Can Class II Biological Safety Cabinet designs be improved to reduce laboratory-associated infections?

An Introduction to Biological Safety Cabinets

Biological Safety Cabinets (BSCs) are a very common piece of equipment used in laboratories by laboratory workers handling infectious agents. The embedded video link below provides a short summary of how BSCs work. BSCs are often referred to as primary containment because they are located at the level of the hazard being handled. Laboratories are ranked in terms of the risk of the organism being used. This ranges from those labs handling low-risk biological agents (Biosafety Level 1 or ‘BSL1’) to those that handle dangerous or exotic biological agents (Biosafety Level 4 or ‘BSL4’).

BSCs can be found in most microbiological and biomedical laboratories around the world and the decision to use one is based on a risk assessment. This looks at the type of biological agents being used and the likelihood of these agents being aerosolised within the laboratory. For example, the Center for Disease Control’s publication for Biosafety in Microbiological and Biomedical Labs (BMBL) suggests that Hepatitis B virus, human immunodeficiency virus (HIV) and Salmonella can be handled on an open bench within a BSL2 facility if:

  • the risk of splashes and aerosolisation is low and
  • the laboratory has appropriate practices and procedures in place.

The World Health Organisation’s Biosafety Manual takes a similar risk-based approach with very little prescriptive guidance on what types of BSCs should be used in which laboratories.

The BMBL also provides a classification scheme for the different types of BSCs that are widely copied internationally with some local variations (refer to Appendix 1). Of the seven different cabinet designs listed in the BMBL, the BSC Class II variant is the most common cabinet used in microbiological and biomedical laboratories as it allows direct manipulation of biological agents held within the cabinet via a front opening of the unit. The basic blueprint for Class II cabinets are shared internationally, despite a number of varying regional standards (e.g. Europe’s EN 12469 :2000) or national standards (e.g. US’s NSF/ANSI 49:2024) which modify their design parameters. In Australia, the design and performance standards for Class II BSCs are documented in the Australian Standards publication AS2252.2:2009

The general guidance on BSC use from both publications could be summarised as follows:

  • BSL1-4: use BSCs when handling biological agents or toxins with the potential for aerosol generation
  • For BSL3 facilities: BSCs should be used when handling infectious agents
  • For BSL4 facilities: Use of a specific type of BSC called an isolator which completely separates the biological agent from the laboratory worker.

Generally speaking, it is up to the laboratory (typically the biosafety officer or a biosafety committee) to undertake a risk assessment to determine the appropriate level of containment used in any situation. These requirements are generally not prescribed in any guidance documents because of the highly varied activities undertaken by laboratories.

Class II BSC design and performance

Class II cabinets are a considerable improvement to working with biological agents on an open benchtop, but they also have their shortcomings – namely the large opening at the front of the cabinet to enable the laboratory technician to work with specimens (Photo 1). In contrast, Class III cabinets (isolators) are fully enclosed, and the laboratory technician manipulates samples via a glove port (Photo 2). The downside to Class III BSCs is that all materials and specimens must pass through a decontamination passbox slowing the experimentation or production process and hindering its use. Class III BSCs are also several times more expensive than the equivalently sized Class II cabinet, as well as being more cumbersome to work in.

Photo 2: Class II cabinet used at the Wuhan Institute of Technology (Source: online image)
Photo 3: NuAire Compounding Isolator and passbox (Source: online image)

The popularity of Class II cabinets is driven by their relative ease of use and flexibility. Both Australian (Australia PC4 Guidelines 2007) and US standards (BMBL) permit their use in BSL4 facilities in combination with positive pressure suits as shown in Photo 1. The cabinets use a combination of fans and High-Efficiency Particulate Air (HEPA) filters to create a negative pressure workspace that limits the escape of biological agents and contains them within the working plenum. The HEPA-filtered air is recirculated into the cabinet or released into the laboratory (Figure 1).

Figure 1 – Cross-sectional view of Class II cabinet showing the movement of contaminated air and HEPA-filtered air; the sash opening at the front of the cabinet and the HEPA filters (Source: BMBL).

