TB Infection Control

Dear Menno,

I am not sure that UV lamps are a substitute for a ventilation system but both contribute to infection control. If you mean for instance negative pressure rooms were patients are accommodated or if you mean laboratory premises, the use and functions of UV lamps itself can be quite different.
Both contribute to IC but are not exclusive.
Best wishes,

Philippe Creac'h, Dr.
The Global Fund to Fight AIDS, TB and Malaria
Chemin de Blandonnet 8
1214 Vernier, Geneva

Dear Menno,

Thanks for your question.

It is not possible to air sample for TB using conventional air sampling methods involving culture - many have tried and failed. The reason is that the airborne concentration of TB, on average, is very, very low whereas every cubic meter of air contains tens of thousands of other, environmentally hardy spores, atypical mycobacteria, and other organisms that will outgrow TB on the culture plate resulting in cultures of mold, etc. but nothing more.

It is possible to air sample for TB under 3 circumstances only:
1) in a test room, by aerosolizing large numbers of BCG or other surrogate test organism so that they can be cultured through conventional air sampling. We and several other groups have done this to prove that upper room UVGI works, but these are not exactly real life conditions - see #3. These studies have been published.
2) in a cough aerosol sampling system, where patients cough directly into a sterile chamber about 1 cubic meter. This is used by Dr. Kevin Fennelly to study the infectiousness of patients, but cannot be used to evaluate UV efficacy in a room.
3) in an experimental hospital using large numbers (hundreds) of ...

Dear Menno,

Thanks for your question.

It is not possible to air sample for TB using conventional air sampling methods involving culture - many have tried and failed. The reason is that the airborne concentration of TB, on average, is very, very low whereas every cubic meter of air contains tens of thousands of other, environmentally hardy spores, atypical mycobacteria, and other organisms that will outgrow TB on the culture plate resulting in cultures of mold, etc. but nothing more.

It is possible to air sample for TB under 3 circumstances only:
1) in a test room, by aerosolizing large numbers of BCG or other surrogate test organism so that they can be cultured through conventional air sampling. We and several other groups have done this to prove that upper room UVGI works, but these are not exactly real life conditions - see #3. These studies have been published.
2) in a cough aerosol sampling system, where patients cough directly into a sterile chamber about 1 cubic meter. This is used by Dr. Kevin Fennelly to study the infectiousness of patients, but cannot be used to evaluate UV efficacy in a room.
3) in an experimental hospital using large numbers (hundreds) of sentinel guinea pigs. GPs are exquisitely sensitive to human TB and ignore all the other environmental airborne organisms. One skin tests the animals every month to prove infection. Under hospital exposure conditions, there is evidence that infection is usually do to the inhalation of a single virulent TB droplet nucleus. This experiment has been done 3 times only - by Richard Riley 50 years ago in Baltimore, by Rod Escombe recently in Lima, and by our group in South Africa. Rod has recently published his UV efficacy study results in PLOS showing something like 74% efficacy overall. Our first UV study just concluded yesterday and so far is showing even higher efficacy. In both cases upper room UV is in real patient rooms and the infectiousness of the air on days when the UV is on compared to infectiousness of the exhaust air when the UV is off. Hundreds of guinea pigs are located in remote exposure chambers and the ward air is delivered by the ventilation system. Riley's work and Rod's is in the literature, and ours is forthcoming.

So, it is difficult to prove a given UV installation is working. What you can do is follow the design principles established by studies in test rooms and on experimental wards and then assume that similar results will be likely. It is essential to assure good room air mixing. Usually, a slow paddle fan pulling upward (for patient comfort) will do that. The design of the fixture is also key.

Please let me know if you require additional details on the application of upper room UVGI.

Ed

Edward A. Nardell, MD
Associate Professor
Harvard Medical School (Medicine; Global Health and Social Medicine)
Harvard School of Public Health (Environmental Health; Immunology and Infectious Diseases)

Brigham and Women's Hospital
Division of Global Health Equity
FXB Building, 709c
651 Huntington Ave.
Boston, MA 02115

No sampling for airborne Mtb has ever been successful due to the scarcity of these organisms in the air and their long culture period that allows common air bacteria and fungi to overgrow the culture plates.

The only procedure that is usable for evaluating the effectiveness of upper-room UVGI is to make radiometer measurements in the upper-room and compare the measured values with laboratory results published by NIOSH, U. of Colorado, and Harvard School of Public Health. More help is on the way from studies ongoing in S. Africa under the direction of Edward Nardell, but the study results are not yet finished.

