Download Detection of Selected Fastidious Bacteria

166
Detection of Selected Fastidious Bacteria
Gary V. Doern
From the Medical Microbiology Division, Department of Pathology,
University of Iowa College of Medicine, Iowa City, Iowa
The intent of this article is to describe the optimal methods for culture recovery of 7 fastidious bacteria: Legionella species, Brucella species, Francisella tularensis, Leptospira species,
Borrelia burgdorferi, Bartonella species, and Bordetella species. These organisms share much
in common beyond the fact that their genus names all end in the letter “a.” Culture recovery
of these organisms, even from adequate clinical specimens, is logistically demanding, often
costly, and lacking in both timeliness and sensitivity. In addition, there is generally no need
to recover culture isolates on which to perform antimicrobial susceptibility tests because these
7 bacteria are nearly uniformly susceptible to specific, clinically useful antimicrobial agents
and because, for some of them, susceptibility tests of proven reliability have not yet been
devised. Perhaps for these reasons, alternative, more rapid, direct diagnostic approaches have
been developed that are based on either immunochemical or nucleic-acid detection methods.
These methods have generally served to supplant culture as a primary diagnostic modality.
Situations exist, however, in which culture may be desirable, if not necessary, to establish a
definitive diagnosis of infection with these 7 organisms. This review attempts to summarize
how best to proceed in those cases.
Determining the precise etiology of infection in individual
patients aids in management decisions, is of prognostic and
epidemiological consequence, and may have profound public
health and infection control ramifications. Culture techniques
traditionally have formed the cornerstone of establishing an
etiologic diagnosis of infection in patients whose disease is due
to bacteria, fungi, mycobacteria, and, in some cases, viruses
and parasites. During the past decade, first immunochemical
techniques that detect microbial antigens and then molecular
methods that detect microbial nucleic-acid sequences have been
developed for infectious disease diagnosis.
By use of these methods, either specific antigens or nucleic
acid segments are detected directly in clinical specimens. These
methods, together with microscopic visualization, serve as adjuncts and, in selected instances, as surrogates to culture methods for establishing the etiology of infection. These methods
have become increasingly popular because of the recognized
limitations of many culture techniques (e.g., technical complexity, lack of timeliness, expense, and relative insensitivity
with certain organisms).
Culture, however, retains 2 distinct advantages over nonculture-based direct detection methods. First, a microorganism
is made available on which to perform antimicrobial susceptibility testing. Second, once a microorganism has been recovReceived 9 February 1999; revised 30 September 1999; electronically published 6 January 2000.
Reprints or correspondence: Dr. Gary V. Doern, Department of Pathology, C606 GH, University of Iowa College of Medicine, Iowa City, IA 52242
([email protected]).
Clinical Infectious Diseases 2000; 30:166–73
q 2000 by the Infectious Diseases Society of America. All rights reserved.
1058-4838/2000/3001-0026$03.00
ered in culture and accurately identified, there can be no mistaking the correctness of the finding; to wit, a positive culture
has absolute specificity.
Simply recovering a microorganism from a high-quality specimen does not necessarily prove that the microorganism is unequivocally responsible for disease. One can conclude with certainty only that the microorganism is present; its clinical
significance is ascertained by taking into account additional
information. In certain cases, however, the mere recovery of
certain microorganisms from a patient by means of culture
nearly always identifies it as the etiology of disease in that
patient. This is true of agents that rarely, if ever, colonize patients asymptomatically.
This axiom applies to the microorganisms under discussion
here, including Legionella species, Brucella species, Francisella
tularensis, Leptospira species, Borrelia burgdorferi, Bartonella
species, and, usually, Bordetella pertussis. All 7 of these organisms have in common the fact that they are generally slowgrowth, fastidious bacteria for which culture methods are logistically complex. In addition, these bacteria are relatively infrequent causes of disease.
Because of the low prevalence of infection due to these bacteria, diagnostic specificity is absolutely mandatory. The predictive value of a positive test result with these organisms becomes unacceptably low if the test that is used for diagnosis
yields even small numbers of false-positive results. For these
reasons, culture remains a useful tool in establishing an etiologic
diagnosis of infection due to these agents. The intent of this
article is to discuss the optimal approach to the laboratory
diagnosis of infections due to Legionella species, Brucella species, F. tularensis, B. burgdorferi, Leptospira species, Bartonella
species, and B. pertussis.
CID 2000;30 (January)
Detection of Fastidious Bacteria
167
Legionella Species
Brucella Species
Legionella, now recognized as an important cause of pneumonia, is a facultative, highly fastidious, gram-negative bacillus
widely distributed in nature, primarily in association with
aquatic environments. Humans become infected after exposure
to airborne droplets and potable water containing the organism
[1]. At least 40 different species have been recognized; however,
1 species, Legionella pneumophila, accounts for the vast majority of human infections [2].
Culture recovery of Legionella species from patients with
pneumonia is best accomplished by use of bronchoscopy or
lung biopsy specimens [3–5]. Expectorated sputa or suctioned
specimens from patients who are intubated should be considered inferior to specimens collected by more invasive means.
Perhaps this is because of the ability of Legionella species to
survive intracellularly and their tendency to cause an interstitial
pulmonary process. Extracellular organisms free in the lower
airways are probably uncommon. Pleural fluid is also a satisfactory specimen.
Several characteristics of Legionella species are relevant to
their recovery in culture [6]. Organisms of this genus require
L-cysteine for growth. In addition, optimum growth occurs
only over a narrow range of pH and is enhanced by iron salts.
Growth is impeded by small amounts of toxic substances, which
may be present in respiratory tract specimens, and the organism
is intrinsically resistant to several antimicrobial agents. In view
of these characteristics, it is not surprising that Legionella species do not grow on standard supplemented media such as
enriched chocolate agar, notwithstanding the published early
finding of growth on Mueller-Hinton agar with IsoVitaleX
(Becton Dickinson Microbiology Systems, Cockeysville, MD)
[7].
A variety of observations [8–12] together define the optimal
medium for culture recovery of Legionella species, namely,
ACES-buffered charcoal yeast extract agar supplemented with
L-cysteine, alpha-ketoglutarate, and iron (BCYE). When this
medium is used for respiratory tract specimens that are probably contaminated with oropharyngeal bacterial flora, a selective version of BCYE should be used. Antibiotics to which
Legionella species are usually resistant (i.e., anisomycin, polymyxin B, cefamandole, and vancomycin) are added to BCYE
to inhibit growth of commensal bacteria [13].
