| 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 |
|||||||||
Dr. Long is a professor in the Pulmonary Division, Department of Medicine, University of Alberta.
Correspondence to: Richard Long, MD, FCCP, Department of Medicine, University of Alberta, Room 2E4.21, Walter C. Mackenzie Center, 8440–112 St, Edmonton, Alberta T6G 2B7, Canada; e-mail: richard.long{at}ualberta.ca
Much attention has recently been paid to the problem of smear-negative pulmonary tuberculosis. Quite appropriately, the discussion has focused on low-income countries, home to the vast majority of individuals with tuberculosis and HIV and where the ability to culture diagnostic specimens may be lacking.1 2 3 4 5 Yet, there remain legitimate questions concerning this group of patients in industrialized countries. In this issue of CHEST (see page 349), Kanaya et al address one of those questions. Is it possible to predict, among patients with suspected active, smear-negative pulmonary tuberculosis, those patients whose culture specimens will ultimately prove to be positive? The object would be to avoid the adverse consequences that might result from withholding treatment in patients with the disease (remaining ill for excessively long periods of time, and possibly infecting others in the community) or introducing treatment in patients without the disease (their actual illness goes untreated, and they are exposed unnecessarily to possible drug toxicity).
The problem of smear-negative pulmonary tuberculosis is not trivial. In acute-care settings, as many as 8 to 10 patients are suspected to have tuberculosis for every one confirmed case.6 7 In our provincial jurisdiction, respiratory specimens from 125 persons are submitted, on average, for every one culture-confirmed case of tuberculosis.
Presently the most important criteria for establishing a presumptive
diagnosis of tuberculosis are the acid-fast bacilli (AFB) smear
(performed on clinical specimens that are adequate in both quantity and
quality) and a case definition, which may be based on radiographic
signs, clinical findings, risk factors, or a combination of these
factors.8
The sensitivity of the AFB smear result is known
to be poor, varying between 30% and 70% depending on a number of
factors relating to how the test is implemented.3
The
sensitivity is improved by concentration of sputum specimens and use of
fluorescent microscopy but reduced in patients with HIV
disease.3
5
The specificity and positive predictive value
of the smear results may be reduced in settings of high HIV prevalence
or low (high nontuberculous mycobacterial [NTM]) tuberculosis
prevalence.9
Yet, the sputum smear should always be
performed because it is quick and easy to activate, provides
preliminary confirmation of the diagnosis, and gives a quantitative
estimate of the number of bacilli being excreted and therefore the
infectivity of the patient (
5,000 bacilli per milliliter of sputum
must usually be present for the smear result to be
positive).9
In the initial evaluation, a nucleic acid
amplification (NAA) test may allow for a distinction between a positive
smear result due to Mycobacterium tuberculosis and one due
to an NTM (see below).9
But if this test is not available,
then the positive smear result should be interpreted to mean the
presence of infectious tuberculosis until proven otherwise.
False-negative smear results may occur (Table 1
), and efforts to limit this error must be ongoing.10
|
A replacement for culture that is equally sensitive and specific, but more rapid, is clearly needed if patients with early-stage disease, low burden infection, and minimal symptoms are to be detected.15 Technologies that allow for the amplification of specific target sequences of nucleic acids that can then be detected through the use of a nucleic acid probe currently offer the greatest promise of direct detection and identification of M tuberculosis in clinical specimens. To paraphrase the recent American Thoracic Society (ATS) statement on Diagnostic Standards and Classification of Tuberculosis in Adults and Children9 : NAA methods can be applied to clinical specimens within hours. In respiratory specimens that are AFB smear positive, their sensitivity is approximately 95% with a specificity of 98%. In specimens that contain fewer organisms and are AFB smear-negative, results are positive in 48 to 53% of patients with culture-positive tuberculosis and the specificity remains approximately 95%.16 An "enhanced" NAA test has been approved by the Food and Drug Administration advisory panel for use on both smear-positive and smear-negative respiratory specimens from patients who are clinically suspected of having tuberculosis. This recommendation is based on a clinical trial in which the "suspicion" of tuberculosis was quantified.17 In patients (AFB smear-positive and smear-negative) where the clinician had an intermediate or high suspicion of tuberculosis disease, the sensitivity of the enhanced test was 75 to 88% and the specificity was 100%. The clinical use of NAA tests in this setting has to be confirmed, and efforts to clarify appropriate uses are underway.16 18
