Controversies in TB Transmission Control: How important is natural ventilation to the spread of tuberculosis?
Started by Edward Nardell, MD on 14 Jul 2008
Last edited by Julia Fischer-Mackey on 12 May 2010
Ever since the 2007 publication of Rod Escombe's measurements of natural ventilation rates in hospitals in Lima, Peru, (Natural Ventilation for the Prevention of Airborne Contagion, February 2007: http://medicine.plosjournals.org/perlserv/?request=get-document&doi=10.1371/journal.pmed.0040068&ct=1) there has been an outpouring of enthusiasm for this approach.
The study in Lima, a costal city, showed very high estimated ventilation rates with windows open, much higher than in normally possible by mechanical ventilation systems. In published commentary, Hal Levin urged caution in not over interpreting those theoretical results.
Although I think natural ventilation has an important role to play, it is mostly a design feature, i.e, in suitable climates, the use of outdoor covered corridors, waiting areas, sputum induction areas, and exhaust stacks etc. I am less enthused about depending on natural ventilation for air disinfection in rooms simply because conditions in most locations vary from minute to minute, night and day, and by season. Neither the direction of airflow nor the numbers of air changes can be assured, but depend on prevailing climatic conditions. Even in Africa windows are often shut tight at night because of low night temperatures and for security.
Finally, Li published a meta analysis of the literature of ventilation and airborne infection, pointing out that while an association is clearly evident, there is no good basis for recommending specific levels of ventilation in specific settings.
---------------------------------------------------------------------------------------------------------------------------------
Escombe AR, Moore DAJ, Friedland JS, Evans CA, Gilman RH (2007) Natural Ventilation for Prevention of Airborne Contagion: Authors' Reply. PLoS Med 4(5): e195
<http://medicine.plosjournals.org/perlserv/?request=read-response&doi=10.1371/journal.pmed.0040068>
---------------------------------------------------------------------------------------------------------------------------------
Li Y, Leung GM, Tang JW, Yang X, Chao CY, Lin JZ, Lu JW, Nielsen PV, Niu J, Qian H, Sleigh AC, Su HJ, Sundell J, Wong TW, Yuen PL. (2007) Role of ventilation in airborne transmission of infectious agents in the built environment - a multidisciplinary systematic review. Indoor Air, Vol. 17, Number 1, pp. 2-18(17)
<http://www.ingentaconnect.com/content/mksg/ina/2007/00000017/00000001/art00002>
Keywords: Engineering Controls, Haiti
Sophie Beauvais
Today’s (August 19) New York Times Global Update features the new Burera district hospital in Rwanda: Hospital’s Design Keeps Fresh Air in Mind. http://www.nytimes.com/2008/08/19/science/19glob.html?_r=2&ref=health&oref=slogin&oref=slogin
The design relies in part on outdoor walkways and waiting rooms, and large windows on opposing walls to keep air circulating and reduce the spread of airborne disease. The article also cites the 2007 publication of Rod Escombe's measurements of natural ventilation rates in hospitals in Lima, Peru, in PLoS.
4:05 PM, 19 Aug 2008 | Permalink
Edward Nardell, MD
This is the PIH hospital in Rwanda that Michael Murphy, the Harvard Architecture student in our recent post-grad course, and his colleagues are designing. The article refers to Rod Escombe's paper, cited above. Rod also taught in the July 14-25 course entitled, "Engineering methods to reduce Airborne Infections". I will ask Michael to post some of the details for discussion purposes.
Ed Nardell
4:26 PM, 19 Aug 2008 | Permalink
Rod Escombe
Whilst Lima is indeed a coastal city with a prevailing breeze off the Pacific Ocean, there are also many days when the breeze is light, or almost absent. I performed over 400 ventilation measurement experiments, over two years, during different seasons, different times of the day, and on windy days and on still days. I measured ventilation in top floor rooms facing the breeze, and in basement rooms tucked away on the leeside of buildings. I was surprised at how much natural ventilation there was even on relatively still days. As may be seen in Figure 2 of the PLoS Medicine article, the median air-changes/hour (ACH) measured on relatively still days was still 20 – not bad in my book. And that 20 ACH was for all rooms, which included the modern rooms with small windows and low ceilings as well as the old fashioned TB sanatoria style rooms with large windows and high ceilings. The shutting of windows at night is an important point raised by Ed. However, it is possible to maintain a few high-positioned, out-of-reach windows open all the time (or to knock the glass out of selected windows) to ensure at least a few ACH at night, without compromising ambient ward temperatures too much. But certainly it is always a compromise between thermal comfort, draughts, noise pollution, privacy, and good infection control through natural ventilation. Security can easily be maintained using metal bars, which do not impede air currents.
Regarding Li’s meta-analysis which points out that there is no good basis for recommending specific levels of ventilation in specific settings, there is no doubt that more ventilation is better than less ventilation, and that fresh air ventilation in a room is one of the main determinants of airborne infection risk in any airborne infection model.
Using natural ventilation in respiratory isolation rooms is obviously a tricky issue, as negative pressure cannot be generated, and air does not always follow the arrows it is supposed to on the pretty diagrams of naturally ventilated buildings and wind direction. However, transmission of airborne infection to adjacent rooms can be mitigated by common sense approaches to the location and orientation of such rooms. The use of anterooms, the use of well ventilated open-ended corridors, the positioning of isolation rooms on the top floor where skylights can be used, or on the corner of a building to make use of cross ventilation through windows on adjacent external walls, can all help to ‘isolate’ a naturally ventilated respiratory isolation room from adjoining parts of a building.
