Basics of UV light

Wavelengths of UV light, a form of electromagnetic radiation from the sun, are divided into 3 categories: UV-C (100–280 nm), UV-B (280–315 nm), and UV-A (315–400 nm). UV-C light is absorbed by the earth’s atmosphere and does not reach the earth. UV can be generated from artificial sources including low-pressure mercury lamps and pulsed xenon lamps. Some mobile devices use low-pressure mercury lamps to produce continuous UV-C with a peak wavelength of 254 nm,Reference Rutala, Gergen and Weber¹⁰ while others use pulsed-xenon lamps that produce light with a much broader spectrum (200–320 nm).^26, Reference Cadnum, Jencson and Gestrich²⁷

UV irradiance (also called fluence rate) is often expressed as microwatts per square centimeter (uW/cm²). The dose of UV received at a surface during a specific time (fluence) is calculated by multiplying the irradiance times the number of seconds of exposure at a surface and is expressed as Joules/m², mJ/cm², or uWsec/cm². For example, 1 J/m² equals 0.1 mJ/cm² or 100 uWsec/cm². Variables that affect the amount of UV delivered to surfaces and the resulting log₁₀ reductions of pathogens achieved include the amount of irradiance generated by the UV lamp(s), the distance from the lamp to the exposed surface, the angle at which the UV strikes the surface, and whether the surface is in direct line of sight of the lamp or receives light that has been reflected off other objects (ie, surfaces in shaded areas).Reference Nerandzic, Fisher and Donskey^7, Reference Boyce, Farrel, Towle, Fekieta and Aniskiewicz^12, Reference Cadnum, Tomas and Sankar^17, Reference Tande, Pringle, Rutala, Gergen and Weber²³ The intensity of UV irradiance is inversely proportional to the square of the distance between the lamp and the exposed surface, based on the “inverse square law.”Reference Simmons, Dale, Holt, Velasquez and Stibich²⁸ Therefore, the UV intensity received by a surface decreases exponentially the further the surface is from the lamp. As the distance between the lamp and the exposed surface increases, the log₁₀ reduction of pathogens achieved decreases significantly.Reference Nerandzic, Fisher and Donskey^7, Reference Boyce, Farrel, Towle, Fekieta and Aniskiewicz^12, Reference Cadnum, Tomas and Sankar¹⁷ One practical implication of the marked effect of distance on UV irradiance is that short cycles of 1–2 minutes or less may be sufficent for devices that place bulbs in close proximity to the item to be decontaminated (eg, computer tablet decontamination devices), whereas longer cycles are required for room decontamination devices.

UV-C irradiance and doses delivered to various surfaces in patient rooms

Figure 1 shows the UV-C doses delivered by a 5-minute cycle with a mobile UV-C device to various locations in a patient room based on measurements taken with a radiometer and the corresponding log₁₀ reductions in methicillin-resistant Staphylococcus aureus (MRSA) and C. difficile spores. The results demonstrate the dramatic impact of distance, orientation of surfaces, and shading (ie, indirect exposure to UV-C) on the UV-C dose delivered. Irradiance levels of ∼1,000 uW/cm² were achieved on vertical surfaces located in the direct line of sight of the device at distances in the range of 1.3 m (4 feet), whereas irradiance of only 3–10 uW/cm² were achieved on horizontal surfaces located in shaded areas at distances in the range of 3.3 m (10 feet) from the device.Reference Boyce, Farrel, Towle, Fekieta and Aniskiewicz^12, Reference Rutala, Gergen, Tande and Weber^22, Reference Tande, Pringle, Rutala, Gergen and Weber²³ Despite the marked dropoff in UV-C dose in shaded areas, 2 log₁₀ reductions were achieved for MRSA at both 1.3 and 3.3 m. In contrast, C. difficile spores are relatively resistant to UV-C. Minimal reductions in C. difficile spores were achieved in shaded areas, and both distance and orientation of the carriers had a more marked impact on the efficacy of UV-C. Tande et alReference Tande, Pringle, Rutala, Gergen and Weber²³ reported similar findings for the UV-C doses delivered to various locations in a room, and other studies have also demonstrated that UV-C doses are lowest when the surface is not in the direct line of sight of the lamp.Reference Boyce, Farrel, Towle, Fekieta and Aniskiewicz^12, Reference Cadnum, Tomas and Sankar¹⁷

Fig. 1.

