Clinical relevance of culture versus screens for the detection of microbial pathogens in urine specimens.

The use of screens to detect "significant levels" of pathogenic microorganisms in urine specimens offers the advantages of both rapidly reporting results and controlling costs. Many of these screens, however, are insensitive at microbial counts below 10(5) colony-forming units (CFU)/ml of urine. It is increasingly apparent that patients with almost any type of urinary tract infection (except for most patients who are asymptomatic or who have pyelonephritis) may have urine concentrations of pathogens as low as 10(2) to 10(3) CFU/ml. This review documents factors that can contribute to diminished concentrations of microorganisms in urine, lists patient populations in whose urine microorganisms in concentrations well below 10(5) CFU/ml have been associated with infection, and makes recommendations for selection of laboratory tests, including rapid screens, for the diagnosis and management of urinary tract infections.

Frequency of low level bacteremia in infants from birth to two months of age.

BACKGROUND: The frequency of low level bacteremia (< or = 10 colony-forming units/ml) in infants from birth to 2 months of age and the optimal volume of blood and number of blood cultures to be collected have not been well-documented. During 1991 guidelines at this hospital for collection of blood for culture from these infants were revised. METHODS: Blood from each infant with suspected bacteremia was usually inoculated into an Isolator 1.5 Microbial Tube (1.5 ml of blood) and a bottle of anaerobic broth (0.5 to 3.0 ml of blood). The use of a second Isolator tube and the total blood volume recommended for culture (2 to 6 ml) depended on the weight and total blood volume of each infant. RESULTS: Forty-four bacterial pathogens were recovered from the blood of 40 (2.5%) of 1589 infants. Of 34 infants from whose blood the concentration of pathogens could be determined, 23 (68%) had low level bacteremia. Of 50 isolates of pathogens recovered from Isolator cultures, 32 (64%) were detected in counts of < or = 10 colony-forming units/ml. When 2 or 3 blood culture devices were inoculated with a total of 2 to 6 ml of blood from each infant, significantly more cases of bacteremia were detected (34 (3.0%) of 1126 infants had positive blood cultures) than when only one culture device containing < or = 1.5 ml of blood was used (2 (0.5%) of 398 infants had positive blood cultures; P = 0.008). However, when 4 or more culture devices were inoculated with a total of > 6 ml of blood from each infant (5 (7.7%) of 65 infants had positive blood cultures), the difference in recovery of pathogens compared with the culturing of from 2 to 6 ml of blood per infant was not significant (P = 0.089). CONCLUSIONS: Low level bacteremia was common in our infants' patient population. The culturing of up to 6 ml of blood which represented up to 4.5% of an infant's total blood volume was required for detection of the pathogens.

Problems encountered in using Cefinase disks to predict in vitro penicillin susceptibility of coagulase-negative staphylococci.

The value of Cefinase (BBL Microbiology Systems, Cockeysville, M.D.) disks in predicting penicillin responses of coagulase-negative staphylococci was studied with the use of clinical isolates. The beta-lactamase activity of each isolate was determined both before and after enzyme induction for 18-24 hours with a 5-micrograms methicillin disk. Penicillin responses were determined in microdilution minimum inhibitory concentration (MIC) panels after 18 hours of incubation at 35 degrees C. Of 485 isolates tested, 338 (70%) were beta-lactamase positive, 168 (35%) before and 170 (35%) after induction. Only two (0.6%) of these isolates had a penicillin MIC of less than or equal to 0.12 microgram/mL (less than or equal to 0.12 mg/L). Of the 147 Cefinase-negative isolates, penicillin MICs of 0.12 microgram/mL or less (less than or equal to 0.12 mg/L) and 0.25 microgram/mL or more (greater than or equal to 0.25 mg/L) were obtained from 110 (75%) and 37 (25%), respectively. Although Staphylococcus saprophyticus accounted for only 30 (6%) of the total isolates, 54% of the organisms that were beta-lactamase negative with penicillin MICs of greater than or equal to 0.25 microgram/mL (greater than or equal to 0.25 mg/L) belonged to that species. A positive Cefinase test accurately predicted elevated penicillin MICs, but a negative test, especially with S. saprophyticus, was of less value.

