| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Harvard Medical School, The Cambridge Health Alliance, Cambridge, MA
Correspondence to: Edward A. Nardell, MD, FCCP, The Cambridge Health Alliance, 1493 Cambridge St, Cambridge, MA 02139; e-mail: edward. nardell@state.ma.us
To the Editor:
The recent article by Mastorides and colleagues in CHEST (January 1999)1 on the use of micropore membrane air sampling and polymerase chain reaction (PCR) to detect airborne Mycobacterium tuberculosis (MTB) in six of seven hospitalized patients is at once both exciting and disappointing. It is exciting because advances in the control of tuberculosis (TB) and other airborne infections have been stagnant, hampered greatly by our inability to quantitatively culture organisms from room air. There have been no successful attempts to quantitatively recover tubercle bacilli from room air under clinical conditions since Richard Riley's classic experiments employing guinea pig air sampling almost 40 years ago.2 Although Riley's technique is a proven methodology, its requirement of a specialized hospital ward with at least several infectious patients, with all exhaust air delivered to hundreds of guinea pigs in special exposure chambers, has discouraged replication. By comparison, the method reported by Mastorides and colleagues,1 employing micropore membrane filters to collect airborne material and PCR to detect mycobacterial DNA, seems relatively simple. However, my enthusiasm is dampened by the severe limitations of their technique, which were acknowledged by the authors in passing and which require emphasis. Unless overcome, these limitations will likely render results uninterpretable, at best, and potentially misleading.
Early experience with air sampling and quantitative cultures of aerosolized tubercle bacilli demonstrated that only about 10% of organisms survived artificial aerosolization.3 More recently, the mechanisms by which aerosolization and rapid dehydration disrupt microbial cell components, and by which airborne organisms succumb to natural irradiation, oxidation, and other environmental stresses, have been elucidated.4 A serious limitation of PCR detection in its current state is that is does not distinguish between living and dead organisms. In sampling just 25 cu ft of air per patient, Mastorides and colleagues1 detected the DNA of MTB in six of seven patients whose cultures were positive for MTB. In contrast, most patients newly admitted to Riley's experimental TB ward did not infect any guinea pigs. The average (well-mixed) air concentration over a 4-year period was less than one infectious dose (presumably one droplet nucleus) in > 10,000 cu ft of air. In an explosive episode of nosocomial transmission in an ICU, in which 10 of 13 susceptible health-care workers became infected with MTB during a 150-min exposure, a concentration of approximately one infectious dose in 70 cu ft was estimated.5 Although PCR may be much more sensitive than guinea pigs or people in detecting the airborne DNA of MTB, it is highly likely that the much higher detection rate reported by Mastorides and colleagues1 represents predominantly dead or dying organisms, of little relevance to determination of infectiousness. Similar nucleic acid amplification techniques, when applied directly to the sputum samples of patients who have completed therapy, may continue to show positive results for months after sputum cultures have become negative for MTB, again reflecting the presence of nonviable DNA. Since such positive results are not interpretable, the use of nucleic acid amplification techniques in treated TB cases is not recommended.
Another concern about the study by Mastorides and colleagues1 is that air sampling was performed at a distance of only 1 m from the head of the bed, with no apparent provision that only particles of respirable size were to be sampled. Close proximity and larger particles would greatly increase the likelihood of positive results, which would have little relevance to the presence of droplet nuclei or to their concentration elsewhere in the room. Finally, the technique as presented is not quantitative. Air sampling either does or does not lead to the detection of DNA, but the number of organisms detected is not known.
Given the rapid progress in molecular methods, it is likely that both quantitative detection and correlation with microbial viability, and possibly with infectiousness, will be forthcoming. In its current state, however, aside from the limited application of making possible a noninvasive bacteriologic diagnosis of TB when sputum is unavailable, it is unclear to me how this approach will advance our understanding of airborne TB transmission and its control. The guinea pig, however, despite some logistical untidiness, remains a remarkable air sampler for TB; it is exquisitely sensitive to as few as one infectious droplet nucleus and is oblivious to dead and dying organisms (and other environmental microbes), while providing quantitative results on infectiousness, the only measurement that is of interest. If such an ideal sampling method did not exist, we would surely try to invent it! Perhaps it is time to go back to the future, to reestablish Riley's experimental ward, and to proceed to answer some of today's nagging questions on airborne infection and its control.
References
H. Lee Moffitt Cancer Center and Research Institute at the University of South Florida Tampa, FL
Correspondence to: Ramon L. Sandin, MD, MS, Associate Professor of Pathology and Internal Medicine, University of South Florida, Room 2071 Pathology, H. Lee Moffitt Cancer Center, 12902 Magnolia Dr, Tampa, FL 33612-9497; e-mail: sandin@moffitt.usf.edu
To the Editor:
We are very grateful for the long and useful commentary provided by Dr. Nardell regarding our study on Mycobacterium tuberculosis (MTB) detection published in CHEST (January 1999).1 It was not surprising, however, to read that the study was found to be "at once both exciting and disappointing"! Indeed, almost every study ever published that explored potential or new applications of emerging technologies for the first time evoked the same two emotions. Initial studies that question dogma may not yet contain the full "hard data" to replace the dogma, thus making those studies controversial, but only through such studies are seeds planted that then bloom into final or definitive investigations. It was in that spirit that this first study was performed and published, and, as I now address the most salient points raised in Dr. Nardell's letter, it shall become quite clear that these studies may hold the key to launching a whole new field investigating airborne microbial infectivity.
The traditional method of polymerase chain reaction (PCR) amplification, as practiced in our study, does indeed amplify microbial DNA, it "does not distinguish between living and dead organisms," and detection of dead or dying organisms may be "of little relevance to infectiousness." But does anybody think for a second that investigators in the field of molecular amplification, the hottest topic within diagnostic microbiology at the moment, would care to have their studies stop at the stage of mere simplistic DNA PCRs? Not at all! Research stops to wait for no one! Reverse-transcriptase (RT)-PCR assays for MTB, and the even more exciting transcription-mediated amplification (TMA) assays such as the one made by Gen-Probe (AMTDT-2, Gen-Probe, San Diego, CA),2 amplify RNA, and detection of RNA is a key to solving the viability issue. In a study in which ribosomal RNA (rRNA) was amplified by TMA in order to assess the treatment responses of patients with pulmonary TB during antimicrobial therapy, Moore and colleagues2 elegantly showed that during successful antimicrobial therapy the presence of MTB was eliminated in sputum samples, as measured by amplification of rRNA, and that no long-term rRNA carrier state was detected! The fact that rRNA became nondetectable in therapy responders before therapy was discontinued makes assays that detect RNA a theoretical possibility for monitoring therapy and as a final "test of cure" in the future.2 A recent in vitro study utilizing another mycobacterium, Mycobacterium smegmatis, also concluded that rRNA decays rapidly after cell death while DNA is stable for longer periods.3 Even more exciting is the recent finding that mitochondrial RNA (mRNA) has an even shorter life than rRNA, averaging only a few minutes, and mRNA has already been used in RT-PCR formats to detect only viable mycobacteria.4 So, the newest adaptations of molecular amplification techniques can push aside the issue of viability as an obstacle to the use of such methods in epidemiologic research. The reader can almost imagine the discoveries that researchers are making this very minute and that will be released for our delight in the coming years!
Dr. Nardell says "... the technique as presented is not quantitative. Air sampling either does or does not lead to the detection of DNA, but the number of organisms detected is not known." This is also true. But if Dr. Nardell can appreciate the contributions that quantitative assays for determining HIV viral load have made to the field of AIDS research in predicting resistance to antiviral therapy and prognosis, then it should be easy to envision that, by the addition of several internal or external calibrators or by the use of curves produced by the amplification of aliquots of cultures of known microbial number, assays such as the one we describe can easily be made quantitative.
"Another concern is that air sampling was done at a distance of only 1 m from the head of the bed with no apparent provision that only particles of respirable size were sampled. Close proximity and larger particles would greatly increase the likelihood of positive results... ". This is also true. But research has to begin somewhere, and we chose to begin by placing the apparatus at 1 m from the head of the bed, while the sampling at various other distances could be carried out during future studies. Many other questions, in fact, arose during this work that will need to be explored further, but none detract in any way from the fact that this original contribution was the first large study reporting on the rapid noninvasive detection of airborne MTB in multiple patients through the combined use of air filtration and PCR.
So, do we go back to the future to using guinea pigs? Indeed, perhaps that would be one approach. Another, somewhat more progressive, approach, however, would be to remain patient and open-minded enough to realize that the field of diagnostic molecular amplification is here to stay, and that it will constitute (albeit with all the fine-tuning, modifications, and future advances mentioned in this response, and then some!) a handsome approach to the elucidation of some of the epidemiologic questions about airborne infectivity that still remain to be answered with many infectious diseases!
References
This article has been cited by other articles:
|
|
 |
|
  E. A. Nardell Catching Droplet Nuclei: Toward a Better Understanding of Tuberculosis Transmission Am. J. Respir. Crit. Care Med., March 1, 2004; 169(5): 553 - 554. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