Facilities using UV-C room decontamination devices can consider several strategies to reduce the impact of distance and shadowing on UV-C delivery to surfaces in patient rooms. Some manufacturers of UV-C devices recommend that 2 cycles be run with device placement in different room locations. A device with 3 vertical towers positioned in 3 locations in the room to reduce shadowed areas and distance from the UV bulbs is also available. A different device has 3 adjustable bulbs that can be oriented to provide closer proximity to the surface of interest. Another option is to run longer cycles to increase the UV-C dose delivered to shadowed areas. One study reported that on horizontal surfaces in shaded areas at a distance of 3.3 m, increasing the cycle time from 5 to 15 minutes increased the mean UV-C dose from 3,395 and 8,880 uWsec/cm².Reference Boyce, Farrel, Towle, Fekieta and Aniskiewicz¹² Finally, coating room walls with nanostructured UV-reflective paint increased UV doses delivered to shaded areas, resulting in greater log₁₀ reductions than with standard wall paint.Reference Tande, Pringle, Rutala, Gergen and Weber^23, Reference Rutala, Gergen, Tande and Weber²⁹

Practical methods to measure UV-C doses delivered to surfaces in patient rooms

One limitation of current UV-C technologies is that practical tools are not available to monitor the efficacy of the devices. Such tools would be useful to provide comparative data for different devices, to assess delivery of UV to different sites in patient rooms, and to confirm that devices are operating correctly. The intensity of UV-C irradiance and the dose received at a point in the room can be measured quantitatively using calibrated radiometric UV-C sensors, which range in price from several hundred dollars to >$1,000 each. However, measurement of irradiance is not practical for routine monitoring. Inexpensive, commercially available, qualitative photochromic UV-C test cards placed at different points in the room can provide rough estimates of the UV-C doses delivered to surfaces.Reference Boyce, Farrel, Towle, Fekieta and Aniskiewicz^12, Reference Masse, Hartley, Edmond and Diekema³⁰ However, additional validation of the accuracy of UV test cards and more detailed instructions on their use are needed. Measuring irradiance and UV doses from pulsed-xenon devices requires the use of a spectrometer.Reference Cadnum, Jencson and Gestrich²⁷

Studies of UV-C doses yielding significant reduction of pathogens on surfaces

Because the extent to which pathogens must be reduced to prevent transmission is unknown, there are no widely accepted criteria regarding the level of pathogen reduction that UV-C devices should achieve. Because high-touch surfaces in patient rooms are often contaminated with <100 colony-forming units (CFU) per replicate organism detecting and counting (RODAC) plate (∼25 cm²) or per swab sample,Reference Rutala, Gergen and Weber^10, Reference Nernandzic, Cadnum, Pultz and Donskey^11, Reference Wong, Woznow and Petrie^31–Reference Boyce, Havill and Dumigan³³ we elected to assess the UV-C doses needed to produce a 99.9% (3 log₁₀) reduction of microorganisms on surfaces, a criterion for effectiveness used by others.Reference Nerandzic, Fisher and Donskey^7, Reference Rutala, Gergen and Weber¹⁰

A literature review revealed few data regarding the doses of UV-C that yield specific log₁₀ reductions of pathogens on dry surfaces. Studies varied substantially in the methods used, and not surprisingly, UV-C doses yielding ≥3 log₁₀ reductions varied considerably (Table 1). For example, in a study in which C. difficile was exposed to a collimated beam of UV, a dose of 16,000 uWs/cm² yielded a 3 log₁₀ reduction.Reference Mitchell, Sifuentes, Wissler, Abd-Elmaksoud and Gerba²⁴ However, in other studies utilizing mobile devices, doses ranging from 67,567 to > 300,000 uWs/cm² produced ≥3 log₁₀ reductions (Table 1).Reference Boyce, Farrel, Towle, Fekieta and Aniskiewicz^12, Reference Alhmidi, Cadnum, Piedrahita, John and Donskey³⁴ The variability may have been due to differences in test methods used, including the strains tested and the presence or absence of an organic soil load (Table 1).

Table 1.

Log₁₀ Reductions of Different Strains of Clostridium difficile, Methicillin-Resistant Staphylococcus aureus (MRSA), and Vancomycin-Resistant Enterococcus (VRE) Achieved Following Exposure to Various Ultraviolet (UV) Doses (uWs/cm²) Under Varying Conditions

^a Organic load.

^b No organic load.

^c Personal communication with J.B. Mitchell.

^d Methicillin-susceptible S. aureus.

Laboratory variables affecting reduction of pathogens achieved with UV-C

UV-C doses that yield a 3 log₁₀ reduction of a specific pathogen vary considerably, due in large part to differences in the in vitro laboratory techniques used by investigators (Table 2). For example, methods for preparing the bacterial inoculum differ, particularly with respect to preparing C. difficile spores for testing.Reference Rutala, Gergen and Weber^10–Reference Boyce, Farrel, Towle, Fekieta and Aniskiewicz^12, Reference Mitchell, Sifuentes, Wissler, Abd-Elmaksoud and Gerba²⁴ Also, to achieve a given log₁₀ reduction, using a higher inoculum (eg, 10⁶) may require greater UV-C doses than a lower inoculum (eg, 10⁴).Reference Moore, Ali and Cloutman-Green²¹ Spreading the inoculum over the entire surface of a carrier may yield greater log₁₀ reductions than if the inoculum is applied as droplet without spreading,Reference Cadnum, Tomas and Sankar¹⁷ and applying test microorganisms onto a carrier 40 mm or 20 mm in diameter may yield greater log₁₀ reductions than using a carrier 10 mm in diameter.Reference Cadnum, Shaikh and Piedrahita³⁵ These issues are likely due in part to clumping or “stacking” of organisms (especially spore-forming organisms) on carriers, which can vary with the method of applying the inoculum to carrier.Reference Raguse, Fiebrandt and Stapelmann³⁶ For a given UV-C dose, placing an inoculum onto glass slides and onto stainless steel disks may yield different log₁₀ reductions if the inoculum is not spread over the surface of the carriers.Reference Cadnum, Tomas and Sankar¹⁷ In at least 1 study, UV-C doses required to achieve a 3 log₁₀ reduction were not the same when stainless steel and Formica carriers were used.Reference Mitchell, Sifuentes, Wissler, Abd-Elmaksoud and Gerba²⁴ The presence of an organic load on carriers results in lower log₁₀ reductions than if no organic load is used.Reference Nerandzic, Fisher and Donskey^7, Reference Cadnum, Tomas and Sankar^17, Reference Moore, Ali and Cloutman-Green²¹

Table 2.

Variables Affecting Ultraviolet C (UV-C)Doses Required to Reduce Microorganism Counts

Strain-specific susceptibility to UV-C may also affect the doses necessary to achieve a given log₁₀ reduction. For example, in studies that exposed microorganisms suspended in liquid, the UV-C doses needed to yield a 3 log₁₀ reduction of different strains of Escherichia coli reported by the same investigator varied from 1 mJ/cm² to 5.5 mJ/cm² (1,000 to 5,500 uWsec/cm²).Reference Sommer, Lhotsky, Haider and Cabaj³⁷ However, investigators who exposed different ATCC strains of E. coli under similar conditions reported that higher UV-C doses of 7 mJ/cm² to 9 mJ/cm² were needed to yield 3 log₁₀ reductions.Reference Malayeri, Mohseni, Cairns and Bolton²⁰ Strain-related differences in the susceptibility of C. difficile to UV-C have also been reported.Reference Nernandzic, Cadnum, Pultz and Donskey^11, Reference Moore, Ali and Cloutman-Green^21, Reference Rutala, Gergen, Tande and Weber^22, Reference Mitchell, Sifuentes, Wissler, Abd-Elmaksoud and Gerba²⁴

Humidity levels also affect the impact of a given UV-C dose on the survival of microorganisms. For example, FletcherReference Fletcher³⁸ reported that with a dose of 5 J/m², only 9% of Burkholderia cepacia survived at a low humidity, whereas 50% survived at higher humidity levels. UV-C is generally less effective at low temperatures.Reference Chen, Chu, Liu and Hu³⁹ The method of recovering viable microorganisms from exposed carriers (eg, carrier submerged in liquid with vortexing, recovery using a swab, or by using a RODAC plate) may also have affected the apparent log₁₀ reductions achieved.

Other factors affecting UV-C efficacy

Jelden et alReference Jelden, Gibbs and Smith⁴⁰ inoculated MRSA or VRE onto carriers made of stainless steel, chrome, and plastic from a bedrail and placed the carriers in 5 sites in a hospital room with standard paint, resulting in carriers receiving direct (ie, line of sight) UV-C doses ranging from 9 to 688 mJ/cm² (9,000– 688,000 uWs/cm²). Carriers placed in a room with UV-reflective paint received doses ranging from 2 to 553 mJ/cm². Because data on log₁₀ reductions were aggregated for the 5 sites and reported for each type of carrier, it was not possible to determine log₁₀ reductions achieved by specific UV-C doses. Greater log₁₀ reductions were achieved when carriers were located in the room with UV-reflective paint, as reported by others.Reference Rutala, Gergen, Tande and Weber^29, Reference Jelden, Gibbs and Smith⁴⁰ Notably, log₁₀ reductions were lower on carriers made of plastic from a bedrail. Tande et alReference Tande, Pringle, Rutala, Gergen and Weber²³ also studied the impact of UV-reflective paint on log₁₀ reductions achieved on 10 different surfaces by a mobile UV-C device. Although their report suggests that overall average UV-C doses required to achieve mean log₁₀ reductions of 4.1 for MRSA and 3.4 for C. difficile on the 10 directly exposed surfaces were very high,Reference Tande, Pringle, Rutala, Gergen and Weber²³ the investigators found that 3.0 log₁₀ reductions were consistently achieved with doses >10,000 uWs/cm² for MRSA and >50,000 uWs/cm² for C. difficile (B.M. Tande, personal communication).Moore et alReference Moore, Ali and Cloutman-Green²¹ exposed 8 types of medical equipment to UV-C in an enclosed cabinet and found that log₁₀ reductions varied with the type of surface. Also, tympanic thermometers and blood pressure cuffs yielded lower log₁₀ reductions than other objects.