Further, there are published papers from Harvard School of Public Health that contain recommendations for effective installation and use of Upper-room UVGI in medical care and other kinds of facilities.

Melvin First

Dear Menno,

As Dr. Nardell pointed out in his email response, the best way to ensure that the UV lights work well is the use a low moving gentle ceiling fan that provides for the upward motion of the air. As CDC noted in their 2005 Guidelines for the Prevention of TB in Healthcare Settings:
"Effective use of UVGI ensures that M. tuberculosis, as contained in an infectious droplet, is exposed to a sufficient dose of ultraviolet-C (UV-C) radiation at 253.7 nanometers (nm) to result in inactivation. Because dose is a function of irradiance and time, the effectiveness of any application is determined by its ability to deliver sufficient irradiance for enough time to result in inactivation of the organism within the infectious droplet. Achieving a sufficient dose can be difficult with airborne inactivation because the exposure time can be substantially limited; therefore, attaining sufficient irradiance is essential."

The second way to ensure efficacy is to with a regular cleaning schedule for the lights to remove a build up of dust which would decrease UV dosage. This is especially true in a hospital setting where the accumulation of dust is higher due to linen and bedding particles. And lastly ...

Dear Menno,

As Dr. Nardell pointed out in his email response, the best way to ensure that the UV lights work well is the use a low moving gentle ceiling fan that provides for the upward motion of the air. As CDC noted in their 2005 Guidelines for the Prevention of TB in Healthcare Settings:
"Effective use of UVGI ensures that M. tuberculosis, as contained in an infectious droplet, is exposed to a sufficient dose of ultraviolet-C (UV-C) radiation at 253.7 nanometers (nm) to result in inactivation. Because dose is a function of irradiance and time, the effectiveness of any application is determined by its ability to deliver sufficient irradiance for enough time to result in inactivation of the organism within the infectious droplet. Achieving a sufficient dose can be difficult with airborne inactivation because the exposure time can be substantially limited; therefore, attaining sufficient irradiance is essential."

The second way to ensure efficacy is to with a regular cleaning schedule for the lights to remove a build up of dust which would decrease UV dosage. This is especially true in a hospital setting where the accumulation of dust is higher due to linen and bedding particles. And lastly, to ensure that the lights are changed out yearly or as per manufacturers recommendations.

Best regards,

Ginny

Ginny Lipke RN, MHA, ACRN, CIC
Infection Control Preventionist (CTR/TSI)
HIV/TB Prevention and Care
Global AIDS Program
Centers for Disease Control and Prevention

Dear All

Speaking from a low resource country perspective, where we have a TB prevalence of 1000/ 100 000, we have to make choices- although UVGI are being installed I am not sure that these are, or can be maintained to provide ideal results. I am still very sceptical of UVGI for low resource countries- there is a real fear that the common and tried and tested methods of containment are abandoned in favour of last resort methods., Sorry, I still need to be convinced of the effectiveness of UVGI in the field-- comparative studies perhaps????

Shaheen

Prof Shaheen Mehtar
MBBS, FRC Path (UK), FCPath (Micro) (SA), MD (Eng)
Head of Academic Unit for Infection Prevention and Control
Tygerberg Hospital & Stellenbosch Uni
PO Box 19063,
Tygerberg 7505, Cape Town

I also am not convinced about their effectiveness. May you please inform me as well if you get anything convincing?

We have experience of 4 years 24/7 use of upper room shielded UVGI fixtures in TB hospital in Vladimir, Russia (designed by Paul Jensen, CDC).

I believe with proper installation (you need UV-radiometer to do that) and inexpensive and easy maintenance, they HAVE to be used in resource limited settings. It is even more important for cold climate countries where we can not rely on adequate natural ventilation. Our study showed that our particulrar fixture combined with air mixing fan can provide up to 7 - 8 equivalent ACH in 18 sq.m patient room.

Dr. Grigory V. Volchenkov
Head Doctor
Vladimir Oblast TB Dispensary
Sudogodskoe shosse, 63
Vladimir 600023
RUSSIA

Thanks for the information.

Dear Professor Mehtar,

I understand your skepticism about upper room germicidal UV. At least with ventilation you can measure something (CO2 dilution or particle clearance) if you want to. As we have been discussing, it is difficult to prove in each instance that upper room germicidal UV is doing what we think it is doing. However, we now have the solid evidence for efficacy that you seek.

