JHI letter damns Wilson’s Deprox test

A letter published on May 31st in the Journal of Hospital Infection shows Prof Peter Wilson’s controversial Deprox test results can be attributed to the very high level of (now illegal) silver nitrate in the Deproxin disinfectant solution. Wilson’s paper had already been strongly criticised by Dr Jon Otter of Imperial College, who suggested that Deprox manufacturer Hygiene Solutions Ltd. had added “A dash of peracetic acid” to the mix in order to achieve the improbable results.

However, as Dr Singh points out in the letter, the explanation is that Deproxin contains a whopping 2000ppm of silver nitrate, forty times as much as the ASP Glosair system that Dr Otter was comparing it to. The spray of silver nitrate solution settles on the BIs (conveniently unpouched and placed face up) and is concentrated by evaporation to highly germicidal levels. Meanwhile the volatile hydrogen peroxide component is dispersed and diluted through the volume of the room, and may play relatively little part in the process.

This substantial silver nitrate content is confirmed by a Daily Mail report from the Royal Liverpool Hospital, where a patient complained of “black grime” on the inside of the windows. The Hospital explained that it was a “sterile residue” from the hydrogen peroxide decontamination process. The hospital was using the Deprox process at the time.

Deprox mail

By an interesting coincidence, the JHI “articles in press” also has another paper on the antimicrobial efficacy of silver nitrate. This paper shows the MIC (Minimum Inhibitory Concentration) of silver nitrate for a range of vegetative bacteria, as below:

“The silver nitrate MIC was tested on a total of 443 isolates, ranging from 16 to 32 mg/L for the majority of the tested strains with or without sil genes. For Enterobacter and Klebsiella spp., elevated MIC (≥64 mg/L) for silver nitrate was recorded in E. cloacae (15/99, 15%), E. aerogenes (2/29, 7%), K. oxytoca (2/59, 3%), and K. pneumoniae (2/95, 2%).”

Note that 1mg/L = 1ppm. These bacterial strains were inhibited by just 16 to 32ppm AgNO3, compared to 2000ppm in Deproxin. No wonder the BIs were sterilised!

Hygiene Solutions Ltd is now between the devil and the deep blue sea. Do they remove the silver nitrate from the Deproxin, in which case their already shaky “validated to achieve a log 6 reduction” claim collapses, or do they continue the ludicrous pretense that the silver nitrate is actually just “silver” – in spite of the obvious point that metallic silver is a powerful catalyst for the decomposition of hydrogen peroxide?

This is a dilemma for hospital staff also, as according to the HSE it is illegal to use a PT2 (i.e. fogging or airborne) product containing silver nitrate. There can be no question that silver nitrate is an “active ingredient” in Deproxin. The Deprox unit contains a palladium catalyst to remove the hydrogen peroxide at the end of the process – however, this catalyst will not remove the silver nitrate, leaving an extremely fine dust or droplets of silver nitrate solution in the air when the room is re-entered. This chemical is highly toxic by inhalation, with a legal limit of just 0.01mg/m³. Certainly it would be risky to re-enter treated rooms without some kind of measurement process to assess the air quality.

Deprox nitrate

For those without access to the JHI, I have reproduced Dr Singh’s letter below.

Sir,

I note with interest the May 2016 article by S. Ali et al. “Efficacy of two hydrogen peroxide vapour aerial decontamination systems for enhanced disinfection of meticillin-resistant Staphylococcus aureus, Klebsiella pneumoniae and Clostridium difficile in single isolation rooms.”[1]

The two systems compared in this study use very different concentrations of hydrogen peroxide, and yet showed almost indistinguishable efficacy in these tests.

This would lead to the conclusion that the bactericidal and sporicidal efficacy of H2O2 is independent of concentration, which seems improbable – indeed, previous comparative evaluations of a high-concentration (30%) hydrogen peroxide system (Bioquell) with a low-concentration (5%) hydrogen peroxide system (ASP Glosair) by Fu et al.[2] Holmdahl et al.[3] and Beswick et al.[4] have demonstrated that the low-concentration fogging only achieved log reduction factors (LRF) of between 2 and 4, which was much smaller that the LRF of 6 generally achieved with the higher hydrogen peroxide concentration.

I would suggest that the unexpectedly high efficacy of the 4.9% hydrogen peroxide system evaluated by S. Ali et al. may be attributable to the relatively high level of silver nitrate in the proprietary Deproxin solution. The Deproxin MSDS[5] states: “CAS: 7761-88-8 Silver <0.2% EINECS: 231-853-9”. While “Silver” is given as the description, the CAS and EINECS numbers show that this is in the form of silver nitrate. In terms of ppm, 0.2% equates to 2000ppm. By contrast, the ASP Glosair system evaluated in the three papers mentioned above contained “<50ppm silver nitrate”. The solution used by S. Ali et al. thus apparently contained around forty times more “silver” than solutions used in systems previously evaluated.

Even at 2000ppm, the silver nitrate in Deproxin is considerably less concentrated than the hydrogen peroxide, by a factor of 25. (0.2% AgNO3: 4.9%H2O2) However, there is an important difference in the mode of distribution for these two active ingredients that may tend to preferentially concentrate the silver nitrate in the vicinity of the biological indicators. The hydrogen peroxide is volatile and unstable, and as the fog droplets evaporate, it is distributed throughout the whole volume of the room (about 60m3 in the example given), where the concentration then drops substantially with elapsed time as it spontaneously decays to oxygen and water. The silver nitrate however is not at all volatile, and is persistent.