Earlier definitive diagnosis of the smear-negative case will impact clinical (treatment) and public health (isolation and contact investigation) decision making. Currently, these decisions must be made in the absence of a definitive diagnosis. They require a sophisticated analysis of all available information and an estimate of relative risk. Prediction rules may help in this regard, but they must be understood to reflect the population under study, both patients and physicians. Factors likely to affect physicians’ estimates of relative risk include the customary prevalence of disease in the practice setting, the clinical spectrum of disease, the specialty or experience of the physician, and the quality of the medical history.19 Prediction rules are also subject to bias, as a more determined effort to seek the diagnosis, and therefore recover the organism, usually attends a higher level of suspicion. At its workshop on rapid diagnostic tests,16 the ATS developed an algorithm in which the decision to treat, isolate, or commence contact investigation was based on an integration of the following variables: high or low clinical suspicion of tuberculosis, AFB smear-positive or smear-negative results, and positive or negative NAA test results. Actions were compared with or without knowledge of the NAA test. The algorithm was not intended to represent an official recommendation from either the Centers for Disease Control and Prevention or the ATS. In a cogent companion editorial, Peter Barnes20 presented guidelines to assist clinicians in using NAA tests and interpreting their results. To the extent that NAA test results lead to deferral of further diagnostic evaluation, they may reduce the yield in culture and the benefits that accrue from recovery of the organism. Ultimately, phage replication systems that can be used for both detection and drug susceptibility testing of live mycobacteria may have greater utility.21 22 23
A decision to treat a patient with smear-negative finding may influence diagnosis (evidence for or against a therapeutic response) and infectivity (a patient with a smear-negative and ultimately culture-positive finding may be considered noninfectious after 2 weeks of effective treatment24 ). A decision to treat may or may not intimate a public health action and vice versa.7 25 26 27 28
In deciding whether or not to introduce treatment, we have found it helpful to classify patients with smear-negative, culture-positive findings into one of three groups: (1) patients with primary pulmonary tuberculosis who are not HIV coinfected (group 1); (2) patients with primary or postprimary pulmonary tuberculosis who are HIV coinfected (group 2); and (3) patients with postprimary tuberculosis who are not HIV coinfected (group 3).
Group 1 patients usually have noncavitary, relatively paucibacillary tuberculosis. Not uncommonly, their specimens are culture-negative.9 Their diagnosis is dependent on a good history of tuberculosis contact, a positive tuberculin skin test (TST) result or possibly a TST conversion, and a compatible clinical and radiographic picture. Group 1 patients are often children, and they should be treated promptly, as there is a risk of progression to more sinister forms of disease (neurotuberculosis or disseminated disease). Likewise, treatment with antituberculosis drugs should be initiated promptly in group 2 patients, as both tuberculosis disease and HIV replication may accelerate in their absence.29 30 Unfortunately, making a presumptive diagnosis of tuberculosis in patients with advanced HIV disease is not a simple task. Their chest radiographs may be atypical, their histopathologic findings misleading, and their TST results falsely negative. The differential diagnosis may be broad.29 The overall characteristics of NAA tests may be uniquely well suited to this population.17 30
From a bacteriologic point of view, patients in group 3 appear to be different from HIV-uninfected patients with smear-positive, culture-positive postprimary pulmonary tuberculosis.10 31 32 33 First, they have been demonstrated to be discharging small numbers of bacilli intermittently.10 Secondly, their infectivity is very much lower than that of patients with smear-positive findings.34 Third, the minimal extent of their disease does not, as one might expect, automatically imply that they are in an early stage of what will become smear-positive cavitary disease.10 31 And fourth, their prognosis is often very good even without treatment.10 Thus, their natural history is usually more benign than that of patients with smear-positive sputum findings. That much is lost by not initiating treatment in some of these patients at the point of suspicion is not at all a given. On the contrary, it may be more appropriate to investigate these patients until a diagnosis is established rather than treating empirically. Certainly if a life-threatening form of nonrespiratory tuberculosis is suspected of being coexistent, then treatment must be instituted promptly. Among group 3 smear-negative, culture-positive patients, it is not established that transmission is greater from those who begin treatment at the point when culture results are found to be positive than those who begin treatment at the point of suspicion. Prudence would dictate that each of these cases be considered individually, with the relative importance of a broad range of determinants, only some of which may be at hand, weighted (Table 2 ).
|
Acknowledgements
The author is very grateful to Lisa Meyers, MLT, and Michele Zielinski, BScN, for technical assistance; Drs. Alex Herbert, Earl Hershfield, Dennis Kunimoto, Vernon Hoeppner, and Bruce Light for their review of the article, and Cassandra Martin and Joanne Leroux for preparation of the article.
References
This article has been cited by other articles:
|
|
 |
|
  American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America: Controlling Tuberculosis in the United States Am. J. Respir. Crit. Care Med., November 1, 2005; 172(9): 1169 - 1227. [Full Text] [PDF]
|
|
|
|
 |
|
 F. Alcaide, N. Gali, J. Dominguez, P. Berlanga, S. Blanco, P. Orus, and R. Martin Usefulness of a New Mycobacteriophage-Based Technique for Rapid Diagnosis of Pulmonary Tuberculosis J. Clin. Microbiol., July 1, 2003; 41(7): 2867 - 2871. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