Plates should be incubated in a humid environment at 357C
in 3%–5% CO2 for at least 5 days [4, 6, 13]. Colonies typically
become apparent on the second or third day of incubation.
Blood cultures need not be performed, except in rare cases of
culture-negative prosthetic valve endocarditis [14]. The fact that
the nature and quality of specimens play a major role in the
sensitivity of culture has been demonstrated by the finding that,
in general, cultures for !50% of patients with legionellosis will
yield positive results, even when adequate methods and media
are employed [6]. Nonculture-based diagnostic methods have
been reviewed in a prior article in this series [15].
Brucella species are associated with a febrile illness in humans
acquired by direct exposure to infected animals or, more commonly today, consumption of contaminated dairy products,
especially fresh goat cheeses [16, 17]. Acquisition of this microorganism by laboratory workers as a result of laboratory
accidents is also common. Four species are most commonly
recognized as causes of human infections: Brucella abortus, Brucella canis, Brucella melitensis, and Brucella suis [17]. The usual
animal reservoirs for these species are cattle, dogs, sheep and
goats, and swine, respectively [17].
The specimens that yield Brucella species most often are
blood and bone marrow from infected patients [18, 19]. In
selected cases, organisms may also be recovered from biopsy
specimens of the liver and lymph nodes [20]. Rarely, normally
sterile body fluids such as CSF and peritoneal fluid have yielded
the organism, as have soft-tissue biopsy specimens and urine
[21].
Although the optimal blood culture method for detecting
brucella bacteremia has not been defined, methods are evolving
and have improved. Traditional methods that employ biphasic
blood culture bottles with brain-heart infusion broth and agar
media, vented and incubated in 5%–7% CO2 at 357C for 30
days [21], are effective but no longer necessary [22]. If used,
however, biphasic bottles should be inverted to inoculate the
agar surface once daily after careful macroscopic examination
[21].
Brucella species have also been recovered with the most
widely used continuous-monitoring automated blood culture
systems [23–27], including the BACTEC NR660 and 9240 (Becton Dickinson Microbiology Systems) and the BacT/Alert (Organon Teknika, Durham, NC) blood culture systems. Moreover, published results, mostly concerning B. melitensis, show
that >95% of isolates are recovered within 2–6 days with the
most recent media and reports [23, 26, 27]. A terminal blind
subculture of instrument-negative bottles is warranted for patients whose epidemiological and clinical circumstances suggest
a high probability of brucellosis.
A third approach to detect Brucella species in blood is use
of lysis-centrifugation (Isolator; Wampole Laboratories, Cranberry, NJ) [25, 27]. Comparative data concerning centrifugation-lysis versus automated blood systems are sparse; however,
in 1 study the BACTEC 9240 enabled recovery of more B.
melitensis strains (28 vs. 22) than did centrifugation-lysis, even
though the numbers were insufficient for rigorous statistical
conclusions.
During the acute stages of the disease, brucella bacteremia
is typically continuous. Therefore, 2–3 blood cultures, each consisting of 10–20 mL of blood, should be sufficient to confirm
the presence of bacteremia. For patients with chronic brucellosis, 12 or 3 blood cultures may be necessary to document
bacteremia.
Fluid specimens, such as bone marrow aspirates, CSF, and
168
Doern
CID 2000;30 (January)
peritoneal fluid, should be cultured in biphasic blood culture
bottles, in continuous-monitoring blood culture devices, or by
the lysis-centrifugation method, as described above for blood
specimens. Biopsy specimens and exudates may be cultured on
standard agar media, such as 5% defibrinated sheep blood agar
and enriched chocolate agar plates [21]. As was the case with
subcultures of blood specimens processed by lysis-centrifugation, because of the slow-growth nature of Brucella species,
plates should be incubated for up to 14 days in 5%–7% CO2
at 357C [20].
previously, current blood culture practices frequently involve
use of a continuous-monitoring device. The media employed
with these systems are typically enriched and probably support
growth of F. tularensis. Indeed, recovery of F. tularensis by
means of instrument-based blood culture methods has been
described in the literature [37–40]. Incubation of blood culture
bottles beyond the usual 5 to 7-day cycle recommended with
these systems may be required; alternatively, a terminal gram
stain and subculture could be done when the index of suspicion
is high.
Francisella tularensis
Borrelia burgdorferi
F. tularensis, an infrequent cause of sporadic cases of the
zoonotic infection tularemia, is a fastidious gram-negative bacillus found in nature in association with a wide variety of
animals and birds [28]. Numerous arthropod vectors play an
important role in maintaining the organism in mammalian and
avian reservoirs [29]. Humans acquire infection by direct contact with infected animals or when bitten by an insect vector
[30]. Lagomorphs represent the most common animal source
for human infection [28]; ticks and (less commonly) deerflies
are the most important arthropod vectors for human disease
[29].
The most definitive specimens for recovery of F. tularensis
are biopsy specimens of infected soft-tissue or lymph nodes. In
addition, blood cultures should be performed, especially when
the septicemic form of tularemia is suspected. F. tularensis is
extremely fastidious and dies rapidly unless specimens are processed expeditiously. F. tularensis usually requires both cysteine
and glucose for growth [29]. Recently, however, clinical isolates
of this organism that lacked the cysteine growth requirement
have been described elsewhere [31].
Traditionally, agar media supplemented with cysteine and
glucose have been used to recover F. tularensis in the laboratory
[29]. However, it has recently been found that enriched chocolate agar and nonselective BCYE adequately support the
growth of F. tularensis and can therefore be recommended for
use in isolating this organism from clinical specimens [32–34].
Plates should be incubated for at least 5 days in 5%–7% CO2
at 357C–377C [29]. Most isolates appear after 2–4 days of incubation [29].
Even when adequate specimens are processed under optimal
culture conditions, recovery of F. tularensis is problematic. As
few as 10% of cases yield the organism [35]. When it is recovered
in the laboratory, however, extreme care should be exercised in
handling F. tularensis. It is a common cause of laboratoryacquired infection, notwithstanding its very infrequent isolation. Biosafety level 2 precautions should be employed [36].
