766

DIAGNOSTIC MICROBIOLOGY UPDATES

General Principles of Specimen Collection and Transport

Michael L. Wilson From the Department of Pathology and Laboratory Services, Denver
Health and Hospitals, and Department of Pathology, University of
Colorado School ofMedicine, Denver, Colorado

In this issue of Clinical Infectious Diseases, we present the first article in a series entitled .'Diagnostic
Microbiology Updates. " Although clinical microbiology is included in the curricula ofvirtually all infectious
diseasefellowships, the degree ofemphasis on this subject varies considerably. Infectious disease physicians-

Downloaded from https://academic.oup.com/cid/article/22/5/766/360981 by guest on 27 April 2024

even those who have direct responsibities or consulting responsibilities for the microbiology laboratories of
the institutions in which they practice-may be hard pressed to keep up with the rapidly changing content
of the primary literature in clinical microbiology. The purpose of this series, therefore, is at least in part to
fill this void and to provide concise updates for clinicians. The first article, written by Dr. Michael 1. Wilson,
reviews current concepts in specimen collection and transport. A key issue for all clinicians (which is not
always sufficiently emphasized) is the quality ofthe specimen submitted to the laboratory. It is an axiom that
if specimens of poor quality are submitted, the results generated by the laboratory will have little or no
clinical utility. Dr. Wilson's article describes some of the methods available to assure that only specimens
of good quality, i.e., those most likely to be useful clinically, are processed in the microbiology laboratory.
Future articles will address specific types of specimens, groups of pathogens, and diagnostic techniques,
including molecular methods. We hope this series will be irformaiive and valuable to the readers of Clinical
Infectious Diseases, and we look forward to your comments.

Melvin P. Weinstein and L. Barth Reller

Departments of Medicine and Pathology, University of Medicine and
Dentistry of New Jersey- Robert Wood Johnson Medical School,
and the Microbiology Laboratory, Robert Wood Johnson University
Hospital, New Brunswick, New Jersey; and Departments of Pathology
and Medicine, Duke University School of Medicine, and the Clinical
Microbiology Laboratory, Duke University Medical Center, Durham.
North Carolina

Specimens submitted for microbiological testing require proper handling from the time of collec-
tion through all stages of transport, storage, and processing. Issues common to all clinical specimens
submitted for microbiological testing include not only proper identification but also collection tech-
niques that maximize recovery of microbial pathogens and minimize contamination. For specimens
such as sputum and urine, the relative proportions of microorganisms present in vivo must be
preserved, or culture results may be misleading. If specimens are handled properly, culture results
are easier to interpret, patient care is improved, and costs are potentially decreased. Although most
guidelines for specimen handling remain unchanged, a recent emphasis has been placed on modifying
traditional practices to decrease or eliminate unnecessary work, increase laboratory efficiency, and
make microbiological testing more cost effective.

Proper handling of specimens is crucial for obtaining micro- important factors-along with appropriate use of tests-in
biological test results that are both timely and clinically rele- maximizing the cost-effectiveness and clinical relevance of
vant. Proper handling of specimens is also one of the most microbiological testing. The purpose of this article is to review
recent changes in specimen handling, particularly specimen
collection, that can be used to modify traditional practices in
clinical microbiology [l].
Received 27 December 1995. General information on the collection, transport, and storage
Reprints or correspondence: Dr. Michael L. Wilson, Medical Laboratories of specimens from different body sites for microbiological test-
#0224, Denver General Hospital, 777 Bannock Street, Denver, Colorado 80204.
ing is presented in table 1. More-specific information is pre-
Clinical Infectious Diseases 1996;22:766-77
© 1996 by The University of Chicago. All rights reserved.
sented in the references from which this table was derived
1058-4838/9612205 -00 18$02.00 [2-6].
ern 1996;22 (May) Microbiological Specimen Collection and Transport 767

General Principles Test requisitions should specify whether separate specimens

Collection. Several axiomatic principles guide the collec- have been submitted for culture and for histopathologic or
tion of specimens for microbiological testing (table 2). The cytopathologic examination or whether the laboratory staff
most self-evident of these principles is that specimens must be needs to divide the specimen. In general, it is preferable for
collected with use of strict aseptic technique from anatomic specimens to be divided in the laboratory, since this provides
sites most likely to yield pathogenic microorganisms. Even so, an opportunity for the staff and the pathologists to examine
it is surprising how often microbiology laboratories receive specimens to determine which portions should be cultured and
specimens collected from sites that are inappropriate for testing. which should be processed for pathological examination; this
Common examples include sinus tract specimens from patients is particularly important in terms of infectious diseases for
with suspected osteomyelitis, surface material from decubitus which histopathologic examination can yield rapid preliminary
ulcers or diabetic foot ulcers, nasal swab specimens from pa- or definitive identification of etiologic agents. Even if microbial
tients with suspected sinusitis, and 24-hour collections of urine. identification is not possible, classification of characteristic tis-

Downloaded from https://academic.oup.com/cid/article/22/5/766/360981 by guest on 27 April 2024