The impact of UV-C on pathogens on solid surfaces has also been studied utilizing a mobile device with 8 sensors. The device automatically turns off when the sensor facing the least reflected area of the room has received a reflective dose of 12,000 uWsec/cm² (for vegetative bacteria) or 22,000 uWs/cm² (for spores).Reference Rutala, Gergen and Weber¹⁰ Importantly, these programmed reflected doses detected by device sensors (a) are not the same as much higher doses (not specified) received by some surfaces in patient rooms that receive high-intensity UV-C from the deviceReference Tande, Pringle, Rutala, Gergen and Weber^23, Reference Rutala, Weber, Gergen, Tande and Sickbert-Bennett⁴¹ and (b) should not be confused with doses needed to reduce specific pathogens by 3 log₁₀.Reference Masse, Hartley, Edmond and Diekema³⁰ Notably, one study that utilized the device found that a reflected dose of 20,000 uWs/cm² yielded a 3 log₁₀ reduction of the NAP1/BI (027) strain of C. difficile.Reference Nerandzic and Donskey⁴²

A few articles have reported the susceptibility of pathogens exposed to UV-C under conditions other than inocula dried onto solid carriers or suspended in liquid.

For example, SharpReference Sharp⁴³ and Cairns et alReference Cairns, Kerr and Beggs⁴⁴ irradiated bacteria inoculated onto solid agar in Petri plates, but it is not clear whether this approach yields the same log₁₀ reductions as irradiating dried pathogens on hard surfaces.

A mobile device that utilizes pulsed-xenon technology instead of UV-C has been used in a number of facilities to reduce environmental contamination and/or HAIs.Reference Haas, Menz, Dusza and Montecalvo^14, Reference Stibich, Stachowiak and Tanner^45, Reference Zeber, Pfeiffer and Baddley⁴⁶ However, published data regarding spectrophotometrically determined doses of 200–320 nm light that yield 3 log₁₀ reductions of specific pathogens are lacking.

Attempts to standardize methods for evaluation of mobile UV devices

A standard method (ASTM E2839-11) for preparing spores of C. difficile strain ATCC 43598 for use in evaluating the efficacy of antimicrobial agents has been endorsed by the Environmental Protection Agency (EPA).^{47, 48} This method could be used as part of a protocol for evaluating log₁₀ reductions achieved by mobile UV devices. However, only 1 of the studies cited in Table 1 prepared spores from that strain of C. difficile, and not all other aspects of spore preparation mirrored the ASTM method.Reference Alhmidi, Cadnum, Piedrahita, John and Donskey³⁴ An ASTM standard method (ASTM E 3135-18) for testing the efficacy of UV (UV-C or pulsed-xenon) against microorganisms on carriers with simulated soil has been published recently.⁴⁹ However, this standard does not specify the pathogens, type of carrier, or culture media that are to be used. The EPA has recommended a different quantitative method for testing antimicrobial products against spores of C. difficile on hard, nonporous surfaces.⁴⁸ However, challenges in obtaining standardized uniform preparations of C. difficile spores raise the question of whether Bacillus subtilis, a nonpathogenic aerobic spore former also relatively resistant to UV-C, could serve as a surrogate when testing the efficacy of UV devices.Reference Raguse, Fiebrandt and Stapelmann^36, Reference Lindsley, McClelland and Neu⁵⁰ Unfortunately, currently recommended protocols do not provide a standardized method for evaluating the microbiological efficacy of mobile UV devices, a need previously expressed by investigators.Reference Cowan^18, Reference Nerandzic and Donskey¹⁹ Notably, a subcommittee of the International Ultraviolet Association (IUVA) is currently in the process of developing proposed standard protocols for testing mobile UV devices.Reference Martinello^51–Reference Miller and Jillavenkatesa⁵⁴

In conclusion, information about the doses of UV needed to achieve 3 log₁₀ reductions of healthcare-associated pathogens that frequently contaminate environmental surfaces in healthcare facilities is scarce, and the available data are inconsistent. Healthcare facilities that are considering the use of mobile UV devices as an adjunct to standard manual disinfection of inanimate surfaces would benefit from knowing the UV doses required to achieve desired log₁₀ reductions of pathogens and from having accurate and inexpensive means of determining whether those doses have been achieved on various surfaces in patient rooms.