Norovirus Cross-contamination with/without Gloves

March 2013

Humans infected with noroviruses can harbor and shed in feces up to 1011 genomic copies of norovirus/g when ill and may also shed the pathogen when asymptomatic. Gloves are often worn as a protective barrier when handling ready-to-eat foods but they cannot prevent cross-contamination if the outside of the gloves become contaminated.

We did a study to determine rates of transfer of norovirus to gloves from bare hands during glove application. Norovirus transfer from contaminated hands or gloves to stainless steel surfaces and to leaves of lettuce were quantified, and the extent to which sequentially touched lettuce leaves can become contaminated was also determined. Transfer of a murine norovirus (MNV-1) surrogate was compared to the transfer of human norovirus (genotypes I [GI] and II [GII]). Porcine skin was used as a model for human fingertips. Quantitative transfer data are important for risk modeling and assessment.

Using a mixture of viruses (human GII and GI viruses and MNV-1) suspended in 10% stool as an inoculum, virus transfer from porcine skin to gloves (thin latex, thick latex, and nitrile) (n=9 each) and from porcine skin and gloves (thin latex) to stainless steel surfaces (n=12 each) was measured after interfacing the two surfaces for 10 seconds using a mechanical transfer device (1,000 g/4.4 cm2). A stool suspension of MNV-1 was used to quantify the extent of contamination after inoculating a donor surface (porcine skin or glove; n=6 each) and sequentially touching ten leaves of fresh iceberg lettuce. Viruses were eluted from recipient surfaces and real-time RT-qPCR was used to determine virus recovery.

The transfer rates of human norovirus from porcine skins to gloves, in many cases, approached 50%, demonstrating the ease with which virus can transfer from contaminated hands to gloves during application. Rates of transfer of GII, GI, and MNV-1 from porcine skin to thin latex were 38%, 33%, and 35%, respectively; to thick latex, 51, 52, and 41%, respectively; and to nitrile gloves, 35, 36 and 29%, respectively. The transfer rates of GII, GI, and MNV-1 from porcine skin to stainless steel were 39, 42, and 43%, respectively, but transfer rates were higher significantly from thin latex gloves to stainless steel (53, 61, and 57% for GII, GI and MNV-1, respectively).

Rates of norovirus transfer to the first iceberg lettuce leaf surface touched by contaminated thick latex gloves, nitrile gloves, and porcine skin were 54, 38, and 42%, respectively. To the second and third lettuce surfaces, 20 to 23% and 8 to 13%, respectively, were transferred from for thick latex gloves, nitrile gloves, and porcine skins. After touching five lettuce leaves, transfer rates were 2 to 3% for all surface types. After touching ten lettuce leaf surfaces, norovirus was still detected on all replicate samples, but rates of transfer decreased to less than 1%.

Observations in this study will be useful for developing quantitative models for estimating the risk of human norovirus contamination of foods and food-contact surfaces by food handlers if hand hygiene is inadequate. The study also highlights the importance of good hand hygiene prior to glove application, as viruses can be transferred from bare hands to the outside surface of gloves.

The lack of a significant difference in transfer rates when comparing human and murine noroviruses suggests MNV-1 is a good surrogate for transfer studies. Similarities between porcine skins and human hands will be investigated in the future. Human volunteers will be recruited for conducting similar norovirus transfer experiments.