Laboratory personnel and clinical staff should define carefully sue reactions yields important clues to diagnosis, may be used
which types of specimens are appropriate for testing. to modify microbiological testing, and can be used to guide or
Specimens should be collected in such a way that contamina- modify empirical antimicrobial therapy. Finally, it is important
tion by indigenous flora is minimized. This is of paramount that the pathologist grossly examine tissue specimens to deter-
importance for cultures of blood, bone, and other tissues or mine if other disease processes are present because the gross
fluids in which infection is often caused by indigenous flora morphological features associated with some infectious dis-
and for specimens collected from sites of putative infection eases overlap those associated with other types of disease.
that are contiguous to, or immediately adjacent to, cutaneous Transport. In general, specimens collected for microbio-
or mucosal surfaces. logical testing can be transported in sterile specimen containers
Sufficient material must be submitted for cultures and other or, in the case of fluid specimens, in the syringe in which the
tests; laboratory staff are often asked to perform routine cul- specimen was collected. This is especially true for specimens
tures, anaerobic cultures, or fungal and mycobacterial cultures collected from hospitalized patients, since transportation to the
on specimens consisting only of swabs or a few drops of fluid. laboratory is generally faster and more reliable than it is for
Volume, while important for all specimens, is crucial for blood specimens collected from physicians' offices, clinics, or off-
and for mycobacterial and fungal cultures of CSF and urine. site facilities. If transport of a specimen will be delayed, use
Whenever possible, tissue or fluid should be submitted for of a transport device may be necessary to optimize testing.
culture; with a few obvious exceptions (e.g., throat cultures Many commercial products are available for transporting tissue
and urethral cultures), swab specimens are unacceptable. While and/or fluid specimens.
swabs have the advantage of being convenient and easy to use, Specimens for culture should be transported to the laboratory
they limit the volume of specimen that can be collected, they as promptly as possible for processing. Health care systems
can compromise a direct gram stain, they become contaminated should have an infrastructure that eliminates systematic delays
easily, and they can adversely affect recovery of certain micro- in either transport or processing; unavoidable delays must be
organisms [5]. Although occasions do occur when collection minimized. Specimen containers must be transported in such
of tissue or fluid is not possible and swabs must therefore be a way as to minimize damage that could result in contamination
used, this should be an infrequent event in routine patient care. of the specimen or exposure ofpersonnel to blood or other body
Persons collecting specimens should provide complete infor- fluids. Most specimens can be transported at room temperature.
mation on specimen requisition forms or in computerized order- Some specimens must be transported on ice (table I). For health
entry systems. Important information includes (I) the specific care systems with off-site facilities, transportation procedures
site(s) from which specimens were collected; (2) whether the should be developed to minimize delays, prevent damage to
patient was receiving antimicrobial therapy prior to specimen specimens during transport, and prevent loss of specimens.
collection or at the time specimens were collected; (3) specific Storage. Most specimens requiring prolonged storage be-
pathogens that are being sought; (4) the methods by which fore processing should be refrigerated. Refrigeration maintains
specimens were collected; and (5) whether the patient may be the viability of pathogens and preserves them in their relative
infected with pathogens known to be hazardous to laboratory proportions. The latter factor is crucial when semiquantitative
personnel (e.g., Brucella or Mycobacterium tuberculosis). Such cultures or quantitative cultures (e.g., cultures of sputum or
information is necessary to ensure that specimens are processed urine) are necessary for interpretation of results. Refrigeration
promptly, that appropriate cultures are performed, that test pro- also minimizes the growth of contaminants. Specimens that
cessing is appropriate for the method of specimen collection should not be refrigerated include blood, which should be kept
(e.g., urine obtained via suprapubic aspiration vs. the clean- at room temperature or in an incubator at 35°C; CSF, which,
catch method), and that laboratory personnel are not inadver- with the exception of that collected for viral cultures, should
tently and unknowingly exposed to highly pathogenic micro- be transported at room temperature; and specimens submitted
organisms. for culture of Neisseria species, which should be transported
768 Wilson em 1996;22 (May)

Table 1. Guidelines for collection, transport, and storage of specimens for microbiological testing.

Transport
No. of (temperature, Storage
Specimen or site Container or method Volume (mL) specimens time) (temperature) Comments

Blood
Type of test
Routine Blood culture vials 20-30 for adults, 2-3 25'C 25'C or 35°C Avoid delays in processing lysis-
(aerobic and anaerobic) 1-5 for centrifugation tubes
or lysis-centrifugation children
tube
Fungal Aerobic blood culture 20-30 for adults 2-3 25°C 25'C or 35°C
vials, BACTEC HBV-
FM* vials, or lysis-
centrifugation tube

Downloaded from https://academic.oup.com/cid/article/22/5/766/360981 by guest on 27 April 2024

Mycobacterial BACTEC 13A or lysis- 10-20 25°C 25'C or 35°C Submit one specimen initially;
centrifugation tube repeat if negative but
mycobacteremia is clinically
suspected
Bone Sterile vial NA NA 25'C 4°C Only infected bone should be
cultured; avoid contamination
from sinus tracts or skin
Catheters
Type of catheter
Urinary NA NA NA NA NA Specimen is inappropriate for
culture
Vascular Catheter tip in sterile vial NA NA 25°C 4'C For detection of line sepsis, draw
peripheral blood for cultures
CNS
Abscess fluid Anaerobic vial 1-5 NA 25°C 4'C
CSF Sterile vial 1-5 NA 25'C WC for 4'C for 0:;;24- Cytomegalovirus loses infectivity
viral 48 h (-70°C if stored at - 20'C and with
cultures) for ;,.48 h freeze-thaw cycles
for viral
cultures)
Shunt/catheter fluid Sterile vial 1-5 NA 25'C 4'C
Tissue Anaerobic vial NA NA 25'C 4'C
Eye Sterile vial NA NA 25'C 4'C For bacterial cultures, handle in
same way as other tissues
Fluid
Abdominal Anaerobic vial 1-10 NA 25'C 4'C
Pericardial Anaerobic vial 1-10 NA 25'C 4'C
Pleural Anaerobic vial 1-10 NA 25'C 4°C
Synovial Anaerobic vial 1-10 NA 25'C 4'C
Other Anaerobic vial As appropriate NA 25'C 4'C
Genitourinary tract
(STDs)
Organism to be
cultured
Candida albicans Swab NA NA NA NA
Chlamydia Swab (transfer contents to NA NA 4°C 4'C Avoid use of cotton swabs
trachoma/is 2-sucrose phosphate
solution)
Haemophilus Swab NA NA NA NA
ducreyi
Herpes simplex Swab NA NA NA 4'C Avoid use of calcium alginate
virus swabs and swabs on wooden
applicator sticks
Mycoplasmal Swab (transfer contents to NA NA NA NA Avoid use of calcium alginate
Ureaplasma 2-sucrose phosphate) swabs and swabs on wooden
applicator sticks
Neisseria Insulate medium NA NA 25'C None; incubate Various transport systems are
gonorrhoeae immediately immediately available; avoid calcium
alginate swabs
Treponema pallidum Scraping or aspirate of NA NA Transport NA Perform darkfield microscopic
lesions on slide immediately examination immediately
Trichomonas Swab for culture, smears NA NA NA NA
vaginalis for rapid tests
em 1996;22 (May) Microbiological Specimen Collection and Transport 769

Table 1. (Continued)

Transport
No. of (temperature, Storage
Specimen or site Container or method Volume (mL) specimens time) (temperature) Comments

Genitourinary tract
(diseases other
than STDs)
Specimen or site of
specimen
collection
Amniotic fluid Anaerobic vial 1-10 NA 25'C 4'C
Cervix Swab NA NA NA NA Specimen is unacceptable for
anaerobic culture
1~5 25'C 4"C

Downloaded from https://academic.oup.com/cid/article/22/5/766/360981 by guest on 27 April 2024

Endometrium Anaerohic vial NA
Pelvic fluid Anaerohic vial 1-5 NA 25"C 4'C
(culdocentesis
fluid/abscess)
Prostate Obtain secretions via 1-5 NA 25"C 4'C Specimen is unacceptable for
prostatic massage (use anaerobic culture
sterile vial)
Vagina Swab NA NA 25"C 4'C Specimen is unacceptable for
anaerobic culture
Hair Sterile vial or Petri dish NA NA NA NA
Oral cavity Anaerobic vial NA NA 25'C 4"C Submit tissue or fluid collected
from site of infection; collect
specimen in such a way as to
eliminate or minimize
contaminationwith oral flora
Respiratory tract
Specimen or site of
specimen
collection
Bronchoscopy fluid Sterile vial NA NA Specimen is unacceptable for
anaerobic culture unless
collected with protected
catheter
Expectorated Sterile vial NA NA 25"C 4'C Specimen is unacceptable for
sputum anaerobic culture; screen for
contamination with saliva
Nasopharynx Swab NA NA 25'C (plate 4'C Specimen is unacceptable for
immediately) anaerobic culture
Sinuses Aspirate transferred to NA NA 25'C 4'C
anaerobic vial
Throat Swab NA NA 25"C (plate 4"C Specimen is unacceptable for
immediately) anaerobic culture
Skin and soft tissues
Site of collection or
organism to be
cultured
Deep wound/abscess Anaerobic vial, syringe NA NA
Dermatophytes Sterile Petri dish NA NA
Superficial wound Anaerobic vial, syringe NA NA Do not submit swabs of
specimens from the surface of
decubitus ulcers, diabetic fool
ulcers. margins of nonviable
amputations, or other wounds
Stool
Type oftest
Culture Sterile screw-capped jar or NA Immediately 4'C for fresh Do not submit specimens from
container specimen patients who develop diarrhea
Ova and parasite Sterile screw-capped jar or NA I (initially) Immediately 4'C for fresh after 3- 4 days of
examination container for fresh specimen hospitalization; submit
specimen; specimen for detectiou of
commercial Clostridium difficile
system for
preserved
specimens
770 Wilson em 1996;22 (May)