An excellent review of the overall effectiveness of Class II BSCs is provided by

Kimmen et al (2008), which found that:

  • There is not a large body of research addressing the performance of Class II BSCs;
  • Laboratory workers can still be infected due to biological agents moving out of the front opening;
  • Workers using BSCs are only fully protected when the unit is fully enclosed (i.e. in a Class III isolator);
  • Sash opening height was a significant variable determining the risk of contamination outside the cabinet.

The paper concludes that BSCs significantly lower LAIs despite the shortcomings in the design caused by the opening of the front sash window.

 

From a functional perspective, the popularity of the Class II design is driven by its flexibility and relative effectiveness in reducing the frequency of LAIs in the absence of any other primary containment methods.

 

In forming a view on the relative performance of Class II BSCs, this report will now look at:

  • The epidemiology of LAIs;
  • The effectiveness of HEPA filters in capturing pathogens.

Epidemiology of Laboratory-Associated Infections (LAIs)

The incidence of reported LAIs is likely to be well below the true number of infections. For example, a 2018 review paper of the Asia Pacific region identified only 27 LAIs reported between 1982 and 2016 of which 59% were from bacterial origin, 33% from viruses and 7% from fungi. More than half of these reports occurred in research laboratories (Siengsanan-Lamont & Blacksell, 2018). Two review papers by Harding and Byres (2006) and Byres and Harding (2016) provide a more comprehensive review of LAIs internationally. Of the 270 publications reviewed in the 2006 review, the authors concluded:

  • The source of exposure for the majority of LAIs is unknown (82%);
  • Bacteria and viruses were 83% of total reported incidences;
  • Tuberculosis was the cause of 33% of bacterial LAIs.

 

The 2016 review by the same authors offered some additional information on the incident rate of infection:

  • 2-4 per 1000 for laboratory workers in the US and Japan;
  • 1 per 1000 for hospital researchers globally;
  • 43 per 1000 for laboratory workers in England/Wales.

Some 51% of LAIs were caused by bacteria and occurred mainly in clinical laboratories. The paper noted that sniffing plates are likely a historical source of infection and that the reported numbers are substantially less than the true situation.

 

The key points to be taken from these publications were that:

  • A substantial proportion of laboratory workers are impacted by LAI annually and this rate is likely to be underreported;
  • The mode of infection in the majority of LAIs (>80%) was unknown.

 

There are no general findings in this literature review on the epidemiology of LAIs associating infection events with BSC use or design. It is noted, however, by Kimman et al (2008) that laboratory workers do get infected by biological agents moving out of the front of Class II cabinet.

HEPA Performance in relation to the capture of biological agents

HEPA filters are an industry standard developed in the 1940s for the US nuclear program and adopted by the biosafety sector during the early development of BSCs in the 1950s and 1960s. HEPA filters are designed to remove 99.97% particulates of the 0.3-micron size which represents the most penetrating particle size. This filtration efficiency is very effective in removing most bacterial or fungal biological agents and many viruses (refer to Figure 2). Abraham et al (1998) examined the test reports for 969 HEPA filters over 13 years and showed filters were exceptionally reliable with replacement rates of less than 3% per year.          

Early BSCs were developed using a HEPA 50% thinner than the nuclear 12-inch filter. There were no actual studies on microbial penetration for these early filters (First 1996). A study by Wen et al (2014) noted that there was a “lack of direct measurements on the efficiency of BSL-3 laboratory HEPA filtration systems against aerosolized bio-logical particles”. This study looked at the HEPA penetration of a small-sized bacteria Serratia marcescens (1µm x 0.5µm) and found 100% capture for correctly installed and undamaged HEPA filters. Two of the eight HEPA filters used in the study failed due to incorrect installation and filter damage.

 

During the COVID epidemic, there was a significant effort in the research literature to demonstrate the effectiveness of portable HEPA filters to reduce the spread of airborne viruses. Curtis (2022) noted that HEPA filters are a convenient and inexpensive way to remove many airborne pathogens in hospitals but that more research was needed for efficacy studies. The paper noted that filters were effective in the 0.012 – 19.81 microns range in reducing the spread of disease. Hammond et al (2021) looking at the use of HEPA filters in all disease-control applications concluded that there were no good studies showing a reduction in respiratory infections associated with HEPA filter use but that filters ‘in theory’ provide real-world solution.

The capacity of biological agents to penetrate a filter is not just a function of the particulate size. Pathogen-spreading viral aerosols are typically enveloped in water molecules rather than moving as individual viral particles. Similarly, while tuberculosis (TB) is only 0.2-0.5 µm in diameter, viable TB aerosols are in the size range of 0.65-4.7 µm and it is well within the capacity of the filters to achieve 100% capture (Fennelly et al 2012).

 

The size of SARS-CoV-2 ranges from 50 to 140 nanometres (nm) in diameter. This small size allows the virus to be easily transmitted through respiratory droplets and aerosols. Studies by Alsved et al (2023) detected COVID RNA in the range of 0.34 to >8.1 µm. Liu et al (2022) found peak concentrations of COVID in the 0.25–0.5 μm range. No published studies on HEPA penetration of viral particles were found as part of this review.

An upgraded Class II BSC design using UVC Air Disinfection

Class II BSCs have remained largely unchanged from their early designs developed in the 1950s-60s relying on HEPA filters and negative air pressures to limit the potential exposure of laboratory workers to biological agents. The popularity of Class II cabinets is a testament to their flexibility and effectiveness and a review of available literature on LAIs and HEPA filter performance suggests that their current design is fit for purpose. Nonetheless, several findings from the literature review suggest that there is a significant opportunity to improve on the Class II design:

  • The negative pressure and air barrier provide only partial protection to the laboratory worker from the opening in the front sash of the BSC (see Kimman et al 2008);
  • Research on HEPA filter effectiveness shows them to be highly effective when correctly installed and undamaged but otherwise ineffective (see Wen et al 2014);
  • Over 80% of LAIs have no explanation (see Harding and Byres 2006).

 

The application of UVC disinfection technology has rapidly evolved over the past 20 years with the more widespread use of UV in air conditioning ducts of hospitals to disinfect supply air to vulnerable patients. This same technology could readily be retrofitted into existing Class II BSCs in the rear plenum of the unit where air is returned from the working surface to the fan plenum (Figure 3). Currently we are not aware of any cabinet supplier offering this additional layer of protection in their cabinet designs.

The installation of this retrofitted UVC system would result in the following calculated mortality rates on a single pass for the following selected pathogens:

  • SARS-CoV-2 – Log 5 (100%);
  • Influenza A – Log 3 (99.97%);
  • Mycobacterium tuberculosis – Log 5 (100%).

 

Based on an internal volume of 2 cubic metres and an airflow rate of 753 m3/hr, it is estimated that any air within the unit will recirculate via the rear plenum over 300 times per hour ensuring 100% air disinfection of any known pathogens. The estimated electricity consumption costs associated with this UV retrofit is 1340 kWh per annum or approximately $450 in additional annual electricity costs. The installation of the lamps would have no impact on the laminar flow performance of the cabinet and cause minimal pressure drop which would cause the fan motors to work harder. A datasheet used for this proposed installation is provided in Appendix 2.

Conclusion

We examined the effectiveness of Class II BSCs to reduce the incidence of LAIs and have identified opportunities to improve their performance involving a relatively straight-forward retrofit of UVC lamps into the rear plenum of the BSC. The UVC installation would protect the user against any damage to HEPA filters or incorrect HEPA installation which could result in leakage of aerosolised pathogens into the laboratory. The report finds that Class II BSCs are a relatively effective means of primary containment of biological agents but have clear limitations if strict hygiene protocols are not followed allowing pathogens to escape into the laboratory.

