TB Infection Control

Please be very very clear on what I have said. The existing data for drug
susceptible TB is very, very clear that infectiousness ends well before the
smear and culture turns negative. Riley's guinea pig data indicates that
patients started on therapy the same day as they entered his experimental
facility were 98% less infectious than those who did not start therapy.
However, treatment must be effective treatment - not just any treatment.
So you cannot say that no precautions are necessary as soon as you start
ATT (antituberculosis therapy) unless you know the drug susceptibility
pattern of the patient's isolate, which is only possible with rapid
diagnostics. If you do not know the drug susceptibility pattern, the
patient could be drug resistant, not on EFFECTIVE therapy, and may be
infectious, so the usual precautions are necessary. I will provide this
data on line here as soon as I am situated in a place with reliable email -
am travelling.

Edward Nardell, MD

Totally agree. Precautions remain in place for two weeks for sensitive TB in S Africa but less so in the UK. MDR-TB should have airborne precautions during hospital admission unless cleared.
Shaheen
Sent via my BlackBerry from Vodacom - let your email find you!

Dear Shaheen,

You are reflecting current, widely held policy. I am suggesting that there
has been excellent evidence for more that 50 years that infectiousness of
drug susceptible patients ceases much sooner than 2 weeks, regardless of
smear or culture status - on EFFECTIVE therapy. There is some evidence
that the same holds for MDR-TB, and I will elaborate on this shortly.
Although sputum culture colony counts decrease rapidly on effective
therapy (early bacteriocidal activity), they do not convert to negative
until 2 months on average. Therefore, we as a community in applying the
"two week rule" we have long accepted the notion that sputum smear is not a
reliable indicator of infectiousness. That is, many patients are smear
positive when they are considered non-infectious. It is true that we also
say that patients should be clinically responding to treatment with
decreasing cough, etc, but these are simply additional indicators of
EFFECTIVE therapy. Now, with the advent of rapid DSTs, we have the
opportunity to avoid INEFFECTIVE therapy from the beginning, allowing us to
have greater certainty (when rapid DSTs are available) that patients are
on effective therapy.

I will post a more complete version of my argument shortly.

Ed Nardell ...

Dear Shaheen,

You are reflecting current, widely held policy. I am suggesting that there
has been excellent evidence for more that 50 years that infectiousness of
drug susceptible patients ceases much sooner than 2 weeks, regardless of
smear or culture status - on EFFECTIVE therapy. There is some evidence
that the same holds for MDR-TB, and I will elaborate on this shortly.
Although sputum culture colony counts decrease rapidly on effective
therapy (early bacteriocidal activity), they do not convert to negative
until 2 months on average. Therefore, we as a community in applying the
"two week rule" we have long accepted the notion that sputum smear is not a
reliable indicator of infectiousness. That is, many patients are smear
positive when they are considered non-infectious. It is true that we also
say that patients should be clinically responding to treatment with
decreasing cough, etc, but these are simply additional indicators of
EFFECTIVE therapy. Now, with the advent of rapid DSTs, we have the
opportunity to avoid INEFFECTIVE therapy from the beginning, allowing us to
have greater certainty (when rapid DSTs are available) that patients are
on effective therapy.

I will post a more complete version of my argument shortly.

Ed Nardell, MD

In response to several comments asking for more information on my assertion
that the EFFECTIVE treatment of drug susceptible TB, and probably drug
resistant TB, occurs almost immediately, I am posting a summary of a talk
given at a TB IC subcommittee meeting in Berlin, November, 2010. I have
edited it to remove data that might be considered a barrier to publishing
this work in a peer-reviewed journal.

Some have asked how rapid is rapid? All I can say is that, as you will
see, Richard Riley started patients on therapy the same day that they
entered his experimental ward and we have done the same. The 98% reduction
in infectiousness to sentinel guinea pigs that Riley reported was not due
to 2 wks or even 2 days treatment, but somewhat less than 24 hrs, conceding
that there is a time lag in the process of transmission. That is, even
untreated patients may not infect guinea pigs on their first day of
exposure. The point is that the effect is rapid and profound, to
paraphrase Riley - quoted in the text, below. My apologies for the length
of this post.

To reiterate one more time, I am not saying that no ...

In response to several comments asking for more information on my assertion
that the EFFECTIVE treatment of drug susceptible TB, and probably drug
resistant TB, occurs almost immediately, I am posting a summary of a talk
given at a TB IC subcommittee meeting in Berlin, November, 2010. I have
edited it to remove data that might be considered a barrier to publishing
this work in a peer-reviewed journal.