Table 1. (Continued)

Transport
No. of (temperature, Storage
Specimen or site Container or method Volume (mL) specimens time) (temperature) Comments

Urine For all specimens, submit>20

Specimen or site mL of urine for mycobacterial
or fungal cultures
Clean-catch Sterile vial 1-20 NA Immediately at 4'C Specimen is unacceptable for
4'C or 25'C anaerobic culture
Indwelling catheter Not acceptable for culture NA NA NA NA
Straight catheter Sterile vial 1-20 NA Immediately at 4'C Specimen is unacceptable for
4'C or 25'C anaerobic culture
Suprapubic aspirate Anaerobic vial 1-20 NA Immediately at 4'C Specimen is acceptable for

Downloaded from https://academic.oup.com/cid/article/22/5/766/360981 by guest on 27 April 2024

4°C or 25°C anaerobic culture

NOTE. Data are from [2-6]. NA = not applicable; STDs = sexually transmitted diseases .
• BACTEC high-blood-volume fungal medium (Becton Dickinson Diagnostic Instrument Systems, Sparks, MD).

in an atmosphere with sufficient CO2 and humidity and in a harm - to patients when specimens that have been improperly
manner that prevents wide temperature fluctuations. collected or improperly transported are processed and test re-
Written guidelines. Clinical microbiology laboratories sults are reported. Correct labeling is of particular importance
should distribute written guidelines for proper specimen collec- for ensuring that patient misidentification does not occur and
tion, transport, and storage. These guidelines should be com- that appropriate testing is performed. In all instances, the physi-
plete, explicit, and up-to-date. Copies should be available to cian who ordered the test, as well as the person who collected
all personnel who handle specimens. Identical guidelines the specimen, should be notified when a specimen is rejected.
should be included in the laboratory procedure manual, with
copies available for laboratory staff who answer telephone que-
Blood
ries. In laboratories with sufficient computer resources, particu-
larly those with a computerized order-entry system, guidelines Accurate and timely detection of bacteremia and fungemia
can appear as a prompt when a test is ordered. In teaching remains one of the most important functions of clinical microbi-
institutions, new house staff should be given written instruc- ology laboratories. For more than any other type of specimen,
tions on handling specimens as part of their orientation. pre-laboratory (pre-analytic) variables affect microbial recov-
Rejection criteria. Accredited laboratories must specify re- ery, contamination rates, and the ability of physicians to inter-
jection criteria for specimens that are collected, transported, or pret test results. Although a subsequent article in this series
stored under improper conditions prior to processing. Examples will be devoted entirely to the subject of blood cultures, it is
of specimens that are unacceptable for processing are listed in important to emphasize that proper collection and transport of
table 3. Clinical laboratories should be nearly inflexible regard- blood for culture are crucial.
ing this issue; there is no benefit-and there is the potential for Collection. The clinical interpretation of blood culture re-
sults, as well as the cost-effectiveness of blood cultures, de-
pends on many variables; of these variables, the most important
is the proportion of blood cultures that are contaminated by
Table 2. Principles of specimen collection for microbiologi- skin flora [7, 8]. Since only 8%-9% of blood cultures yield
cal testing. microorganisms that are ultimately judged to be the cause of
an episode of bacteremia or fungemia, it is imperative that
• To minimize contamination, use strict aseptic technique when
collecting specimens
• Collect specimens from anatomic sites most likely to yield
pathogens and least likely to yield contaminants Table 3. Specimens that are unacceptable for microbiologi-
• Tissue or fluid submitted for culture is always superior to material cal testing.
on swabs
• Submit adequate volumes of specimens • Unlabeled or improperly labeled specimens
• Provide complete information on specimen requisition forms or • Specimens received in leaking, cracked, or broken containers
during entry of electronic orders
• Specimens with obvious (visually apparent) contamination
• Notify microbiology laboratory and surgical pathology laboratory
• Unpreserved specimens received> 12 hours after being collected
when there is a need for both culture and histopathologic
• Specimens not appropriate for a particular test
examination
em 1996; 22 (May) Microbiological Specimen Collection and Transport 771

Table 4. Comparisons of disinfectants used for blood cultures. For patients with suspected infective endocarditis that is likely
to be caused by indigenous bacteria, the performance of three
No. of No. (%) of No. (%) of
or four cultures may be necessary before the physician can be
Reference Disinfectant cultures pathogens contaminants
certain of the clinical importance of any isolates that are recov-
[13] Isopropyl alcohol 1,609 159 (9.9) 18 (1.1) ered [7]. As cultures ofblood drawn from patients with continu-
(applied twice) ous bacteremia almost always yield microorganisms, it is rarely
Tincture iodine! 179 15 (8.4) 2 (1.1) necessary to perform more than three or four cultures. The
isopropyl alcohol
once-common practice of ordering "blood cultures times six"
[15] Isopropyl alcohol! 181 12 (3.3)- 8 (4.4)
povidone iodine in cases of suspected infective endocarditis should be aban-
Isopropyl alcohol 181 6 (3.3) doned, as this practice contributes nothing to patient care, and
[16] Povidone iodine 4,139 626 (7.4)- 259 (6.3) it is costly and wasteful of resources and needlessly contributes
Tincture iodine 4,328 162 (3.7)' to nosocomial anemia. On the other hand, drawing only a single

Downloaded from https://academic.oup.com/cid/article/22/5/766/360981 by guest on 27 April 2024