Impact of variation in endocervical specimen collection and testing techniques on frequency of false-positive and false-negative Chlamydia detection results.

Rapid tests for detection of Chlamydia trachomatis are considerably more likely to provide accurate, reliable results if high quality endocervical specimens containing large quantities of the pathogen are submitted for testing, and if laboratories routinely detect and confirm Chlamydia at levels below the test manufacturer's recommended cut-off using previously published, well-documented guidelines that have been verified by in-house testing. Routine or periodic microscopic analysis of endocervical specimen quality may be necessary both to ensure the adequacy of specimens and to help motivate personnel performing the specimen collection procedures. False-positive test results can be significantly reduced or eliminated by confirming positive results with the use of an independent assay. Clinical laboratories currently have the opportunity to substantially improve both the sensitivity and the specificity of many currently available rapid assays for the detection of Chlamydia trachomatis.

Inability to control selected drugs on commercially obtained microdilution MIC panels.

As more broad spectrum antimicrobics continue to appear, many manufacturers have added drugs to their microdilution minimal inhibitory concentration (MIC) test panels at the expense of testing each drug over a broad range of twofold dilutions. Micro Scan (Mahwah, NJ) gram-positive and gram-negative 96-well microdilution panels were tested with the suggested control organisms over a period of 60 days to determine if on-scale MIC endpoints could be achieved with at least one organism per drug. Of 13 antimicrobics in the gram-positive panel, seven (clindamycin, tetracycline, cephalothin, gentamicin, nitrofurantoin, trimethaprim-sulfa, and streptomycin) could not be controlled effectively. Of 18 drugs in the gram-negative panel, six (tobramycin, trimethaprim-sulfa, cefoperazone, mezlocillin, ticarcillin, and kanamycin) similarly could not be controlled. The accuracy of results obtained when patient isolates were tested with these 13 antimicrobics could not be substantiated by control procedures as stated in the Micro Scan instructions.

Improved detection of beta-lactamase activity in isolates of Staphylococcus saprophyticus with the use of a modified Cefinase disk procedure.

The Cefinase disk (BBL Microbiology Systems, Cockeysville, MD) technique has been reported to be relatively insensitive for detection of beta-lactamase production by Staphylococcus saprophyticus. The effects of time (up to 24 hours) and temperature (25 degrees C vs. 35 degrees C) of test incubation as well as enzyme induction on Cefinase detection of beta-lactamase activity in clinical isolates of S. saprophyticus was investigated. Results were compared to beta-lactamase detection with the use of conventional nitrocefin and cloverleaf methods and with penicillin minimal inhibitory concentrations (MICs). Of 85 S. saprophyticus isolates, 84 (98.8%) had penicillin MIC end points of greater than or equal to 0.25 microgram/mL. Beta-lactamase was detected by at least one method in 73 (85.9%) of the isolates: 63 (74.1%) by nitrocefin, 55 (64.7%) by cloverleaf, and 63 (74.1%) by Cefinase. Of Cefinase-positive isolates, 56 (88.9%) and 63 (100%) were detected at 25 degrees C and 35 degrees C, respectively, and none required enzyme induction. At 35 degrees C, 4 (6.3%) and 19 (30.2%) of the Cefinase-positive isolates were detected at 1 hour and 24 hours, respectively. Incubation of Cefinase tests at 35 degrees C for 24 hours provides a satisfactory alternative to the conventional nitrocefin assay for detection of beta-lactamase activity in S. saprophyticus.

Suitability of a throat culture method for evaluation of group A streptococcal antigen detection kits.

Previous reports have indicated a wide variation in observed sensitivity of antigen-detection kits for group A streptococci. Before undertaking an evaluation of these new kits, the sensitivity of the throat culture technic routinely used by this laboratory was reexamined. Each throat swab was directly inoculated to sheep blood agar containing trimethoprim-sulfamethoxazole (SXT X BA) and drug-free sheep blood agar (SBA) plates. Swabs were then washed in saline and the saline used to inoculate one more of each type of medium. SXT X BA cultures were incubated aerobically (5 to 10% CO2), and SBA cultures were incubated anaerobically, both for two days at 35 degrees C. From 726 patients, 164 (22.6%) of the specimens contained group A streptococci, 99% detected on directly inoculated cultures and 100% on cultures inoculated with the saline wash. Either an aerobically (CO2) incubated SXT X BA or an anaerobically incubated SBA, directly inoculated and held for two days, appears to offer a satisfactory reference culture method for the recovery of group A streptococci.

