Do Hand Sanitizer and Liquid Hand Soap Remove Viruses?

While reading my back issues of Applied and Environmental Microbiology (AEM), I came across an interesting paper that detailed an in-depth study on the effectiveness of hand cleaners to remove Norwalk virus (NV) from intentionally contaminated hands.

Yes that’s right – intentionally contaminated, and how. The study volunteers allowed a 20% stool suspension containing Norwalk virus to be pipetted onto their fingers and then examined how well the different hand sanitizers and liquid soap removed the virus. Nice.

Because of the uniqueness of the study and because hand santizers are everywhere now, even at the bank to disinfect hands after touching the “community pen”, I thought it would be interesting to review this paper and summarize the results.

The paper is called Effectiveness of Liquid Soap and Hand Sanitizer against Norwalk Virus on Contaminated Hands by Pengbo Liu, Yvonne Yuen, Hui-Mien Hsiao, Lee-Ann Jaykus, and Christine Moe. All authors were from the Rollins School of Public Health at Emory University in Atlanta except for Lee-Ann Jaykus who is from the Dept. of Food Science at North Carolina State University in Raleigh, NC. The paper was published January 2010, vol. 76, No. 2, p. 394-399.


Norwalk virus infects 23 million people a year and is the cause of 81% of non-bacterial gastroenteritis outbreaks. While the virus can contaminate inanimate surfaces, hand to hand transmission is thought to be the primary vehicle for spread of infection. Hand sanitizers make no virucidal claims against noroviruses because proving it is difficult. The virus can’t be cultured so experiments need to be performed with clinical samples.  And that is what this research team did. The authors infected the fingers (called “finger pads”) of volunteers with human Norowalk virus and then used RT-qPCR to evaluate the effectiveness of liquid hand soap, sodium hypochlorite (bleach), ethanol, and water for inactivation of the virus and removal from the surface of skin.


Human volunteers: Norwalk virus was obtained from the stool of an experimentally infected human volunteer (who would volunteer for that?) and then their stool was collected, tested for viral load, and diluted 20% (w/v) in RNase-free water prior to inoculation onto the finger pads of volunteers.

Ten volunteers enrolled in the study to be infected with Norwalk virus on their hands. Five volunteers participated in two trials on separate days and the right and left hand served as duplicate treatments.

Suspension assays to test for virucidal activity: were performed using ethanol at 3%, 17%, 31%, 47%, 62%, and 95% strength and hypochlorite solution (bleach) at 3 ppm, 22 ppm, 51 ppm, 160 ppm, and 1,600 ppm.  Ten microliters of the 20% suspension was added to 990 ul of each solution and after neutralization, virus was recovered for RT-qPCR analysis.

Disinfectants: Hand sanitizer was purchased locally which contained 62% ethyl alcohol and antibacterial liquid hand soap from Fisher Scientific contained 0.5% triclosan. Both of these reagents were pre-tested for PCR inhibition properties.

Inoculation of human volunteers: Volunteers first cleaned their hands with soap and ethanol and then allowed them to dry. Ten microliters of the 20% Norwalk virus infected stool solution was placed on the fingers of the volunteers.

Recovery of NV from the fingers of volunteers: A 2 ml plastic vial containing 1 ml of HBSS was used to cover the area on the finger where the NV was inoculated and inverted 20 times to elute the virus from the finger.

Treatments of NV on human finger pads: After inoculation of the fingers with the 10 ul of NV stool suspension, fingers were air dried and then exposed to 1 ml of hand sanitizer or liquid hand soap in the 2 ml vial for 20 seconds. Excess cleaner was scraped off into the vial. Following this, the same finger was then placed over the 2 ml tube containing 1 ml HBSS to elute remaining virus into the test vial. For liquid hand soap, the finger pad was rinsed in 1 ml of water for 10 seconds before eluting remaining virus into the HBSS vial.

The pinky fingers were used to determine the effectiveness of removing NV with water alone. Water in a 2 ml vial was used to wash the finger pad by inversion and then the HBSS was used to collect eluates. All finger pads samples were stored at -80C until assay.

Hydrolysis Probe assays: Samples were tested with and without RNase treatment (to differentiate live, infectious virus vs. dead virus) and analyzed using a one-step RT-PCR kit on the Mx3000P instrument. Quantification was done using the full-length cDNA cloned into a plasmid used for a standard curve. The assay was linear in the range of 1.3 copies to 4.3 Log 10 copies of NV DNA.


Virucidal suspension assays: Sodium hypochlorite showed a concentration dependent inactivation of virus between 22 ppm and 160 ppm after 30 seconds. The virus was reduced 5 logs. Ethanol, on the other hand, did not show a concentration dependent effect of inactivation of virus. The most reduction in virus was 0.5 Log 10 at 17% ethanol, but this sample had the least consistency between replicates.

Results of hand cleaners: The results reported were for the RNase treated samples only, so this data reflects the effects of the cleaners on live infectious virus. Compared to the baseline control, the liquid hand soap had the greatest reduction in NV cDNA (0.67 Log 10), followed by the reduction from just a simple water rinse (0.58 Log 10). The alcohol based sanitizer reduced the virus the least, only 0.27 Log 10.

To determine if rubbing the test reagent on the finger pads helps reduce the presence of virus, volunteers repeated the experiment with rubbing for 10 seconds of the hand soap, hand sanitizer, or water on the finger pad before eluting the left over virus into HBSS. This did increase the Log 10 reduction of NV. The greatest reduction was seen with just the water rinse (1.38 Log 10 reduction) followed by antibacterial soap with 1.10 Log 10 reduction). The ethanol based hand sanitizer was the least effective with only a .34 Log 10 reduction of virus with rubbing and was not statistically significant compared to the baseline control.