[14] Isopropyl alcohol! 763 104 (13.6) 35 (4.6)
sample of blood for routine bacterial and fungal cultures is
povidone iodine
PREP) 783 114 (14.6) 17 (2.2)! inappropriate [19]; single blood cultures will not reliably result
in detection of all septic episodes, and such a practice compro-
- Total no. (%) of pathogens recovered during both phases of each study. mises the ability of physicians to interpret blood culture results.
t p < .00001.
) Consists of 2% iodine tincture and isopropyl alcohol swabs (Mediflex
Receipt of single blood samples in the laboratory should result
Hospital Systems, Overland Park, KS). in notification of the physician that additional samples are
§ P < .01. needed [19].
In contrast, when blood for cultures is drawn for the purpose
of detecting disseminated Mycobacterium avium complex
contamination rates be minimized. Even with good collection (MAC) infection, single samples should be drawn initially [20].
technique, 1%- 3% of blood cultures are found to be contami- Although drawing two samples has been shown to increase
nated. In some clinical settings (e.g., teaching hospitals and yield [20, 21], a second blood sample is unnecessary since the
emergency departments), blood culture contamination rates are first culture yields the result (positive or negative) for 98% of
much higher, compromising the ability of physicians to distin- specimens [20]. Lack of a final result for the other 2% of
guish between contaminants and clinically important isolates. specimens is generally acceptable in terms of patient care, since
Blood culture contamination rates can be minimized by strict the clinical course of MAC bacteremia is more indolent than
adherence to aseptic collection technique and, whenever possi- that of bacteremia caused by other bacteria or fungi; a second
ble, collection of peripheral blood via venipuncture rather than blood culture is not needed to interpret mycobacterial blood
via indwelling vascular catheters [9]. culture results (i.e., mycobacteria are virtually never contami-
Although the results of a recent meta-analysis [9a] suggest nants); and second specimens can be collected easily. There-
that changing needles prior to inoculating blood culture bottles fore, a second specimen should be drawn only if the first culture
results in a statistically significant decrease in contamination is negative and the clinical findings continue to suggest dissem-
rates, the results of several other studies do not support this inated MAC infection.
conclusion [10-12]. Moreover, switching needles increases the Volume of blood cultured. Because of the low number of
likelihood of a needle-stick injury [12]. In the absence of an microorganisms present in the blood of adults who are bacter-
unequivocal benefit that outweighs the potential risk of needle- emic or fungemic, the most important variable in recovering
stick injury, this practice should be proscribed. bacteria or fungi from adults is the volume of blood cultured
At most institutions, an iodophor or tincture of iodine, alone [22, 23]. For adults, the recommended volume of blood to be
or in combination with isopropyl alcohol, is used for disin- drawn for each blood culture (i.e., from each venipuncture site)
fecting skin. Despite the clinical importance of the choice of is 20-30 mL. For infants and small children, the number of
disinfectant, few controlled comparisons of different disinfec- organisms in the blood can be, but is not always, higher [24].
tants have been performed (table 4) [13-16]. Even though two Therefore, although bacteremia or fungemia can be reliably
studies [14, 16] have shown statistically significant differences detected when small volumes of blood (~l mL) are cultured,
in the effectiveness of disinfectants, taken together the results microbial recovery is enhanced when larger volumes of blood
of all the evaluations suggest that the specific disinfectant used are cultured [25]. For infants and small children, the recom-
may be less important with regard to contamination of blood mended volume is 1-5 mL. For older children, the volume of
cultures than is good disinfection technique. blood to be drawn per culture should be appropriate for the
Number of cultures. If sufficient blood (20-30 mL) is age and weight of the patient.
drawn from each venipuncture site, virtually all septic episodes Culturing an adequate volume of blood also ensures that the
can be detected with two-to-three blood cultures [17, 18]. This proper ratio of blood to broth medium is attained within each
is true both for septic episodes characterized by intermittent bottle. Although this ratio is probably less important than vol-
bacteremia and those characterized by continuous bacteremia. ume per se in optimizing microbial recovery, maintaining a
772 Wilson eID 1996;22 (May)

blood-to-broth ratio of between 1:5 and 1:10 enhances micro- disinfection. The risk of contamination is higher when CSF is
bial recovery [26, 27]. collectedfrom cathetersor shunts.Such contaminationis problem-
Timing of collection. It has long been common practice to atic, since contaminating microorganismsare likely to be the same
separate collection of blood specimens from a given patient by microorganisms (e.g., coagulase-negative staphylococci) that
arbitrary time intervals. It is surprising that until recently no cause many CSF catheter and/or shunt infections.
one had studied this practice systematically. In 1994, Li et al. The volume ofCSF needed for culture depends on the patho-
[28] showed that drawing blood for cultures either simultane- gens being sought. For routine bacterial cultures, a few milli-
ously or over a 24-hour period resulted in similar microbial liters of CSF is adequate. In contrast, for fungal and mycobacte-
recovery rates. Since there is no benefit in obtaining blood rial cultures, microbial yield is more proportional to the volume
samples at intervals and it is more practical to draw blood for ofCSF cultured. This has led to the concept of "large-volume"
a set of cultures at the same time than it is to return to the cultures for which as much as 10-20 mL ofCSF is processed.
bedside to draw additional specimens, there appears to be little In 1988, Albright et al. [30] described a procedure for storing

Downloaded from https://academic.oup.com/cid/article/22/5/766/360981 by guest on 27 April 2024

reason to continue this practice. Moreover, drawing blood spec- an aliquot of CSF for future testing. This procedure, known by
imens simultaneously may help ensure that the blood is drawn the acronym TRAP (transport, rapid accessioning for additional
before antimicrobial agents are administered. One possible ex- procedures), has been used successfully to provide a mecha-
ception would be the collection of blood for cultures over a nism by which testing of specimens is delayed pending results
24-hour period to determine whether a patient has continuous of initial screening.
bacteremia. Although this is not a necessary practice for most Two studies have demonstrated the usefulness of the TRAP
patients with infective endocarditis, it may be useful when the method. In both of these studies, three strategies for decreasing
clinical importance of isolates is unclear. inappropriate testing were evaluated: physician education, op-
Transport and storage. Blood culture bottles should be tional use of screening tests prior to the testing of CSF, and
transported immediately to the laboratory. Ifthis is not possible, mandatory use of screening tests prior to the testing of CSF
bottles can be kept at room temperature or in an incubator at [31, 32]. In the first study, CSF specimens submitted for CSF-
a temperature of 35°C to 37°C. These bottles should never VDRL (Venereal Disease Research Laboratory) testing were
be refrigerated. Blood collected in Isolator tubes (Wampole evaluated; the initial screening tests were serological tests for
Laboratories, Cranbury, NJ) should be processed within 8 syphilis [32]. In the second study, CSF specimens submitted
hours, since delays in processing may decrease the microbial for smears and cultures for acid-fast bacilli were evaluated; the
yield [29]. initial screening tests were determinations of the cell counts
Summary. For adults, two or three blood specimens (20- and levels of glucose and protein in CSF [33].
30 mL each) should be drawn simultaneously from different The results of the first study underscore the observation of
venipuncture sites as soon as possible following the clinical Dans et al. [33] that the use of many, if not most, CSF VDRL
events that prompted performance of cultures. For patients with tests is inappropriate. In both studies, significant reductions in
suspected infective endocarditis (particularly those who have CSF testing occurred with mandatory use of screening tests
prosthetic valves and whose normal skin flora may be the prior to CSF testing but did not occur with optional screening.
source of infection), the results of three or four blood cultures Physician education alone had no effect in the first study [31];
may establish the presence of continuous bacteremia and help although this strategy was effective in the second study, the
the physician determine the clinical relevance of the isolates decrease in testing was apparently related to a decrease in all
that are recovered. For children, two or three blood specimens types of testing for the presence of mycobacteria [32].
(at volumes appropriate for the age and weight of the patient) Transport and storage. CSF specimens should be trans-
should be drawn. Single blood cultures for the detection of ported immediately to the laboratory. Systematic delays in trans-
pathogens other than mycobacteria should not be done. port should be identified and eliminated. Laboratories should
strive to report the results of initial tests within 30 minutes of
receipt of the specimen in the laboratory. From collection
Cerebrospinal Fluid
through processing, CSF specimens (except aliquots collected
The prompt, accurate diagnosis of bacterial meningitis is for viral cultures) should not be refrigerated until initial pro-
among the most important tasks confronting clinical microbiol- cessing is completed. Laboratorians should consider using se-
ogy laboratories. Clinical personnel and laboratory staff should quential testing to reduce the number of unnecessary CSF tests.
carefully coordinate the handling of specimens from the time
of collection through processing.
Respiratory Tract Specimens
Collection. CSF must be collected by means of strict aseptic
technique, both to minimize specimen contaminationand to pre- Collection. Expectorated sputum continues to be the most
vent introduction ofbacteria into the CNS. The risk of contamina- commonly collected respiratory specimen for bacterial cultures.
tion is low when the skin is adequately disinfectedprior to lumbar Expectorated sputum specimens should be screened by gram
puncture; either an iodophor or chlorhexidine can be used for staining for contamination with saliva; results should be inter-
ern 1996;22 (May) Microbiological Specimen Collection and Transport 773

Table 5. Criteria for rejecting specimens of expectorated sputum. refrigerated. For fungal and mycobacterial cultures, prompt
processing and refrigeration help prevent overgrowth of normal
flora in the specimens, which complicates the recovery of
Bartlett*
pathogens.
Neutrophils per field
(magnification, X10) Grade Stool
<10 o
10-25 +1 The laboratory diagnosis of enteric infections is challenging.
>25 +2 Problems include the number of potential pathogens; the bio-
Presence of mucus +1
logic diversity ofthese organisms; the emergence ofnew patho-
gens; and the fact that accurate, reliable, and practical diagnos-
Squamous epithelial cells per field
(magnification, x 10) tic tests have yet to be developed for many pathogens [5].

Downloaded from https://academic.oup.com/cid/article/22/5/766/360981 by guest on 27 April 2024

10-25 -I Moreover, international travel has become so common that, in
>25 -2 some instances, epidemiological clues as to likely etiologic
Total! agents may not be as helpful as they once were. At the least,
there are many more potential causes of diarrhea in travelers
Murray and Washington l
than in patients who have not traveled. Consequently, microbi-
Squamous epithelial Neutrophils ology laboratorians must have the expertise, experience, and
cells per low-power per low-power Perform resources to recover and identify a variety of potential patho-
field field culture
gens. For many laboratories, the problem is the provision of
Group I >25 <10 No such a service in a cost-effective manner. This problem is
Group 2 >25 10-25 No best solved by close collaboration between laboratorians and
Group 3 >25 >25 No clinical staff, with development of test utilization strategies
Group 4 10-25 >25 No§
that are appropriate for the patient population being served.
Group 5 <10 >25 Yes
Collection. Numerous studies have verified the observation
that there is minimal value in routinely performing stool cul-
* Data are from [34]. tures or microscopic examinations of stool for ova and parasites
t The nwnbers of neutrophils and squamous epithelial cells are averaged in patients who develop diarrhea after 3 or 4 days ofhospitaliza-
based on examination of 20- 30 separate fields (magnification, x 10). The total
tion [39-48]. This observation holds true for both adults and
is then calculated; final scores of ,,;0 suggest contamination with saliva and!
or absence of acute infammation. children. Although the exact cutoff time (3 days vs. 4 days of
I Data are from [35]. hospitalization) may vary slightly depending on the specific
§ Some laboratories set up cultures of specimens in group 4.
health care setting, it is neither beneficial to patients nor cost-
effective to routinely process specimens for these tests after
the 4th day of hospitalization. Stool specimens collected from
preted with use of criteria such as those shown in table 5 [34, patients who develop diarrhea in the hospital should be tested
35]. If a sputum specimen is rejected, another specimen should for the presence of Clostridium difficile. Manabe et a1. [49]
be collected and screened in the same manner. Specimens sub- have published guidelines for using clinical data and laboratory
mitted for mycobacterial culture should not be screened with data to guide evaluation of patients with suspected C. difficile
use of these criteria, as the results do not reflect the likelihood disease.
that mycobacteria will be recovered [36]. Similarly, specimens Similarly, there is little value in routinely testing three or
submitted for culture should not be screened on the basis of more stool specimens as part of an evaluation of acute diarrhea,
the relative numbers of neutrophils and alveolar macrophages as the majority of published studies indicate that most patho-
[37]. Morris et al. [38] studied the use of gram staining in gens are detected in the first specimen [41,45,47,48,50] (this
screening endotracheal aspirates; if a gram stain reveals no is true for both cultures and examinations for ova and parasites).
bacteria or reveals > 10 squamous epithelial cells per low- However, some studies have not supported this observation, a
power field, the specimen should be rejected. Other respiratory finding suggesting that in some settings it may be appropriate
tract specimens (e.g., bronchial lavage fluid) should not be to routinely collect two or three specimens [51, 52].
rejected on the basis of criteria used for other specimens such This issue is a difficult one to resolve because the failure to
as sputum. detect pathogens in stool specimens relates partly to intermit-
Transport and storage. Because most respiratory tract tent shedding of some intestinal pathogens as well as to current
specimens are likely to contain at least a few contaminating diagnostic limitations [5]. Although the testing of multiple
microorganisms, specimens should be transported quickly to specimens may eventually overcome the former circumstance,
the laboratory to minimize overgrowth of contaminants. If it cannot overcome the latter. Rather than submitting many
transportation or processing is delayed, specimens should be specimens from the same patient, clinicians should consult the
774 Wilson em 1996;22 (May)

laboratory for alternative diagnostic tests for specific patho- viruses survive well at ambient temperature while in common
gens. For example, a significant number of cases of strongyloi- transport media, recovery of viruses from specimens containing
diasis are missed when stool specimens are examined for the low numbers of viruses may be decreased following prolonged
presence of Strongyloides with use of traditional methods (di- holding under these conditions. Therefore, it is advisable to
rect fecal smear, formalin-ether concentration, or culture on a transport specimens on ice or to keep them refrigerated. Speci-
filter-paper strip); this organism can be detected more reliably mens should never be exposed to temperatures higher than
by coproculture with use of agar plates [53]. room temperature. Johnson [55] has reviewed the details re-
More widespread use of newer diagnostic products such as garding collection and transport of specimens for recovery of
enzyme immunoassays for Giardia lamblia may also obviate specific viruses. A subsequent article in this series will include
the need to routinely test more than one stool specimen. In a more detailed review of diagnostic virology.
most settings-and for detection of most common enteric
pathogens-awaiting results for the first specimen before col-

