Download Nucleic Acid Amplification Testing for the Diagnosis of Tuberculosis

Table 3. Data on Patients with Complete Blood Counts Who Had Positive Results for Influenza B by Rapid
Influenza Test Admitted to the Emergency Department Early in the H1N1 Pandemic—April and May 2009
a
d
Age,
years
WBC count,
1 ⫻ 103
cells/mm3
Relative lymphopenia,
lymphocytes as
b
% of WBC
Atypical lymphocytes as
c
% of lymphocytes
Platelet count,
1 ⫻ 103
cells/mm3
Adults
1
23
4.1
26
0
168
2
3
28
36
3.1
3.1
26
23
0
0
246
152
Case
identifier
Children
1
2
1
4.8
40
0
201
3
NA
NA
0
NA
NOTE. The rapid influenza test was Quick-S Influ A/B kit (Denka Seiken). NA, not applicable; WBC, white blood cells.
a
b
c
d
Leucopenia defined as WBC ⭐3.9 ⫻ 103 cells/mm3.
Relative lymphopenia defined by lymphoctes as ⭐21% of WBC.
n p 0%–5%.
Thrombocytopenia defined as ⭐160 ⫻ 103 platelets/mm3.
diagnostic and prognostic significance of relative lymphopenia in adult patients with influenza A. Am J Med 2004; 117:710–1.
7. Criswell BS, Couch RB, Greenberg SB, et al.
The lymphocyte response to influenza in humans. Am Rev Respir Dis 1979; 120:700–4.
8. Van Campen H, Easterday BC, Hinshaw VS.
Destruction of lymphocytes by a virulent avian
influenza A virus. J Gen Virol 1989; 70:467–72.
9. Tumpey TM, Lu X, Morken T, et al. Depletion
of lymphocytes and diminished cytokine production in mice infected with a highly virulent
influenza A (H5N1) virus isolated from humans. J Virol 2000; 74:6105–16.
Reprints or correspondence: Dr Burke A. Cunha, Infectious
Disease Division, Winthrop-University Hospital, Mineola, NY
11501.
Clinical Infectious Diseases 2009; 49:1454–6
2009 by the Infectious Diseases Society of America. All
rights reserved. 1058-4838/2009/4909-0028$15.00
DOI: 10.1086/644496
Nucleic Acid Amplification
Testing for the Diagnosis
of Tuberculosis: Not for All
To the Editor—There is a problem in
extrapolating the results of Laraque et al
[1] into the Centers for Disease Control
and Prevention (CDC) recommendation regarding nucleic acid amplification
(NAA) testing, that “NAA testing should
be performed from at least one respiratory specimen from each patient with
signs and symptoms of pulmonary TB
from whom a diagnosis of TB is being
considered but has not yet been established” [2, p. 7]. The area of dispute does
not involve those cases in which the acid-
fast bacilli (AFB) smear result is positive
but those cases in which the result is negative. The value of using NAA testing in
these cases will depend on the true incidence of tuberculosis-positive cultures
in the population being tested.
Using the approximate numbers given
in the article by Laraque et al [1], ∼30%
of samples submitted for AFB testing had
positive culture results. Even in this highprevalence scenario, 70% of the samples
submitted will have negative culture results. Of the culture positive samples,
∼50% (15 of the original 100 samples) had
positive smear results. This would result
in ∼15 culture-positive samples, of 100
samples received, that would also be tested
by NAA. An 80% sensitivity of NAA
would lead to 12 culture-positive cases being detected among 85 samples with negative smear results. In most hospital laboratories, the rate of cultures positive for
tuberculosis is much lower than this. In
my laboratory, only 6 (2%) of 299 specimens submitted for tuberculosis culture
had positive results within the past year.
The use of NAA testing for all specimens submitted for culture would not lead
to an earlier diagnosis in a significant
number of cases, but it would lead to a
substantial increase in cost. In the “real
world,” samples are received for testing
from patients with a wide range of tuberculosis probability. It would be unrealistic
1456 • CID 2009:49 (1 November) • CORRESPONDENCE
to expect the laboratory alone to assess
which specimen should be assigned to
NAA testing in the event of a negative AFB
smear result. In locations in which the rate
of cultures positive for tuberculosis is low,
it might be more cost effective to restrict
NAA testing to cases with positive smear
results and to those cases with negative
smear results in which an expert clinician
considers there to be a high likelihood of
tuberculosis. Although the CDC recommendations suggest restricting NAA testing to this targeted population, others may
decide to use NAA testing on all samples
received for tuberculosis testing.
Acknowledgments
Potential conflicts of interest.
conflicts.
J.D.: no
Joe Dylewski
Department of Microbiology and Infectious
Diseases, St. Mary’s Hospital, Montreal,
Quebec, Canada
References
1. Laraque F, Griggs A, Slopen M, Munsiff SS.
Performance of nucleic acid amplication tests
for diagnosis of tubeculosis in a large urban
setting. Clin Infect Dis 2009; 49:46–54.
2. Centers for Disease Control and Prevention.
Updated guidelines for the use of nucleic acid
amplification tests in the diagnosis of tuberculosis. MMWR Morb Mortal Wkly Rep
2009; 58:7–10.
Reprints or correspondence: Dr. Joe Dylewski, Director of
Microbiology and Infectious Diseases, St. Mary’s Hospital, Montreal, Quebec, H3T 1M5 Canada (joedylewski
@hotmail.com).