Cross-reaction of clinical isolates of bacteria and yeasts with the chlamydiazyme test for chlamydial antigen, before and after use of a blocking reagent.

More than 1,000 clinical isolates of bacteria and yeasts were identified, subcultured, and tested at 10(8) colony-forming units per milliliter with the Chlamydiazyme assay to determine the variety of microorganisms which could cause false-positive results for Chlamydial antigen. False-positive Chlamydiazyme results were obtained from 27 of 465 (5.8%) gram-negative, aerobic, or facultatively anaerobic bacterial isolates (including 8 of 39 [20.5%] Neisseria gonorrhoeae and 8 of 149 [5.4%] Escherichia coli isolates) and also from 2 of 46 (4.3%) Bacteroides species isolates. No false-positive results were obtained either from 373 gram-positive bacterial isolates or from 153 yeasts. Results from 21 of 29 (72.4%) isolates that cross-reacted with Chlamydiazyme antibodies were repeatedly positive, but all 21 were confirmed as false-positive results using a blocking antibody. When an initial Chlamydiazyme result is positive, repeating the test, with and without use of the blocking antibody, appears to be effective in identifying those results (more likely from poorly collected endocervical specimens) that are falsely positive, even in the presence of high concentrations of cross-reacting bacteria. Microscopic determination of endocervical specimen adequacy also may help to minimize false-positive (and false-negative) Chlamydiazyme results.

Comparison of the TestPack Strep A enzyme immunoassay system with anaerobically incubated cultures for detection of group A streptococci from oropharyngeal swabs.

To find a rapid and sensitive test for detection of Group A streptococci (GABS), the performance of the TestPack Strep A (TPSA; Abbott) was compared with culture with the use of rayon-tipped throat swabs from symptomatic patients six months to 90 years of age. Each swab was first inoculated to a 5% sheep blood agar plate and then tested for GABS antigen with the use of the TPSA and the manufacturer's instructions. Cultures were incubated anaerobically at 35 degrees C for 36-48 hours unless positive results were obtained after one night. GABS were identified with a fluorescent antibody method or a latex antibody test. From 1,616 throat swabs, 296 (18.3%) of the cultures contained GABS. The sensitivity and specificity of the TPSA were 73.3% and 94.8%, respectively, whereas the predictive values of positive and negative results were 75.9% and 94.1%, respectively. Results varied significantly, however, with different production lots of TPSA. The TPSA does not appear to provide a sensitive alternative to an anaerobic culture for detection of GABS.

Occurrence and documentation of low-level bacteremia in a community hospital's patient population.

To document the incidence of low-level bacteremia in the patient population of this study, two blood culture sets were collected from symptomatic patients weighing more than 80 pounds. Each blood culture set consisted of a lysis-centrifugation tube and three bottles containing different culture broths, each inoculated with 10 mL blood. Pathogens from 63 (26.4%) and 48 (20.1%) of the 239 culture-positive patients were recovered from only one and two of the eight culture devices, respectively, representing low-level bacteremia. Isolates from another 60 (25.1%) of the 239 patients were recovered from all eight of the culture devices, representing high-level bacteremia. Whether patients had low-level or high-level bacteremia, there were mostly insignificant differences in the types of species recovered, in the percentages of patients for whom therapy was initiated or changed following the laboratory's reports, and in the clinical signs, symptoms, and characteristics of the patients. Clinically documented, low-level bacteremia is relatively common in this community hospital's patient population. Culturing of up to 80 mL of blood was required for detection of all pathogens from patients weighing more than 80 pounds.

Predictive values of species' identifications from the Vitek Gram-Positive Identification Card using clinical isolates of coagulase-negative staphylococci.