Downloaded from https://academic.oup.com/cid/article/22/5/766/360981 by guest on 27 April 2024

lecting subsequent specimens reduces the amount of unneces- Specimens Submitted for Detection of Microorganisms by
sary testing. Molecular Diagnostic Techniques
Transport. Optimal test results are obtained when microbi-
A variety of molecular diagnostic assays have been devel-
ological testing is performed on fresh stool specimens. Because
oped for use in clinical microbiology laboratories. Despite the
testing fresh specimens is impractical in most clinical settings
potential for improving the diagnosis of infectious diseases,
(particularly outpatient settings), most stool specimens are col-
several important issues need to be resolved before many of
lected and then placed in vials containing different transport
these assays can be recommended for routine use. First, it is not
media and fixatives. A variety of such products are commer-
known whether specimens submitted for molecular diagnostic
cially available; each system typically includes a transport me-
testing should be screened with use of the same methods and
dium for culture and 10% neutral-buffered formalin and polyvi-
criteria as are used for specimens submitted for culture. Second,
nyl alcohol for ova and parasite examination. For most
even though molecular diagnostic assays are analytically more
purposes, these transport systems work well, are convenient
sensitive (i.e., they detect smaller quantities of analyte), it is
and easy to use, and are relatively inexpensive [54].
not known whether they are diagnostically more sensitive (i.e.,
Storage. Stool specimens submitted for culture typically
they generate fewer false-negative test results). Third, many of
are not stored for any length of time, since most laboratories
these tests, particularly nucleic acid amplification assays, are
set up all appropriate cultures at the time of receipt of the
not yet commercially available and thus have not been stan-
specimen. Only rarely is it necessary to retrieve a specimen
dardized. "Home-brew" assays, in particular, remain poorly
for additional testing; such specimens should be refrigerated.
standardized. Last, the clinical relevance of results obtained
Stool specimens submitted for ova and parasite examinations
from many assays has yet to be defined.
are typically stored at room temperature in a fixative. Speci-
Controlled clinical trials to establish performance character-
mens stored in 10% neutral-buffered formalin remain stable
istics will be possible once these assays become commercially
for many months, even when tested with some enzyme immu-
available and are more widely used. Until that time, it would
noassays for G. Lamblia. Because trophozoites can deteriorate
be prudent to test only those specimens that are appropriate
quickly in stool, even when refrigerated, fresh stool specimens
for culture, since they have already been shown to yield more
submitted for ova and parasite examination should be examined
pathogens with fewer contaminants. A subsequent article in
within 2 hours.
this series will include a more-detailed review of molecular
Summary. For patients with acute diarrhea, one specimen
diagnostic techniques.
should be submitted for culture and examination for ova and
parasites. If these tests are negative and symptoms persist,
additional specimens should be submitted for testing. For pa-
Specimens Collected for Serological Diagnosis of
tients who develop diarrhea after the third or fourth day of Infectious Diseases
hospitalization, a stool specimen should be tested for the pres-
ence of C. difficile. For most diseases, serological testing is not a surrogate for
culture or other diagnostic tests. Physicians should order sero-
logical tests sparingly, since many serological assays have limi-
Specimens Collected for Viral Cultures
tations that often are not appreciated. The most important of
Many commercial products are available for the collection these limitations include technical issues such as cross-reactiv-
and transport of viral culture specimens. For most of these ity, turnaround time (for some assays, test results are not avail-
systems, specimens are collected on swabs that are then rinsed able in a clinically relevant time frame), and the inability to
in a broth medium (viral transport medium). Calcium alginate distinguish between acute disease and past exposure to infec-
swabs should not be used, since they are known to adversely tious agents on the basis of single assays. Serological testing
affect recovery of herpes simplex virus [6]. Although most certainly plays an important role in the treatment of patients
em 1996;22 (May) Microbiological Specimen Collection and Transport 775

with infectious diseases, but only when the tests are used appro- Transport. Specimens should be transported with use of
priately. standard precautions. Additional measures should be taken to
Collection. Specimens obtained for serological analysis ensure that specimens are not damaged during transportation,
should be collected either in sterile evacuated tubes or in serum which can result in contamination or leakage of the specimen.
separator tubes. Strict aseptic technique should be used during In particular, specimens in glass containers should be trans-
venipuncture. Adequate volumes of blood should be drawn for ported in such a way that the risk of breakage is minimized.
anticipated tests. In many cases it is prudent to draw a small This is especially important when pneumatic tube systems are
additional volume of blood that can be stored for future testing. used for transport, as cleanup of leaked specimens within these
Most clinical laboratories perform limited serological test- systems is difficult and expensive.
ing, forwarding many specimens to commercial reference labo- Microbiological specimens transported via mail or other in-
ratories, state laboratories, the Centers for Disease Control and terstate couriers is subject to federal regulations; McVicar and
Prevention, and other public health laboratories. Because of Suen [58] have recently reviewed these regulations. If labora-

Downloaded from https://academic.oup.com/cid/article/22/5/766/360981 by guest on 27 April 2024