Some have asked how rapid is rapid? All I can say is that, as you will
see, Richard Riley started patients on therapy the same day that they
entered his experimental ward and we have done the same. The 98% reduction
in infectiousness to sentinel guinea pigs that Riley reported was not due
to 2 wks or even 2 days treatment, but somewhat less than 24 hrs, conceding
that there is a time lag in the process of transmission. That is, even
untreated patients may not infect guinea pigs on their first day of
exposure. The point is that the effect is rapid and profound, to
paraphrase Riley - quoted in the text, below. My apologies for the length
of this post.

To reiterate one more time, I am not saying that no TB IC precautions are
needed once treatment is started. There are several limitations with that
notion: 1) there are inherent delays in suspecting and diagnosing TB, 2)
there are major delays in diagnosing drug resistance, and 3) there are
delays in initiating effective therapy even when drug susceptibility is
known. Because these logistical delays exist and because rapid, reliable
molecular testing are not widely available, usual TB IC precautions are
still needed. Since there will always be a pre-diagnostic period,
conventional TB IC will always be needed. I am suggesting a change
in priorities to !) active case finding, 2) rapid molecular diagnostics,
including DSTs, and 3) rapid institution of DST-based effective therapy. We
call this approach, F-A-S-T: Find cases Actively, Separate safely, and
Treat effectively.


Ed Nardell, MD





TB Infection Control Sub-Working Group

Berlin, November, 2010



*Impact of Treatment on MDR-TB Transmission
Preliminary Data from the Airborne Infections Research (AIR) Facility*



Edward A. Nardell, MD

Harvard Medical School/PIH

Brigham & Women’s Hospital



According to the WHO, of the estimated 500,000 new MDR-TB cases per year,
more than half result from transmission* *Multidrug and extensively
drug-resistant TB (M/XDR-TB). Most MDR cases are treated in hospitals for
first 6 months – until culture conversion. In some high incidence areas of
MDR TB, such as South Africa, there is a growing appreciation of the gap
between the growing need for MDR treatment and the availability of hospital
beds. South Africa is therefore planning to switch to community-based
treatment, as practiced in many sites around the world, but there is
lingering concern about transmission in the community by MDR-TB patients on
therapy.



Exactly how long MDR-TB patients remain infectious on therapy is not
known. A recent CDC task force reviewed the medical literature and found
little or no evidence to support any policy on treatment duration and
infectivity of MDR-TB. Most national policies on TB treatment of drug
susceptible TB indicate that infectiousness ends after about 2 weeks of
effective treatment. However, during the 1985-1992 outbreaks of MDR-TB in
NYC, transmission was associated with undiagnosed MDR-TB patients who were
released from isolation after 2 weeks therapy. In response, many national
programs recommend treatment until smear or culture conversion – at least 6
months, as previously noted.



In examining the evidence base for the 2-week treatment recommendation for
drug susceptible TB, the following citations were reviewed:



• Andrews RH. Bull WHO. 1960 (Madras, India)

• Crofton J. Bull IUAT. 1962 (Edinburg, Scotland)

• Brooks S. Am Rev Resp Dis. 1973 (Ohio)

• Riley R. Am Rev Resp Dis. 1974 (Baltimore)

• Gunnels J. Am Rev Resp Dis. 1974 (Arkansas)

• Rouillon A. Tubercle. 1976 (Review)



The latter review of the literature indicated that it appeared that smear
and culture correlate with infectivity only in untreated cases, and that
there appeared to be a discordance between the effect of treatment on smear
and culture on the one hand and infectivity. The primary evidence is that
smear and culture positive TB patients on therapy generally do not infect
skin test negative close contacts. However, Menzies reviewed the effect of
treatment on the contagiousness of patients with active pulmonary
tuberculosis (Infect Control Hops Epidemiol 1997; 18:582-586) and found
methodological flaws in several of these studies. He questioned, for
example, whether there might have been selection bias in the household
contact studies. Perhaps the sickest patients remained in the hospital
while less infectious patients were sent home.



Despite these reservations, Rouillon concluded the following about the
impact of effective treatment on transmission:



“There is an ever-increasing amount of evidence in support of the idea that
abolition of the patient’s infectiousness *– a different matter from
‘cure,’ which takes months and from negative results of bacteriological
examinations, direct and culture, which may take weeks –* is very probably
obtained after less than 2 weeks of treatment”.



“These facts seem to indicate very rapid and powerful action by the drugs
on infectivity…”



These studies greatly influenced official isolation policies by CDC, ATS,
and other national and international agencies until the mid 70’s, such as:



• 1969 ATS – Guidelines for the general hospital for the admission
and care of tuberculosis patients.