You will eventually see WHO guidelines that will report the findings of a several systematic reviews of the literature on the efficacy of all of the traditional TB transmission control interventions. As expected, there are very few controlled clinical trials of any interventions to hang one's hat on. Recommendations, therefore, are based on common sense and extrapolations of other kinds of studies. For UV, for example, we have had many good studies where test organisms such as BCG are aerosolized into test rooms and their inactivation rate measured by quantitative culture with and without upper room UV. The results in many labs have shown consistently good efficacy, but these are not naturally generated Mtb aerosols - just test organisms quantified by culture, not infections.

However, Rod Escombe has recently published (PLOS, 2008) the ...

Dear Professor Mehtar,

I understand your skepticism about upper room germicidal UV. At least with ventilation you can measure something (CO2 dilution or particle clearance) if you want to. As we have been discussing, it is difficult to prove in each instance that upper room germicidal UV is doing what we think it is doing. However, we now have the solid evidence for efficacy that you seek.

You will eventually see WHO guidelines that will report the findings of a several systematic reviews of the literature on the efficacy of all of the traditional TB transmission control interventions. As expected, there are very few controlled clinical trials of any interventions to hang one's hat on. Recommendations, therefore, are based on common sense and extrapolations of other kinds of studies. For UV, for example, we have had many good studies where test organisms such as BCG are aerosolized into test rooms and their inactivation rate measured by quantitative culture with and without upper room UV. The results in many labs have shown consistently good efficacy, but these are not naturally generated Mtb aerosols - just test organisms quantified by culture, not infections.

However, Rod Escombe has recently published (PLOS, 2008) the results of his study of upper room germicidal UV in Peru, and we have just completed our first study of upper room UV in South Africa. The study design for both experiments was almost the same. In both experiments, air from real TB wards in 2 very different resource-limited countries was vented on alternate days to one of 2 exposure chambers containing large numbers of sentinel guinea pigs (GPs). On the even days air from the wards went to one GP chamber (A) and on odd days it went to the other chamber (B). This is an ideal controlled experiment since the source of infection, the ward, is identical on odd and even days. At the end of several months, the difference in infection rate of GPs in the two chambers represents the efficacy of the upper room UV in the patient rooms. Rod Escombe reported 74% efficacy, I believe, even under conditions of high humidity. We have just completed our first UV study and the results are also excellent. The study was only 3 months long, much shorter than Escombe's, but of 9 GP infections, all occurred in the GP chamber receiving air from the ward on the days that upper room UVGI was off. Our point estimate of efficacy, therefore, was 100%! We plan to repeat this study during the more humid summer months.

The conditions on the wards were identical to those in hospitals all over the world. Patients moved about freely. There were no experimental conditions that make the results hard to transfer to real world conditions. In both cases the UVGI installations were expertly planned, and very likely more effective than poorly planned installations. In both settings room air mixing fans were used to assure good mixing between the upper irradiated zone and the lower breathing zone. Otherwise, Dr. Shaheen, controlled studies don't get any better than these. Yes, we must be sure that UVGI is properly planned and properly installed, for both safety and efficacy, but no, there is no longer any doubt about the effectiveness if properly installed and maintained. Moreover, there is no better evidence for any other alternative strategies.

As Dr. Volchenkov states in his two recent emails, upper room UVGI is especially critical in countries with cold or very cool climates where it has been shown to be as effective, and much more cost effective, than mechanical ventilation. Natural ventilation is not a realistic option.

Sincerely,

Ed Nardell



Dr. Edward A. Nardell
Associate Professor
Harvard Medical School (Medicine; Global Health and Social Medicine)
Harvard School of Public Health (Environmental Health; Immunology and Infectious Diseases)

Brigham and Women's Hospital
Division of Global Health Equity
FXB Building, 709c
651 Huntington Ave.
Boston, MA 02115

Dear Ed

We always have this discussion and I must admit the evidence from your end at least, is becoming stronger. My concerns are that where resources are limited I really do not want to see protective equipment and admin controls being put aside for engineering controls.

I did see, and comment on the new WHO TB guidelines and there too, expressed similar concerns. There should be some sort of a cascade or flow chart which helps to establish priorities. Would you agree?
Regards
Shaheen

Prof Shaheen Mehtar
MBBS, FRC Path (UK), FCPath (Micro) (SA), MD (Eng)
Head of Academic Unit for Infection Prevention and Control
Tygerberg Hospital & Stellenbosch Uni
PO Box 19063,
Tygerberg 7505, Cape Town

Dear Professor Mehtar and others

I tend to agree with Prof. Mehtar that in resource constraint settings administrative controls and protective measure might be set aside in favour of engineering controls as people seem to have more faith in technolgy. I encounter this faith in technology frequently in Mozambique where I work.

In my opnion it can not be stressed enough that whatever measures are taken need to be based on a risk assessment. We tried to work like that in the Dutch TB Services where I worked before. It would be very good if the new WHO guidelines (which I have not seen) give clear guidance on this. The cascade or flow chart that Prof. Mehtar was speaking about would be very helpful, especially if it also takes into account the different level of the health facilities (health post or dispensary, district, provincial, referral hospital).

Miranda Brouwer
Technical Advisor TB/HIV
Mozambique
Technical Advisor

Dear All,

Thanks for your reaction,

I'm well aware that UVC is not a panacea for environmental Infection control.

However the winter situation in Northern Uzbekistan does not make it easy to ventilate with minus 25°C ambient. As heating the building is already a problem.

For building a balanced ventilation system where incoming cold air is preheated by outgoing warm air, resources are too limited.

Hence the idea of measuring the myco bacteria load in the air, also knowing that it will be an air sample take of a relative short time and with a limited amount of air.

I well understand now that the mycobacteria grow very slow and will be overwhelmed by other bacteria. So we'll drop the idea.

How it is done now:

(Embedded image moved to file: pic09514.jpg)

This is an example of our UVC fixture in Uzbekistan explaining its function to national health and logistics staff.

Upper-air irradiation UVC how it works:

In a room where people are present, there will always be a natural circulation of air; a person generates heat, warmed air will rise to the ceiling, cool off, drop down and when warmed up rise again.

TB bacteria ...

Dear All,

Thanks for your reaction,

I'm well aware that UVC is not a panacea for environmental Infection control.

However the winter situation in Northern Uzbekistan does not make it easy to ventilate with minus 25°C ambient. As heating the building is already a problem.

For building a balanced ventilation system where incoming cold air is preheated by outgoing warm air, resources are too limited.

Hence the idea of measuring the myco bacteria load in the air, also knowing that it will be an air sample take of a relative short time and with a limited amount of air.

I well understand now that the mycobacteria grow very slow and will be overwhelmed by other bacteria. So we'll drop the idea.

How it is done now:

(Embedded image moved to file: pic09514.jpg)

This is an example of our UVC fixture in Uzbekistan explaining its function to national health and logistics staff.

Upper-air irradiation UVC how it works:

In a room where people are present, there will always be a natural circulation of air; a person generates heat, warmed air will rise to the ceiling, cool off, drop down and when warmed up rise again.

TB bacteria traveling with the air stream to the upper room area, will be exposed to UVC radiation and killed.
The effectiveness of the UVC lamps depends on the time TB bacteria are exposed to radiation and the strength of the radiation.
In winter when air circulation in the room is much slower and may even stop, a standing fan has to be used to start air circulation (set slow). You can test the movement of the air by holding two lighted incense sticks together and check the moving smoke.

The fixtures radiate UV-C at 360° parallel to the ceiling.

The UVC radiation is measured by an IL 1400A Radiometer, for effectiveness and safety. The approx. 350 Fixtures have 4 Philips TUV PL-S 9W/2P tubes p/p.

kind regards,

-----------------------------------------------------------------
Menno Goedhart ,
Field Support Unit
MSF OCA (Operational Centre Amsterdam)
Plantage Middenlaan 14, 1018 DD Amsterdam

Ed Nardell notes a that "Rod [Escombe] has recently published his UV efficacy study results in PLOS showing something like 74% efficacy overall". But, I cannot find the paper. Maybe it is still in review. Can anyone provide a link? Thanks, Don

A couple of days ago I received an eMail from Rod that, in part, said his paper has been accepted for publication.

Thanks Miranda
It is evident from the various discussants that there are "horses for courses" and not one size fits all. So, we should be able to prioritise in our own situations. For example in colder countries, the UVGI may be an important adjunct while in warm (usually low resourced) countries it may only be used in selective areas while other measures are emphasised.
Keeping things in perspective is ever so important.
Regards
Shaheen

Prof Shaheen Mehtar
MBBS, FRC Path (UK), FCPath (Micro) (SA), MD (Eng)
Head of Academic Unit for Infection Prevention and Control
Tygerberg Hospital & Stellenbosch Uni
PO Box 19063,
Tygerberg 7505, Cape Town

Tel: +27 21 938 5051
Fax: + 27 21 938 5065
Mobile: +27 82 852 3697