The Gram-Positive Identification Card (GPI, bioMerieux Vitek) was compared to conventional tests for identification of 616 clinical isolates of coagulase-negative staphylococci. All tests were inoculated with subcultures from single isolated colonies. The predictive values of GPI species' identifications (the number of the system's correct calls divided by the total number of calls for each species) were 100% (12 of 12) for Staphylococcus capitis, 100% (79 of 79) for S saprophyticus, 98.4% (246 of 250) for S epidermidis and 96.0% (120 of 125) for S haemolyticus, but only 64.5% (69 of 107) for the other species analyzed. When an infrequently encountered Staphylococcus species is named using the GPI, the identification should be confirmed using additional tests.

Culture vs direct antigen assays for detection of microbial pathogens from lower respiratory tract specimens suspected of containing the respiratory syncytial virus.

Following the introduction of effective antiviral chemotherapy, rapid antigen assays have been utilized increasingly, instead of cell cultures, for detection of the respiratory syncytial virus from lower respiratory tract specimens. Because antigen assays, unlike cell culture, cannot amplify low levels of the virus to a detectable level, assay sensitivity is especially dependent on high-quality specimens. In addition, the assays are unable to detect other viruses or bacteria with which the patient may be infected. This review summarizes results from clinical studies of the performance of cell cultures and the more commonly used antigen assays, describes factors that may lead to false-positive or false-negative test results, and makes recommendations for the selection of procedures for the reliable detection of microbial pathogens from patients suspected of being infected with respiratory syncytial virus.

Clinical and laboratory considerations of culture vs antigen assays for detection of Chlamydia trachomatis from genital specimens.

Assays for Chlamydia trachomatis antigen are being increasingly used in lieu of traditional cell culture methods for detection of the organism in patient specimens. While results from these assays are available considerably earlier than are those from culture, the assays have been less sensitive and specific than culture when used in some patient populations. This review summarizes results from investigations into the performance of culture and the two most widely used antigen assays, documents factors that can contribute to false-positive or false-negative results, and, in light of these factors, makes recommendations for the selection of methods for detection of C trachomatis from genital specimens.

Impact of endocervical specimen quality on apparent prevalence of Chlamydia trachomatis infections diagnosed using an enzyme-linked immunosorbent assay method.

Duplicate endocervical swabs were collected from 1824 patients for detection of Chlamydia trachomatis. Specimen pairs were combined into 400 microL of 0.9% saline solution. After vortexing, a 40-microL sample was smeared and stained with Papanicolaou's method for detection of endocervical and/or metaplastic (E-M) cells. The remaining specimen was tested for C trachomatis antigen with the use of an enzyme-linked immunosorbent assay (ELISA) procedure (Chlamydiazyme, Abbott Laboratories, North Chicago, Ill). Chlamydia trachomatis antigen was detected and confirmed (with the use of a blocking antibody [Abbott Laboratories]) in only 16 (1.7%) of 918 specimens that lacked detectable E-M cells, but it was detected significantly more frequently not only in 88 (13.3%) of 661 specimens that contained detectable E-M cells but also in 32 (13.1%) of 245 specimens that contained too many red blood cells to analyze microscopically. Of the initially positive ELISA results, none of 37 were falsely positive from specimens that contained 11 or more E-M cells, but significantly more (six [27.3%] of 22) were falsely positive from specimens that lacked detectable E-M cells. Variations in specimen quality had a significant impact on the incidence of both true-positive and false-positive ELISA results and could significantly influence understanding of the prevalence of chlamydial infections in women.

Inability to adequately control antimicrobial agents on AutoMicrobic System Gram-Positive and Gram-Negative Susceptibility Cards.

The Vitek Gram-Positive Susceptibility Card (GPS) and Gram-Negative General Susceptibility Plus Card (GSC Plus) were tested on the AutoMicrobic System (AMS) 50 times each with the recommended control organisms. Only 1 drug (chloramphenicol) of 11 on the GPS and 1 (gentamicin) of 10 on the GSC Plus could be adequately controlled, leaving unsubstantiated the results obtained with patient isolates on the remaining 19 antimicrobial agents.

Comparison of GonoGen, GonoGen II, and MicroTrak direct fluorescent-antibody test with carbohydrate fermentation for confirmation of culture isolates of Neisseria gonorrhoeae.