Appendix 1 – BSC Classification Scheme (Source: BMBL 2020)

Appendix 2 – UV Disinfection Calculation (Sanuvox Fitting Software Version 1.6.2)

References

Abraham, G., Smith, P. M. L. B., & McCabe, P. (1998). HEPA filter replacement experience in a biological laboratory. Journal of the American Biological Safety Association, 3(4), 134-142.

Alsved, M., Nygren, D., Thuresson, S., Fraenkel, C. J., Medstrand, P., & Löndahl, J. (2023). Size distribution of exhaled aerosol particles containing SARS-CoV-2 RNA. Infectious Diseases, 55(2), 158-163.

Australian BC4 Guidelines 2007. Guidelines for Certification of a Physical Containment Level 4 Facility. Office of the Gene Technology Regulator Retrieved from: https://www.ogtr.gov.au/resources/publications/guidelines-certification-physical-containment-level-4-facility

Byers, K. and Harding L. (2020). Laboratory-Associated Infections. Chapter 4 in

Wooley, D. P., & Byers, K. B. (Eds.). (2020). Biological safety: principles and practices. John Wiley & Sons.

CDC Biosafety in Microbiological and Biomedical Labs (6th Edition 2020) NIH – very similar to WHO guidelines. Nothing terribly prescriptive around BSCs. Goes through many of the BSC types in US. Eg BSL 2 – aerosol generating activities should be in BSC. Same with BSL 3 – when handling infectious agents. BSL 4 to use an isolator.

Curtis, L. (2022). HEPA filters and airborne viruses, bacteria, and fungi. Otolaryngology–Head and Neck Surgery, 166(5), 1005-1005.

Fennelly, K. P., Jones-López, E. C., Ayakaka, I., Kim, S., Menyha, H., Kirenga, B., … & Ellner, J. J. (2012). Variability of infectious aerosols produced during coughing by patients with pulmonary tuberculosis. American journal of respiratory and critical care medicine, 186(5), 450-457.

First, M. W. (1996). Aging of HEPA filters in service and in storage. Journal of the American Biological Safety Association, 1(1), 52-62.

Hammond, A., Khalid, T., Thornton, H. V., Woodall, C. A., & Hay, A. D. (2021). Should homes and workplaces purchase portable air filters to reduce the transmission of SARS-CoV-2 and other respiratory infections? A systematic review. PLoS One, 16(4), e0251049.

Harding, A. L., & Byers, K. B. (2006). Epidemiology of laboratory‐associated infections. Biological safety: Principles and practices, 53-77.

Kimman, T. G., Smit, E., & Klein, M. R. (2008). Evidence-based biosafety: a review of the principles and effectiveness of microbiological containment measures. Clinical microbiology reviews, 21(3), 403-425.

Liu, Y., Ning, Z., Chen, Y., Guo, M., Liu, Y., Gali, N. K., … & Lan, K. (2020). Aerodynamic analysis of SARS-CoV-2 in two Wuhan hospitals. Nature, 582(7813), 557-560.

Siengsanan-Lamont, J., & Blacksell, S. D. (2018). A review of laboratory-acquired infections in the Asia-Pacific: understanding risk and the need for improved biosafety for veterinary and zoonotic diseases. Tropical medicine and infectious disease, 3(2), 36.

Wen, Z., Yang, W., Li, N., Wang, J., Hu, L., Li, J., … & Dong, X. (2014). Assessment of the risk of infectious aerosols leaking to the environment from BSL-3 laboratory HEPA air filtration systems using model bacterial aerosols. Particuology, 13, 82-87.

WHO biosafety manual reviewed  – 4th edition. 2020. Risk-based. Very little prescriptive focus. Prefers Class 1 to Class 2 – simpler. Recommends Class 3 in heightened control measures.

About the Author

Gregor Riese is an occupational hygienist and Director of Opira Group, a leading Australian pharmaceutical and laboratory supply and testing company. Since 2023, Gregor has served as a Director of the Board of the Association of Biosafety in Australia and New Zealand (ABSANZ). Opira distributes AS2252-complaint BSCs in the Australian market and is also the distributor of Sanuvox UV products.

Air Quality Testing Sydney Technician