The final distribution of the silver nitrate is hard to predict, but it may be assumed that much of it eventually drops to the floor or other horizontal surfaces in the room, either as solid particles or as droplets of solution that have been concentrated by partial evaporation. In the tests performed by S. Ali et al., biological indicators(BIs) were placed in horizontal, upwards facing orientation. If these BIs became saturated with a film of Deproxin solution from the fogging process, it can be expected that as the water evaporated during the “deactivation” cycle, the concentration of the silver nitrate would rise from the initial figure of around 2000ppm to substantially higher levels. It is of note that according to S.R.K. Pandian et al.[6] the MIC (Minimum Inhibitory Concentration) of silver nitrate for the spore-forming Bacillus licheniformis is only 5mM, which is equivalent to 850ppm.

There is some circumstantial evidence to support this explanation. S Ali et al. referring to the Deprox system state; “When rooms were disinfected using HPS2, C. difficile persisted most frequently underneath the bed and window frame in 6/21 cases (28.6%).” The “window frame” position is described as being “approximately 2m above floor” where the room height is given as 2.7m. The two positions pointed up as showing the lowest LRFs for the fogging system were those positions with the most restricted headspace – 0.7m in one case and presumably the same or less under the bed. These positions would have received the least precipitation from settling fog droplets or dust, so may have received proportionally less of the silver nitrate.

It would be very instructive to repeat the experiment without the hydrogen peroxide, and thus determine the log reduction attributable to a 2000ppm silver nitrate solution alone.

Conflict of interest

None.

References

  1. S. Ali, M. Muzslay, M. Bruce, A. Jeanes ,G. Moore, A.P.R. Wilson et al. Efficacy of two hydrogen peroxide vapour aerial decontamination systems for enhanced disinfection of meticillin-resistant Staphylococcus aureus, Klebsiella pneumoniae and Clostridium difficile in single isolation rooms. J Hosp Infect. 2016; 93: 70–77
  2. Fu, T.Y., Gent, P., and Kumar, V. Efficacy, efficiency and safety aspects of hydrogen peroxide vapour and aerosolized hydrogen peroxide room disinfection systems. J Hosp Infect. 2012; 80: 199–205
  3. Holmdahl, T., Lanbeck, P., Wullt, M., and Walder, M.H. A head-to-head comparison of hydrogen peroxide vapor and aerosol room decontamination systems. Infect Control Hosp Epidemiol. 2011; 32: 831–836 Beswick, A.J., Farrant, J., Makison, C. et al. Comparison of Multiple Systems for Laboratory Whole Room Fumigation. Applied Biosafety. 2011; 16
  4. Sevron Safety Solutions. http://sevron.co.uk/msds/deproxin-msds-download-2/[accessed 11.05.17]
  5. Pandian, S.R.K., Deepak, V., Kalishwaralal, K. et al. Mechanism of bactericidal activity of Silver Nitrate – a concentration dependent bi-functional molecule. Braz J Microbiol. 2010; 41: 805–809

Freedom of Information request reveals exact C. difficile/Deprox correlation.

C. difficile 1

A Freedom of Information request¹ to UCLH disclosed the starting and finishing dates of their disastrous Deprox decontamination contract with Hygiene Solutions Ltd. The Deprox operations started in June 2013 and ran continuously, 7 days per week through to October 2016. The contract called for at least 4 Deprox units to be at the hospital, and 6 or more processes to be completed daily.

However, due to frequent breakdowns, Hygiene Solutions struggled to meet their obligations, and on occasion asked engineers to put a non-functional Deprox unit in a room, tape up the door and “run” a process – thus not only defrauding the NHS but leaving a dangerously contaminated room that the staff believed to have been sterilized.

Plotting the contract dates against the quarterly UCLH C. difficile data² (extended through 2016 with mandatory government reporting data)³  reveals an exact correlation between the period of Deprox deployment and a substantial step change in the number of C.difficile infections  – approximately 70 extra cases over the 29 month period.

According to the March 2016 government report on C. difficile mortality, the 30 day mortality rates for the London area were about 17%. –  suggesting that approximately 12 deaths in this period could be attributed to Deprox operations. Any patients who acquired C. difficile or any other Heathcare Associated Infection, (HAI) in the UCLH hospitals between June 2013 and October 2015 should consider contacting a medical negligence solicitor and seeking compensation.

Deprox pushed UCLH in to high risk category for C. diff – CQC reports.

The CQC (Care Quality Commission) makes regular evaluations of NHS trusts using a list of critical indicators. These are the Intelligent Monitoring reports. The extracts from a series of these reports below show how the “Incidence of C. difficile” indicator moved from “No evidence of risk” to “Elevated Risk” when the Deprox program was implemented. The complete reports can be found at: http://www.cqc.org.uk/provider/RRV/reports

Deprox UCLH

[1] https://www.uclh.nhs.uk/aboutus/wwd/Annual%20reviews%20plans%20and%20reports%20archive/Infection%20Control%20Annual%20Report%202015-16.pdf  (See graph, p.17)

[2] https://www.gov.uk/government/statistics/clostridium-difficile-infection-monthly-data-by-nhs-acute-trust

[3] https://www.uclh.nhs.uk/aboutus/FOI/FOI%20disclosure%20list/FOI2017271Response.pdf

https://www.dropbox.com/s/71u3j3fcdqwqfnj/ResponseFOI2017271.xlsx%20%28~13%20KB%29.URL?dl=0