• 1970 ATS – Bacteriologic standards for discharge of patients

• 1973 ATS – Guidelines for work for patients with tuberculosis

• 1974 CDC – Recommendation for health department supervision of
tuberculosis patients



However, more direct evidence on the impact of treatment on TB transmission
was obtained by the classic experimental studies by Riley and colleagues
using air sampling by sentinel guinea pigs on a 6-bed experimental
tuberculosis ward at the Baltimore VA Hospital between 1958 and 1962 (Am J
Hyg 1959; 70:185-196, reprinted as “classic” Am J Epidemiol 1995; 142:3-14)



Riley reported that although an average of 3 guinea pigs were infected per
month, infections were highly variable. During one 4-month period
infections ceased completely. In investigating the reasons, it was
discovered that during that period only drug susceptible cases had been
admitted to the ward. Prior to that time, and afterwards, when drug
resistant cases were again admitted to the ward, transmission to guinea
resumed.



Recognizing the importance of treatment to transmission, Riley exposed
guinea pigs to patients with and without treatment during the second 2 year
study. The result was a direct comparison of the relative effects of
treatment on drug susceptible and drug resistant TB. Note that there was
no MDR TB at the time, since this was many years before the introduction of
rifampin. On the other hand, please keep in mind that the remarkable
effects on transmission shown before were accomplished with just INH, PAS,
and SM, and in the case of DR disease, just one or two of these drugs.
Finally, and most importantly*, Riley started therapy on the day that the
patients were admitted to the ward, so these results represent very, very
little treatment*.



*Relative infectivity of patients*:*

– Susceptible TB

• 61 *Untreated* (29 GPs) *100%*

• 29 Treated (1 GP)
*2%*

– Drug-resistant TB

• 6 *Untreated * (14 GPs) *
28%*

• 11 Treated (6 GPs) *
5%*

* *

*all smear positive patients, relative to the amount of time on the ward



Riley concluded (Am Rev Resp Dis 1962; 85:511-525):



“The treated patients were *admitted to the ward at the time treatment was
initiated* and were generally removed before the sputum became completely
negative. Hence the decrease in infectiousness preceded the elimination of
the organisms from the sputum, indicating that *the effect was prompt as
well as striking.” *

* *

*“Drug therapy appeared to be effective in reducing the infectivity of
patients with drug resistant (H, SM, PAS only) organisms, but the data do
not permit detailed analysis of the problem”. *



In another similar, more recent study of the infectivity of TB patients
using guinea pig air sampling in Lima, Peru, Escombe (Plos Medicine 2008;
5:e188) noted that virtually all transmission was from inadequately treated
drug resistant strains. In that study, 97 HIV+ pulmonary TB patients
exposed 292 guinea pigs over 505 days. Of these patients, 66 cult were
culture positive and 35 smear were sputum smear positive. Of 125 infected
guinea pigs, 122 (98%) were due to 9 MDR patients who were inadequately
treated or who’s treatment was delayed. It should be noted that 108 or
the 125 infections were due to just 1 MDR patient. Three drug susceptible
patients infected one guinea pig each – two had had delayed treatment and
one had had treatment stopped for side effects.



Finally, in South Africa, in yet another contemporary facility designed for
sentinel guinea pig air sampling, researchers from the South African MRC,
CDC, and Harvard University have studied the transmission of MDR TB and the
effects of various interventions, such as ultraviolet air disinfection and
masks on patients. The, as yet unpublished, studies mentioned here were
not intended to examine the effects of treatment on transmission, although
such investigations were planned. However, observations made in the course
of a pilot study and 3 subsequent intervention studies independently shed
light on the impact of treatment on MDR and XDR TB transmission.



Because these data are being prepared for publication, I cannot report them
in detail here. Suffice it to say that cohorts of patients caused highly
variable rates of transmission in 6 separate experiments including 109
patients and groups of 90 – 360 guinea pigs. In an already published
pilot study (Dharmadhikari AS, Basaraba RJ, Van Der Walt ML, Weyer K,
Mphahlele M, Venter K, Jensen PA, First MW, Parsons S, McMurray, DN, Orme
IM, *Nardell EA*. *Natural infection of guinea pigs exposed to patients
with highly drug-resistant tuberculosis.* *Tuberculosis*, 2011. *91*(4): p.
329-38.) , 26 patients referred with MDR-TB (selected for pos smear,
coughing, cavitary) treated with standard South African regimen (started on
admission) infected 74% of 360 sentinel guinea pigs over a 4 month
exposure. In contrast, 27 patients selected for the same criteria infected
just 1 of 90 exposed guinea pigs over a 3 month exposure. As indicated in
the above publication, isolates from guinea pigs in the first study matched
by spoligotype with 3 XDR patients among the 26 who occupied the ward.
These XDR patients were unsuspected and not being treated for XDR TB. In
contrast, in the unpublished study where only 1 guinea pig was infected,
there were no XDR patients by MIGIT or LPA among the patient isolates. To
reinterate, 27 patients slected for cough, lung cavitation, smear
positivity, and not on MDR treatment (until they were admitted to the
hospital) infected only 1 of 90 exposed guinea pigs. In 4 other
experiments XDR patients retrospectively identified and transmission
occurred to guinea pigs at rates varying from 10 to 77% of exposed
animals. These data appear to substantiate the observations of Riley and
colleagues some 60 years earlier. Effective treatment appears to have an
extremely rapid and profound effect on transmission from humans to sentinel
guinea pigs. The mechanism of this phenomenon has been considered.