the multiplicity of assays and testing laboratories, serological tory staff are uncertain as to whether shipping a given agent
testing is best facilitated when the laboratory procedure manual is regulated or as to what constitutes appropriate packaging
contains the following data for each reference laboratory: de- and labeling, they should consult with the appropriate agency
tailed information regarding infectious agents and/or diseases before packaging the agent.
for which tests are available; specific assays for each agent Storage and processing. Once received in the laboratory,
and/or disease; types of specimens that will be accepted for specimens should continue to be handled with use of standard
testing; minimum and optimal volumes of specimens for test- precautions. Strict adherence to such precautions decreases the
ing; reference ranges; limitations of assays; recommendations likelihood of exposure to blood-borne pathogens as well as
for specimen collection; guidelines for test interpretation; and pathogens being sought in specimens and those that are clini-
special considerations for specimen collection and transport. cally not suspected. Although most cultures can be plated safely
It is inadvisable to collect specimens other than serum for on a standard laboratory bench, many microbiologists prefer
most serological tests. Since most commercial serological tests setting up cultures in a biological safety cabinet. This is manda-
were not developed to test specimens other than serum, stan- tory for specimens that may contain M. tuberculosis. Once
dardized controls are not available, reference ranges have not pathogens are propagated in the laboratory, there is further
been defined, and the performance characteristics (i.e., sensitiv- risk of developing laboratory-acquired infection [56]; etiologic
ity, specificity, and positive and negative predictive values) of agents of particular risk include M. tuberculosis, Brucella,
the tests are unknown. In addition to these technical limitations, Francisella tularensis, Yersinia pestis, Histoplasma capsula-
the clinical relevance of test results for specimens other than tum, and Coccidioides immitis [56, 59]. Cultures containing (or
serum is usually unknown. suspected of containing) one of these agents should be pro-
Transport and storage. To prevent loss of immunologic cessed only in Class II biological safety cabinets under Biosa-
reactivity and growth of contaminating microorganisms, speci- fety Level 3 conditions [59].
mens should be transported promptly to the laboratory and
centrifuged, and the serum should be poured off and refriger-
ated or frozen immediately. Serum specimens that will be used Summary
within 1 week after collection can be refrigerated. Specimens Obtaining accurate and cost-effective microbiological test
that need longer storage should be frozen at - 70°C. Storage results is possible only when specimens are collected, trans-
of specimens at - 20°C is not recommended, since some anti- ported, and stored properly. When proper procedures are fol-
bodies deteriorate at an unpredictable rate at that temperature. lowed, cultures of specimens are less likely to be contaminated
Most assays of specimens that are properly stored at -70°C and more likely to yield pathogens. Not only does this make
yield accurate results for many months. interpretation of test results easier, but it also reduces unneces-
sary work and, as documented for some specimens, reduces
Safety During Handling of Specimens Collected for health care costs. Proper collection includes submitting the
Microbiological Testing appropriate number of specimens. It is increasingly evident that
for most specimens, submission ofmore than the recommended
Collection. To minimize the potential exposure of person- number of specimens does not improve the physician's ability
nel to infectious agents, specimens should be collected with to interpret test results.
use of standard (universal) precautions [56]. In particular, blood
collection should be performed in strict accordance with guide-
lines to prevent needle-stick injuries [57]. Collection of speci- References
mens from patients with communicable diseases, particularly 1. Mangels JI. Cost containment in the clinical microbiology laboratory. In:
those transmitted via the respiratory route, should be done Helman EZ, ed. Cost containment in the clinical laboratory. Berkeley,
under appropriate isolation conditions. California: Berkeley Scientific Publications, 1995: 142-226.
776 Wilson CID 1996;22 (May)

2. Isenberg RD, Schoenknecht FD, von Graevenitz A. Collection and pro- 22. Wilson ML, Weinstein MP. General principles in the laboratory detection
cessing of bacteriological specimens. In: Rubin SJ, coordinating ed. of bacteremia and fungemia. Clin Lab Med 1994; 14:69-82.
Cumitech 9. Washington, DC: American Society for Microbiology, 23. Mermel LA, Maid DG. Detection of bacteremia in adults: consequences
1979. of culturing an inadequate volume of blood. Ann Intern Med
3. Miller JM. Handbood of specimen collection and handling in microbiol- 1993; 119:270-2.
ogy. CDC laboratory manual. Atlanta: U.S. Department of Health and 24. La Scolea LJ Jr, Dryja D, Sullivan TD, Mosovich L, Ellerstein N, Neter
Human Services, Public Health Service, CDC, 1985. E. Diagnosis of bacteremia in children by quantitative direct plating
4. Miller JM, Holmes HT. Specimen collection, transport, and storage. In: and a radiometric procedure. J Clin Microbiol 1981; 13:478-82.
Murray PR, Baron EJ, Pfaller MA, Tenover FC, Yolken RH, eds. Man- 25. Szymczak EG, Barr IT, Durbin WA, Goldmann DA. Evaluation of blood
ual of clinical microbiology. 6th ed. Washington, DC: American Society culture procedures in a pediatric hospital. J Clin Microbiol 1979; 9:
for Microbiology, 1995:19-32. 88-92.
5. Shea YR. Specimen collection and transport. In: Isenberg RD, ed. Clinical 26. Reller LB, Lichtenstein KA, Mirrett S, Wang W-LL. Controlled evaluation
microbiology procedures handbook. Vol I. Washington, DC: American of the ratio of blood to broth in the detection of bacteremia by blood
Society for Microbiology, 1992:1.1.1-1.1.30. culture [abstract 177]. In: Abstracts of the 1978 Annual Meeting of

Downloaded from https://academic.oup.com/cid/article/22/5/766/360981 by guest on 27 April 2024