Clinical Infectious Diseases 2009; 49:1456–7
2009 by the Infectious Diseases Society of America. All
rights reserved. 1058-4838/2009/4909-0029$15.00
DOI: 10.1086/644497
Reply to Dylewski
To the Editor—We thank Dr. Dylewski
[1] for the feedback on our article [2]. It
is understood that the usefulness of nucleic acid amplification (NAA) tests varies
with the prevalence of the disease among
the population being tested. In our article,
we do not advocate NAA testing for all
specimens submitted for culture. For patients whose specimens have negative
acid–fast bacilli (AFB) smear results, our
recommendation is to order NAA testing
for those patients for whom there is a high
suspicion of tuberculosis (TB) disease. The
recommendation of the Centers for Disease Control and Prevention (CDC) is
“that NAA testing be performed on at least
one respiratory specimen from each patient with signs and symptoms of pulmonary TB for whom a diagnosis of TB
is being considered” [3, pp. 7–8]. In addition, CDC guidelines state that “NAA
tests should not be ordered routinely when
the clinical suspicion of TB is low” [3, p.
9]. Furthermore, we do not recommend
that laboratories assess which specimens
should receive NAA testing. Clinical judgment is essential to optimize the use of
these tests. In New York City, we ask that
health care providers request NAA testing
for patients who have specimens that have
negative AFB smear results and in whom
there is a high suspicion of TB.
Acknowledgments
Potential conflicts of interest. All authors: no
conflicts.
Fabienne Laraque,1 Sonal S. Munsiff,3
Anne Griggs,4 and Meredith Slopen2
Bureau of HIV Prevention and Control and 2Bureau
of Epidemiology Services, New York City
Department of Health and Mental Hygiene,
New York, and 3Alice Hyde Medical Center, Malone,
New York; and 4Malaria Consortium, Maputo,
Mozambique
1
References
1. Dylewski J. Nucleic acid amplification testing
for the diagnosis of tuberculosis: not for all.
Clin Infect Dis 2009; 49:1456–7 (in this issue).
2. Laraque F, Griggs A, Slopen M, Munsiff SS.
Performance of nucleic acid amplication tests
for diagnosis of tubeculosis in a large urban
setting. Clin Infect Dis 2009; 49:46–54.
3. Centers for Disease Control and Prevention.
Updated guidelines for the use of nucleic acid
amplification tests in the diagnosis of tuberculosis. MMWR Morb Mortal Wkly Rep
2009;58:7–10.
Reprints or correspondence: Dr. Fabienne Laraque, Bureau of
HIV Prevention and Control, New York City Dept. of Health
and Mental Hygiene, 40 Worth St, Rm. 1502, New York, NY
10013 ([email protected]).
Clinical Infectious Diseases 2009; 49:1457
2009 by the Infectious Diseases Society of America. All
rights reserved. 1058-4838/2009/4909-0030$15.00
DOI: 10.1086/644498
Serum Cross-Reactivity with
Aspergillus Galactomannan
and Cryptococcal Antigen
during Fatal Disseminated
Trichosporon dermatis
Infection
To the Editor—We read with great interest the article by Ruan et al [1] describing a cohort of patients with trichosporonosis. We recently treated a patient who
developed fatal disseminated Trichosporon
dermatis infection after cord blood transplant with aplasia. The yeast was successively isolated from 6 blood cultures, fluid
from 2 bronchoalveolar lavages, a colic biopsy sample, and a skin lesion. Identification was confirmed at the French National Reference Center for Mycoses and
Antifungals by sequencing the internal
transcribed spacer.
The day the first positive hemoculture
result was observed, antifungal prophylaxis with caspofungin (50 mg/day for 14
weeks) was replaced with voriconazole because of a strongly positive serum galactomannan index (7.2 by the Platelia Aspergillus assay; Bio-Rad). Additionally,
during the 6 days following this switch 2
serum samples were positive for both galactomannan (index, 7.6 and 6.9) and
cryptococcal antigen (positive at a dilution
of 1:40) (CALAS; Meridian Bioscience).
Data on this patient are shown in Figure 1.
As identified by Ruan and colleagues,
the blood and lungs are the primary sites
of Trichosporon infection. In our patient,
clinical and biological features were indisputably consistent with invasive trichosporonosis. Thoracic computed tomography showed nonspecific ground-glass
opacities (suggesting either an infectious
process or a hemorrhagic infarction), but
no characteristic features of aspergillosis
(ie, dense and well-circumscribed abnormalities with or without a halo sign, aircrescent sign, or cavities) were observed.
Criteria for proven, probable, or possible
invasive aspergillosis or cryptococcosis
were not met [2]. Caspofungin, which is
inactive against Trichosporon species, is an
effective salvage treatment for invasive aspergillosis [3]; breakthrough aspergillosis
during caspofungin therapy is exceptional. Importantly, cerebrospinal fluid and
urine, 2 major targets of Cryptococcus neoformans, were negative for cryptococcal
infection by both antigen testing and culture. Ruan and colleagues reported a rate
of candidemia of 32% during the same
hospital course (which was not always simultaneous with Trichosporon infection)
but found neither aspergillosis nor cryptococcosis. We thus consider it extremely
unlikely that a simultaneous triple infection with Trichosporon, Aspergillus, and
Cryptococcus would have occurred in the
patient whose history is reported here.
Some Trichosporon antigens share determinants with the capsular polysaccharide of C. neoformans [4]. Dalle et al [5]
have reported that both C. neoformans and
C. laurentii antigens could cross-react with
Aspergillus galactomannan determination.
We propose that T. dermatis harbors antigens that share common epitopes with
both cryptococcal antigen and galactomannan. To assess this hypothesis, we
tested the supernatant from the T. dermatis culture for its reactivity in the Platelia Aspergillus assay and in the determination of cryptococcal antigen titers
[5]. The galactomannan index was 2.9 and
the cryptococcal antigen titer was positive
at 1:40, thereby supporting the notion that
CORRESPONDENCE • CID 2009:49 (1 November) • 1457