Appropriate Collection Techniques

Specimens should be collected during the acute (early) phase of an illness (or within 2 to 3 days for viral infections) and before antibiotics are administered, if possible. Swabs generally are poor specimens if tissue or needle aspirates can be obtained. It is the microbiologist’s responsibility to provide clinicians with a collection manual or instruction cards listing optimal specimen collection techniques and transport information. Information for the nursing staff and clinicians should include the following:

Instructions should be written so that specimens collected by the patient (e.g., urine, sputum, or stool) are handled properly. Most urine or stool collection kits contain instructions in several languages, but nothing substitutes for a concise set of verbal instructions. Similarly, when distributing kits for sputum collection, the microbiologist should be able to explain to the patient the difference between spitting in a cup (saliva) and producing good lower respiratory secretions from a deep cough (sputum). General collection information is shown in Table 5-1. An in-depth discussion of each type of specimen is found in Part VII.

Specimen Transport

Ideally, specimens should be transported to the laboratory within 2 hours of collection. All specimen containers should be leak-proof, and the specimens should be transported within sealable, leak-proof, plastic bags with a separate section for paperwork; resealable bags or bags with a permanent seal are common for this purpose. Bags should be marked with a biohazard label (Figure 5-1). Many microorganisms are susceptible to environmental conditions such as the presence of oxygen (anaerobic bacteria), changes in temperature (Neisseria meningitidis), or changes in pH (Shigella). Thus, use of special preservatives or holding media for transportation of specimens delayed for more than 2 hours is important to ensure organism viability (survival).

Specimen Labeling

Specimens should be labeled with the patient’s name, identifying number (hospital number) or birth date, date and time of collection, and source. Enough information must be provided on the specimen label so that the specimen can be matched up with the requisition when it is received in the laboratory.

Specimen Requisition

The specimen (or test) requisition is an order form that is sent to the laboratory along with a specimen. Often the requisition is a hard (paper) copy of the physician’s orders and the patient’s demographic information (e.g., name and hospital number). Sometimes, however, if a hospital information system offers computerized order entry, the requisition is transported to the laboratory electronically. The requisition should contain as much information as possible regarding the patient history and diagnosis. This information helps the microbiologist to work up the specimen and determine which organisms are significant in the culture. A complete requisition should include the following:

Rejection of Unacceptable Specimens

Criteria for specimen rejection should be set up and distributed to all clinical practitioners. In general, specimens are unacceptable if any of the following conditions apply:

It is an important rule to always talk to the requesting physician or another member of the health care team before discarding unacceptable specimens. In some cases, such as mislabeling of a specimen or requisition, the person who collected the specimen and filled out the paperwork can come to the laboratory and correct the problem; a mislabeled specimen or requisition should not be identified over the telephone. However, correction of mislabeled specimens must be completed at the discretion of the laboratories standard operating procedures. Frequently, it may be necessary to do the best possible job on a less than optimal specimen, if it would be impossible to collect the specimen again because the patient is taking antibiotics, the tissue was collected at surgery, or the patient would have to undergo a second invasive procedure (bone marrow or spinal tap). A notation regarding improper collection should be added to the final report in this instance, because only the primary caregiver is able to determine the validity of the results.

Specimen Processing

Depending on the site of testing (hospital, independent lab, physician’s office lab) and how the specimens are transported to the laboratory (in-house, courier, or driver), microbiology samples may arrive in the laboratory in large numbers or as single tests. Although batch processing may be possible in large independent laboratories, most often hospital testing is performed as specimens arrive. When multiple specimens arrive at the same time, priority should be given to those that are most critical, such as cerebrospinal fluid (CSF), tissue, blood, and sterile fluids. Urine, throat, sputa, stool, or wound drainage specimens can be saved for later. Acid-fast, viral, and fungal specimens are usually batched for processing at one time. When a specimen is received with multiple requests but the amount of specimen is insufficient to do all of them, the microbiologist should call the clinician to prioritize the testing. Anytime a laboratory staff member contacts the physician or nurse, the conversation and agreed-upon information should be documented to ensure proper follow-up. On arrival in the laboratory, the time and date received should be recorded.

Gross Examination of Specimen

All processing should begin with a gross examination of the specimen. Areas with blood or mucus should be located and sampled for culture and direct examination. Stool should be examined for evidence of barium (i.e., chalky white color), which would preclude O&P examination. Notations should be made on the handwritten or electronic work card regarding the status of the specimen (e.g., bloody, cloudy, clotted) so that if more than one person works on the sample, the results of the gross examination are available for consultation.