In addition to biopsy-specimen cultures, blood cultures, especially for patients with the septicemic form of tularemia, may
be appropriate even though they are rarely positive and the
optimal detection system has not been delineated. As noted
B. burgdorferi is the etiologic agent of Lyme borreliosis [41,
42]. Attempts to culture this organism are rarely necessary;
however, if culture is to be performed, it should be restricted
to the acute, primary stage of infection. The specimens of choice
are skin lesion biopsy specimens, blood, and CSF from patients
with clinical evidence of meningeal involvement [43]. Optimal
sample sizes and transport conditions have not been defined.
In light of the absence of such information, transport of specimens directly to the laboratory, followed by immediate inoculation and incubation of media, are recommended.
Several different media have been proposed for culturing B.
burgdorferi from clinical specimens. All are derived from Kelly’s
medium, originally described in 1971 as a means for propagating Borrelia hermsii, the cause of tick-borne relapsing fever
in North America [44]. Stoenner’s modification, referred to as
“fortified Kelly’s medium” was described in 1982 [45] and modified by Barbour, giving rise to BSK-I (Barbour-Stoenner-Kelly)
medium [46] and later BSK-II medium, a semisolid liquid medium [47]. The principal ingredients of BSK-II medium are Nacetylglucosamine, peptone, bovine serum albumin, yeast extract, a supplement (CMRL, consisting of amino acids,
vitamins, nucleotides, and other growth factors), glucose, rabbit
serum, gelatin, and HEPES buffer [47].
Several additional modifications of BSK-II medium have
subsequently been described [48–50], but use of BSK-II medium
remains the cornerstone of efforts to recover B. burgdorferi from
clinical specimens. Supplementation of BSK-II medium with
various antimicrobial agents has been recommended as a means
of enhancing recovery of B. burgdorferi from specimens such
as skin biopsy specimens, which are potentially contaminated
with nonfastidious bacteria [43, 51–54].
Tubes containing BSK-II medium with or without antimicrobial agents should be tightly closed after specimen inoculation and then incubated at 327C–347C in ambient air for up
to 6 weeks before being considered negative and discarded.
Cultures should be examined visually every 2–3 days for macroscopic evidence of growth (e.g., turbidity, often occurring near
the bottom of the medium because of the microaerophilic nature
of B. burgdorferi). If growth is observed, a drop of turbid medium should be removed and examined for the presence of
CID 2000;30 (January)
Detection of Fastidious Bacteria
organisms morphologically compatible with B. burgdorferi (i.e.
spirochetes 10 to 30-mm long, with loose, irregular coils). Either
phase-contrast or dark-field microscopy is the preferred method
of visualizing B. burgdorferi microscopically.
B. burgdorferi stains inconsistently in gram preparations.
Giemsa or silver stains are preferred but often are not available
in clinical microbiology laboratories. At least once weekly in
macroscopically negative cultures, a drop of medium from the
bottom of culture tubes should be stained blindly for the presence of organisms morphologically compatible with B.
burgdorferi.
The likelihood of recovering B. burgdorferi from human clinical specimens depends on the quality and nature of the specimen, the stage of the disease, and the expertise of the laboratory. Recovery from blood has been reported for only 3.1%
and 5.6% of patients in 2 large studies [55, 56]. In contrast, up
to 45% recovery rates have been reported for skin biopsies in
patients with erythema chronicum migrans [48, 55, 57, 58]. Obviously, isolation rates are variable and tend to be low even
when care is taken to maximize recovery. Irrespective of the
specimen processed, recovery of B. burgdorferi probably occurs
during the early stages of the disease, especially before antibiotic
therapy is initiated.
Leptospira Species
The Leptospiraceae family is generally divided into 2 species,
Leptospira interrogans and Leptospira biflexa, the latter species
encompassing free-living, nonpathogenic forms [59]. More than
200 serovars of L. interrogans have been recognized and, in
turn, are categorized into about 19 serogroups based on crossreacting antigens. Organisms in the L. interrogans group are
responsible for human infections. Leptospira species are harbored by numerous wild and domesticated animals [60] and are
excreted in the urine.
The diagnosis of leptospirosis usually is based on the results
of serological tests. Recovery of the organism in culture is
largely restricted to reference and public health laboratories.
However, occasionally, general clinical microbiology laboratories are justified in attempting to culture Leptospira species
from human specimens. Blood should be cultured during the
acute stage of the disease, CSF and urine later in the illness.
Numerous different semisolid media, dispensed in 5 to 10mL aliquots in sterile screw-capped tubes, may be used to propagate Leptospira species. These include Ellinghausen-McCullough medium as modified by Johnson and Harrison
(EMJH); supplemented with bovine serum albumin and polysorbate 80; and Fletcher’s, Stuart’s, and Korthof’s media
[59–63], the latter 3 being supplemented with rabbit serum. The
optimal medium for culture of Leptospira species has not been
defined; it may be appropriate when performing Leptospira
cultures to inoculate 11 medium.
Blood is cultured by placing 1–4 drops of specimen into each
169
of 3–5 culture tubes. With CSF specimens, 0.5 mL of fluid is
cultured per tube. Because of the possibility of inhibitory substances and bacterial contamination, urine specimens should
be cultured undiluted and in serial 10-fold dilutions up to 1023
or 1024, 1 drop per culture tube. Five-fluorouracil (200 mg/mL)
[63] or neomycin (6-mm disk containing 30 mg) [59] can be
added to leptospiral culture media to suppress growth of contaminants present in urine specimens. Whenever an antibioticcontaining medium is inoculated, a companion tube lacking
antibiotic should also be inoculated.
Culture tubes should be incubated at 307C in ambient air in
the dark for up to 4 months [59] and examined once weekly
for evidence of growth of leptospiras. By use of aseptic technique, a drop of medium 1–3 cm below the surface is aspirated
and examined with a dark-field or phase-contrast microscope
for the presence of spirochetes with characteristic morphology
and motility. Leptospira species organisms are typically ∼0.1
mm in diameter and 6–12 mm in length. Organisms are tightly
coiled (118 coils per cell) and have conspicuous hooks at one
or both ends.
