to the bulk flow of the aerosol generated in our COVID-19 theatre seemed unjustifiable even if this were diluted. There is currently little evidence on adapting theatre ventilation systems to reduce the risks of nosocomial spread of SARS-CoV-2. We therefore reviewed the techniques used to produce respiratory isolation rooms described in the literature from previous respiratory disease outbreaks.⁸ We outline these options below in the hope of providing potential solutions to colleagues faced with similar challenges.

Though contrary to national guidelines,¹ we initially trialled turning off the theatre ventilation for the duration of any COVID-19 case, thereby reducing airflow and with it the risk of disease transmission to adjacent areas. However, this resulted in unbearable temperatures in the operating room, especially for staff wearing personal protective equipment (PPE). The increased risk of wound infections was difficult to justify, as was the increased viral aerosol load to which staff would be exposed. Further, the absence of aerosol removal by the ventilation system necessitated the use of prolonged airborne precautions, significantly slowing patient flow.

A seemingly simple way of creating a respiratory isolation area is to modify positive pressure theatre ventilation systems to maintain negative pressure by rebalancing the air input and extraction.² Unfortunately, our obstetric theatre does not contain any extraction vents and the airflow in the input vents cannot be reversed, rendering this adaption impossible in our case. Allocating a physically separate theatre complex for COVID-19 patients to reduce the risk of contamination of adjacent areas^{3 , 4} was also impossible in our small obstetric theatre setup.

We considered sealing the ventilation grills between theatre and recovery and installing an exhaust pipe to vent theatre air to the outside to create a make-shift negative pressure room as described in the context of the 2003 severe acute respiratory syndrome (SARS-CoV-1) outbreak.⁸ However, this solution required significant structural work, rendering it impracticable.

In the absence of temporary negative pressure respiratory isolation rooms, airborne infectious disease outbreaks have resulted in the use of novel techniques to limit nosocomial disease spread. Amongst these, high-efficacy particulate air (HEPA) filters have been shown to be effective.⁸ HEPA filters are composed of mats of randomly arranged fibres that intercept passing particles by a combination of diffusion, interception, and inertial impaction. This confers their filtration capacity which is effective to 99.97% of 0.1 micron particles.⁹ Aerosol droplets are 1–5 microns in size. Though research on HEPA filtration of SARS-CoV-2 is yet to be published, their efficacy in capture and containment of diseases of similar particle size is well documented.⁹ Because of the low cost and ease of installation, we concluded that fitting a HEPA filter over the ventilation grills to filter the air passing to recovery was the best solution for our case. We designed a housing for the filters to ensure a good seal.

After installation, theatre ventilation function testing was repeated. In view of the increased resistance to air flow created by the filter, we assessed for a potential air leak, which was found to be significant. The smoke seal on the theatre doors was refitted. Further testing showed that the theatre air changes remained above the recommended minimum 25 per hour with negligible leak and contamination of adjacent areas.

It is important to highlight that in a modified airborne infection isolation room such as this, the mechanical elements degenerate over time, rendering them inefficient and placing staff and patients at risk. They therefore require regular testing and maintenance to ensure they are operating effectively, the frequency of which should be guided by the manufacturer. Usual precautions required in respiratory isolation, such as keeping doors closed and using appropriate protective equipment, must also be adhered to.

As our obstetric theatre setup is not unusual, we considered that we may not be alone in facing this infection control challenge. With the COVID-19 crisis potentially continuing for some time, we wished to share our experience in the hope that this will prove useful to colleagues in other hospitals and trusts.

Authors' contributions

Engineering and technical aspects: LMD

Infection control and prevention consulting: IH

Clinical aspects and research and drafting manuscript: SY

Manuscript proof reading and editing: all authors

Declarations of interest

The authors declare that they have no conflicts of interest.