When testing 248 clinical isolates of Neisseria gonorrhoeae, the sensitivity was 100% with GonoGen (Becton Dickinson Microbiology Systems), 99.6% (247 of 248) with GonoGen II (Becton Dickinson), 97.2% (241 of 248) with the MicroTrak direct fluorescent-antibody test (Syva), and 97.6% (242 of 248) with Rapid Fermentation Agar carbohydrates (Remel). Of 62 isolates of other Neisseria species, none was misidentified as N. gonorrhoeae by GonoGen, MicroTrak, or Rapid Fermentation Agar carbohydrates but 7 (31.8%) of 22 isolates of N. meningitidis gave strong, repeatedly false-positive results with GonoGen II. The sensitivity of all four assays was good to excellent, but all positive GonoGen II results should be confirmed with an independent assay, especially when isolates are recovered from sites where N. meningitidis is likely. Positive results from any of the assays should be routinely confirmed when dictated by specific clinical, legal, or microbiological circumstances.

Detection of group A streptococci in the laboratory or physician's office. Culture vs antibody methods.

Much controversy has existed concerning the clinical significance of small numbers of group A streptococci (eg, one to ten or even 50 colonies) recovered in culture. The relative quantity of streptococci recovered is, in part, technique dependent. This review documents the need for sensitive throat culture technology, addresses technical problems associated with the culture procedure, and, in light of these problems, explores the need for very carefully performed studies on streptococcal antigen detection kits before their implementation as a routine test either in physicians' offices or in microbiology laboratories.

Distribution of bacteria and yeasts within the 10-ml Isolator during the processing of seeded blood samples.

Forty-five organisms consisting of stock cultures and clinical isolates of bacteria and yeast were separately inoculated into outdated blood bank blood to achieve a concentration of approximately 100 CFU/ml. Blood with each organism was introduced into groups of four Isolators (E. I. du Pont de Nemours & Co., Inc., Wilmington, Del.), which were then processed according to the Isostat instructions of the manufacturer. The supernatant, sediment, and wash (material removed from the surface of the slanted stopper after sediment removal) were inoculated onto 5% sheep blood agar plates. Cultures were incubated aerobically (5 to 10% CO2) at 35 degrees C for 48 to 72 h. From the 180 Isolators, the mean recovery was 6% (range, 0 to 48%) for the supernatant, 87% (range, 47 to 98%) for the sediment, and 8% (range, 3 to 23%) for the wash. Neither variation among technologists nor intentional misalignment of additional Isolators in the centrifuge could explain all of the losses of microorganisms from the sediment. The manual nature of the Isolator procedure, which led to the loss of significant amounts of organisms from the sediment, may help to explain false-negative Isolator results obtained from blood of patients, particularly when small numbers of pathogens are present.

Justification for use of a single trichrome stain as the sole means for routine detection of intestinal parasites in concentrated stool specimens.

Of 12,321 stool samples analyzed over a 6-year interval, 870 (7.1%) were positive for a total of 1,019 parasites, of which 1,011 (99.2%) were found in trichrome-stained smears of unconcentrated specimens while only 479 (47.0%) were detected in iodine-stained smears of concentrated samples. Stool specimens were next analyzed by trichrome staining of both unconcentrated and concentrated specimens preserved in either mercury-polyvinyl alcohol (PVA) or cupric PVA. Of 2,198 specimens, 171 (7.8%) were positive for a total of 208 parasites, 192 (92.3%) and 204 (98.1%) of which were found in the unconcentrated and concentrated specimens, respectively (P < 0.05). In our patient population, examination of a single trichrome-stained smear of a concentrated stool specimen is a cost-effective alternative to routinely analyzing both concentrated and unconcentrated specimens for parasites.

Direct fluorescent-antibody confirmation of chlamydial antigen below the detection threshold of the chlamydiazyme enzyme-linked immunosorbent assay.

Of 4,000 endocervical specimens tested with the Chlamydiazyme enzyme-linked immunosorbent assay (Abbott Laboratories), 233 (5.8%) gave positive results (A492 above the cutoff), which were confirmed with a blocking reagent (Abbott). An additional 34 specimens (14.6%) with chlamydial antigen were detected and confirmed with the direct fluorescent-antibody test (Syva) from among those 66 Chlamydiazyme-negative specimens which had A492s that ranged from 0.030 to the cutoff and that could be blocked by > or = 50% with the blocking reagent.