6. Woods GL, Washington JA. The clinician and the microbiology laboratory. the American Society for Microbiology. Washington, DC: American
In: Mandell GL, Bennett JE, Dolin R, eds. Mandell, Douglas and Ben- Society for Microbiology, 1978.
nett's principles and practice of infectious diseases. Vol I. 4th ed. New 27. Salventi JF, Davies TA, Randall EL, Whitaker S, Waters JR. Effect of
York: Churchill Livingstone, 1995:169-99. blood dilution on recovery of organisms from clinical blood cultures
7. Aronson MD, Bor DH. Blood cultures. Ann Intern Med 1987; 106: in medium containing sodium polyanethol sulfonate. J Clin Microbiol
246-53. 1979;9:248-52.
8. Bates DW, Goldman L, Lee TH. Contaminant blood cultures and resource 28. Li J, Plorde Jl, Carlson LG. Effects of volume and periodicity on blood
utilization. The true consequences of false-positive results. JAMA cultures. J Clin Microbiol 1994; 32:2829-31.
1991;265:365-9. 29. Stockman L, Roberts GD, Ilstrup DM. Effect of storage of the Du Pont
9. Bryant JK, Strand CL. Reliability of blood cultures collected from intravas- lysis-centrifugation system on recovery of bacteria and fungi in a pro-
cular catheter versus venipuncture. Am J Clin PathoI1987;88:113-6. spective clinical trial. J Clin Microbiol 1984; 19:283-5.
9a.Spitalnic SJ, Woolard RH, Mermel LA. The significance of changing 30. Albright RE Jr, Christenson RH, Habig RL, Mears TP, Schneider KA.
needles when inoculating blood cultures: a meta-analysis. Clin Infect Cerebrospinal fluid (CSF) TRAP: a method to improve CSF laboratory
Dis 1995;21:1103-6. efficiency. Am J Clin PathoI1988;90:707-10.
10. Isaacman DJ, Karasic RE. Lack of effect of changing needles on contami- 31. Albright RE Jr, Christenson RH, Emlet JL, et al. Issues in cerebrospinal
nation of blood cultures. Pediatr Infect Dis J 1990; 9:274-8. fluid management: CSF Venereal Disease Research Laboratory testing.
II. Krumholz HM, Cummings S, York M. Blood culture phlebotomy: switch- Am J Clin Patholl991;95:397-401.
ing needles does not prevent contamination. Ann Intern Med 32. Albright RE Jr, Graham CB III, Christenson RH, et al. Issues in cerebrospi-
1990; 113:290-2. nal fluid management: acid-fast bacillus smear and culture. Am J Clin
12. Leisure MK, Moore DM, Schwartzman JD, Hayden GF, Donowitz LG. Patholl991;95:418-23.
Changing the needle when inoculating blood cultures: a no-benefit and 33. Dans PE, Cafferty L, Otter SE, Johnson RJ. Inappropriate use of the
high-risk procedure. JAMA 1990;264:2111-2. cerebrospinal fluid Venereal Disease Research Laboratory (VDRL) test
13. Lee S, Schoen I, Malkin A. Comparison of use of alcohol with that of to exclude neurosyphilis. Ann Intern Med 1986; 104:86-9.
iodine for skin antisepsis in obtaining blood cultures. Am J Clin Pathol 34. Bartlett RC. Medical microbiology: quality, cost and clinical relevance.
1967;47:646-8. New York: Wiley, 1974.
14. Schifinan RE, Pindur A. The effect of skin disinfection materials on reduc- 35. Murray PR, Washington JA. Microscopic and bacteriologic analysis of
ing blood culture contamination. Am J Clin Pathol 1993;99:536-8. expectorated sputum. Mayo Clin Proc 1975;50:339-44.
15. Shahar E, Wohl-Gottesman B, Shenkman L. Contamination of blood cul- 36. Havlik D, Woods GL. Screening sputum specimens for mycobacterial
tures during venepuncture: fact or myth? Postgrad Med J culture. Laboratory Medicine 1995;26:411-3.
1990;66:1053-8. 37. Flournoy DJ, Beal LM, Smith MD. What constitutes an adequate sputum
16. Strand CL, Wajsbort RR, Sturmann K. Effect ofiodophor vs iodine tincture specimen? Laboratory Medicine 1994;25:456-9.
skin preparation on blood culture contamination rate. JAMA 38. Morris AJ, Tanner DC, Reller LE. Rejection criteria for endotracheal
1993; 269: I004-6. aspirates from adults. J Clin Microbioll993;31:1027-9.
17. Washington JA II. Blood cultures: principles and techniques. Mayo Clin 39. Siegel DL, Edelstein PH, Nachamk:in I. Inappropriate testing for diarrheal
Proc 1975;50:91-8. diseases in the hospital. JAMA 1990;263:979-82.
18. Weinstein MP, Reller LB, Murphy JR, Lichtenstein KA. The clinical 40. Asnis DS, Bresciani A, Ryan M, McArdle P, Mollura JL, Ilardi CF. Cost-
significance of positive blood cultures: a comprehensive analysis of effective approach to evaluation of diarrheal illness in hospitals [letter].
500 episodes of bacteremia and fungemia in adults. I. Laboratory and J Clin Microbiol 1993;31:1675.
epidemiologic observations. Rev Infect Dis 1983; 5:35-53. 41. Barbut F, Leluan P, Antoniotti G, Collignon A, Sedallian A, Petit JC.
19. Kellogg JA, Ferrentino FL, Liss J, Shapiro SL, Bankert DA. Justification Value of routine stool cultures in hospitalized patients with diarrhea.
and implementation of a policy requiring two blood cultures when one Eur J Clin Microbiol Infect Dis 1995; 14:346-9.
is ordered. Laboratory Medicine 1994;25:323-30. 42. Bowman RA, Bowman JM, Arrow SA, Riley TV. Selective criteria for
20. Stone BL, Cohn DL, Kane MS, Hildred MV, Wilson ML, Reves RR. the microbiological examination of faecal specimens. J Clin Pathol
Utility of paired blood cultures and smears in diagnosis of disseminated 1992;45:838-9.
Mycobacterium avium complex infections in AIDS patients. J Clin Mi- 43. Brady MT, Pacini DL, Budde CT, Connell MJ. Diagnostic studies of
crobiol 1994; 32:841-2. nosocomial diarrhea in children: assessing their use and value. Am J
21. Yagupsky P, Menegus MA. Cumulative positivity rates of multiple blood Infect Control 1989; 17:77-82.
cultures for Mycobacterium avium-sintracellulare and Cryptococcus 44. Church DL, Cadrain G, Kabani A, Jadavji T, Trevenen C. Practice guide-
neoformans in patients with the acquired immunodeficiency syndrome. lines for ordering stool cultures in a pediatric population. Am J Clin
Arch Pathol Lab Med 1990; 114:923-5. Patholl995; 103:149-53.
em 1996;22 (May) Microbiological Specimen Collection and Transport 777

45. Fan K, Morris AJ, Reller LB. Application of rejection criteria for stool 53. Sato Y, Kobayashi J, Toma H, Shiroma Y. Efficacy of stool examination
cultures for bacterial enteric pathogens. J Clin Microbiol 1993; 31: for detection of Strongyloides infection. Am J Trop Med Hyg
2233-5. 1995;53:248-50.
46. Kabani A, Cadrain G, Trevenen C, Jadavji T, Church DL. Practice guide- 54. Garcia LS. Pros and cons of using preservatives for O&P fecal specimens.
lines for ordering stool ova and parasite testing in a pediatric population. Clinical Microbiology Newsletter 1995; 17:164-7.
Am J Clin Patholl995; 104:272-8. 55. Johnson FB. Transport of viral specimens. Clin Microbiol Rev
47. Morris AJ, Wilson ML, Reller LB. Application of rejection criteria for stool 1990;3:120 31.
ovum and parasite examinations. J Clin Microbioll992;30:3213-6. 56. Wilson ML, Reller LB. Clinical laboratory-acquired infections. In: Bennett
48. Yannelli B, Gurevich I, Schoch PE, Cunha BA. Yield of stool cultures, ova JV, Brachman PS, eds. Hospital infections. 3rd ed. Boston: Little,
and parasite tests, and Clostridium difficile determinations in nosocomial Brown, 1992:359-74.
diarrheas. Am J Infect Control 1988; 16:246-9.
57. U. S. Department of Labor. Occupational Safety and Health Administra-
49. Manabe Ye, Vinetz JM, Moore RD, Merz C, Charache P, Bartlett JG.
tion. Occupational exposure to bloodborne pathogens; final rule (29
Clostridium difficile colitis: an efficient clinical approach to diagnosis.
CFR Part 1910.1030). Federal Register 1991; 56:64175--82.
Ann Intern Med 1995; 123:835-40.
50. Tamboli P, Mezger E. Is the ex.amination of multiple stool specimens for 58. McVicar JW, Suen 1. Packaging and shipping biological materials. In:

Downloaded from https://academic.oup.com/cid/article/22/5/766/360981 by guest on 27 April 2024

ova and parasites justifiable? [abstract 136]. Am J Clin Pathol Fleming DO, Richardson JH, Tulis 11,Vesley D, eds, Laboratory safety:
1993; 100:343. principles and practices. 2nd ed. Washington, DC: American Society
51. Hiatt RA, Markell EK, Ng E. How many stool examinations are necessary for Microbiology, 1995:239-46.
to detect pathogenic intestinal protozoa? Am J Trop Med Hyg 59. U.S. Department of Health and Human Services, Public Health Ser-
1995; 53:36-9. vice, Centers for Disease Control and Prevention, and National
52. Nazer H, Greer W, Donnelly K, et al. The need for three stool specimens Institutes of Health. Biosafety in microbiological and biomedical
in routine laboratory examinations for intestinal parasites. Br J Clin laboratories. 3rd ed. Washington, DC: U.S. Government Printing
Pract 1993;47:76-8. Office, 1993.