A positive control culture should be inoculated with a known
viable stock culture of Leptospira species at the time all clinical
specimens are processed. This process attests to the adequacy
of what is invariably a little-used culture routine and serves as
a source of organisms from which microscopic comparisons
can be made. Alternative leptospiral culture methods have been
described elsewhere but remain investigational [64, 65].
Given the technical complexity and difficulties in culturing
Leptospira, one can understand why such requests should virtually always be referred to public health or reference
laboratories.
Bartonella Species
Bartonella species are small, curved, highly fastidious gramnegative bacilli. Five species have been delineated: Bartonella
bacilliformis, Bartonella vinsonii, Bartonella quintana, Bartonella
henselae, and Bartonella elizabethae [66, 67]. B. bacilliformis is
the causative agent of a life-threatening bacteremic illness
known as Oroya fever, which occurs in the Andean mountain
regions of Colombia, Equador, and Peru. B. vinsonii is not
known to cause human infection. B. quintana is the principal
etiologic agent of louse-borne trench fever.
B. henselae is recognized as a cause of bacillary angiomatosis,
parenchymal bacillary peliosis, endocarditis, and fever with
bacteremia, all of which occur most commonly but not exclusively in patients who are infected with HIV [68–74]. Uncommonly, B. quintana and the most recently described species of
Bartonella, B. elizabethae [67], may be associated with some of
these same infections. In addition, B. henselae is now thought
to be the principal etiologic agent of cat-scratch disease, in some
cases presumably in conjunction with a related organism, Afipia
felis [73].
170
Doern
For patients suspected of having bartonella infections, blood
or biopsy specimens of involved tissue offer the best opportunity for culture recovery of the organism [66]. Most recent
experience with isolation of Bartonella species in the laboratory
has been with B. henselae. Although B. henselae has been propagated on standard media, such as 5% sheep blood and enriched
chocolate agar, the optimal solid culture medium for growth
of this organism appears to be freshly prepared heart infusion
agar containing 5%–10% defibrinated rabbit or horse blood
[66]. Recently, a chemically defined liquid medium has been
described that yielded excellent growth of several clinical isolates of B. henselae [75]. The utility of this liquid medium for
processing clinical specimens from patients with bartonella infections needs to be further explored.
Tissue specimens should be transported to the laboratory
expeditiously and then, after homogenization, inoculated directly onto solid medium. Blood specimens should be collected
in lysis-centrifugation (Isolator) tubes and transported directly
to the laboratory, and concentrates should be subcultured
promptly to solid media [66]. The optimal volume of blood per
culture, the preferred number of cultures, and the timing of
collection for maximum recovery of Bartonella species have not
been defined.
Similarly, the optimal subculture routine is unknown. Plates,
whether inoculated with tissue specimens or lysis-centrifugation
concentrates, should be incubated at 357C–377C in a humidified
atmosphere of 5%–10% CO2 for up to 4 weeks before they are
considered negative and discarded. Recovery of Bartonella species from instrument-based broth blood cultures has been described elsewhere [76, 77]; however, the optimal system has not
been defined.
The culture routines described above are probably also applicable to non-henselae Bartonella species, in particular, B.
quintana and B. elizabethae. Growth of B. bacilliformis and A.
felis is facilitated at lower temperatures of incubation
(257C–307C).
The yield from cultures for Bartonella species is unknown.
Even when cultures are performed under optimal conditions,
isolation rates are very low. As a result, serology [66] and nonculture-based molecular detection methods such as PCR [78]
are important adjuncts to establishing an etiologic diagnosis of
bartonella infections.
Bordetella pertussis
Because of the resurgence of pertussis as a clinical problem
in the United States [79], there is renewed interest in the recovery of B. pertussis in culture. Bordetella parapertussis may
cause a similar albeit less-severe illness and may not have the
same epidemiological implications as B. pertussis. This discussion will focus on the culture recovery of B. pertussis; however,
the recommendations stated below may also be considered applicable to B. parapertussis.
CID 2000;30 (January)
Two other species within this genus, Bordetella bronchiseptica
and Bordetella avium, pathogens of dogs and turkeys, respectively, have only rarely been isolated from humans and will not
be considered herein.
B. pertussis is an extremely fastidious gram-negative bacillus
that typically fails to grow even on enriched chocolate agar, at
least on primary isolation. Optimal recovery is achieved by
obtaining samples from the nasopharynx, either by swabbing
with calcium alginate or synthetic-polyester swabs on a flexible
wire or by aspiration [80–86]. Pharyngeal swab specimens
should be avoided [83]. Cotton swabs should not be used because they contain toxic substances such as fatty acids on the
cotton fibers. The swab should be inserted well into the nasopharynx, rotated several times, and left in place for 30–60 s
[81]. Upon removal, nasopharyngeal swabs should be immediately inoculated to suitable agar medium at the patient’s location or placed directly into transport medium.
Numerous different transport media have been recommended [81, 86–90], but Regan-Lowe transport medium containing half-strength charcoal agar, 10% defibrinated horse
blood, and cephalexin (40 mg/mL) is probably the most useful
[81, 91] because of its long shelf-life, its commercial availability,
and the extensive experience with its use. In circumstances
where transport of nasopharyngeal swab specimens to the laboratory will be delayed, several studies have suggested that
incubation of Regan-Lowe transport media at the collection
site for 1–2 days prior to transport will increase culture recovery, presumably owing to inhibition of contaminants present
in the specimen with simultaneous initial growth of B. pertussis
[85, 87, 91, 92]. In addition, maintaining specimens at 47C rather
than at 257C prior to and during transport may enhance recovery [86, 93]. When transport to the laboratory can be accomplished within a few hours or less, swabs in transport media
should be transported immediately at room temperature.
Numerous different media have been advocated for use in
the culture recovery of B. pertussis from clinical specimens [81,
82, 89, 92, 94–100]. Bordet-Gengou medium, consisting of
starch, glycerol, NaCl, and 5%–20% defibrinated sheep or horse
blood, supports luxuriant growth of B. pertussis, but it has a
very short shelf-life and does not effectively suppress growth
of contaminants [98]. As a result, the presence of B. pertussis,
a slow-growth organism, can be obscured by contaminants.
Addition of antimicrobials to Bordet-Gengou agar limits this
problem.
In 1953 Mishulow et al. recommended use of charcoal agar
for removing toxic substances from clinical specimens that interfere with growth of B. pertussis [99]. This substitution also
substantially lengthened the shelf-life of the medium. Further
modifications included the addition of penicillin at a concentration of 0.3 mg/mL [94] and later cephalexin (40 mg/mL) [96]
to suppress contaminating flora, as well as 10% defibrinated
horse blood to encourage growth [96]. It is this medium, often
CID 2000;30 (January)
Detection of Fastidious Bacteria
referred to as charcoal–horse blood agar or Regan-Lowe agar,
that is most commonly used today.
Because of the possibility of inhibition of the growth of some
strains of B. pertussis by the high concentration of cephalexin
in charcoal–horse blood agar, it has been recommended that a
plate lacking cephalexin or containing different antimicrobial
agent(s) be inoculated along with the cephalexin-containing
plate [93]. Assuming fresh medium is used, this very conservative culture approach appears to be unnecessary in most cases
[81, 101].
Plates should be incubated in a humidified environment for
7–10 days in ambient atmospheric air [101] at 357C [81] and
examined daily for the appearance of colony growth morphologically consistent with B. pertussis. Many laboratories have
traditionally incubated B. pertussis cultures in an elevated CO2
atmosphere of 5%–10%. There is now evidence, however, that
ambient atmospheric air is superior to 5%–10% CO2 incubation
[101]. The use of broth enrichment prior to culture is of no
definable value [102].
The rate of culture-positivity among patients with pertussis
has been shown to vary markedly (e.g., 20%–83% [81,
103–105]). This wide variation undoubtedly is due to differences
in patient ages; antibiotic administration; organism load at the
time specimens were collected; adequacy of the specimens;
transport medium; temperature and time; culture medium used;
incubation conditions; and the expertise and familiarity of laboratory personnel with B. pertussis cultures. Notwithstanding
these considerations, culture remains an important vehicle for
establishing an etiologic diagnosis of pertussis. The significance
of culture recovery of B. pertussis is underscored by the recent
recognition of resistance to erythromycin in a clinical isolate
of this organism [106].
Conclusions
This discussion has centered around optimizing culture routines for recovery of 7 fastidious bacteria. As noted in the introductory paragraphs, assuming representative clinical specimens have been submitted, the culture recovery of Legionella
species, Brucella species, F. tularensis, B. burgdorferi, Leptospira
species, or Bartonella species unequivocally defines a patient as
having disease due to that organism. In most cases, the same
conclusion can be drawn from the culture recovery of B. pertussis. Furthermore, the prevalence of infection due to all of
these organisms, again with the possible exception of B. pertussis, is also very low. As a result, many clinical laboratories
often do not have the requisite expertise for performing cultures, and the cost of performing cultures may be prohibitive
because the necessary media and supplies become outdated before they can be used.
For these reasons, cultures for Legionella species, Brucella
species, F. tularensis, B. burgdorferi, Leptospira species, and
Bartonella species, except in cases in which blood cultures are
171
indicated and can be performed by use of a continuous-monitoring device, should probably be restricted to clinical microbiology laboratories that handle a large amount of referral
testing.
Acknowledgment
I am very grateful for the excellent secretarial assistance of Kay
Meyer, who typed the manuscript.
References
1. Stout J, Yu VL, Vikers RM, et al. Ubiquitousness of Legionella pneumophila
in the water supply of a hospital with endemic Legionnaires’ disease. N
Engl J Med 1982; 306:466–8.
2. Muder RR, Yu VL. Legionnaires’ disease and related pneumonias. In: Gorbach SL, Bartlett JG, Blacklow NR, eds. Infectious diseases. Philadelphia: WB Saunders, 1992:505–12.
3. Edelstein PH, Meyer RD, Finegold SM. Laboratory diagnosis of Legionnaires’ disease. Am Rev Respir Dis 1980; 121:317–27.
4. Winn WD Jr, Pasculle AW. Laboratory diagnosis of infections caused by
Legionella species. Clin Lab Med 1982; 2:343–69.
5. Zuravleff JJ, Yu VL, Shourard JW, Davis BK, Rihs JD. Diagnosis of Legionnaires’ disease: an update of laboratory methods with new emphasis
on isolation by culture. JAMA 1983; 250:1981–5.
6. Winn WC Jr. Legionnaires disease: historical perspective. Clin Microbiol
Rev 1988; 1:60–81.
7. Feeley JC, Gorman GW, Weaver RE, Mackel DC, Smith HW. Primary
isolation media for Legionnaires disease bacterium. J Clin Microbiol
1978; 8:320–5.
8. Feeley JC, Gibson RJ, Gorman GW, et al. Charcoal–yeast extract agar:
primary isolation medium for Legionella pneumophila. J Clin Microbiol
1979; 10:437–41.
9. Pasculle AW, Feeley JC, Gibson RJ, et al. Pittsburgh pneumonia agent:
direct isolation from human lung tissue. J Infect Dis 1980; 141:727–32.
10. Edelstein PH. Comparative study of selective media for isolation of Legionella pneumophila from potable water. J Clin Microbiol 1982; 16:697–9.
11. Pine L, Franzus MJ, Malcolm GB. Guanine is a growth factor for Legionella
species. J Clin Microbiol 1986; 23:163–9.
12. Hoffman, PS, Pine L, Bell S. Production of superoxide and hydrogen peroxide in medium used to culture Legionella pneumophila: catalytic decomposition by charcoal. Appl Environ Microbiol 1983; 45:784–91.
13. Winn WC Jr. Legionella. In: Murray PA, Baron EJ, Pfaller MA, Tenover
FC, Yolken RH, eds. Manual of clinical microbiology. 6th ed. Washington, DC: American Society for Microbiology, 1995:533–44.
14. Tompkins LS, Roessler BJ, Redd SC, Markowitz LE, Cohen ML. Legionella
prosthetic-valve endocarditis. N Engl J Med 1988; 318:530–5.
15. Reimer LG, Carroll KC. Role of the microbiology laboratory in the diagnosis of lower respiratory tract infections. Clin Infect Dis 1998; 26:
742–8.
16. Taylor JP, Perdue JN. The changing epidemiology of human brucellosis in
Texas, 1977–1986. Am J Epidemiol 1989; 130:160–5.
17. Chomel BB, DeBess EE, Mangiamele DM, et al. Changing trends in the
epidemiology of human brucellosis in California from 1973 to 1992: a
shift toward foodborne transmission. J Infect Dis 1994; 170:1216–23.
18. Young EJ. Brucella species. In: Mandell GL, Bennett JE, Dolin R, eds.
Principles and practice of infectious disease. 4th ed. New York: Churchill
Livingstone, 1995:253–60.
19. Gotuzzo E, Carrillo C, Guerra J, Llosa L. An evaluation of diagnostic
methods for brucellosis—the value of bone marrow culture. J Infect Dis
1986; 153:122–5.
20. Daugherty MP, Dolter J, Evans GC, et al. Processing of specimens for
172
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
Doern
isolation of unusual organisms. Part 6. Brucella spp. In: Isenberg HD,
ed. Clinical microbiology procedures handbook. Washington, DC:
American Society for Microbiology, 1992:1.18.23–7.
Mayer NP, Holcomb LA. Brucella. In: Murray PA, Baron EJ, Pfaller MA,
Tenover FC, Yolken RH, eds. Manual of clinical microbiology. 6th ed.
Washington, DC: American Society for Microbiology, 1995:549–55.
Doern GV, Davaro R, George M, Campognone P. Lack of requirement for
prolonged incubation of Septi-Chek blood culture bottles in patients
with bacteremia due to fastidious bacteria. Diagn Microbiol Infect Dis
1996; 24:141–4.
Yagupsky P, Peled N, Press J, Abu-Rashid M, Abramson O. Rapid detection
of Brucella melitensis from blood cultures by a commercial system. Eur
J Clin Microbiol Infect Dis 1997; 16:605–7.
Solomon HM, Jackson D. Rapid diagnosis of Brucella melitensis in blood:
some operational characteristics of the BACT/Alert. J Clin Microbiol
1992; 30:222–4.
Navas E, Guerrero A, Cobo J, Loza E. Faster isolation of Brucella spp.
from blood by Isolator compared with BACTEC NR. Diagn Microbiol
Infect Dis 1993; 16:79–81.
Bannatyne RM, Jackson MC, Memish Z. Rapid diagnosis of Brucella bacteremia by using the BACTEC 9240 system. J Clin Microbiol 1997; 35:
2673–4.
Yagupsky P, Peled N, Press J, Abramson O, Abu-Rashid M. Comparison
of BACTEC 9240 Peds Plus medium and Isolator 1.5 microbial tube
for detection of Brucella melitensis from blood cultures. J Clin Microbiol
1997; 35:1382–4.
Hopla CE. The ecology of tularemia. Adv Vet Sci Comp Med 1974; 18:
25–53.
Stewart SJ. Francisella. In: Murray PA, Baron EJ, Pfaller MA, Tenover FC,
Yolken RH, eds. Manual of clinical microbiology. 6th ed. Washington,
DC: American Society for Microbiology, 1995:545–8.
Penn RL. Francisella tularensis (tularemia). In: Mandell GL, Bennett JE,
Dolin R, eds. Principles and practice of infectious diseases. 4th ed. New
York: Churchill Livingstone, 1995:2060–8.
Bernard K, Tessier S, Winstanley J, Chang D, Borczyk A. Early recognition
of atypical Francisella tularensis strains lacking a cysteine requirement.
J Clin Microbiol 1994; 32:551–3.
Baker CN, Hollis DG, Thornsberry C. Antimicrobial susceptibility testing
of Francisella tularensis with a modified Mueller-Hinton broth. J Clin
Microbiol 1985; 22:212–5.
Clark WA, Hollis DG, Weaver RE, Riley P. Identification of unusual pathogenic gram-negative aerobic and facultatively anaerobic bacteria. US
Department of Health and Human Services publication no. 017-02300149. Washington, DC: US Government Printing Office, 1984:164–5.
Westerman EL, McDonald J. Tularemia pneumonia mimicking Legionnaires’ disease: isolation of organism on CYE agar and successful treatment with erythromycin. South Med J 1983; 76:1169–70.
Taylor JP, Istre GR, McChesney TC, Satalowich FT, Parker RL, McFarland
LM. Epidemiologic characteristics of human tularemia in the southwestcentral states, 1981–1987. Am J Epidemiol 1991; 133:1032–8.
US Department of Health and Human Services. Biosafety in microbiological
and biomedical laboratories. 2d ed. US Department of Health and Human Services publication no. 88-8395. Washington, DC: US Government
Printing Office, 1988.
Provenza JM, Klotz SA, Penn RL. Isolation of Francisella tularensis from
blood. J Clin Microbiol 1986; 24:453–5.
Centers for Disease Control. Tularemic pneumonia—Tennessee. MMWR
Morb Mortal Wkly Rep 1983; 32:363–9.
Evans ME, Gregory DW, Schaffner W, McGee ZA. Tularemia: a 30-year
experience with 88 cases. Medicine (Baltimore) 1985; 64:251–69.
Kaiser AB, Rieves D, Price AH, et al. Tularemia and rhabdomyolysis.
JAMA 1985; 253:241–3.
Steere AC, Broderick TE, Malwista SE. Erythema chronicum migrans and
CID 2000;30 (January)
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
Lyme arthritis: epidemiologic evidence for a tick vector. Am J Epidemiol
1978; 108:312.
Steere AC. Borrelia burgdorferi (Lyme disease, Lyme borreliosis). In: Mandell GL, Bennett JE, Dolin R, eds. Principles and practice of infectious
diseases. 4th ed. New York: Churchill Livingstone, 1995:2143–55.
Barbour AG. Laboratory aspects of Lyme borreliosis. Clin Microbiol Rev
1988; 1:399–414.
Kelly R. Cultivation of Borrelia hermsii. Science 1971; 173:443.
Stoenner HG, Dodd T, Larsen C. Antigenic variation of Borrelia hermsii.
J Exp Med 1982; 156:1297–311.
Barbour AG, Burgdorfer W, Hayes SF, Peter O, Aeschlimann A. Isolation
of a cultivable spirochete from Ixodes ricinus ticks of Switzerland. Curr
Microbiol 1983; 8:123–6.
Barbour AG. Isolation and cultivation of Lyme disease spirochetes. Yale J
Biol Med 1984; 57:521–5.
Rawlings JA, Fournier PV, Teltow GA. Isolation of Borrelia spirochetes
from patients in Texas. J Clin Microbiol 1987; 25:1148–50.
Pollack RJ, Telford SR III, Spielman A. Standardization of medium for
culturing Lyme disease spirochetes. J Clin Microbiol 1993; 31;1251–5.
Callister SM, Case KL, Agger WA, Schell RF, Johnson RC, Ellingson JLE.
Effects of bovine serum albumin on the ability of Barbour-StoennerKelly medium to detect Borrelia burgdorferi. J Clin Microbiol 1990; 28:
363–5.
Johnson SE, Klein GC, Schmid GP, Bowen GS, Feeley JC, Schulze T. Lyme
disease: a selective medium for isolation of the suspected etiological
agent, a spirochete. J Clin Microbiol 1984; 19:81–2.
Steere AC, Grodzicki RL, Kornblatt AN, et al. The spirochetal etiology of
Lyme disease. N Engl J Med 1983; 308:733–40.
Burgdorfer WR, Lane RS, Barbour AG, Gresbrink RA, Anderson JR. The
western black-legged tick, Ixodes pacificus: a vector of Borrelia burgdorferi. Am J Trop Med Hyg 1985; 34:925–30.
Burgdorfer W, Gage KL. Susceptibility of the hispid cotton rat (Sigmodon
hispidus) to the Lyme disease spirochete (Borrelia burgdorferi). Am J
Trop Med Hyg 1987; 37:624–8.
Steere AC, Grodzicki RL, Craft JE, Shrestra M, Kornblatt AN, Malawista
SE. Recovery of Lyme disease spirochetes from patients. Yale J Biol Med
1984; 57:557–60.
Benach JL, Bosler EM, Hanrahan JP, et al. Spirochetes isolated from the
blood of two patients with Lyme disease. N Engl J Med 1983; 308:740–2.
Berger BW, Kaplan MH, Rothenberg IR, Barbour AG. Isolation and characterization of the Lyme disease spirochete from the skin of patients
with erythema chronicum migrans. J Am Acad Dermatol 1985; 13:444–9.
Preac-Mursic V, Wilske B, Schierz G, Pfister HW, Einhaupl K. Repeated
isolation of spirochetes from the cerebrospinal fluid of a patient with
meningoradiculitis Bannwart. Eur J Clin Microbiol 1984; 3:564–5.
Kauffmann AF, Weyant RS. Leptospiraceae. In: Murray PA, Baron EJ,
Pfaller MA, Tenover FC, Yolken RH, eds. Manual of clinical microbiology. 6th ed. Washington, DC: American Society for Microbiology,
1995:621–5.
Farrar WE. Leptospira species (leptospirosis). In: Mandell GL, Bennett JE,
Dolin R, eds. Principles and practice of infectious disease. 4th ed. New
York: Churchill Livingstone, 1995:2137–41.
Cole JR. Spirochetes. In: Carter GR, Cole JR, eds. Diagnostic procedures
in veterinary bacteriology and mycology. 5th ed. New York: Academic
Press, 1990:41–60.
Saubolle MA. Leptospirosis. In: Wentworth BB, ed. Diagnostic procedures
for bacterial infections. 7th ed. Washington, DC: American Public Health
Association, 1987:335–46.
Baron EJ, Peterson LR, Finegold SM, eds. Spirochetes and other spiralshaped organisms. In: Bailey and Scott’s diagnostic microbiology. 9th
ed. St. Louis: Mosby, 1994:445–50.
Manca N, Verardi R, Colombrita D, Ravizzola G, Savoldi E, Turano A.
A radiometric method for the rapid detection of Leptospira organisms.
J Clin Microbiol 1986; 23:401–3.
CID 2000;30 (January)
Detection of Fastidious Bacteria
65. Rule PL, Alexander AD. Gellan gum as a substitute for agar in leptospiral
media. J Clin Microbiol 1986; 23:500–4.
66. Slater LN, Welch DF. Bartonella. In: Murray PA, Baron EJ, Pfaller MA,
Tenover FC, Yolken RH, eds. Manual of clinical microbiology. 6th ed.
Washington, DC: American Society for Microbiology, 1995:690–5.
67. Daly JS, Worthington MG, Brenner DJ, et al. Rochalimaea elizabethae sp.
nov. isolated from a patient with endocarditis. J Clin Microbiol 1993;
31:872–81.
68. Regnery RL, Anderson BE, Clarridge JE III, Rodriguez-Barradas MC,
Jones DC, Carr JH. Characterization of a novel Rochalimaea species,
R. henselae sp. nov., isolated from blood of a febrile, human immunodeficiency virus–positive patient. J Clin Microbiol 1992; 30:265–74.
69. Welch DF, Hensel DM, Pickett DA, San Joaquin VH, Robinson A, Slater
LN. Bacteremia due to Rochalimaea henselae in a child: practical identification of isolates in the clinical laboratory. J Clin Microbiol 1993; 31:
2381–6.
70. Welch DF, Pickett DA, Slater LN, Steigerwalt AG, Brenner DJ. Rochalimaea henselae sp. nov., a cause of septicemia, bacillary angiomatosis,
and parenchymal bacillary peliosis. J Clin Microbiol 1992; 30:275–80.
71. Koehler JE, Quinn FD, Berger TG, LeBoit PE, Tappero JW. Isolation of
Rochalimaea species from cutaneous and osseous lesions of bacillary
angiomatosis. N Engl J Med 1992; 327:1625–32.
72. Reed J, Brigati DJ, Flynn SD, et al. Immunocytochemical identification of
Rochalimaeae henselae in bacillary (epithelioid) angiomatosis, parenchymal bacillary peliosis, and persistent fever with bacteremia. Am J Surg
Pathol 1992; 16:650–7.
73. Schwartzman WA. Infections due to Rochalimaea: the expanding clinical
spectrum. Clin Infect Dis 1992; 15:893–902.
74. Slater LN, Welch DF, Min KW. Rochalimaea henselae causes bacillary angiomatosis and peliosis hepatitis. Arch Intern Med 1992; 152:602–6.
75. Wong MT, Thornton DC, Kennedy RC, Dolan MJ. A chemically defined
liquid medium that supports primary isolation of Rochalimaea (Bartonella) henselae from blood and tissue specimens. J Clin Microbiol
1995; 33:742–4.
76. Larson AM, Dougherty MJ, Nowowiejski DJ, et al. Detection of Bartonella
(Rochalimaea) quintana by routine acridine orange staining of broth
blood cultures. J Clin Microbiol 1994; 32:1492–6.
77. Spach DH, Callis KP, Paauw DS, et al. Endocarditis caused by Rochalimaea
quintana in a patient infected with human immunodeficiency virus. J
Clin Microbiol 1993; 31:692–4.
78. Anderson B, Sims K, Regnery R, et al. Detection of Rochalimaea henselae
DNA in specimens from cat scratch disease patients by PCR. J Clin
Microbiol 1994; 32:942–8.
79. Centers for Disease Control and Prevention. Resurgence of pertussis—United States, 1993. MMWR Morb Mortal Wkly Rep 1993; 42:
952–60.
80. Regan J. The laboratory diagnosis of whooping cough. Clin Microbiol
Newsl 1980; 2:78–82.
81. Marcon MJ. Bordetella. In: Murray PA, Baron EJ, Pfaller MA, Tenover
FC, Yolken RH, eds. Manual of clinical microbiology. 6th ed. Washington, DC: American Society for Microbiology, 1995:566–73.
82. Freidman RL. Pertussis: the disease and new diagnostic methods. Clin Microbiol Rev 1988; 1:365–76.
83. Marcon MJ, Hamoudi AC, Cannon HJ, Hribar MM. Comparison of throat
and nasopharyngeal swab specimens for culture diagnosis of Bordetella
pertussis infection. J Clin Microbiol 1987; 25:1109–10.
84. Hallander HO, Reizenstein E, Renemar B, Rasmuson G, Mardin L, Olin
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
173
P. Comparison of nasopharyngeal aspirates with swabs for culture of
Bordetella pertussis. J Clin Microbiol 1993; 31:50–2.
Hoppe JE, Weib A. Recovery of Bordetella pertussis from four kinds of
swabs. Eur J Clin Microbiol 1987; 6:203–5.
Ross PW, Cumming CG. Isolation of Bordetella pertussis from swabs. Br
Med J 1981; 283:403–4.
Hoppe JE, Woerz S, Botzenhart K. Comparison of specimen transport systems for Bordetella pertussis. Eur J Clin Microbiol 1986; 5:671–3.
Hunter PR. Survival of Bordetella pertussis in transport media. J Clin Pathol
1986; 39:119–20.
Regan J, Lowe F. Enrichment medium for the isolation of Bordetella. J Clin
Microbiol 1977; 6:303–9.
Gilligan P. Laboratory diagnosis of Bordetella pertussis infection. Clin Microbiol Newsl 1983; 5:115–7.
Hoppe JE. Methods for isolation of Bordetella pertussis from patients with
whooping cough. Eur J Clin Microbiol Infect Dis 1988; 7:616–20.
Kurzynski TA, Boehm DM, Rott-Petri JA, Schell RF, Allison PE. Comparison of modified Bordet-Gengou and modified Regan-Lowe media
for the isolation of Bordetella pertussis and Bordetella parapertussis. J
Clin Microbiol 1988; 26:2661–3.
Morrill WE, Barbaree JM, Fields BS, Sanden GN, Martin WT. Effects of
transport temperature and medium on recovery of Bordetella pertussis
from nasopharyngeal swabs. J Clin Microbiol 1988; 26:1814–7.
Jones GL, Kendrick PL. Study of a blood-free medium for transport and
growth of Bordetella pertussis. Health Lab Sci 1969; 6:40–5.
Stauffer LR, Brown DR, Sandstrom RD. Cephalexin-supplemented JonesKendrick charcoal agar for selective isolation of Bordetella pertussis:
comparison with previously described media. J Clin Microbiol 1983; 17:
60–2.
Sutcliffe EM, Abbott JD. Selective medium for the isolation of Bordetella
pertussis and parapertussis. J Clin Pathol 1972; 25:732–3.
Aoyama T, Murase Y, Iwata T, Imaizumi A, Suzuki Y, Sato Y. Comparison
of blood-free medium (cyclodextrin solid medium) with Bordet-Gengou
medium for clinical isolation for Bordetella pertussis. J Clin Microbiol
1986; 23:1046–8.
Bordet J, Gengou O. Le microbe de la coqueluche. Ann Inst Pasteur (Paris)
1906; 20:731–41.
Mishulow L, Sharpe LS, Cohen L. Beef heart charcoal agar for the preparation of pertussis vaccine. Am J Public Health 1953; 43:1466–72.
Hoppe JE, Schwaderer J. Comparison of four charcoal media for the isolation of Bordetella pertussis. J Clin Microbiol 1989; 27:1097–8.
Hoppe JE, Schlagenhaur M. Comparison of three kinds of blood and two
incubation atmospheres for cultivation of Bordetella pertussis on charcoal
agar. J Clin Microbiol 1989; 27:2115–7.
Hoppe JE, Weiss A, Woerz S. Failure of charcoal-horse blood broth with
cephalexin to significantly increase the rate of Bordetella isolation from
clinical specimens. J Clin Microbiol 1988; 26:1248–9.
Halperin SA, Bortolussi R, Wort AJ. Evaluation of culture, immunofluorescence and serology for the diagnosis of pertussis. J Clin Microbiol
1989; 27:752–7.
Cruickshank R. A combined Scottish study. Diagnosis of whooping cough:
comparison of serological tests with isolation of Bordetella pertussis. Br
Med J 1970; 4:637–9.
Lewis FA, Gust ID, Bennet NK. On the aetiology of whooping cough. J
Hyg (Lond) 1973; 71:139–44.
Lewis K, Saubolle MA, Tenover FC, Rudinsky MF, Barbour SD, Cherry
JD. Pertussis caused by an erythromycin-resistant strain of Bordetella
pertussis. Pediatr Infect Dis J 1995; 14:388–91.