10:30 AM, 27 Feb 2009 | Permalink
S. Mehtar
Dear Rod
Thank you for this comprehensive review. I too, feel there should be adequate ventilation with the system you have described- I think the work done by you and your team is elegant, simple and "do-able" in developing countries- whether we like it or not, we just do not have the infra-structure to maintain a sophisticated mechanical air handling system with UVGI- I see it almost every day!
I also note with interest the severe weather conditions noted in Eastern Europe and colder climes. Could I put a hypothesis forward? In the days of teepees. there was a fire in the centre of the tent which allowed air to be drawn from the sides and drawn up through the opening in the tent to the outside- based on the concept of hot air rising and carrying aerosols etc with it. Has anyone done any work on this? It seemed to work quite well in confined spaces but I am not sure about the larger hospital wards.
Regards
Shaheen
1:46 AM, 2 Mar 2009 | Permalink
Philippe Creach
Dear All,
Prof Shaheen Mehtar mentioned the cold climates of Eastern Europe which is right. Medical staff there is maintaining the infra-structure of UVGI with simple brushes, pens and pieces of papers and think to replace bulbs when needed. So far, so good!
As to transforming TB wards in "teepees", I believe that the principle of hot air going up is precisely the one calling for the UVGI be placed up and not down. So, in a way, the proposition of Prof Shaheen Mehtar is already implemented.
Best wishes,
Philippe Creac'h, Dr.
4:31 AM, 2 Mar 2009 | Permalink
Edward Nardell, MD
I think the following facts about ventilation, upper room UV and TB
transmission control are becoming quite clear as our discussions
progress;
1) Natural ventilation can be a cost-effective solution to indoor
transmission of airborne contaminants including infectious aerosols when
it is possible, but it is not always reliable because of changing
climatic conditions from hour to hour, day to day, and season to season.
There are notable geographic exceptions where the climate is conducive
most of the time. However, I recall a hospital in Soweto, South Africa,
where the patients were sitting outside during the warm day (optimal
natural ventilation) but where at night the windows were shut tight both
for the cool night air and for security. Such practices are difficult,
but perhaps not impossible to change. In a clinic with the windows
open, for example, one needs to do a smoke test under many different
conditions to be sure that air direction is not from patient to
provider.
2) Although many facilities lack resources for full mechanical
ventilation systems, simple extraction fans if properly sized and placed
can provide both air changes and assure favorable directional airflow.
These too must be checked with smoke tubes and other techniques. A
hospital in Haiti I visited, for example, had a large extraction fan
behind a patient waiting area, assuring that air from the clinic exposed
everyone sitting there before being extracted. Such interventions
require planning with a knowledge of airborne transmission. First and
foremost, in my mind, is allocation of appropriate space for the
required function, including patient flow patterns, etc. The
availability of GFATM and other funding sources is making possible
building renovations, including proper use of natural ventilation,
mechanical systems, and germicidal UV.
3) Properly designed, installed, and tested, upper room UVGI can provide
the air disinfection equivalent of high levels of natural or mechanical
ventilation. It does not provide odor control and other key functions
of ventilation, so that in that sense only it is supplementary.
Those who dismiss any of these 3 modalities out of hand, without
considering local conditions and new funding opportunities are depriving
patients of the benefits of a complete range of potentially viable
interventions to prevent the airborne transmission of TB, flu and other
pathogens. There are bad examples of natural ventilation, mechanical
ventilation and UVGI, and there are examples where each of these
interventions, and their combination, have been carefully applied and
maintained. In my view we need to stop thinking of these as competing
technologies and begin to view them as a range of tools to be properly
applied under a range of conditions. Finally, we need more tools because
these alone do not meet the needs of all situations.
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
8:00 AM, 2 Mar 2009 | Permalink
S. Mehtar
An excellent synopsis, Ed. thank you. I agree totally- these should not be competing technologies but complimentary. My only reservation is that one system is implemented which is less effective in given geographical locations- copycat if you will- and the other aspects required to reduce transmission are totally ignored! Is there any way one can "prioritise" PPE, triage, transmission based precautions and ventilation or is that too difficult? I believe these are equally important to reduce the burden of disease if one follows a systematic flow- this seems to escape most administrators I'm afraid.
I hope someone will challenge me- it does for good debate.
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
6:22 AM, 3 Mar 2009 | Permalink
Hans Mulder
Interesting discussion.
Just back from a mission in Kenya and a mission to Ethiopia. I fully agree on the fact that we should see these technologies as complementary. There might be one or two problems with this as I have noticed on my visits to other hospitals.
The more systems available within one building / room / waiting area the more you run the risk of having them clashing with each other. Most of the people don't understand what the influence of a free standing fan / desktop fan can be if it is used in connection with mechanical ventilation.
For example in one of the bronchoscope rooms we had three systems "fighting" each other to provide ventilation. The doors to support rooms were open, the air supply and extractor were on full force trying to supply fresh air and extract infected air from the room and a desktop fan was trying to mix everything in the room at full speed. The ultimate complementary solution I would say. After doing some smoke test we could clearly see that the effect of the extractor fan was almost nil, unless you are 20 cm. below it. The air supply would leaf through the open doors into the control room for the nurse depending if doors in the passage were closed or opened. After closing all the doors and switching of the desktop fan, air movement stabilized and the mechanical ventilation was able to do its work in a more proper way.
Conclusion, multiple ventilation systems come with a good education towards the people using it. Restricting certain items, locks on windows for example, no desktop fans, etc. might be a thought. The line between user friendliness and effective ventilation is a fine one and should be considered for every project / situation and not become a standard for all.
Best regards,
Hans Mulder
for STAUCH + PARTNERS ARCHITECTS
5:53 AM, 4 Mar 2009 | Permalink