Direct Microscopic Examination

All appropriate specimens should have a direct microscopic examination. The direct examination serves several purposes. First, the quality of the specimen can be assessed; for example, sputa can be rejected that represent saliva and not lower respiratory tract secretions by quantitation of white blood cells or squamous epithelial cells present in the specimen. Second, the microbiologist and clinician can be given an early indication of what may be wrong with the patient (e.g., 4+ gram-positive cocci in clusters in an exudate). Third, the workup of the specimen can be guided by comparing what grows in culture to what was seen on the original smear. A situation in which three different morphotypes (cellular types) are seen on direct Gram stain but only two grow out in culture, for example, alerts the microbiologist to the fact that the third organism may be an anaerobic bacterium. Or there are more than three organisms on the culture plate that were not visible on Gram stain, indicating possible contamination. Gram stains are also layered with cells and debris. Organisms that appear on the surface of white blood cells may actually be ingested organisms that are no longer viable or capable of growth. It is imperative that the Gram stain results and specimen culture correlate to the type of specimen to ensure accurate information is provided to the clinician.

Direct examinations are usually not performed on throat, nasopharyngeal, or stool specimens but are indicated from most other sources due to the presence of abundant normal microbiota.

The most common stain in bacteriology is the Gram stain, which helps the clinician to visualize rods, cocci, white blood cells, red blood cells, or squamous epithelial cells present in the sample. The most common direct fungal stains are KOH (potassium hydroxide), PAS (periodic-acid Schiff), GMS (Grocott’s methenamine silver stain), and calcofluor white. The most common direct acid-fast stains are AR (auramine rhodamine), ZN (Ziehl-Neelsen), and Kinyoun. Chapter 6 describes the use of microscopy in clinical diagnosis in more detail.

Selection of Culture Media

Primary culture media are divided into several categories. The first are nutritive media, such as blood or chocolate agars. Nutritive media support the growth of a wide range of microorganisms and are considered nonselective because, theoretically, the growth of most organisms is supported. Nutritive media can also be differential, in that microorganisms can be distinguished on the basis of certain growth characteristics evident on the medium. Blood agar is considered both a nutritive and differential medium because it differentiates organisms based on whether they are alpha (α)-, beta (β)-, or gamma (γ)- hemolytic (Figure 5-2). Selective media support the growth of one group of organisms, but not another, by adding antimicrobials, dyes, or alcohol to a particular medium. MacConkey agar, for example, contains the dye crystal violet, which inhibits gram-positive organisms. Columbia agar with colistin and nalidixic acid (CNA) is a selective medium for gram-positive organisms because the antimicrobials colistin and nalidixic acid inhibit gram-negative organisms. Selective media can also be differential media if, in addition to their inhibitory activity, they differentiate between groups of organisms. MacConkey agar, for example, differentiates between lactose-fermenting and nonfermenting gram-negative rods by the color of the colonial growth (pink or clear, respectively); this is shown in Figure 5-3. In some cases (sterile body fluids, tissues, or deep abscesses in a patient on antimicrobial therapy), backup broth (also called supplemental or enrichment broth) medium is inoculated, along with primary solid (agar) media, so small numbers of organisms present may be detected; this allows detection of anaerobes in aerobic cultures and organisms that may be damaged by either previous or concurrent antimicrobial therapy. Thioglycollate (thio) broth, brain-heart infusion broth (BHIB), and tryptic soy broth (TSB) are common backup broths.

Selection of media to inoculate for any given specimen is usually based on the organisms most likely to be involved in the disease process. For example, in determining what to set up for a CSF specimen, one considers the most likely pathogens that cause meningitis (Streptococcus pneumoniae, Haemophilus influenzae, Neisseria meningitidis, Escherichia coli, group B Streptococcus) and selects media that will support the growth of these organisms (blood and chocolate agar at a minimum). Likewise, if a specimen is collected from a source likely to be contaminated with normal flora, for example, an anal fistula (an opening of the surface of the skin near the anus that may communicate with the rectum), one might want to add a selective medium, such as CNA, to suppress gram-negative bacteria and allow gram-positive bacteria and yeast to be recovered.

Routine primary plating media and direct examinations for specimens commonly submitted to the microbiology laboratory are shown in Table 5-1. Chapter 7 on bacterial cultivation reemphasizes the strategies described here for selection and use of bacterial media.

Specimen Preparation

Many specimens require some form of initial treatment before inoculation onto primary plating media. Such procedures include homogenization (grinding) of tissue; concentration by centrifugation or filtration of large volumes of sterile fluids, such as ascites (peritoneal) or pleural (lung) fluids; or decontamination of specimens, such as those for legionellae or mycobacteria. Swab specimens are often vortexed (mixed) in 0.5 to 1 mL of saline or broth for 10 to 20 seconds to dislodge material from the fibers.

Inoculation on Solid Media

Specimens can be inoculated (plated) onto solid media either quantitatively by a dilution procedure or by means of a quantitative loop, or semiquantitatively using an ordinary inoculating loop. Urine cultures and tissues from burn victims are plated quantitatively; everything else is usually plated semiquantitatively. Plates inoculated for quantitation are usually streaked with a 1 : 100 or 1 : 1000 loop. Plates inoculated for semiquantitation are usually streaked out in four quadrants. Detailed methods for streaking solid media are provided in Chapter 7, Figure 7-9. Semiquantitation is referred to as streaking for isolation, because the microorganisms present in the specimen are successively diluted out as each quadrant is streaked until finally each morphotype is present as a single colony. Numbers of organisms present can subsequently be graded as 4+ (many, heavy growth) if growth is out to the fourth quadrant, 3+ (moderate growth) if growth is out to the third quadrant, 2+ (few or light growth) if growth is in the second quadrant, and 1+ (rare) if growth is in the first quadrant. This tells the clinician the relative numbers of different organisms present in the specimen; such semiquantitative information is usually sufficient for the physician to be able to treat the patient.