Using a suspension assay to analyze for virucidal effects, the sodium hypochlorite solution demonstrated a concentration dependent increase in viral killing with virus elimination at >160 ppm, while ethanol had no effect, even at 95% strength.  For enveloped viruses, such as influenza, hepatitis B, and herpes simplex 1, ethanol-based sanitizers are known to be effective. However, the enteric nonenveloped viruses including hepatitis A, poliovirus, and FCV, are resistant to the effects of ethanol.

Additional research by others has shown that enteric viruses can be inactivated using a combination of solutions that include isopropanol, citric acid, or phosphoric acid along with ethanol. Because this virus is typically shed in very high titers by infected individuals and predominantly affects the food service industry, commercial development of a virucidal sanitizer designed specific for the enteric viruses is needed.


My first thought was that of course the liquid hand soap performed better because it was not only soap but also water rinsed to help remove the left over soap before eluting into HBSS. These fingers received the benefit of a double cleaning. We saw that water alone was very effective so perhaps the water following soap helped to solubilize more virus. However, water alone with rubbing has the best reduction in virus and the rubbing with soap followed by water does not have an additive effect, so this may not be a factor.

A good control would have been to perform the ethanol treatment followed by a water rinse, exactly the way they did the liquid hand soap samples, to negate any amplification in virus killing due to the combination.  But, it makes sense that if you use liquid hand soap, you will always rinse in water too. Whereas if you use a liquid hand sanitizer, you will not.

As for the commercial development of better hand sanitizers that protect against viruses spread by fecally contaminated sources, perhaps the cruise industry would be interested in providing research funding for the development of anti-Norwalk virus reagents that are safe for both food handlers and for food preparation surfaces.



  1. Kurt Lager on May 28, 2010 at 4:39 am

    Although animals such as pigs can be infected by norovirus, it’s not the same strains that infect humans, and that’s also one of the reasons it’s not possible to grow the virus. There’s been speculation of asymptomatic people that would act as a reservoir, but no such evidence has been found, water is an important reservoir. The cases where oysters concentrate virus particles and give infections show that, otherwise infections is probably a lot more common than officially reported, so that virus is constantly circulating, only causing real outbreaks when people are more crowed, in hotels, cruise ships and seasonally in wintertime when people are more indoor .

  2. Suzanne Kennedy on May 27, 2010 at 2:55 pm

    I know. I don’t know if we want to know.

    Kurt- do you know what is the reservoir for the Norwalk virus?

  3. Roberto Rosati on May 27, 2010 at 1:22 pm

    The fingertip inoculum is curious, but even more curious to me is… if the virus cannot be cultured, by which method was the human volunteer infected with the virus? (…ew?)

  4. Suzanne Kennedy on May 27, 2010 at 4:52 am

    Ah- so it all starts with one person having it right? So is that a crew member or a passenger? Is the person who starts it all asymptomatic?

  5. Kurt Lager on May 27, 2010 at 4:13 am

    The cruise ships have very many people in a small area. And I think it’s mainly spreading between people there, and from surfaces to people. If you have textiles and other soft surfaces it’s very difficult to disinfect these if there’s a lot of virus particles there, from people vomiting! A classical case of norovirus outbreak was during a a college football game in Florida, where the two teams shared no food, all transmission was from person-to-person .

    Becker KM, Moe CL, Southwick KL, MacCormack JN. Transmission of Norwalk virus
    during football game. N Engl J Med. 2000 Oct 26;343(17):1223-7. PubMed PMID:

  6. Suzanne Kennedy on May 26, 2010 at 2:30 pm

    Hi Kurt,
    I am at ASM this week and spoke with a scientist working with raspberries and norovirus yesterday. Interesting- the hand picked foods may be the more susceptible ones.
    But what happens on the cruise ships? It’s in the water, right?

    Gardo- I totally agree about gloves. Seeing food servers wearing gloves does not make me feel any more safe. I think the protection is more for them than for the customers.

    All: Maybe there needs to be a fast norovirus test so that employees that are positive are not working in the food industry until they are well. If it is shed in high titers, it should be detectable some how. I don’t want my food exposed to a whole lot of chemicals either.

    What do you think?

  7. Edgardo Rivera Delgado on May 26, 2010 at 1:40 pm

    Interesting. What would be a good portable idea to solve this problem? A higher content of water in Hand Sanitizers or perhaps a high water content soap based hand sanitizer with non-sticky soap? The tradeoff will be that more bacteria might survive under these conditions. The best option still seems to be washing your hands with plenty of warm water and hand soap. Yet nowadays the trend in restaurants is people serving food with plastic gloves to make the customer think they are being clean. I’ve seen those same gloves go into their pockets and back to their hands. I always wonder if because of gloves in restaurants people have started cutting back on washing their hands. To my understanding properly washed bare clean hands are more antiseptic than plastic ‘sterile’ gloves for serving food. In lab the same thing applies, clean hands plus clean gloves.

  8. Kurt Lager on May 26, 2010 at 4:59 am

    The 23 million annually infected people is just within US, and not the world. A funny thing with norovirus is food borne infection by norovirus and how common raspberries are as source! Raspberries have a very big surface and they have to be hand picked! Certain human genotypes have been associated with genetic resistance to symptomatic norovirus infections, but now it seems as if the virus has evolved to also infect these people.

  9. Suzanne Kennedy on May 26, 2010 at 4:29 am

    I like how the citation on that headline image is “brain malfunction”. That picture does look like someone had one. 🙂

Leave a Comment