Writing about the rapid effects of treatment on transmission based on early
household contact and experimental data in the 60’s, Loudon and colleagues
(Am Rev Resp Dis 1969; 100:172-176) had speculated that as larger
respiratory droplets evaporated into droplet nuclei, drug concentrations
surrounding airborne organisms increased dramatically, blocking
transmission to animal (or human) hosts armed with some innate immunity,
whereas organisms in sputum could easily contain organisms that are visible
on concentrated smear and which might grow under culture conditions. In
collected sputum there is no aerosolization damage to organisms and no
concentration of antibiotic. Recall also Roullion’s conviction, quoted
earlier, of the discordance between culture and smear on the one hand and
infectivity in the treated patient.



We conclude that airborne transmission may be the weak link in TB
propagation. Only about 1/3 of untreated pulmonary TB patients infect any
close contacts. In Fennelly’s cough aerosol studies, *Mtb* is cultured
from only approximately 1/3 of smear positive patients. Based on both the
historical and recent animal data, we suggest that it may be possible that
very little *effective* treatment may tip the balance against transmission.



We further conclude a restatement of the dogma that effective treatment is
the best form of TB infection control. However, in the era of MDR and XDR
TB, this implies the need for a rapid, point of care diagnostic that can
identify drug resistance and lead to proper treatment within hours of
presentation. The advent of Gene Xpert is rapidly making this goal
possible. Together with cough-based, active case finding in hospital and
clinic admission areas, a diagnostic with a 2 hour turn-around time could,
in theory, assure that almost all TB patients will be identified and placed
on effective therapy, thereby stopping transmission.



Some questions for further research are apparent. What is the mechanism for
the rapid effect of treatment on transmission? The escalating drug
concentration hypothesis should be proven or disproven. The effects of
specific antimicrobials on transmission should be explored, including the
likelihood that the *fluoroquinolones (rather than the injectable drugs)
account for the effectiveness of MDR treatment on transmission. *

colleagues of other problems related to the infectiousness of MDR-TB patients is the malabsorption syndromes are so frequent in HIV + patients and keeping their infectivity for longer. Given that the monitoring of all these aspects is fundamenrtal with DOT

Thanks Elizabeth for raising that issue. About 60+% of our MDR
patients are co-infected with HIV at the Mpumalanga MDR Treatment
Center in South Africa, but still effective therapy in a cohort of 27
mostly smear positive, coughing, mostly cavitary MDR-TB cases appeared
to almost completely suppress transmission to guinea pigs at the AIR
Facilty over a 3 month period (1 gp infected). Additional confirmatory
studies are needed and planned for the near future.

Ed

I may be grossly wrong but can we think about inhalation therapy as well
for the infectiousness to become less? Has it been tried ?
Thanks

You are right on, Dr. Ghatak. We are about to begin a trial in South
Africa of a dry-powder inhaled antibiotic specifically to reduce TB
transmission, including a novel drug never used for TB that has
anti-mycobacterial activity and may be effective in reducing XDR
transmission.

There is one study in the literature showing that inhaled
aminoglycosides resulted in sputum conversion in chronic patients, but
they did not have the ability to measure transmission that we have
with the guinea pig model. Presumably, any effective inhaled drug
should help, although for drug susceptible TB, systemic therapy is
amazingly fast.

Ed Nardell, MD

Can we have a control subgroup of COPD patients as well? More and more
deaths are occuring in adult males who have healed TB lesions but have not
survived the lung damage by COPD .
And there is no national level control program for COPD anywhere? The WHO
also has not been very proactive about COPDs and in truth with all the
environmental /pollution /air particulate screenings .....bottomlin
eremains they are dying from COPD. Just a thought .....

Thanks !

and inhalation would be faster in reducing the infectiousness and spread of
the disease.
thats what matters more ?

transmissibility of XDR is less ? because of the morphological fragility ?
Is this based on hard evidence? This was being discussed by a lab expert ?
How much truth is there it is hard to verify but it is evident that
without injudicious use of drugs, XDR wouldn't have emerged this way.
blame game is on DOT , government , private sector ...OTC sales....but
only POOR DRUG USAGE has given us the XDR