Incubation Conditions

Inoculated media are incubated under various temperatures and environmental conditions, depending on the organisms suspected, for example, 28° to 30° C for fungi and 35° to 37° C for most bacteria, viruses, and acid-fast bacillus. A number of different environmental conditions exist. Aerobes grow in ambient air, which contains 21% oxygen (O₂) and a small amount (0.03%) of carbon dioxide (CO₂). Anaerobes usually cannot grow in the presence of O₂, and the atmosphere in anaerobe jars, bags, or chambers is composed of 5% to 10% hydrogen (H₂), 5% to 10% CO₂, 80% to 90% nitrogen (N₂), and 0% O₂. Capnophiles, such as Haemophilus influenzae and Neisseria gonorrhoeae, require increased concentrations of CO₂ (5% to 10%) and approximately 15% O₂. This atmosphere can be achieved by a candle jar (3% CO₂) or a CO₂ incubator, jar, or bag. Microaerophiles (Campylobacter jejuni, Helicobacter pylori) grow under reduced O₂ (5% to 10%) and increased CO₂ (8% to 10%). This environment can also be obtained in specially designed jars or bags.

Extent of Identification Required

As health care continues to change, one of the most problematic issues for microbiologists is the extent of culture workup. Microbiologists still rely heavily on definitive identification, although shortcuts, including the use of limited identification procedures in some cases, are becoming commonplace in most clinical laboratories (see CLSI document M35-A2 for information on abbreviated identification of organisms). Careful application of knowledge of the significance of various organisms in specific situations and thoughtful use of limited approaches will keep microbiology testing cost effective and the laboratory’s workload manageable, while providing for optimum patient care.

Complete identification of a blood culture isolate, such as Clostridium septicum as opposed to a genus identification of Clostridium spp., will alert the clinician to the possibility of malignancy or other disease of the colon. At the same time, a presumptive identification of Escherichia coli if a gram-negative, spot indole-positive rod is recovered with appropriate colony morphology on MacConkey agar (flat, lactose-fermenting colony that is precipitating bile salts) is probably permissible from an uncomplicated urinary tract infection. In the final analysis, culture results should always be compared with the suspected diagnosis. The clinician should be encouraged to supply the microbiologist with all pertinent information (e.g., recent travel history, pet exposure, pertinent radiograph findings) so that the microbiologist can use the information to interpret culture results and plan appropriate strategies for workup.

Communication of Laboratory Findings

To fulfill their professional obligation to the patient, microbiologists must communicate their findings to those health care professionals responsible for treating the patient. This task is not as easy as it may seem. This is nicely illustrated in a study in which a group of physicians was asked whether they would treat a patient with a sore throat given two separate laboratory reports—that is, one that stated, “many group A Streptococcus,” and one that stated, “few group A Streptococcus.” Although group A Streptococcus (Streptococcus pyogenes) is considered significant in any numbers in a symptomatic individual, the physicians said that they would treat the patient with many organisms but not the one with few organisms. Thus, although a pathogen (group A Streptococcus) was isolated in both cases, one word on the report (either many or few) made a difference in how the patient would be handled.

In communicating with the physician, the microbiologist can avoid confusion and misunderstanding by not using jargon or abbreviations and by providing reports with clear-cut conclusions. The microbiologist should not assume that the clinician is fully familiar with laboratory procedures or the latest microbial taxonomic schemes. Thus, when appropriate, interpretive statements should be included in the written report along with the specific results. One example would be the addition of a statement, such as “suggests contamination at collection,” when more than three organisms are isolated from a clean-voided midstream urine specimen.

Laboratory newsletters should be used to provide physicians with material such as details of new procedures, nomenclature changes, and changes in usual antimicrobial susceptibility patterns of frequently isolated organisms. This last information, discussed in more detail in Chapter 12 is very useful to clinicians when selecting empiric therapy. Empiric therapy is based on the physician determining the most likely organism causing a patient’s clinical symptoms and then selecting an antimicrobial that, in the past, has worked against that organism in a particular hospital or geographic area. Empiric therapy is used to start patients on treatment before the results of the patient’s culture are known and may be critical to the patient’s well-being in cases of life-threatening illnesses.

Positive findings should be communicated to the clinician in a timely manner, and all verbal reports should be followed by written confirmation of results. Results should be legibly handwritten or generated electronically in the laboratory information system (LIS).

Critical (Panic) Values

Certain critical results must be communicated to the clinician immediately. Each clinical microbiology laboratory, in consultation with its medical staff, should prepare a list of these so-called panic values. Common panic values include the following: