Povidone-iodine hand wash and hand rub products demonstrated excellent in vitro virucidal efficacy against Ebola virus and modified vaccinia virus Ankara, the new European test virus for enveloped viruses

Four PVP-I antiseptic products were tested in this study. These included 4 % PVP-I skin cleanser, 7.5 % PVP-I surgical scrub and 10 % PVP-I solution, all with the brand name Betadine, manufactured by Mundipharma (Limburg, Germany), and a 3.2 % PVP-I/alcohol solution containing 78 % alcohol (2-propanol and ethanol), with the brand name Betaseptic, also manufactured by Mundipharma.

When using cell culture to evaluate disinfectant activity on viruses, the target cells are often very sensitive to the active ingredient. However this cytotoxic effect seen in vitro does not affect the way the products behave in vivo. The cytotoxicity of test products in disinfectant testing is commonly overcome by dilution of the virus-disinfectant mixture. In order to determine cytotoxicity, the test products were serially diluted ten-fold in MEM up to a dilution of 10⁻⁸. One part by volume of water of standardised hardness (instead of virus suspension) was mixed with one part by volume of organic load [clean conditions (0.3 g/l bovine serum albumin; BSA) or dirty conditions (3.0 g/L BSA + 3.0 mL/L erythrocytes)] and eight parts by volume of the test product. Aliquots of 100 μL from each test product at each dilution were then inoculated into six wells of a 96-well microtitre plate containing 200 μL cell suspension BHK-21 cells. The cell cultures were observed for cytotoxic effects for the same incubation time as was later used for the suspension tests.

Tests were carried out once, in accordance with EN14476:2013/FprA1:2015 at 20 °C ± 1 °C [7]. One part by volume of MVA virus suspension (titre of at least 10⁷–10⁸ tissue culture infectious dose 50 % [TCID₅₀]/mL) and one part by volume of the organic load were mixed with eight parts by volume of the PVP-I hand wash or hand rub product. The test products were examined undiluted and as 1:10 and 1:100 solutions. After the specified contact time (15 s, 30 s and 60 s), the virucidal activity was immediately suppressed by dilution with nine volumes of ice-cold medium (MEM + 2.0 % FCS) and without delay the assay was serially diluted 10-fold. Due to the immediate titration, no after-effect of the product could occur. Infectivity was determined by means of end point dilution titration in microtitre plates. Aliquots of 100 μL from each dilution were placed in six wells of a sterile polystyrene flat-bottomed 96-well microtitre plate containing 200 μL BHK-21 cells. Cultures were observed for cytopathic effects after eight days of inoculation. For the virus control, doubly distilled water was applied instead of the test product. All tests were performed under clean conditions (0.3 g/L BSA) and dirty conditions (3.0 g/L BSA + 3.0 mL/L erythrocytes) as interfering substance.

The virus titres were determined using the Spearman-Kärber method [24, 25] and expressed as TCID₅₀/mL, including standard deviation. Titre reduction is presented as the difference between the virus titre after contact with the test product and the control virus titre. This difference is given as a reduction factor including its 95 % confidence interval. A reduction in virus titre of ≥4 log₁₀ (corresponding to an inactivation of ≥99.99 %) was regarded as evidence of sufficient virucidal activity. The calculation was performed according to EN14476 [7].

Usually, only low titres of EBOV can be harvested in cell culture, resulting in a range of 5.00 to 7.70 log₁₀ TCID₅₀/mL. To demonstrate at least a 4 log₁₀ reduction in virus titre, it is necessary for test mixtures containing low virus titres to undergo detoxification by molecular sieving, or to use a more sensitive assay such as LVP [8]. Molecular sieving with Microspin S 400 HR columns is not the method of choice for EBOV because the process reduces the titres of these large, filamentous viruses by 1.00 to 2.00 log₁₀. In LVP, a high volume of the lowest apparently non-cytotoxic dilution of the inactivation assay test mixture is added to the detector cell line and the cultures are monitored for virus-specific effects. Using this method, larger reductions in virus titre can be shown even at lower viral loads.

The inactivation tests were conducted once at 20 °C ± 1 °C. One part EBOV suspension (100 μL) was mixed with 100 μL of either 0.3 g/L BSA (clean conditions) or 3.0 g/L BSA + 3.0 mL/L erythrocytes (dirty conditions) as the interfering substance. The virus-protein mixture was added to 8 parts (800 μL) of the 1:10 diluted test product, with the exception of 4 % PVP-I skin cleanser which was diluted to 1:100. After a contact time of 15 s, 600 μL of the test mixture for the 10 % PVP-I solution, 7.5 % PVP-I surgical scrub, and 3.2 % PVP-I/alcohol solution were added to 57.0 mL ice-cold medium resulting in a 1:96 dilution of the test product. For the 4 % PVP-I skin cleanser, 120 μL of the test mixture was added to 57.6 mL which corresponds to a 1:481 dilution. After dilution, the samples were added in 200 μL aliquots to microtitre plates (288 wells) containing the indicator cells in 100 μL cell culture medium. The cells were cultivated for an incubation period of 6 days, and inspected microscopically for virus foci (virus-induced changes in cell morphology) (Fig. 1). The viral titre was calculated as follows:

The test concentrations and contact periods were chosen in order to observe the point at which each test preparation produced efficient virus inactivation.

To demonstrate virucidal efficacy, disinfectant and antiseptic products are required to produce a log₁₀ reduction in virus titre of at least 4 [7]. The log₁₀ reduction factors produced by each test product under clean and dirty conditions at each time point are shown in Table 1.

Viral titres of MVA were reduced by >99.99 % to >99.999 % under clean and dirty conditions after application of undiluted and 1:10 dilutions of 10 % PVP-I solution, 7.5 % PVP-I surgical scrub and 4 % PVP-I skin cleanser for as little as 15 s. The 3.2 % PVP-I/alcohol solution had a similar effect, except under dirty conditions at the 1:10 dilution for 15 s, when the reduction factor was 3 (99.9 % reduction).

Under both clean and dirty conditions, a concentration-dependent virucidal activity was seen with the PVP-I only products (Table 1). Under dirty conditions, the minimum concentration of PVP-I needed for a 4 log₁₀ reduction of MVA was 5.3 times higher at a 30 s contact time (0.4 g/L versus 0.075 g/L) and 4 times higher at a 15 s contact time (0.4 g/L versus 0.1 g/L), than under clean conditions (Table 1).

The titres of EBOV present in the control samples were comparable between the test runs, with values around 7 log₁₀ TCID₅₀/mL. All of the PVP-I preparations tested reduced the EBOV viral titres by a log₁₀ reduction factor of between 5.66 and 6.84 after 15 s (Fig. 2). This corresponds to a reduction in EBOV viral titre of >99.999 % for all products tested.

Until recently, the European guidelines were lacking an enveloped test virus for use when evaluating hygienic hand rub and hand wash. Due to the recent Ebola outbreak, MVA was introduced as a reference virus for all enveloped viruses. However, there was still uncertainty about whether this surrogate would be suitable for dangerous viruses such as EBOV. In this study, we investigated the antiviral efficacy of PVP-I against MVA and then verified the concentration and contact time findings with EBOV. We found that PVP-I was as effective against EBOV as MVA with similar concentration and contact time parameters. In fact, when comparing the results with the 4 % PVP-I skin cleanser, which was pre-diluted to 1:100 for the EBOV test, MVA was even more resistant to PVP-I under dirty conditions than EBOV. Our results suggest that MVA is a suitable surrogate test virus that can be safely used for testing the virucidal efficacy of antiseptic products against EBOV.

Our study confirmed the excellent virucidal activity of PVP-I described in the literature. In one study by Sauerbrei and Wutzler [23], PVP-I solution and PVP-I/alcohol solution showed good efficacy against enveloped viruses (vaccinia virus and bovine viral diarrhoea virus [BVDV]) within 30 s. With the non-enveloped viruses tested, the PVP-I solution showed good efficacy within 5 min against polyomavirus SV40 and adenovirus, but took ≥60 min to inactivate poliovirus type 1. The PVP-I/alcohol solution showed better overall efficacy, especially against the non-enveloped viruses, causing significant inactivation of all test viruses within 5 min, and highlighting the synergistic virucidal activity of PVP-I and alcohol.

In another study by Steinmann et al. [27], a hand wash containing 7.5 % PVP-I was active against vaccinia virus and BVDV within 30 s, but not against the non-enveloped viruses tested. However, in a fingerpad test simulating practical conditions, the PVP-I containing soap was superior to other sanitisers against the non-enveloped viruses.

Although the minimum contact time for hygienic hand rub and hand wash defined in EN14476:2013 [7] is 30 s, we investigated the four PVP-I products with a minimum contact time of 15 s. We did this to study the kinetics of the PVP-I products against MVA over time, and because we assumed that a sufficient degree of effectiveness would not be shown after 15 s. However, all four undiluted preparations efficiently inactivated the European test virus MVA after as little as 15 s.

Protein load (e.g. blood or sputum) may have an influence on the activity of disinfectants, including PVP-I products [28, 29]. In our study, there was a relatively small difference in the virucidal activity of PVP-I when comparing clean and dirty conditions, and then only after 1:100 dilution of the products. Under dirty conditions, a 5-fold higher concentration of PVP-I was needed to inactivate MVA than under clean conditions.

The results of suspension tests only allow the virucidal efficacy of a product to be predicted, and do not give information about its effectiveness in practice. In order to further improve our knowledge of the time/concentration relationships of hand rubs and hand wash products which reflect daily needs in clinical surroundings, more practical testing would be necessary. EN1500 [30] describes a bactericidal test simulating practical conditions in volunteers, however, there is no enveloped harmless virus available that can be used to inoculate the volunteers. More than 30 years ago, tests with vaccinia virus on the fingertips of volunteers were performed [31], however today this can no longer be justified ethically.

Hand hygiene is an infection control procedure with clearly demonstrated efficacy, and remains the cornerstone of efforts to reduce the spread of infection. According to WHO guidelines, hand hygiene is the most important infection prevention and control measure against EBOV [6]. Generally, there are two methods of hand hygiene, social hand washing and hygienic hand disinfection [32, 33]. Socially clean hands are achieved using soap and water, which removes transient microorganisms. Social hand washing should be practised before and after performing routine tasks in all clinical areas. Hygienic hand disinfection is carried out with alcohol-based hand rubs toto remove or destroy all transient microorganisms, and may have a prolonged effect. It is recommended, for example, during outbreaks of infections, and after contact with body fluids or infectious material.

Since the skin is a routine source of infection transmission, the use of PVP-I for hand disinfection may be an alternative to WHO recommended alcohol-based hand rubs. Medicated soaps containing PVP-I are available. Thus, the hygiene measure of hand washing is supported by the virucidal effect of PVP-I. This could be extended beyond surgical use. Healthcare professionals, patients, family members and professionals at risk of infection could use these products to protect themselves after exposure to patients.

PVP-I soap has shown similar efficacy to alcohol-based hand sanitisers against norovirus in a hand wash model [27]. Chlorhexidine- and triclosan-based hand washes, however, were not effective against norovirus in suspension tests and practical application tests [27].

In another study in a neonatal ICU setting, a PVP-I based hand wash showed antimicrobial efficacy comparable to an alcohol-based hand rub, with both being superior to use of soap and water alone [34]. In relation to EBOV and similar viruses, PVP-I has been recommended as a hand disinfectant in laboratories dealing with samples collected from patients with suspected viral haemorrhagic fevers [35].

The results of our study support these findings, and suggest that hand washing with PVP-I based products can help to prevent infection with EBOV. In addition to hand washing, PVP-I preparations may have other applications in reducing the spread of EBOV, as they are not limited to use on intact skin but can also be administered to the mucosa of the mouth, nose, eye, urinary and genital tract [36, 37]. Treatment of these potential entry points for EBOV and other viruses may not be appropriate for other preparations, such as more highly concentrated alcohol-based products.